RE: Request for Interview- Academic Project

Arenguseire Keskus | Lossi plats 1a,15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

1

2021

Arenguseire Keskus

Platvormitöö

Eestis 2021 Küsitlusuuringu tulemused

Johanna Vallistu

Magnus Piirits

2021

Platvormitöö Eestis 2021

Küsitlusuuringu tulemused

2.06.2021

Autorid:

Johanna Vallistu, Magnus Piirits

Uuringus sisalduva teabe kasutamisel palume viidata allikale: Vallistu, J., Piirits, M.,

(2021) Platvormitöö Eestis 2021. Küsitlusuuringu tulemused. Arenguseire Keskus.

Uuringus sisalduva teabe kasutamisel palume viidata lisaks allikale ka tellijale.

Arenguseire Keskus on Riigikogu juures tegutsev mõttekoda, mille ülesandeks on

ühiskonna pikaajaliste arengute analüüsimine, uute trendide ja arengusuundade

avastamine ning arengustsenaariumide koostamine.

Sisukord

1. Sissejuhatus ....................................................................................................... 7

2. Platvormitöö küsitluse tulemused .................................................................... 8

2.1. Uuringu meetod ............................................................................................. 8

2.2. Platvormitöö Eestis – üldine vaade ............................................................. 10

2.3. Fookuses: platvormitöötajate toimetulek ja kriisi mõju ................................. 27

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

4

Uuringu lühikokkuvõte

• Küsitlusuuringu eesmärk oli saada aimu platvormitöö ulatusest Eestis ning

mõista, kui tihti ja millist platvormitööd tehakse. Uuriti ka platvormitöötajate

sissetuleku ja tööaja kohta. Kus võimalik, võrreldi tulemusi 2018. aastal

korraldatud platvormitööd puudutava küsitluse tulemustega.

• Iganädalaselt teeb platvormitööd 7,0% esindusliku valimiga küsitlusele

vastanutest, mis tööealise elanikkonna (vanuses 18–64) arvule laiendades on

ligikaudu 56 000 inimest. 2018. aastal oli vastav osakaal 8,1%, seega ei ole

regulaarsete platvormitöötajate osakaal kasvanud. Suurenenud on aga

vähemalt kord kuus platvormitööd tegevate inimeste osakaal (11,9% vs. 10,3%

2018. aastal). Vähemalt kord kuus teeb platvormitööd 11,9% vastanutest (u

95 000 inimest), vähemalt kord aastas ligikaudu viiendik (u 161 000 inimest)

ning kunagi on platvormitööd teinud enam kui veerand tööealisest

elanikkonnast (u 208 000 inimest).

• Platvormitöö on lisatöö muu põhitegevuse kõrval. Ligi pool regulaarselt

platvormitööd tegevatest inimestest teeb platvormitööd kuni 10 tundi nädalas.

Veerand regulaarsetest platvormitöötajatest teeb platvormitööd 11–25 tundi,

veidi rohkem kui viiendik 25–40 tundi ning väike osa (7,2%) rohkem kui 40 tundi

nädalas.

• Ainus sissetulekuallikas on platvormitöö 4,4% platvormitöötajate jaoks (u

2500 inimest) ning see ei ole võrreldes 2018. aastaga kasvanud.

• Platvormitöötajad, kes on viimase aasta jooksul vähemalt ühe korra

platvormitööd teinud (20,2% tööealisest elanikkonnast ehk u 161 000 inimest),

teenivad platvormitööd tehes keskmiselt 18,4% oma kuu netosissetulekust.

Seega on platvormitööst teenitava tulu osakaal märkimisväärne.

• Koroonaajal on kasvanud digikeskkonnas tehtava platvormitöö maht, seejuures

eriti eestikeelsete vastajate seas. Sellest järeldub, et platvormi ärimudel laieneb

professionaalsete teenuste sektorites.

• Võrreldes 2018. aastaga on kasvanud enamiku küsitluses uuritud platvormitöö

tüüpide osakaal. Enim on suurenenud sõidujagamise osakaal (+7,6%) ja

veebipõhise töö osakaal (keerukam +4%, lihtsam +3,4%), aga ka

professionaalsete teenuste osakaal (+3,2%). Samale tasemele on jäänud

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

5

kontoritöölaadse platvormitöö tegemine ning vähenenud on

isikuhooldusteenuste osakaal (–5,5%) ja ülesannete täitmine teiste kodudes (–

2,6%). Kulleriteenuse osakaalu muutus ei ole eelmise küsitluslainega võrreldav.

• Kokku kuni 40 tundi nädalas töötavate platvormitöötajate puhul on platvormitöö

osakaal selles 14,5 tundi ning platvormitöötaja keskmine töökoormus ei erine

platvormitööd mitte tegeva inimese töökoormusest. Inimestel, kes töötavad

kokku rohkem kui 40 tundi nädalas (u 20% regulaarsetest platvormitöötajatest),

on oluliselt suurem ka platvormitöö tundide arv – keskmiselt 26 tundi nädalas.

See viitab ebamõistlikult suurele summaarsele töökoormusele, mis võib olla

ajendatud probleemidest igapäevase toimetulekuga.

• Platvormitööd tegevate inimeste keskmine netosissetulek on väiksem kui neil,

kes platvormitööd ei tee. Vähemalt kord elus platvormitööd teinud inimeste

netosissetulek märtsis 2021 oli keskmiselt 1017 eurot. Eesti keskmine netopalk

samas kuus oli 1255 eurot. Suurim osa (19,3%) on regulaarsete

platvormitöötajate hulgas neid, kelle igakuine netosissetulek jääb vaid 400–700

euro vahele.

• Kui ankeedile vastanud eestlastest tegi iganädalaselt platvormitööd 6,3%, siis

venekeelsele ankeedile vastanutest teeb platvormitööd iganädalaselt 8,5%.

Sarnaselt jaotuvad ka igakuiseid platvormitöötajad: eestikeelsele ankeedile

vastanutest teeb igakuiselt platvormitööd 11% ja venekeelsele ankeedile

vastanutest 13,9%. See viitab, et platvormitöö kipub olema tööturu duaalsust

arvestades pigem n-ö väheste valikuvõimalustega töötegijate pärusmaa. Kuigi

platvormitöö avardab selliste töötegijate võimalusi tööturul osaleda, on

probleemiks sotsiaalkaitse madal tase töölepingulise töösuhtega võrreldes.

• Veebipõhise keeruka töö tegijate näitel selgus, et töid, mille puhul tavapäraselt

eeldatakse erialast haridust, teeb platvormide vahendusel tööd ka suur hulk

põhi- (12,3%) ja keskharidusega (54,5%) inimesi. See vastab akadeemilises

kirjanduses esile kerkinud tähelepanekule, et platvormid võimaldavad keerulisi

ülesandeid osadeks jagada ning standardiseerida.

• Keskmine regulaarne (igal nädalal vähemalt korra) platvormitöötaja on pigem

mees, pigem nooremaealine, elab pigem Põhja- või Kirde-Eestis, on pigem

vene keelt kõnelev ning pigem põhitöökohaga kontoris või teenindussektoris.

Tulemused viitavad aga sellele, et sõltuvalt platvormitöö tüübist on

platvormitöötajate profiil väga erinev.

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

6

• Platvormitöö mustritest nähtub süvenev segregatsioon tööturul. Venekeelsete

vastajate seas on just füüsiliste platvormide kaudu vahendatavate teenuste

osutamise tase kõrgem (venekeelsele ankeedile vastanutest 9,5% osutab

vähemalt kord aastas sõidujagamisteenust ja 11,7% osutab vähemalt kord

aastas kulleriteenust. Eestikeelsele ankeedile vastanute vastav osakaal oli

4,1% ja 6,5%). Sellest võib järeldada, et venekeelsetel vastajatel on lihtsaid

oskusi eeldavat platvormitööd sissetulekute suurendamiseks rohkem vaja kui

eestikeelsetel.

• Euroopa Komisjon on välja töötamas platvormitöö regulatsiooni, kaaludes

töölepingulise töösuhtega kaasneva sotsiaalkaitse laiendamist platvormitööle.

Eestis on platvormitöötajate sotsiaalkaitse üheks lahenduseks seni peetud

ettevõtluskontot (füüsilisest isikust kauba müüjale või teenuse osutajale

kohalduv lihtsustatud maksurežiim koos ravikindlustuse võimalusega). Samas

kasutab ettevõtluskontot praegu üksnes 2% platvormitöötajatest.

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

7

1. Sissejuhatus

Platvormitöö kui töösuhteid ja töökorraldust muutev nähtus kerkis esile 2010. aastate

alguses sõidujagamisplatvormi Uber turuletulekul. Järgnenud on erinevaid töid

vahendavad platvormid. Platvormitöö tekkis oluliste tehnoloogiliste ja sotsiaalsete

muutuste ristumisel – ühelt poolt võrgustikuefekti oskuslikult enda kasuks pööravad

globaalsed digiplatvormid, kes on platvormitöö puhul rõhutanud oma rolli töö

vahendajana, ning teiselt poolt tarbijad ja teenusepakkujad, kes tänu nutitelefonile ja

mobiilse interneti levikule on teenusemudeli muutusega kaasa läinud.

Kui alguses seostati platvormitöö mõistet jagamismajandusega, sai üsna ruttu selgeks,

et pigem räägime tavapärase töö ümberkorraldamisest uudsel viisil. Tarbijad saavad

teenust kasutada nüüd ja kohe – toit või teenus on kliki kaugusel. Varem töövõtjana

tegutsenud inimesed leiavad end tööplatvormil teenust pakkudes iseseisva

lepingupartneri rollist, vastutades ise oma tööaja, töötingimuste ning tervise ja

pensionipõlve kindlustamise, aga ka maksude maksmise eest.

Ligi 10 aastat on Euroopas ja ka mujal maailmas toimunud elav arutelu, kas

platvormitöötaja on ikka tõepoolest ettevõtja, kui tööd vahendav platvorm määrab nii

töö hinna kui ka kirjutab ette, mil viisil tuleks teenust pakkuda. Suurbritannias ja

Ameerika Ühendriikides on jõutud esimeste kohtulahenditeni, kus on leitud, et

platvormidel on tööandja tunnused.1

Eestis on platvormitööd puudutavad arutelud jäänud seni tagasihoidlikuks. 2019.

aastal Arenguseire Keskuse osalusel valminud platvormitöö uurimiseks korraldatud

küsitlus näitas, et platvormitöö on saanud oluliseks osaks Eesti tööturul, isegi kui see

on peamiselt lisatöö. Kui kaks aastat tagasi võisime eeldada, et tööjõupuuduse korral

on pall töötegija väravas ning ebasobivate töötingimustega seoses on inimesel

võimalik tööturul ise sobivam valik teha, siis COVID-19 kriisi olukorras ja

toimetulekuraskuste süvenemisel on asjakohane teemasse tõsisemalt suhtuda.

1Allikad: BBC, 11.08.2020, Uber and Lyft drivers are employees, says US judge; CNBC, 18.03.2021, Uber employment rights setback is a ‘gut punch’ to its prospects in the UK.

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

8

2. Platvormitöö küsitluse tulemused

2.1. Uuringu meetod

Arenguseire Keskuse tellimusel korraldas uuringufirma SAAR POLL 2021. aasta

märtsis üle-eestilise platvormitööd puudutava küsitlusuuringu. Esindusliku valimiga

veebipõhises paneelküsitluses osales 2000 inimest vanuses 18–64. Küsitlus viidi läbi

eesti ja vene keeles. Uuringu meetod valiti võimalikult sarnane 2018. aasta lõpus

tehtud platvormitöö küsitluses kasutatud meetodiga. Statistilise mudeli ja

küsitlustulemi erisuste vähendamiseks tulemit kaaluti. Sotsiaaldemograafiliste

tunnuste mudeli loomisel kasutati rahvastikustatistika andmeid seisuga 01.01.2021.

Haridust ja rahvust kaaluti 2019. aasta andmeid kasutades. Maksimaalse valimivea

suurus jääb 95% usaldusnivool 2,19% piiridesse (väiksemate alagruppide vaatlemisel

võib viga olla suurem). Elukohatüübi puhul on lähtutud Statistikaameti määratlusest.

Linnalises asustusüksuses on rahvastiku tihedus suurem kui 1000 elanikku km² kohta,

väikelinnalises asustusüksuses on rahvastiku tihedus 200–1000 elanikku km² kohta ja

maalises asustusüksuses vähem kui 200 elanikku km² kohta. Põhja-Eesti alla kuulub

Harjumaa (sh Tallinn), Lääne-Eesti alla Hiiumaa, Läänemaa, Pärnumaa ja Saaremaa,

Kesk-Eesti alla kuuluvad Järvamaa, Lääne-Virumaa ja Raplamaa, Kirde-Eesti alla Ida-

Virumaa, Lõuna-Eesti alla Jõgevamaa, Põlvamaa, Tartumaa, Valgamaa, Viljandimaa

ja Võrumaa.

Platvormitöö küsitlusuuringu eesmärk oli võrrelda platvormitöö levimust Eestis 2018.

aastal tehtud uuringu tulemustega ning leida, millises ulatuses ja millist platvormitööd

kõige enam tehakse, kui suurt sissetulekut platvormitöö pakub ning kuidas on

korraldatud platvormitöötajate sotsiaalkaitse. 2021. aasta küsitlusuuringu lähtekohaks

oli 2018. aastal kasutatud küsimustik, kuid täielik võrreldavus kahe uuringu vahel ei

olnud eesmärk omaette. Värskem küsitlus keskendus enam platvormitöötajate

sissetuleku ja tööaja küsimustele ning ka kriisi mõjudele. Kõrvale jäid laiemad

ühiskonna ja majanduse platvormistumisega seotud teemad, mis olid hõlmatud

eelmises küsitluses.

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

9

Mis on platvormitöö?

Platvormitöö on uuringus defineeritud kui digitaalse töövahendusplatvormi abil

leitud tööotste tegemine erinevatele klientidele. Digitaalsed töövahendusplatvormid

on näiteks Uber, Bolt, Yandex, Upwork, CareMate. Digitaalsete platvormide

vahendatav töö võib olla nii füüsilises keskkonnas tehtav, näiteks sõidujagamine ja

koristustöö, kui ka veebipõhine, näiteks graafiline disain. Uuringu fookuses ei olnud

igasugune veebi kaudu leitud ühekordne juhutöö (nt ühekordne Facebooki

platvormil leitud tööots) ega püsiv töölepinguline töö, mille korraldamiseks

kasutatakse digiplatvormi (nt veebipõhine broneerimissüsteem). Ei uuritud ka

Airbnb või mõne muu lehekülje kaudu kinnisvara renditeenuse pakkumist, kuigi

mõnikord loetakse ka seda platvormitöö tüübiks.

Uuringus jagame tulemused järgneva kaheksa platvormitöö tüübi vahel:

• sõidujagamisteenuse pakkumine nt Bolti, Uberi, Yandexi vahendusel;

• toidu kohaletoomise või muude kulleriteenuste pakkumine nt Wolti, Bolti

vahendusel;

• erinevate tööde tegemine teiste inimeste kodudes (ehitustööd, remont,

koristamine, lapsehoid, aiatöö jmt) nt Taskus, GoWorkaBiti, Handiesi

vahendusel;

• isikuhooldusteenuste pakkumine (juukselõikus, massaaž, maniküür jmt) nt

Taskus, CareMate’i vahendusel;

• jooksvate asjatoimetuste või bürootööga tegelemine kliendi ruumides nt

GoWorkaBiti vahendusel;

• administratiivse töö tegemine või väiksemate ülesannete täitmine (nn click

work) nt Amazon Mechanical Turki, Clickworkeri, Upworki vahendusel;

• loomingulise või IT-alase töö tegemine (kujundamine, toimetamine,

tarkvaraarendus või tõlkimine jmt) nt Upworki, DesignCrowdi, Freelanceri,

Fiverri vahendusel;

• professionaalsete teenuste pakkumine (õigusabi, raamatupidamine jmt) nt

UpWorki, Freelanceri, Fiverri vahendusel.

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

10

2.2. Platvormitöö Eestis – üldine vaade

2021. aasta märtsis tegi platvormitööd igapäevaselt 1,7%, peaaegu iga päev 2,1%

ning umbes kord või kaks nädalas 4,1% küsitlusele vastanud 18–64-aastastest Eesti

elanikest. Neis numbrites kajastuvad ka inimesed, kes teevad erinevaid platvormitöid

erineva sagedusega, mistõttu võib üks platvormitöötaja olla arvesse võetud mitmel

korral.

Kuni kord nädalas platvormitöö tegijaid loeme regulaarseteks platvormitöötajateks.

Kokku oli vastajate seas regulaarseid platvormitöötajaid erinevate platvormitööde

üleselt 7% (tulemus pärast kaalumist). Laiendades tulemust tööealise elanikkonna

arvule, teeb platvormitööd iganädalaselt ligikaudu 56 000 inimest.

Võrreldes 2018. aasta tulemustega on regulaarsete platvormitöötajate osakaal

tööealise elanikkonna hulgas mõnevõrra vähenenud. 2018. aastal tegi platvormitööd

regulaarselt 8,1% elanikkonnast.

Joonis 1 Platvormitöö tegemise sagedus 2018 ja 2021 (% kõikidest vastanutest)

Varasemast enam on neid, kes teevad platvormitööd aeg-ajalt või on kunagi

platvormitööd teinud. Kui jagame platvormitöötajad platvormitöö tegemise sageduse

järgi gruppidesse (arvestades iga platvormitöötaja vaid ühte sagedusgruppi), siis

näeme, et vähemalt kord kuus teeb platvormitööd 11,9% vastanutest, vähemalt kord

1,70%

2,10%

4,10%

6,90%

4,50%

4,70%

6,30%

8,90%

0,00% 10,00%

Iga päev

Peaaegu iga päev

Umbes kord või kaks nädalas

Umbes kord või kaks kuus

Umbes kord kolme kuu tagant

Umbes kord kuue kuu tagant

Umbes kord aastas

Vähem kui kord aastas

Platvormitöö tegemise sagedus

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

11

aastas ligikaudu viiendik ning kunagi on platvormitööd teinud enam kui veerand

tööealisest elanikkonnast (vastavalt 96 000 inimest, 160 000 inimest ning 208 000

inimest) (Tabel 1).

Platvormitöö sagedus 2021 2018

Vähemalt kord nädalas 7% 8,1%

Vähemalt kord kuus 11,9% 10,3%

Vähemalt kord aastas 20,2% 16,9%

On kunagi üldse platvormitööd teinud (Kokku) 26% 19,5%

Tabel 1 Platvormitöö sagedus 2021 ja 2018 võrdluses (% kõikidest vastanutest)

Millist platvormitööd tehakse?

Eesti platvormitöötajatest teeb 24% iganädalaselt keerukamaid oskuseid eeldavat

veebipõhist tööd (loominguline töö ja IT-alane töö, näiteks graafiline disain,

tarkvaraarendus, tõlkimine) ja 18,9% teeb lihtsamaid veebipõhiseid töid

(administratiivne töö ja lihtsamad ülesanded).

Võrreldes 2018. aastaga tehakse sarnasel määral töid teiste inimeste kodudes, näiteks

ehitustööd, remont, koristus (21,6% 2021. aastal ja 24,2% 2018. aastal), ning samal

määral (13,7%) kontoritöö laadi töid. Vähem osutatakse digiplatvormide vahendusel

isikuhooldusteenuseid, näiteks juuksuriteenust (6,9% 2021. aastal ja 12,4% 2018.

aastal). Sõidujagamisteenuse osakaal on koroonakriisist hoolimata kasvanud. Kui

2018. aastal pakkus sõidujagamisteenust iganädalaselt 11,4% platvormitöötajatest,

siis 2021. aasta kevadel juba 19%. On võimalik, et enne koroonakriisi algust oli

sõidujagamise osakaal veelgi suurem.

Kulleriteenust osutas 2021. aastal ligikaudu veerand ehk 26,5% platvormitöötajatest.

Kulleriteenuse osutamise näitajat ei saa 2018. aasta tulemusega muutunud

küsitlusmeetodi tõttu võrrelda. Küsitluse valimisse sattusid vaid eesti ja vene keelt

kõnelevad inimesed – on tõenäoline, et ajutiselt Eestis viibivate kulleritööd tegevate

inimeste tõttu on kulleritöö osakaal kogu platvormitöös oluliselt suurem.

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

12

Üks inimene võib teha erinevat tüüpi platvormitöid erineva sagedusega. Seetõttu on

küsitluses märgitud sageduste summa suurem kui 100%. Näiteks võib selle põhjuseks

olla soov suurendada oma sissetulekut erinevate allikate kaudu. Iganädalaselt teeb

ühte tüüpi platvormitööd 5,2% ning kahte ja enamat platvormitööd 1,8% vastanutest.

Igakuiselt teeb ühte platvormitööd 8,1% vastanutest, kahte tüüpi platvormitööd 2,5%

ning 1,3% vastanutest teeb kolme või enamat tüüpi platvormitööd.

Joonis 2 Platvormitöö tüübid regulaarsete platvormitöötajate seas 2018 vs 2021 (%)

Platvormitöö kogemus aastates

Kõige rohkem on platvormitöötajate hulgas neid, kes on platvormitööd teinud alla aasta (39%) ning 1–2 aastat (26%). Veidi enam kui veerand vastanutest oli platvormitööd teinud 3–4 aastat ja 19% kauem kui 4 aastat (

Joonis 3).

18,9%

24,3%

15,1%

13,7%

21,6%

19,0%

6,90%

26,50%

0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%

Veebipõhine lihtsam töö

Keerukamaid oskuseid eeldav veebipõhine töö

Professionaalsed teenused

Kontoritöö laadi tööd

Ülesannete täitmine teiste kodudes

Sõidujagamisteenus

Isikuhooldusteenused

Kullerteenused

Millist platvormitööd tehakse?

2018 2021

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

13

Joonis 3 Platvormitöö kogemus aastates, platvormitööd kunagi teinud inimeste %

Platvormitöötajate vanus

Platvormitöötajate hulgas on esindatud kõik tööealise elanikkonna grupid, kuid

võrreldes kogu Eesti elanikkonnaga on platvormitöötajate seas enam nooremas

vanusgrupis inimesi (Joonis 4). Iganädalaselt platvormitööd tegevatest inimestest on

enim 35–44-aastaseid ja 25–34-aastaseid. Kõige vähem on regulaarselt platvormitööd

tegevate inimeste hulgas 18–24-aastaseid (11,5%) ja 55–64-aastaseid (14,4%).

Kunagi platvormitööd teinud inimeste vanuseline jaotus on sarnane. Kõikidest

platvormitöötajatest on 17,4% vanuses 18–24, 25,9% vanuses 25–34, 21% vanuses

35–44, 18,7% vanuses 45–54 ja 17% vanuses 55–64. Kunagi platvormitööd teinute

suur osakaal noorimas vanusegrupis iganädalaselt platvormitööd tegevate inimestega

võrreldes viitab asjaolule, et noored on kunagi platvormitööd proovinud või teevad

seda aeg-ajalt, kuid pigem on siin nende jaoks tegu juhusliku ja harva tööotsaga.

39,3%

25,6%

16,6%

18,5%

Kui mitu aastat Te olete digiplatvormide vahendusel teinud tööd?

Alla aasta 1–2 aastat 3–4 aastat Kauem

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

14

Joonis 4 Igal nädalal platvormitööd tegevate inimeste vanuseline jaotus kogu valimi vanusjaotusega võrreldes

Platvormitöötajate sugu

Platvormitöötajate seas on mõnevõrra rohkem mehi. Igal nädalal platvormitööd

tegevatest inimestest on 51,4% mehed ja 48,6% naised. Kõikidest platvormitöötajatest

on 52,3% mehed ja 47,7% naised. Vastanud meestest on kunagi platvormitööd teinud

27% ja naistest 25%. Võrreldes 2018. aasta küsitluse tulemustega on platvormitöö

sooline jaotus siiski oluliselt rohkem tasakaalus. Tookord oli platvormitööd kunagi

teinud vaid 13% naistest ning 26% meestest. Märgatavad on erinevused platvormitöö

tüüpides. Oluliselt rohkem on mehi sõidujagajate ja klikitöö tegijate hulgas. Naisi on

rohkem või samal määral isikuhooldusteenuste, loomingulise töö ja IT-töö tegijate

hulgas (Joonis 5). Siiski oli vastajate arv platvormitöö tüüpide jaotuse osas niivõrd

väike, et ei saa öelda, et tulemuste erinevus oleks statistiliselt oluline.

11,5%

26,6% 28,1%

19,4%

14,4%

10,9%

22,7% 23,1% 21,8% 21,5%

0,0%

5,0%

10,0%

15,0%

20,0%

25,0%

30,0%

18-24 25-34 35-44 45-54 55-64

Iganädalane Kogu valim

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

15

Joonis 5 Platvormitöö tegijate sooline jaotus

Platvormitöötajate tegutsemispiirkond

Platvormitöö on koondunud suurte tõmbekeskuste ja eriti pealinna ümbrusesse.

Iganädalastest platvormitöötajatest 75% elab linnas, 8% väikelinnas ning 17% maal.

Joonis 6 Regulaarsete platvormitöötajate elukoht Eesti piirkonna lõikes

0,0% 50,0% 100,0%

Sõidujagamine

Kullerteenused

Töö teiste kodudes

Isikuhooldusteenused

Asjatoimetused kliendi ruumides

Klikitöö

Loominguline ja IT-töö

Professionaalsed teenused

Vähemalt kord aastas platvormitööd tegevate inimeste

sooline jaotus (%)

Mehi Naisi

48,6%

8,6% 4,3%

13,6%

25,0%

Regulaarsete platvormitöötajate elukoht

Põhja-Eesti Lääne-Eesti Kesk-Eesti Kirde-Eesti Lõuna-Eesti

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

16

Iganädalaselt platvormitööd tegevatest inimestest ligi pooled tegutsevad Põhja-Eestis

(Joonis 6). Eriti märgatav on Põhja-Eesti ehk Tallinna lähikonna päritolu sõidujagajate

hulgas, kellest 67% on märkinud piirkonna oma elukohaks. Iganädalasi

platvormitöötajaid on palju ka Lõuna-Eestis ehk Tartu lähiümbruses.

Platvormitöötajate hulgas on kõige vähem Lääne- ja Kesk-Eestis elavaid inimesi

(Joonis 7).

Joonis 7 Iganädalase platvormitöö piirkondlik jaotus

Platvormitöö koondumine linnadesse ning tõmbekeskuste ümber on loomulik, sest

need on piirkonnad, kus elabki rohkem inimesi ja on piisavalt suur turg. Huvitavam on

jälgida, kas mõne piirkonna elanike seas on vastanute arvuga võrreldes rohkem või

vähem platvormitöötajaid kui mujal. Andmetest on näha, et iganädalaste

platvormitöötajate osakaal on kõige suurem Narva, Kohtla-Järve, Sillamäe ja Jõhvi

linnades (kokku 11,6%). Järgnevad Tartu (9,9%), seejärel Pärnu (8,9%) ja alles siis

Tallinn (8,2%). Muude asulate elanikest teeb regulaarselt platvormitööd 4,8%.

Platvormitöötajate rahvus

Ankeedile vastajad said valida küsimustele vastamiseks eesti või vene keele. Vene

keelt vastamiskeelena eelistanud inimeste seas on regulaarsete platvormitöötajate

osakaal märgatavalt suurem kui eesti keelt eelistavate vastajate hulgas. Kui ankeedile

0,0% 20,0% 40,0% 60,0% 80,0% 100,0% 120,0%

Iganädalane kokku

Sõidujagamine

Kullerteenus

Töö teiste inimeste kodudes

Isikuhooldusteenused

Jooksvad asjatoimetused kliendi ruumides

Administratiivne töö veebis

Loominguline või IT-alane töö veebis

Professionaalne töö

Iganädalase platvormitöö piirkondlik jaotus

Põhja-Eesti Lääne-Eesti Kesk-Eesti Kirde-Eesti Lõuna-Eesti

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

17

vastanud eestlastest tegi iganädalaselt platvormitööd 6,3%, siis venekeelsele

ankeedile vastanutest tegi platvormitööd iganädalaselt 8,5%. Samamoodi on

igakuiseid platvormitöötajaid 11% eestikeelsele ankeedile vastanutest ja 13,9%

venekeelsele ankeedile vastanutest.

Mõned platvormitöö tüübid (kulleriteenus, sõidujagamine) eeldavad vähest kohaliku

keele oskust ja sealne sisenemisbarjäär on madal, sest need tööd ei nõua ei keerulisi

oskusi ega tule läbida ka nõudlikku värbamisprotsessi. Kuivõrd ankeedile sai vastata

üksnes eesti või vene keeles, võib eeldada, et mitte-eesti päritolu inimeste osakaal

platvormitöö tegijate seas on alahinnatud.

Platvormitöötajate haridustase

Platvormitöötajate haridustase platvormitöö tegemise sageduse lõikes oluliselt ei erine

ning on sarnane kogu elanikkonna haridustaseme jaotusega. Suurem osa (54,7%)

platvormitöötajatest on keskharidusega (valimis keskmiselt 55,8%), 15,1% on

põhiharidusega (keskmiselt 14,1%) ning 30,2% on kõrgharidusega (keskmine on

30,1%) (Joonis 7).

Joonis 8 Platvormitöötajate haridustase

15,1%

54,7%

30,2%

Regulaarselt platvormitööd tegevate inimeste haridustase (%)

Põhiharidus Keskharidus Kõrgharidus

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

18

Kas platvormitöötajad erinevad sõltuvalt platvormitöö tüübist?

Seni oleme vaadelnud platvormitöötajaid kõikide uuritud platvormitöö tüüpide kohta

kokku. Kuid nagu pole olemas keskmist inimest, ei ole olemas ka keskmist

platvormitöötajat, sest platvormitöö mõiste tähistab väga erinevaid töid. Erinevad tööd

eeldavad erinevaid oskuseid ning nii töökorraldus kui ka töötasu on erinev.

Otstarbekas on jagada peamised platvormitöö tüübid kohapõhisteks ja

veebipõhisteks töödeks ning väheseid oskuseid eeldavateks ja

professionaalseid oskuseid eeldavateks töödeks (vt Joonis 9). Uurisime

sõidujagamise ning veebipõhise loome- ja IT-töö näitel, kas ja millistes aspektides

platvormitöötajad üksteisest erinevad.

Joonis 9 Platvormitöö jaotus: kohapõhine-veebipõhine ning väheseid oskuseid – professionaalseid oskuseid eeldavad tööd jaotuses

Mõlemat tüüpi platvormitööd teeb võrreldes 2018. aastaga regulaarselt suurem hulk

inimesi – kui 2018. aastal tegeles 11,4% iganädalastest platvormitöötajatest

sõidujagamisega ja 20,3% keerukamaid oskusi eeldava veebipõhise tööga, siis 2021.

aastal tegeles 19% iganädalastest platvormitöötajatest sõidujagamisega ning 24,3%

veebipõhise keerukama tööga. Suurendamaks tulemuste usaldusväärsust, kaasasime

Kohapõhine ja professionaalseid oskuseid eeldav (nt kontoritöö,

õpetamine, hooldus,

professionaalsed teenused)

Veebipõhine ja professionaalseid oskuseid eeldav

(nt IT, disain)

Kohapõhine ja väheseid oskuseid

eeldav (kullerteenus,

sõidujagamine,

Veebipõhine ja väheseid oskuseid eeldav (nt klikitöö)

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

19

võrdlusesse kõik vastavat tüüpi platvormitööd tegevad inimesed, kes teevad

platvormitööd vähemalt kord aastas.

Esmalt on näha, et platvormitöötajate sooline jaotus sõidujagajate ja keeruka

veebipõhise töö vahel on erinev. Kui loomingulise või IT-töö tegijate seas on võrdselt

mehi ja naisi (mehi 50,5% ja naisi 49,5%), siis sõidujagajate hulgas on 57,3% mehi ja

42,7% naisi (Joonis 10). Viimane tulemus võib muidugi olla tingitud Eesti tööturu

ajaloolisest soolisest segregeerumisest taksoteenuse valdkonnas.

Joonis 10 S Sõidujagajate ja keeruka veebipõhise töö tegijate sooline võrdlus

Erinevust on märgata ka platvormitöötajate rahvuses. Eelnevalt nägime, et keskmiselt

on suurem platvormitöötajate osakaal nende seas, kes vastasid venekeelsele

ankeedile ja väiksem nende seas, kes vastasid eestikeelsele ankeedile. Võrreldes

veebipõhist ja kohapõhist platvormitööd, on keele-eelistuse erinevus silmatorkav –

loomingulise või IT-alase veebipõhise platvormitöö tegijate seas on u 67% eestlasi ja

33% mitte-eestlasi (ehk inimesi, kes vastasid venekeelsele ankeedile), sõidujagajate

seas on võrdselt eesti ja vene keeles ankeedile vastanuid. Valimis oli eestlasi

keskmiselt 69% ja mitte-eestlasi 31% (Joonis 11).

57,3%

50,5%

42,7%

49,5%

0,0% 20,0% 40,0% 60,0% 80,0% 100,0% 120,0%

Sõidujagamine

Loominguline või IT-alane töö veebi kaudu

Veebipõhiste ja kohapõhiste platvormitöötajate

sooline võrdlus

Mees Naine

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

20

Joonis 11 Sõidujagajate ja keeruka veebipõhise töö tegijate võrdlus rahvuste lõikes

Erineb ka sõidujagajate ja veebipõhise keeruka töö tegijate vanuseline jaotus (Joonis

12). Kui sõdujagajate hulgas on suurim 35–44-aastaste inimeste osakaal (30,2%) ning

sellele järgneb 45–54-aastaste vanusgrupp (24,9%), siis keeruka veebipõhise töö

tegijad on nooremad – enim oli neid vanuses 25–34 (28,4%) ning 18–24 (21,9%).

Joonis 12 Sõidujagajate ja keeruka veebipõhise töö tegijate vanuseline võrdlus

Mõneti üllatuslikult ei ole olulisi erinevusi platvormitöötajate haridustasemes. Kui

mõlemas platvormitöötajate kategoorias on põhiharidusega inimesi veidi üle 12%

(sõidujagajate seas 12,7% ja veebipõhise töö tegijate seas 12,3%), siis põhiliselt on

nii sõidujagajad kui ka veebipõhise töö tegijad keskharidusega (57,5% sõidujagajatest

49,5%

66,7%

50,5%

33,3%

0,0% 50,0% 100,0%

Sõidujagamine

Loominguline või IT-alane töö veebi kaudu

Veebipõhise ja kohapõhise platvormitöötaja rahvuse võrdlus

Eestlane Mitte-eestlane

12,5%

18,7%

30,2%

24,9%

13,7%

21,9%

28,4%

20,0%

15,7% 14,0%

0,0%

5,0%

10,0%

15,0%

20,0%

25,0%

30,0%

35,0%

18-24 25-34 35-44 45-54 55-64

Veebipõhiste ja kohapõhiste platvormitöötajate vanuse võrdlus

Sõidujagamine Loominguline või IT-alane töö veebi kaudu

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

21

ja 54,5% veebipõhistest töötajatest), mis võib viidata sellele, et IT-alaseid või muid

professionaalseid oskuseid on võimalik omandada ka näiteks täienduskursustel ega

pea kindlasti ülikoolis käima. Alternatiiv paistab olevat selles, et need ülesanded, mis

on leitavad veebipõhiste platvormide vahendusel, ei ole võrreldavad keeruka

tavatööga. Varasemas erialakirjanduses on leitud, et keerukamaid töid vahendavatel

veebiplatvormidel jagatakse tööd paljudeks eri osadeks ning ühele ülesandele

spetsialiseerunud platvormitöötaja täidab järjest ning korduvalt just üht kindlat

ülesannet erinevate klientide jaoks. Üks võimalus on see, et veebipõhise platvormitöö

tegijad alles õpivad ülikoolis (mida kinnitaks ka nende keskmisest noorem iga), aga

põhitegevusalade võrdlus seda väidet ei tõenda.

Joonis 13 Sõidujagajate ja keeruka veebipõhise töö tegijate haridustaseme võrdlus

Ka tööalase staatuse poolest on sõidujagajad ja veebipõhise töö tegijad pigem

sarnased. Keerukat veebipõhist platvormitööd tegevate inimeste seas on enam

spetsialiste (26,7%, sõidujagajate seas 22,5%). Sõidujagajate hulgas on rohkem

oskus- või lihttöölisi (26,6% vs. 23%). Sarnane on mõlemas grupis mittetöötavate

inimeste, kontori- või teenindussektori töötajate ja juhtide osakaal.

12,7%

57,5%

29,8%

12,3%

54,5%

33,2%

0,0%

10,0%

20,0%

30,0%

40,0%

50,0%

60,0%

70,0%

Põhiharidus Keskharidus Kõrgharidus

Veebipõhiste ja kohapõhiste platvormitöötajate haridustaseme

võrdlus

Sõidujagamine Loominguline või IT-alane töö veebi kaudu

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

22

Joonis 14 Sõidujagajate ja keeruka veebipõhise töö tegijate elukoha võrdlus tööalase staatuse lõikes

Igakuiste (neto)sissetulekute jaotuses paistab silma, et mõlemas grupis moodustavad

suurema osa inimesed, kelle kogusissetulek jääb alla Eesti keskmise palga või ulatub

veidi üle selle2 (Joonis 15). Veebipõhise keerukama platvormitöö tegijate seas on

suurim nende osakaal, kelle sissetulek jääb vahemikku 701–1000 eurot (30,4%),

sellele järgneb netopalk 1001–1200 eurot (9,4%) ja 401–700 eurot (7,3%).

Sõidujagajate seas on enim neid, kelle netopalk on vahemikus 401–700 eurot kuus

(24,5%) ja 1001–1200 eurot kuus (23,6%).

2 Eesti keskmine brutopalk 2020. aasta keskmisena oli 1448 eurot (allikas: Statistikaamet). Kui arvestada maksuvaba tulu, kohustuslikke makseid ning II pensionisamba makseid, oli vastav netopalk 1189 eurot kuus.

19,9%

22,5%

25,4%

26,6%

2,8%

2,8%

18,9%

26,7%

24,4%

23,0%

3,5%

3,5%

0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%

Ettevõtte juht, keskastme juht, üksikettevõtja, FIE

Spetsialist

Kontori- või teenindussektori töötaja

Oskus- või lihttööline

Pensionär (mittetöötav), töövõimetuspensionär (mittetöötav)

Kodune, õpilane, töötu

Veebipõhiste ja kohapõhiste platvormitöötajate tööalane

staatus

Loominguline või IT-alane töö veebi kaudu Sõidujagamine

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

23

Joonis 15 Sõidujagajate ja keeruka veebipõhise töö tegijate võrdlus (neto)sissetuleku lõikes kõikide sissetulekuallikate peale kokku

Sõidujagajate ja keerukat veebipõhist tööd tegevate inimeste elukoha erinevus on

märgatav. Sõidujagajate hulgas on oluliselt rohkem neid, kes elavad Põhja-Eestis ehk

Tallinna lähiümbruses (Joonis 16). Piirkondlikust erinevusest veelgi olulisem on aga

piirkonna tüüp. Mõlema grupi esindajad elavad pigem linnas, kuid veebipõhise töö

tegijate hulgas on rohkem neid, kes elavad maal (Joonis 17). Ka ainult linnu vaadeldes

on näha, et veebipõhise töö tegijad on Eesti suuremate linnade vahel ühtlasemalt

jaotunud, samal ajal kui sõidujagajate hulgas on suurem osa inimesi Tallinnast või Ida-

Virumaa linnadest (Joonis 18). See tulemus oli ka eeldatav, sest sõidujagamise

teenust saab pakkuda ainult valitud Eesti linnades.

Joonis 16 Sõidujagajate ja keeruka veebipõhise töö tegijate piirkondlik võrdlus

0,0%

5,0%

10,0%

15,0%

20,0%

25,0%

30,0%

35,0%

Kuni 400 eurot

401-700 eurot 701-1000 eurot

1001-1200 eurot

1201-1400 eurot

1401-2000 eurot

Üle 2000 euro

Veebipõhiste ja kohapõhiste platvormitöötajate sissetuleku

jaotus võrreldes kogu valimiga (neto)

Sõidujagamine Loominguline või IT-alane töö veebi kaudu Kogu valim

0,0%

10,0%

20,0%

30,0%

40,0%

50,0%

60,0%

Põhja-Eesti Lääne-Eesti Kesk-Eesti Kirde-Eesti Lõuna-Eesti

Veebipõhiste ja kohapõhiste platvormitöötajate elukoha

võrdlus (piirkond)

Sõidujagamine Loominguline või IT-alane töö veebi kaudu

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

24

Joonis 17 Sõidujagajate ja keeruka veebipõhise töö tegijate piirkondlik võrdlus linn-maa lõikes

Joonis 18 Sõidujagajate ja keeruka veebipõhise töö tegijate elukoha võrdlus linnade lõikes

Eelneva analüüsi põhjal saame üldistada, et keerukama veebipõhise töö tegijad on

sõidujagajatega võrreldes nooremad, nende seas on rohkem naisi, nad räägivad

emakeelena pigem eesti keelt ning elavad sagedamini maapiirkonnas, kuid neil ei ole

ei kõrgemat haridustaset ega ka oluliselt suuremat sissetulekut. See võib viidata

asjaolule, et kuigi sõidujagamine on madala sisenemisbarjääriga töö, teevad seda hobi

korras ka kõrgharitud inimesed. Teisalt võib olla nii, et platvormide vahendusel tehtav

92,6%

86,5%

7,4%

13,5%

75,0% 80,0% 85,0% 90,0% 95,0% 100,0% 105,0%

Sõidujagamine

Loominguline või IT-alane töö veebi kaudu

Veebipõhiste ja kohapõhiste platvormitöötajate elukoha

võrdlus (linn-maa)

Linn Maa

0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0% 35,0% 40,0% 45,0%

Tallinn

Tartu

Pärnu

Narva, Kohtla-Järve, Sillamäe, Jõhvi

Muu asula

Veebipõhiste ja kohapõhiste platvormitöötajate elukoha

võrdlus (linn)

Sõidujagamine Loominguline või IT-alane töö veebi kaudu

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

25

töö on jagatud osadeks või vahendatakse seal vähem keerukaid töid ning sellega

saavad hakkama ka inimesed, kellel ei ole antud valdkonnas ei pikka kogemust ega

kõrgharidust.

Kokkuvõttes saime kinnitust, et platvormitöötajate profiil on olenevalt tehtavast tööst

erinev ning seega tuleks ka näiteks poliitikakujunduses erinevaid gruppe eraldi

käsitleda.

Platvormide roll professionaalsete teenuste turul suureneb

Käesolev platvormitöö küsitluslaine tõi esile, et suurenenud on nii professionaalsete

kui ka veebipõhiste platvormitööde osakaal. Võime eeldada, et tulevikus see

suundumus jätkub. Esiteks need valdkonnad, kus koroonakriisi ajal on toimunud

oluline nihe – inimesed on harjunud üha rohkem iseseisvalt ja virtuaalselt töötama.

See harjumus on sageli hüppelaud iseendale tööandjana lisasissetuleku

otsimiseks. Teisest küljest võimendab suundumust see, et digiplatvormide

vahendusel töötamine on muutunud harjumuspärasemaks ja tekkinud on ka

keerukamaid spetsiaalseid oskustöid vahendavaid platvorme – näiteks Hugo Legal,

mis vahendab juristide teenuseid.

Mis iseloomustab regulaarset, vähest ja harva platvormitöötajat?

Jagasime kõik kunagi platvormitööd teinud inimesed nelja rühma – igal nädalal

platvormitööd tegevad inimesed ehk regulaarsed platvormitöötajad, ligikaudu kord

kuus platvormitööd tegevad inimesed, vähemalt kord aastas platvormitööd tegevad ja

kunagi varem või üldse mitte platvormitööd teinud inimesed. Mis paistab selles

võrdluses silma?

Näeme, et iganädalaste platvormitöötajate seas on veidi enam mehi, pigem on nad

vanusegrupis 35–44, kuid 25–34-aastaste osakaal on samuti suur. Pigem elavad nad

linnas ja eriti palju on Tallinnas ja Tartus elavaid inimesi. Kirde-Eesti linnade kõikide

elanike osakaal regulaarsete platvormitöötajate seas on suurem kui teistes

piirkondades keskmiselt. Mitte-eestlased teevad platvormitööd võrreldes eestlastega

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

26

suuremas mahus, aga absoluutarvuna on neid siiski vähem. Gruppide võrdluses on

siin kõige suurem vene keelt rääkivate inimeste osakaal. Suurim osa regulaarsetest

platvormitöötajatest teenib igakuiselt kokku 401–700 eurot ning eriti paistab see silma

mitte-eestlaste puhul. Võrreldes teiste gruppidega on siinses grupis enim iseendale

tööandjaid ja ettevõtete juhte (siia alla kuulub ka isiklik OÜ ja FIE-na tegutsemine), kuid

kõige rohkem on regulaarsete platvormitöötajate hulgas spetsialiste. Iganädalaselt

platvormitööd tegevate inimeste hariduslik jaotus ei erine oluliselt kogu valimi

keskmisest.

Igakuiselt ehk natuke platvormitööd tegevate inimeste puhul paistab silma see, et enim

on selles grupis noori inimesi – vanuses 18–24 on siin rohkem inimesi kui teistes

platvormitöötajate gruppides. Pigem elavad nad Põhja-Eestis ja eriti Tallinnas. Selles

grupis on rohkem ka keskharidusega inimesi, kuid nende keskmine sissetulek on

analüüsitavate gruppide võrdluses suurim. Põhitegevuse plaanis paistab silma suur

oskus- ja lihttööliste osakaal.

Inimesed, kes teevad platvormitööd vähemalt kord aastas, kuid harvemini kui kord

kuus ehk harvad platvormitöötajad, paistavad silma keskmisest kõrgema vanusega

– enam on inimesi vanusegrupis 55–64 eluaastat. Pigem elab harv platvormitöötaja

Tallinnast väljaspool – Lõuna-, Lääne- ja Kirde-Eestis, ka maal – ning on

põhiharidusega. Võrreldavate gruppide lõikes on harvade platvormitöötajate sissetulek

kõige väiksem.

Kui võrrelda regulaarselt, veidi ja harva platvormitööd tegevate inimeste gruppe kunagi

platvormitööd teinud või üldse mitte platvormitööd teinud inimestega, on näha, et

viimase grupi keskmine sissetulek on kõige suurem. Seega on platvormitöötajate

keskmine igakuine netosissetulek kõikide sissetulekuallikate peale kokku

väiksem kui platvormitööd mitte tegevatel inimestel.

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

27

2.3. Fookuses: platvormitöötajate

toimetulek ja kriisi mõju

Tihtipeale räägitakse platvormitööst kui võimalusest paindlikult lisasissetulekut

teenida. On inimesi, kes teevad platvormitööd hobi korras, näiteks teiste inimestega

suhtlemiseks või uue oskuse omandamiseks. Teisalt on leitud, et Eesti tööturul on suur

hulk palgavaesuses inimesi, kelle jaoks võib platvormitöö pakkuda hädavajalikku

lisasissetulekut ots otsaga kokkutulekul. See muudab nad vähem nõudlikuks

töötingimuste ja sotsiaalkaitse osas. 2021. aasta küsitlusvoorus keskendusime

platvormitöötajate sissetulekule ja viiruskriisi mõjule. Lisasime ankeeti küsimused

tööaja, sissetulekute ja sotsiaalkaitse kohta.

Platvormitöötajate töökoormus

Platvormitöö keskmine tundide arv nädalas näitab, et üldjuhul on platvormitöö

kõrvaltöö. Ligi pool regulaarselt platvormitööd tegevatest inimestest teeb platvormitööd

nädalas kuni 10 tundi. Veerand vastajatest teeb platvormitööd 11–25 tundi nädalas,

veidi rohkem kui viiendik 25–40 tundi nädalas ning väike osa (7,2%) rohkem kui 40

tundi nädalas. Mõned üksikud viimase kategooria vastajad märkisid, et nad teevad

platvormitööd kõvasti üle tavapärase töönädala pikkuse – 50, 60 ja isegi 80 tundi

(Joonis 19).

Regulaarselt platvormitööd tegevate inimeste keskmine töökoormus ei erine nende

inimeste töökoormusest, kes platvormitööd ei tee (keskeltläbi 38,5 töötundi nädalas).

Kokku kuni 40 tundi nädalas töötavate platvormitöötajate platvormitöö osakaal on 14,5

tundi. Kuid inimestel, kes töötavad kokku rohkem kui 40 tundi nädalas, on oluliselt

suurem ka platvormitöö tundide arv – keskmiselt 26 tundi nädalas.

Kuni 40 tundi nädalas töötavate platvormitöötajate puhul on platvormitöö

osakaal kogu töös keskmiselt 14,5 tundi. Inimestel, kes töötavad kokku

rohkem kui 40 tundi nädalas, on oluliselt suurem ka platvormitöö tundide

arv – keskmiselt 26 tundi nädalas.

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

28

Joonis 19 Keskmine digiplatvormi(de) vahendusel töötatud tundide arv nädalas

Nädala keskmise digiplatvormide vahendusel töötatud aja osas paistavad silma Ida-

Virumaa platvormitöötajad. Ida-Virumaa linnade platvormitöötajate keskmine

platvormitöö tundide arv nädalas on regulaarselt 23,8 tundi (keskmine 18,2 tundi).

Kõige väiksem on Pärnu linna regulaarsete platvormitöötajate keskmine töötundide arv

– 12,9 tundi. Sealjuures on Pärnu linna elanike keskmine töötundide arv kõiki töid

arvestades 40,2 tundi nädalas, aga Ida-Virumaa linnade (Narva, Kohtla-Järve, Jõhvi,

Sillamäe) elanikel vaid 37,9 tundi nädalas.

Platvormitöö osakaal kogu sissetulekus

Enamiku platvormitöötajate jaoks on platvormitöö lisasissetuleku allikas.

Platvormitöötajad, kes on viimase aasta jooksul platvormitööd teinud (ca 160 000

inimest), teenivad platvormitööd tehes keskmiselt 18,4% oma kuu netosissetulekust.

Veidi enam teenivad mehed (18,8%) ja vähem naised (17,8%). 81,3% vastajatest

ütles, et nad teenivad sel viisil kuni poole oma sissetulekust (2018. aastal 76,4%).

Vähemalt poole sissetulekust moodustab platvormitöö 18,7% platvormitöötajatest

(2018. aastal 23,6%). Platvormitöö on ainus sissetulekuallikas 4,4%

platvormitöötajatest (2018. aastal 4,2%). Kõige suurema osa sissetulekust moodustab

platvormitöö vanuses 35–44 (21,1%) ja kõige väiksema vanusegrupis 25–34 (15,5%).

46,6%

25,0%

21,2%

7,2%

Keskmine töötatud tundide arv digiplatvormi(de) vahendusel

nädalas

Kuni 10 tundi 11-25 tundi 25-40 tundi 40+ tundi

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

29

Vähemalt kord aastas platvormitööd teinud inimeste platvormitööst teenitavas

sissetulekus on olulised erinevused sõltuvalt nende põhitegevusest. Kõige suurema

osa teenivad platvormitööst ettevõtjad (sh FIE-d) (27,6%), neile järgnevad kontori- või

teenindussektori töötajad (20,9%), kodused inimesed, õpilased või töötud (18,9%) ja

spetsialistid (18,4%).

Küsimusele ei osanud või ei soovinud vastata 40% platvormitöötajatest.

Joonis 20 Platvormitöö osakaal platvormitöötajate netosissetulekus põhitegevuse järgi (%)

Võrreldes regulaarsete platvormitöötajate kogusissetulekut (mh palk, platvormitööst

saadud sissetulek, stipendiumid, toetused) kogu elanikkonna jaotusega, näeme, et

suurem osa regulaarsetest platvormitöötajatest on need, kelle kogusissetulek on

vahemikus 401–1200 eurot. Kõrgemates sissetulekugruppides on platvormitööd mitte

tegevate inimeste osakaal suurem kui platvormitöötajate osakaal.

0,0 5,0 10,0 15,0 20,0 25,0 30,0

Ettevõtte juht, keskastme juht, üksikettevõtja, FIE

Spetsialist

Kontori- või teenindussektori töötaja

Oskus- või lihttööline

Pensionär (mittetöötav), töövõimetuspensionär (mittetöötav)

Kodune, õpilane, töötu

Platvormitöö osakaal platvormitöötajate

netosissetulekus põhitegevuse järgi (%)?

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

30

Joonis 21 Regulaarsete platvormitöötajate jaotumine isikliku netosissetuleku järgi kogu elanikkonnaga võrreldes

Platvormitöötajate põhitegevus

Regulaarselt (vähemalt kord nädalas) platvormitööd tegevate inimeste hulgas paistab

silma suur ettevõtjate (ka iseendale tööandjate) ja kontori- või teenindussektori

töötajate osakaal nende inimestega võrreldes, kes platvormitööd ei tee. Oskus- ja

lihttöölisi ning mittetöötavaid inimesi (pensionärid, kodused, töötud) on

platvormitöötajate hulgas vähem kui platvormitööd mitte tegevate inimeste hulgas.

7,9%

19,3%

16,4% 15,7%

9,3% 10,0%

6,4%

0,0%

5,0%

10,0%

15,0%

20,0%

25,0%

Kuni 400 eurot

401-700 eurot 701-1000 eurot

1001-1200 eurot

1201-1400 eurot

1401-2000 eurot

Üle 2000 euro

Sissetuleku võrdlus

Iganädalane Kogu valim

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

31

Joonis 22 Platvormitöötajate põhitegevus

Tulude laekumine

Esmakordselt küsisime platvormitöötajatelt, millisele kontole laekuvad nende tulud.

Küsimuse mõte oli mõista, kui paljud platvormitöötajad tegutsevad ettevõtja vormis

ning mil määral on platvormitöötajate seas ettevõtluskonto kasutajaid. Suurem osa

regulaarsetest platvormitöötajatest märkis, et nende sissetulekud laekuvad isiklikule

pangakontole Eestis. See tähendab, et platvormitöötajad käsitlevad platvormitööd

juhutuluna. 8,2% märkis, et tulu laekub nende ettevõttega seotud kontole ning alla 5%

märkis, et platvormitöö tulu laekub kas ettevõtluskontole (2,1%), muule pangakontole,

muul viisil (2,3%) või välismaisele pangakontole (1,3%).

On märkimisväärne, et ligikaudu 25% ei osanud küsimusele vastata. See tähendab, et

platvormitöötajad ei pruugi teada, kuhu nende tulud laekuvad, või ei soovinud seda

informatsiooni avaldada.

0,0% 5,0% 10,0% 15,0% 20,0% 25,0% 30,0%

Ettevõtte juht, keskastme juht, üksikettevõtja, FIE

Spetsialist

Kontori- või teenindussektori töötaja

Oskus- või lihttööline

Pensionär (mittetöötav), töövõimetuspensionär (mittetöötav)

Kodune, õpilane, töötu

Platvormitöötajate põhitegevus

Ei tee platvormitööd Iganädalane platvormitöö

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

32

Joonis 23. Konto, millele laekub digiplatvormidel teenitud tulu

näiteks poliitikakujunduses erinevaid gruppe eraldi käsitleda.

Platvormitöö vôiks tuua 68 miljonit maksutulu.

Kui kõik platvormitöötajad kasutaksid ettevõtluskontot, kes senimaani pole

kasutanud või pole olnud ettevõtjad või FIE-d, siis nende aastane tulu küündiks 340

miljoni euroni ja maksude summa 68 miljoni euroni.

Eeldatud on, et vähemalt korra aastas tegutsenud platvormitöötajaid on 20,2%

rahvastikust vanuses 18–64, mis vastab 161 000 inimesele. Nendest 88,1% ei

kasuta ettevõtluskontot või ei ole ettevõtjad ega FIE-d. Nende platvormitöö

keskmine kuine netotulu on 201 eurot, eeldatud on 20% maksumäära kohaldumist

(ettevõtluskontot kasutades kehtib maksumäär 20% tuludele, mis ei ületa 25 000

eurot aastas ning 40% seda summat ületavale tulule).

Koroonakriisi mõju platvormitööle

Koroonakriis on kaasa toonud turbulentsi tööturul ning seadnud paljud inimesed

ebakindlasse olukorda. Tavapärase töö kaotanud inimesed võivad hakata rohkem

platvormitööd tegema, et kompenseerida vähenenud töömahtu. Seepärast küsisime,

kas koroonakriisi algusest alates on platvormitöötajad tajunud platvormitöö töömahu

55,8%

25,7%

8,2%

4,5%

2,3%

2,1%

1,3%

0,0% 10,0% 20,0% 30,0% 40,0% 50,0% 60,0%

Isiklikule pangakontole Eestis

Ei oska öelda

Oma ettevõtte (AS, OÜ, FIE) pangakontole

Rahvusvahelise makseteenuse pakkuja (nt PayPal, Revolut) kontole

Muule kontole või muul viisil

Ettevõtluskontole

Välismaisele pangakontole

Kui Te teete digiplatvormi(de) vahendusel tööd, siis millisele

kontole peamiselt Teie teenitud tulud laekuvad?

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

33

suurenemist. Platvormitöö suurenev osakaal võib viidata sellele, et inimesed otsivad

aktiivselt täiendavat lisasissetulekut.

Rohkem kui 60% platvormitöö tegijatest leiab, et platvormitöö maht on pärast

viiruskriisi algust tõusnud. 15% leidis, et platvormitöö maht on langenud.

Joonis 24 Koroonaviiruse kriisi tulem platvormitöötajate töökoormusele

Küsitlusele vastanutelt küsiti ka seda, kas nad on saanud töötukassa töötasuhüvitist

ajavahemikus märts – juuni 2020. 20% vastajatest oli hüvitist saanud ning 80% ei

olnud.

20,2%

40,6%

15,9%

6,7%

8,8%

7,9%

Platvormitöö maht pärast koroonakriisi algust

Oluliselt tõusnud Pigem tõusnud Samaks jäänud

Pigem langenud Oluliselt langenud EI OSKA ÖELDA

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

34

Lisa 1. Vastajate profiil

Arv

Sugu Mees 1006 50,3%

Naine 994 49,7%

Vanus

18-24 218 10,9%

25-34 454 22,7%

35-44 462 23,1%

45-54 436 21,8%

55-64 430 21,5%

Rahvus Eestlane 1388 69,4%

Mitte-eestlane 612 30,6%

Haridus

Põhiharidus 284 14,2%

Keskharidus 1114 55,7%

Kõrgharidus 602 30,1%

Linn-maa Linn 1392 69,6%

Maa 608 30,4%

Linn-väikelinn-maa

Linn 1224 61,2%

Väikelinn 168 8,4%

Maa 608 30,4%

Regioon

Põhja-Eesti 928 46,4%

Lääne-Eesti 220 11,0%

Kesk-Eesti 178 8,9%

Kirde-Eesti 194 9,7%

Lõuna-Eesti 480 24,0%

Isiklik keskmine netosissetulek kuus

Kuni 400 eurot 177 8,8%

401-700 eurot 311 15,5%

701-1000 eurot 327 16,3%

1001-1200 eurot 273 13,7%

1201-1400 eurot 183 9,2%

1401-2000 eurot 255 12,7%

Üle 2000 euro 150 7,5%

EI OSKA ÖELDA 77 3,8%

KEELDUS VASTAMAST 248 12,4%

Põhitegevus

Ettevõtte juht, keskastmejuht, üksikettevõtja, FIE

266 13,3%

Spetsialist 466 23,3%

Kontori- või teenindussektori töötaja 400 20,0%

Oskus- või lihttööline 437 21,9%

Pensionär (mitte-töötav), töövõimetuspensionär (mitte-töötav)

78 3,9%

Kodune, õpilane, töötu 352 17,6%

Põhitegevus

Ettevõtte juht, keskastmejuht, üksikettevõtja, FIE

266 13,3%

Spetsialist 466 23,3%

Kontori- või teenindussektori töötaja/Oskus- või lihttööline

837 41,9%

Pensionär (mitte-töötav), töövõimetuspensionär (mitte-töötav)

78 3,9%

Kodune, õpilane, töötu 352 17,6%

Põhitegevus Töötav 1570 78,5%

Mitte-töötav 430 21,5%

KOKKU 2000 100,0%

Arenguseire Keskus | Lossi plats 1a, 15165 Tallinn | riigikogu.ee/arenguseire | [email protected]

35

ei saa eelviimane leht (ehk dokumendi sisekaan) tühjaks jäädaendi sisekaanele jätta

tek.

Arenguseire Keskus

Lossi plats 1a, 15165 Tallinn

[email protected]

riigikogu.ee/arenguseire

T A L L I N N A Ü L I K O O L I R A H V U S V A H E L I S T E S O T S I A A L U U R I N G U T E K E S K U S

P L A T F O R M I S A T I O N

O F T A L L I N N ’ S T A X I

I N D U S T R Y : R E S U L T S

F R O M T H E P L U S

P R O J E C T

RASI toimetised nr. 13

Kairit Kall Triin Roosalu

Marge Unt Liis Ojamäe

November 2021

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

2

Rahvusvaheliste Sotsiaaluuringute Keskus (RASI) on Tallinna Ülikooli Ühiskonnateaduste

Instituudi sotsiaalteaduslik interdistsiplinaarne teadus- ja arenduskeskus, mis teostab

teadusprojekte. RASI teadurid on tegevad ekspertidena ühiskonnaelu analüüsimisel ja

kujundamisel. TLÜ RASI uurimisteemad hõlmavad ühiskondliku ebavõrdsuse (või ka

kihistumise) erinevaid tahke – sugu, rahvus, vanus, põlvkond, haridus, ametipositsioon. Viimastel

aastatel on hakatud suurt tähelepanu pöörama elukestva õppe problemaatikale kui eluteed

kujundavale ja sotsiaalset sidusust Eestis ning laiemalt kogu Euroopa Liidus tagavale tegurile.

Teine uuem temaatika osakonna uurimistöös on seotud aktiivse vananemise küsimustega.

This report is prepared as part of project Platform Labour in Urban Spaces (PLUS).

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 822638.

The European Commission support for the production of this publication does not constitute an endorsement of the contents which reflects the views only of the authors, and the Commission cannot be held responsible for any use which may be made of the information contained therein.

Kall, K., Roosalu, T., Unt, M., & Ojamäe, L. (2021). Platformisation of Tallinn’s taxi industry: Results from the PLUS Project. RASI toimetised nr 13. Tallinn: Tallinna Ülikool.

ISSN 2613-733X ISBN 978-9949-29-589-0 (pdf)

Kontaktandmed

Rahvusvaheliste sotsiaaluuringute keskus Uus-Sadama 5, 10120 Tallinn Koduleht: http://www.tlu.ee/rasi Telefon: 619 9860 E-post: [email protected]

RASI sotsiaalmeedias

Twitter https://twitter.com/rasi_iiss Facebook https://www.facebook.com/TLURASI/ The research team would like to express their gratitude to all of the interviewees and participants of various events of the PLUS project who have devoted their time to share their knowledge, ideas and experiences with us. We are also thankful to Johanna Vallistu for reviewing the earlier version of the report and providing us valuable comments and suggestions on how to improve it.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

3

Contents

EXECUTIVE SUMMARY .................................................................................................................................................. 4

INTRODUCTION ................................................................................................................................................................ 9

METHODOLOGY ............................................................................................................................................................. 11

PART I: INSTITUTIONAL CONTEXT ....................................................................................................................... 16

1.1. Platformisation of the taxi industry: Competitive pressures and changing regulations

16

1.2. Working conditions, labour standards and collective mobilisation ..................................... 18

1.3. Employment status and tax compliance ........................................................................................... 20

1.4. Social protection ......................................................................................................................................... 23

1.5. Regulations ‘in becoming’ ....................................................................................................................... 26

1.6. City-level regulations for the taxi industry in Tallinn ................................................................. 28

1.7. Impact of the 2020 Covid-19 pandemic ............................................................................................ 29

PART II: PLATFORM WORKERS’ PERSPECTIVE ............................................................................................... 31

2.1 Labour process ............................................................................................................................................ 31

2.2 Skills ................................................................................................................................................................. 37

2.3 Social protection ......................................................................................................................................... 42

2.4 Perception of the impact of the pandemic ....................................................................................... 45

2.5 Alternatives to the current status quo ............................................................................................... 46

References ........................................................................................................................................................................ 48

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

4

EXECUTIVE SUMMARY

The executive summary highlights the main conclusions based on the empirical research

conducted in Tallinn between 2019 and 2021 for a European Union-funded project Platform Labour in Urban Spaces: Fairness, Welfare, Development (PLUS). The project aimed to map the

situation and changes in the platform economy in general and in the ride-hailing/taxi industry

and accommodation/hospitality sector in particular. The current summary focuses on the ride-

hailing/taxi industry in Tallinn. The comparative research conducted for the PLUS project

highlighted that Tallinn – when compared to the six European cities of Barcelona, Berlin, Bologna,

Lisbon, London and Paris – stands out as the least regulated when it comes to platform work

(Altenried et al., 2021). Our current analysis sheds light on the reasons behind this, and also

shows how this impacts the livelihoods of platform labour.

Several analyses that open up the Estonian context of the platform economy are available, for

example, by focusing on the “sharing economy” in particular or on the future of work more

generally, especially those written from the legal/macroeconomic point of view (e.g., Eljas-Taal

et al., 2016; Erikson & Rosin, 2018; Holts, 2018; Vallistu et al., 2017; Vallistu, 2018a).

Furthermore, a few survey-based quantitative reports that depict and analyse the group of

platform workers also exist (Huws et al., 2019; Vallistu & Piirits, 2021). Our analysis, however,

focuses on the different perspectives of both stakeholders and workers in the platform economy

and platform taxi driving. By studying the topics of working conditions, social security, and skills,

we aim to understand what it is like to work for a platform in a loosely regulated market situation,

including how people can experience it differently depending on their position, worldview and

identity.

The focus on studying working conditions, skills and social security derives from the fact that

these are the aspects that platform work has had a considerable impact on. The logic of how (lean)

platforms operate regarding their workers directly aims to change the traditional functioning of

the employee-employer relationship; platform workers themselves are responsible for making

sure they are earning a living wage, providing themselves social security coverage, and making

sure they are competent enough to safely provide the services. However, thus far, not much

research on these aspects has been conducted in Estonia, and the current analysis aims to address

this gap.

By studying these issues, we also aim to contribute to several ongoing policy debates both at the

national and European level. As platform work poses considerable challenges to the European

Pillar of Social Rights (European Commission, 2021a; Kampouri & Walsh, 2021) – a guiding

document promoting decent working conditions – the European Union has started to propose

policy initiatives for bringing platform work in line with it. Most importantly, in February 2021,

the European Commission launched a consultation process with European social partners on the

issues of improving the working conditions and social protection of platform workers (European

Commission, 2021b). This process will end in a possible legislative instrument that might become

applicable to the national labour markets as well. Therefore, it is important to understand the

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

5

current functioning of platform-based work and both the positive and negative sides it might pose

to actors involved in it.

The empirical research our results are based on includes secondary literature, available statistics,

expert interviews conducted between 2019 and 2021 (both individual and focus groups with

researchers, policymakers, enforcement agents, labour representatives, taxi company

representatives). This has helped us map the institutional setting influencing the ride-hailing/taxi

industry. In addition, we highlight the perspectives of platform workers, more precisely that of

platform taxi drivers, with whom we conducted individual in-depth interviews and one focus

group interview. The topics discussed included working conditions and labour processes, their

social security situation and what kinds of skills are necessary for this line of work and how

platform taxi driving impacts skill development. The individual interviews (15) were conducted

between November 2019 and March 2020, thus before the Covid-19 pandemic became

prominent. To gain an understanding of workers’ experiences during the pandemic, we also

concluded a focus group interview on 6 November 2020. It comprised four participants,

representing 2 types of platforms (namely Uber/Bolt and Airbnb).

Platforms entered Tallinn’s taxi industry in the mid-2010s, with the Estonian Taxify (now Bolt)

leading the way in 2013, followed by Uber in 2015. Initially, “ride-sharing” as a service was not

regulated and thus providing it was much more cost-efficient than providing regular taxi services

(where licences etc. were mandatory). Although taxi companies started using applications in

providing their services already before Uber and Bolt entered the market, the latter players have

been more successful in gaining market share, and also developing other app-based services like

food delivery. Taxi platforms started by engaging in price-dumping, providing services with lower

prices than traditional taxi companies, leading to the lowering of prices, as well as remuneration

for providing taxi services, on the one hand, and making the taxi service more

attractive/affordable to new customer groups, thus expanding the market (until the Covid-19

pandemic), on the other.

De-regulation of the taxi industry due to the inclusion of ride-hailing as a taxi service at the end

of 2017 has also had a considerable impact on the sector, including lowering the barriers to

becoming a taxi driver and also causing a de-professionalisation of the job. On the other hand,

there are still some regulations that are applicable to traditional taxis, but not platform-based

taxis, and control activities are difficult if not impossible to conduct on platform taxis. This, in

effect, caused unfair competition and made the sector also more insecure and prone to a grey

economy.

Important impacts of the platformisation of the industry on the working lives of platform taxi

drivers include the following. First, although the traditional taxi sector has relied mostly on (false)

self-employment (own-account workers) when hiring taxi drivers (e.g. taxi services are provided

via a one-person company who signs a contract with a larger one), and companies do not provide

employment contracts to drivers, platforms introduced a new practice that made the situation

more precarious: working as a natural person without any kind of contract with the platform.

Therefore, platform taxi drivers are totally responsible for covering their own taxes and social

security payments. This can lead to tax avoidance and a lack of social security. Second, operating

under (non-transparent) algorithmic management (often for multiple apps simultaneously) has

its own specific qualities. Although arguably you should be your own boss, in practice you have

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

6

to conform to platform standards (that drivers have no say over), otherwise you might even get

blocked by the algorithm without due process. In addition, your flexibility as a driver is limited

by existing bonus systems and available clients.

Unlike many other European cities, collective organization against platforms has been relatively

modest and no worker cooperatives/ community-owned platforms exist. The reasons why

collective mobilisation seems unfruitful to the interviewees relate to the huge pool of available

drivers, general (perceived) weakness of trade unions in Estonia, and the possibility of being

blocked by the app if drivers “act out”. Generally, it seems that both individual and collective

agency is constrained by the (perceived) control the taxi-apps exercise over the drivers. The

pandemic seems to have made these tendencies even more intense, as competition among drivers

has increased.

How the current situation is perceived by different actors, including platform taxi drivers,

however, is rather multifaceted – depending, for example, on the reason this line of work is chosen

(or was there really a choice), if it is the main income source or an additional one, and what kind

of values/identity the person has. Regarding important differences, we encountered both drivers

who considered taxi apps to be de facto employers (who should be given more responsibility) on

the one hand, and those who saw platforms as the ultimate expression of the free market, and

thought the situation should be even less regulated.

However, in general, Estonian society tends to be techno-optimistic and have trust in the free market (e.g., European citizen’s …, 2021; Lanamäki & Tuvikene, 2021), and this was also the case

with our interviewees. Peoples' work and life in general are becoming more and more intertwined

with technology, and Estonia has the image of being technologically progressive, especially in the

fields of information technology and e-services. Innovative technological solutions and related

know-how developed here are acknowledged and sometimes exported worldwide, but the

aforementioned image is even more shared and popular inside the country. The discourse of

Estonia being a “little but smart country” is frequently used by local politicians, entrepreneurs,

educational institutions and the media. Therefore, it has become the basis and explanation for

several political decisions – including the deregulation of the taxi industry – and functions sort of

as a narrative for national identity.

That in turn could partly explain why some interviewees warmly welcome and praise Uber, Bolt

and others, and emphasise the possibilities these platforms offer them. In addition, trust and a

belief that platforms know better seems to be connected with these tendencies. Those who have

a more negative attitude towards platforms, nevertheless, display a prevailing acquiescence; in

other words, a feeling that there is nothing much to do and the only way is to accept the

technological changes and algorithmic control in one's life. Although some interviewees

perceived the management strategies of the platform to be somewhat constraining and to

generate precariousness, most of them mentioned that this kind of work offers them freedom.

That fact can also be connected with local mentalities. Namely, after gaining re-independence

from the Soviet Union, Estonia has firmly chosen the road of a liberal market economy. This

course is vividly expressed in a slogan “return to the West” which affirms Estonia's wish to

dissociate itself from anything considered socialist or overtly left-wing (Helemäe & Saar, 2015;

Lauristin, 2003). Therefore, the idea of the free market and the aspect of individual choice is seen

in a particularly positive light. As the interviews indicate, being able to be, so to speak, your own

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

7

boss (switch application on or off whenever one pleases) is felt to go inevitably hand in hand with

deregulation; that is, keeping the government from intervening in the platform business models.

Furthermore, although platforms like Uber and Bolt collect huge amounts of data from drivers and

riders (to use as competitive advantage in the services market, but also for building alternative

services and infrastructure), in Estonia this is not really problematized either by experts or the

drivers themselves. Nevertheless, in Europe and elsewhere at the academic, activist and policy

levels, multiple issues regarding data collection, especially privacy concerns, but also

concentration of capital and monopolisation of certain services, have been highlighted (see, e.g.,

Booth, 2020; Monahan, 2020; Srnicek, 2017; Zhao et al., 2019).

In conclusion, taxi platforms have had a considerable impact on the industry and on the

profession of taxi drivers, including lowering service prices (as well as remuneration for

providing the taxi service), setting in motion the deregulation of the industry and causing a de-

professionalization and loss of job security. These tendencies, however, have been generally

accepted or even celebrated in Estonian society, without much resistance and elaboration on the

possible negative side-effects these developments can entail for the workers. Besides revealing

Estonia's contextual forces and local discourses, the interviews show us their negative impact. On

several occasions, the same interviewees who expressed sympathy towards the freedom that

platforms give them, also spoke about – in addition to the already mentioned aspects like the

feeling of being controlled by the app and additional precariousness – low wages and the lack of

social guarantees. Therefore, unregulated technological and entrepreneurial ambition leave their

mark on workers' well-being rather directly. At the same time, the popularity of such ambition

often inhibits workers to recognize that. Hence, it is clear that Estonia's overall (neo)liberal and

techno-optimistic mentality provides workers with narrative resources (Gubrium & Holstein,

2008); that is, topics and ways of talking and thinking that are presumed to be legitimate and

appropriate when discussing and framing the situation they find themselves in. Instead of

drawing connections between the negative side of their working conditions and the dominant

economic ideas, many interviewees saw these two as separate things and talked positively about

the latter.

Negative aspects that need to be tackled by a platform taxi driver daily can be divided into three

types: those related to coping with the poor image of an app-driver (one needs to do identity work

to overcome this); those related to coping with high-intensity competition and low prices; and

those related to coping with regular inherent dangers in this kind of job (danger from traffic,

customers, or greedy apps, is dealt with by being calm, patient, using good self-control, and,

ultimately, by turning down offers). Therefore, the drivers do not encourage seeing this as “easy

money” – it can be precarious, dangerous, there is a risk of overwork, and there are problems

with clients. Once a driver can deal with the challenges, among the positive aspects there is the

possibility to work in a field that has always been of interest; to work a schedule that is most

suitable; to prove yourself as an entrepreneur; to meet interesting people; and a chance to “get

lucky” (e.g. to find a (real) job). For some, the negative effects of taxi-apps are stronger than the

positive, and they search for a professional full-time job so they can quit. However, no one, even

among the most optimistic, sees the app as leading to a career path, and it is rather portrayed as

a dead-end job that does not contribute to their professional CV in positive ways. At the same

time, it is difficult to find a professional job while driving, but the skills and degrees may lose their

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

8

competitive advantage during this period. There is also uncertainty related to the present day and

the ability to cope with uncertainty thus is very important.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

9

INTRODUCTION

The impact of digital platforms and the accompanying platformisation of parts of the economy

are increasingly important topics for understanding current developments in capitalism,

including how these influence the working lives of a growing number of people. Although

terminological ambiguity exists, at the general level digital/online platforms can be defined as

“digital infrastructure that enable two or more groups to interact”, bringing together different

users, like service providers and customers (Srnicek, 2017, p. 43). The current report focuses on

a specific type of digital platforms, namely lean platforms: platforms that own software and data

analytics, but have outsourced other aspects of their business: workers, fixed assets, training etc.

(Srnicek, 2017). In the case of Uber and Bolt, for example, workers (i.e. drivers) are themselves

responsible for their employment, training, social security and working conditions.

Platform work1 can be defined as “the matching of the supply of and demand for paid work

through an online platform”. We will focus on the more specific type of work, namely on-location

(as opposed to online) platform work that is delivered in person (Eurofound, 2018, p. 1), as

driving the taxi via the platform would fall under this category. Sometimes it is also referred to as

ride-sharing (and the platform economy as the sharing economy), but as market relations

dominate and not all drivers actually use their own car, but rather rent one, the “sharing” aspect

seems not to be the central element for this line of work, and can be misleading. Furthermore,

taxi platforms exercise quite a lot of control over their drivers (Maffie, 2020). Therefore, we

prefer to use the term ride-hailing, or indicate that a person is driving a platform taxi.

Several analyses are available that open up the Estonian platform economy context by focusing

on the “sharing economy” in particular or on the future of work more generally, especially those

written from the legal/macroeconomic point of view (e.g., Eljas-Taal et al., 2016; Erikson & Rosin,

2018; Holts, 2018; Vallistu et al., 2017; Vallistu, 2018a). Furthermore, a few survey-based

quantitative reports that depict and analyse the group of platform workers also exist (Huws et

al., 2019; Vallistu & Piirits, 2021). Our analysis, however, focuses on the different perspectives of

both stakeholders and workers in the platform economy (specifically platform taxi driving). By

studying the topics of working conditions, social security and skills, we aim to understand what it is like to work for a platform in a loosely regulated market situation, including how people can

experience it differently depending on their position, worldview and identity.

The focus on studying working conditions, skills and social security derives from the fact that

these are the aspects that platform work has had a considerable impact on. The logic of how (lean)

platforms operate regarding their workers directly aims to change the traditional functioning of

the employee-employer relationship; platform workers themselves are responsible for making

sure they are earning a living wage, providing themselves social security coverage, and making

sure they are competent enough to safely provide the services. However, thus far, not much

research on these aspects has been conducted in Estonia and the current analysis aims to address

this gap.

1 Sometimes also called work in the sharing economy, gig work, crowd work or on-demand work.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

10

By studying these issues, we aim also to contribute to several ongoing policy debates both at the

national and European level. As platform work poses considerable challenges to the European

Pillar of Social Rights (European Commission, 2021a; Kampouri & Walsh, 2021) – a guiding

document promoting decent working conditions – the European Union has started to propose

policy initiatives for bringing platform work in line with it. Most importantly, in February 2021,

the European Commission launched a consultation process with European social partners on the

issues of improving the working conditions and social protection of platform workers (European

Commission, 2021b). This process will end in a possible legislative instrument that might also

become applicable to the national labour markets. Therefore, it is important to understand the

current functioning of platform-based work and both the positive and negative sides it might pose

for the actors involved in it.

The first part of the report is dedicated to the institutional setting that influences the ride-

hailing/taxi industry. It is based on secondary literature, available statistics and expert interviews

conducted in 2019 with key stakeholders, including experts in the field of social security, taxation,

public transport, and platform work in general. In the later stage of the project we also conducted

two group interviews (one in November 2020 focusing on platform work in general and the other

in March 2021 focusing on the taxi industry) with different stakeholders (researchers,

policymakers, enforcement agents, labour representatives, taxi company representatives) to

further elaborate the developments in the field, including discuss how the Covid-19 pandemic has

affected the situation of platform work and workers, and elaborate on possible policy changes

that might be necessary to enhance the situation of platform workers.

The second part of the report is devoted to the perspectives of platform workers, more precisely

that of platform taxi drivers. Although we specifically targeted Uber drivers, they mostly also used

other taxi applications (especially Bolt) for providing their services, making their experiences

with taxi platforms wider than just Uber. The main topics that were covered during the interviews

were related to working conditions and labour process, social security and skills. The fifteen

individual interviews were conducted between November 2019 and March 2020, thus before the

Covid-19 pandemic became prominent. To gain an understanding of workers’ experiences during

the pandemic, we also concluded a focus group interview on 6 November 2020.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

11

METHODOLOGY

This report is based on empirical research conducted in Tallinn between 2019 and 2021 for a

European Union-funded project Platform Labour in Urban Spaces: Fairness, Welfare, Development (PLUS). The project aimed to map the situation and changes in the platform economy in general

and in the ride-hailing/taxi industry and accommodation/hospitality sector in particular. The

current report focuses on the ride-hailing/taxi industry in Tallinn.

The first part of the report is dedicated to the institutional setting that influences the ride-

hailing/taxi industry. It is based on secondary literature and available statistics – including

publicly available state administrative statistics, survey data, newspaper articles, research

reports/articles, legislation (acts), and homepages of some relevant companies/organisations.

Furthermore, we conducted five semi-structured face-to-face and three e-mail interviews

(conducted between 2019 and 2021) with key stakeholders, including experts in the field of social

security, taxation, public transport, labour rights and platform work in general. The main aim of

expert interviews was to elaborate on the role and impact of platforms. The face-to-face

interviews lasted between 25 to 60 minutes. In the later stage of the project, we also conducted

two group interviews – one in November 2020 focusing on platform work in general and the other

in March 2021 focusing on the taxi industry in particular – with different stakeholders

(researchers, policy makers, enforcement agents, labour representatives, taxi company

representatives). These lasted 87 and 107 minutes respectively, and aimed to further elaborate

on the developments in the field, including how the Covid-19 pandemic has affected platform

work and workers, and to elaborate on possible policy changes that might be necessary to

enhance the situation for platform workers.

The second part of the report is devoted to the perspectives of platform workers, more precisely

that of platform taxi drivers. Although due to project specifics we targeted Uber drivers, they

mostly also used other taxi applications (especially Bolt) for providing their services, making

their experience with taxi platforms wider than just Uber. The main topics that were covered

during the interviews were related to working conditions and labour processes, social security

and skills. The individual interviews (15) were conducted between November 2019 and March

2020, thus before the Covid-19 pandemic became prominent. To gain an understanding of

workers’ experiences during the pandemic, we also concluded a focus group interview on

6 November 2020. It comprised four participants, representing 2 types of platforms (namely

Uber/Bolt and Airbnb). All participants had experienced platform work during pre-pandemic

times as well as during the pandemic.

Individual in-depth interviews with platform taxi drivers

To find Uber drivers who would be willing to give an interview, a variety of strategies were

employed. There was an attempt at convenience sampling, when two of our team members rode

with Uber and recruited four people this way – this strategy seemed to be rather difficult, as many

declined to be interviewed. In addition, access was sought through institutional settings, as we

posted a call for participation in two of the Uber drivers’ Facebook groups (one group is in English

and one in Estonian) and also advertised during a few classes at Tallinn University – this strategy

brought us eight interviewees. Finally, a snowballing design was applied, as we also used personal

contacts and recruited two acquaintances through this method. One of the interviewees was

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

12

found by asking relevant contacts from another interviewee. There were also a few cancellations;

these occurred with people we found via riding with Uber, and their excuse was not having

enough time (even when they initially agreed to be interviewed). This indicates that our sample

might be inclined toward the group of drivers who were not struggling the hardest to make ends

meet.

Our aim was to have as varied a sample as possible at least regarding age, gender, nationality

(mother tongue), educational level, experience driving a traditional taxi and work situation.

While considering some of these characteristics (age, experience driving a traditional taxi and

work situation), the final pool of interviewees was indeed rather diverse, men (13 out of 15),

people with higher education (10 out of 15) and Estonians (whose mother tongue is also

Estonian, 11 out of 15) were clearly over-represented (see Table 1).

Table 1: Characteristics of the interviewed Uber drivers Number of interviewees Sex Male 13 Female 2 Age group Below 30 5 30–50 7 Over 50 3 Nationality/majority/ minority

Estonian 11 Estonian-Russian 2 New immigrants 2

Education level Secondary 5 Higher 10 Monthly working hours on all platforms

Up to 40 2 40–100 4 More than 100 9

Main source of income Uber 1 Regular employment 5 Multiple taxi platforms 5 Own company (not taxi) 3 Pension/scholarship/other allowances 1

Basis for social security Platform labour 1 Other 12 None 2*

Source: own calculations Note: * One person was not clear about their social security status. They mentioned that they pay only income tax from Uber-earnings, but sometimes they also pay wages through their own company to provide health insurance (but not pension insurance).

The age ranged from 24–66 years and some of the interviewees had previously driven traditional

taxis, some still did that as well, and some had no experience of it. There were no respondents for

whom Uber was the only source of income, most of them combined Uber with other taxi platforms.

In Tallinn working only for Uber does not seem to be a viable strategy, as Uber does not have a

lot of work to provide. Some interviewees were also employed outside the taxi industry, using

Uber as an additional income source. The number of hours worked for Uber also ranged

considerably (from 3 to 70 per week), but not all could make a clear separation between working

for Uber and for other platforms, as when multi-apping, people often have several applications

open in parallel while waiting for clients and the amount worked for one or the other app changed

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

13

from week to week, depending on several external factors, such as seasonality, but also the agency

of the drivers choosing the platform based on available bonus systems. As the Uber sample was

quite diverse, we were able to gain insight into different reasons people had for doing this line of

work, understand varying experiences with the app and clients, and gain insight into the different

expectations workers can have towards platforms.

Our sample shows the diversity of the population of the drivers: there are those from the generally

most privileged socio-demographic group – male, with Estonian as their first language – but there

are also women drivers, those from an ethnic minority with Estonian as their second language,

and recent immigrants. Migrants mostly claimed to be occupied with other apps, less with Uber

as entry barriers are higher with Uber. The ages of the drivers we interviewed vary from 24 to 66

years, but with some gaps in certain decades: six are in the age group 24–31, seven in the age

group of 40–51, and two beyond 60. This seems to be an important differentiator of what the

driver’s motivation for driving is, but not that much when their experience as a whole is

concerned.

We can assume that the self-selection of the study made it more likely that nearly all the study

participants are from the dominant social group (men, Estonian speakers) who might have been

a more advantaged group among drivers as they could afford the time to be interviewed and were

more reflective towards the platformisation of the taxi industry. We succeeded in interviewing

Estonian-Russians and new immigrants, but it is possible that we still missed enough interviews

to cover the whole spectrum of different social groups, especially those who do not speak either

Estonian, Russian or English.

The women in our sample did not report many gender-specific difficulties, such as harassment,

but this may come from a background of low expectations – as one of the women drivers put it,

she often gets asked if it is safe for women to drive a taxi, but she feels that Uber clients are

generally decent and very rarely have there been some unpleasant situations and she is not worried. This means that she has had some unpleasant situations, and this may explain why in

general women are relatively underrepresented in the less-regulated ride-hailing business: there

is the expectancy that this is less safe for them, and also there are some occasions that may make

those who consider this too worried to continue. The other just stated that she does not feel more

threatened by the clients because she is a woman, alluding that this business is inherently

somewhat more dangerous. It is maybe important to point out here that both of the female drivers

in our sample belong to the medium age group, so they may already have varied experience with

the matter and have been able to prepare their own strategies to respond or prevent these. On

the other hand, their primary socialisation as well as socialisation into work contexts took place

in the pre-me-too era and perhaps their cultural capital allows this risk to be seen as irrelevant.

Those with recent migration background in our sample were from third countries and were of

visible minority. Our interviewees were not refugees or labour migrants but rather engaged in

ride-hailing as a side activity parallel to their graduate studies. Still, they confirmed they had

witnessed racially motivated insults addressing them and had even experienced violence.

All interviewees were able to choose the place and time that suited them for having the interview.

Most of the interviews took place at the premises of Tallinn University. The rooms used for the

interviews were spacious and quiet with no distractions, thereby creating a comfortable and

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

14

trusting atmosphere. One of the Uber interviews, however, took place at a café chosen by the

respondent. This proved to be a more difficult environment because of the noise and surrounding

people. However, the interviewee seemed to be comfortable enough and the recording quality

was good. During the interview session, interviewees were offered refreshments and, as a thank-

you, a gift card to the value of €20 was given after the interview. This also helped to make the

interview more appealing and compensate them for their time. Generally, the interview situation

was rather relaxed, and people seemed to be willing to discuss their experiences and opinions

quite openly.

The length of the interviews varied between 48 to 119 minutes. A few of the Uber respondents

seemed to be in a hurry during the interview (one had parked their car for only an hour, and a

few said they had to hurry back to work). Some interviews, where respondents expressed

dissatisfaction with their work situation, were mentally rather difficult for the interviewers. In a

few cases, it also seemed that the interviewees were satisfied that finally someone is asking their

opinion about platform work and related issues. The most uncomfortable topics for the

interviewer (and sometimes it seemed also for interviewee) were topics related to family and

social activism outside the taxi industry, as these seemed to be out of the main focus and thus too

personal, whereas participants expected the interview to be about work-related matters. Some

respondents were also vague regarding paying taxes, and that might be related to a lack of

knowledge about tax issues, of them not being sure of the legality of their tax situation, or that

they were engaged in semi-legal tax practices.

Focus group interview

An invitation to join the focus group was circulated amongst all Uber drivers who were previously

interviewed. During the period preceding the interviews, the Estonian situation with Covid-19

still meant it was possible to organise face-to-face meetings, so the decision was made to use that

opportunity. Having a face-to-face interview proved to be the right choice in terms of active

engagement and positive group dynamics. However, as society experienced an increase in Covid-

19 numbers during the days preceding the focus group session, there were some cancellations

before the interview. Finally, the group consisted of 4 participants (3 Uber and 1 Airbnb), all the

Uber drivers were also interviewed earlier, the Airbnb host was a new participant in the project.

The interview was conducted within the premises of the university, with the moderation of the

PLUS team members. The duration of the interview was 1 hour and 30 minutes.

The focus group guidelines were constructed with the aim of introducing the participants to the

first results of the analysis of the interviews conducted in Tallinn, and to ask for additional

reflections, comments and further experiences related to the topics. Some slides with the main

results were prepared and presented to the participants as a stimulus for the discussion. After a

short overview of the PLUS project aims and activities so far, the interviewees were provided

with the initial research results for the key issues of the labour process and social protection for

both platforms; the discussion on skills was postponed for the following training session. Time

was also allocated to discuss Covid-19 period experiences. Since the group composition was

dominated by Uber drivers, the focus group produced somewhat more information about Uber,

especially for Covid-19 influences.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

15

Analysis of the interviews

All interviews were recorded with the permission of the respondent and all interviewees signed

an informed consent form that was introduced to them before the interview. There was no

notable concern about the fact that interviews were recorded or about how the collected data are

going to be used. The interviews were verbatim transcribed by a trusted subcontractor.

The analysis followed the principles of thematic analysis. The structure of the interview guide

divided the discussion into three broad topics, namely labour process, skills, and social protection

but we allowed people to freely reflect their perceptions and topics related to platform work

important to them. The analytical summaries were compiled on all interviews, following the main

research questions and aims of the PLUS project as the central themes. We applied the inductive

approach keeping an open eye for unexpected topics, aspects and nuances from the interviews.

Then, based on the analytical summaries as well as the interviews as raw data, the central themes

found in the material were developed and structured into the report. The analytical findings in

the report are supported by direct quotations from the interviews.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

16

PART I: INSTITUTIONAL CONTEXT

1.1. Platformisation of the taxi industry: Competitive pressures and changing

regulations

Digital platforms entered Tallinn’s taxi industry in the mid-2010s, with the Estonian Taxify (now

Bolt) leading the way as it started providing ride-hailing services in 2013, followed by Uber in

2015. Beyond these, other apps used in Tallinn to order taxis include Taxigo, Taxofon and

Yandex.taxi, and the city is well covered by a more traditional taxi service network with more

than 30 taxi companies offering services, which means the competition in the sector is rather

fierce. In 2017, Taxify (now Bolt, the most popular taxi platform in Estonia) had the second

biggest revenue (€2,619,703) among taxi companies in Estonia, Uber’s was 533,426 euros

(Kranich, 2018). Taxi platforms are widely used by the general population. According to a survey

by Statistics Estonia in 2018, 25% of all internet users (aged 16–74) reported that they have ‘used

any website or app to arrange a transport service from another individual’. The percentage is

higher among younger age groups, especially concerning ride-hailing (Statistics Estonia, 2021).

Another survey conducted in 2018 indicates that in Tallinn, out of those who have used taxi

services within the previous six months, 75% have used Taxify, 35% Uber and 31% Yandex.taxi

(Kantar Emor, 2018).

A clear direction in Tallinn’s taxi business has been the platformisation of services in general, as

the use of applications through which clients can order a taxi (in parallel to calling on the phone)

started happening already before platforms like Bolt and Uber came onto the market. However,

no other taxi app has experienced the success that platforms like Uber and Bolt have, who have

also started developing an app-based infrastructure for other, related services like food delivery,

which requires considerable investment that traditional taxi companies are lacking: ‘/…/ companies like Uber and Bolt have a huge advantage because they have the possibility to lose billions [to develop something] without any hassle. We do not.’ (Tallinn taxi industry focus group,

March 2021). Similarly, taxi platforms started providing services with considerably lower prices

than traditional taxi companies (possibly to build their client base and secure their market

position), leading to a downward spiral in service prices, as well as remuneration for providing

taxi services.

Deregulation of the taxi industry

In several European countries (e.g. Germany, Portugal), specific regulations for providing taxi

services via platforms have been established, such as the necessity to have some kind of

(alternative) employment arrangement or status (Kampouri & Walsh, 2021). Usually, this has

happened in relation to the collective mobilisation of platform workers, including workers taking

matters to court. In Estonia, the taxi industry is the only sector that has been re-regulated due to

the impact of platform work. Although the taxi industry in Tallinn was already relatively lightly

regulated before platforms entered the market, their entrance still brought along further

deregulation at the national level, related to the amendments to the Public Transport Act (Riigi

Teataja, 2021a) applicable since the end of 2017.

The Act unified most requirements for traditional and platform-based taxis, loosening them for

traditional taxis. For example, language and training requirements specifically applicable to taxi

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

17

drivers were eliminated. Before platform taxis entered the market, taxi driving was considered a

regulated profession requiring specific knowledge and skills. In 2016, the requirements were

described by interviewed taxi company representatives as involving compulsory training and

obtaining a qualification certificate, but also a service provider card and licence to provide a taxi

service (Võsoberg, 2016). State mandated driving schools were regularly providing additional

vocational education and training courses for taxi drivers. Such dedicated courses lasted around

35 hours and usually included insights to a number of relevant fields.

In the beginning, Uber and other taxi platforms did not fall under any regulations, but since the

amendments to the Public Transport Act, their drivers have to apply for a service provider card

and a vehicle card granted by the city government following the same (now simplified) procedure

as traditional taxi drivers. Although the amendments to the Public Transport Act unified most

requirements for traditional and platform-based taxis, there are still some distinctions. Platform

taxis do not have to have a taximeter, printer, printed price list, or an illuminated sign.2

Furthermore, local municipalities can set local rules that can only be extended to traditional taxis

(currently local governments cannot do that for platform taxis), including maximum prices for

rides and the requirement to have less-polluting cars (Riigi Teataja, 2021b). These distinctions

still create quite a lot of inequality between traditional versus platform taxis. For example,

traditional taxis are not allowed to use dynamic pricing (where price depends on demand and

can get quite high during peak periods) and platforms can, and some local municipalities (e.g.

Tallinn) have set the upper price limit for taxis with taximeters. Furthermore, it is more difficult

to monitor whether platform taxis follow the rules compared to traditional taxis (Tallinn taxi

industry focus group, March 2021). These differences put traditional taxi companies in a

disadvantageous position compared to platform taxis.

Currently, regular taxi companies advertise the driver’s position to those who have at least 3

years of driving experience; fluency in the Estonian language and a working knowledge of

Russian, Finnish and English; and agree to provide a formal, polite, customer friendly service (e.g.

Tulika Takso, 2021). The relevance of languages might deserve a special mention here: the Public

Transport Act (Riigi Teataja, 2021a) does not mention the need for any language skills in order

to obtain the right to operate a taxi, and neither do local level regulations in Tallinn. However,

according to the Estonian Language Act (2011) drivers of public transport vehicles are required

to demonstrate knowledge of Estonian at the level B1, and as taxis are among public transport,

this should apply to them. The Public Transport Act only sets requirements for good reputation,

as defined by the absence of a poor track record or criminal record, and attaining of driving

license. It appears then that by law, the one main requirement for drivers in general, as well as

for taxi drivers and for ride sharers is a driver’s licence.

Related to the changes in the law, an important development is the specific taxi insurance

introduced by insurance companies after vehicle cards (mandatory for taxi drivers) became

publicly available. Although insurance companies previously also wanted to know if the vehicle

is used as a taxi, the insurance premium did not differ much. Currently, the difference is huge, as

2 Public Transport Act: (2) Where the illuminated taxi sign has been installed on the vehicle and the name of the carrier

or the carrier’s trade mark enjoying protection for the purposes of § 5 of the Trade Marks Act or the trade mark granted

use of to the carrier under a contract is visible on the outer right side of the vehicle, the passenger may be serviced

from a taxi stop and the lane designated for public transport may be used upon provision of taxi services.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

18

a taxi company representative explains: “…in extreme cases the car insurance payment for taxis

can be 100 times higher” (Tallinn taxi industry focus group, March 2021). This is also an aspect

that might push some drivers into the grey economy.

It was highlighted by the taxi industry focus group participants throughout the interview that the

discrepancy between quite extensive control exercised over traditional taxi companies versus

lack of control over platform taxis is the crucial issue that makes the situation in the industry so

unfair (Tallinn taxi industry focus group, March 2021). A recent decision by the Supreme Court

of Estonia (Riigikohus, 2021) stated that the Public Transport Act does not provide the possibility

for authorities to order a taxi via a mobile app in order to monitor the driver (e.g. to see if their

licences are in order). The court concluded that the current legislation does not provide any

possibility to effectively monitor platform taxi drivers (see more from section 1.6 regarding the

situation in Tallinn).

Another recent court decision (in the court of first instance) that might have an impact on the

industry concerns Bolt. A person who ordered a taxi via the Bolt application got into an accident

and received minor injuries. They sued Bolt as Bolt was the company who owned the licence to

provide the taxi service that was related to the driver’s service provider’s card. The court ordered

Bolt to pay damages (ERR, 2021a). The taxi company representative present in the focus group

felt that the decision was unfair and taxi companies should not be responsible for their drivers'

accidents. The representative was worried that this decision might put pressure on taxi

companies to insure drivers against such lawsuits, but taxi companies do not have the finances

for that (Tallinn taxi industry focus group, March 2021). As Bolt and the plaintiff reached a

compromise, the decision did not enter into force (Riispapp, 2021).

1.2. Working conditions, labour standards and collective mobilisation

Next, we will highlight how the platformisation of the taxi industry relates to the changing

working conditions – income, training, working time, safety at work and autonomy – for taxi

drivers.3 Second, we will elaborate what kind of collective reactions taxi drivers have had to the

changes in the business.

Our research suggests that as platform taxis entered the taxi business, the income of taxi drivers

has stagnated compared to the rest of the economy. The market, however, has expanded (until

Covid-19 hit), as new customer groups started using taxi services. Generally, the Estonian taxi

industry has been and is a low-income industry, so much, indeed, that according to one taxi

company representative (Tallinn taxi industry focus group, March 2021) it is not even possible

to get decent compensation for your work if you declare all your income and pay taxes from that:

accordingly, tax avoidance is written into the business model (not only regarding platforms, but

traditional taxis as well). The constant supply of new drivers enhances the situation. What is also

related to the entrance of platform taxis is that taxi drivers have started using multiple taxi

companies and platforms simultaneously and, according to a representative of a taxi company,

usually this is not restricted by the taxi companies (although some have non-competition clauses

3 Little is known about the actual number of workers taxi platforms have, or how much, on average, they work, earn

and so on, as platforms do not share this information with public /state authorities.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

19

with their drivers). This also raises the question of which taxi company can be held responsible

for this driver’s licences, security etc.

Taxi driver occupation has also changed in the sense of de-professionalization. A representative

of a taxi company argued that very few taxi companies train their drivers at all (training was

mandatory before the changes to the Public Transport Act, and now it is voluntary, see previous

section). On the one hand, as an expert from the Tallinn Transport Department we interviewed

in 2019 argued, ride-hailing introduced new and better taxi standards for customers. Service has

become more comfortable, simple and accessible. There is more competition, so taxi companies

try harder. He concluded that “I do not see any negative aspects related to ridesharing. People [=clients] win.” (TA-EX-4, 2019). However, the impact of platforms may have been in two

directions: first the quality of the service indeed increased, as new drivers and more competition

came onto the market. But when Bolt started dominating the market (as it does now), it started

to decrease, reaching maybe even worse levels than before. Now “the situation is so that no self- respecting person will choose to drive a taxi. You go drive a taxi under Bolt when you are in trouble.” (Tallinn taxi industry focus group, March 2021). The diversity of drivers has also

increased. According to one representative of a taxi company, Bolt’s dominance is clear and other

taxi companies have had to considerably decrease the number of taxis they operate and the taxi

drivers they have.

The high level of flexibility and autonomy are the main issues that are emphasised in relation to

the essence of platform work – be it in the media, by the experts we have interviewed during the

PLUS project or by the workers themselves. Combining platform work with other

work/study/private life obligations is an important reason why some people choose this line of

work. However, some workers and experts have raised doubts about whether the autonomy and

flexibility that the apps arguably provide – including setting your own working time – can be

realized (especially after the pandemic hit). First, platforms provide clients, set rates, sanction

non-compliance and more, and second, if you want to earn decent money, you have to be active

during specific times, and so the schedule is not actually so flexible.

The job might have also become more unsafe. Regarding health and safety at work, providing this

mostly falls on the workers themselves, as health and safety regulations are the responsibility of

the employer / service provider (Riigi Teataja, 2021c). According to the statistics of the Estonian

Motor Insurance Bureau, in 2019 taxis that are operated via an application caused 1.8 times more

traffic accidents than traditional taxis, and seven times more than regular cars (Eesti

Liikluskindlustuse Fond, 2020). The focus group participants highlighted that this might partly

be because platform taxi drivers spend more hours in the traffic, so obviously have a higher

chance of getting into an accident. A Municipal Police (MUPO) representative also elaborated that

platform taxi drivers often have their “eyes on the app”, so they pay less attention to the traffic.

In addition, as platforms sometimes offer bonuses for faster/more rides, drivers might hurry in

the traffic.

Unlike many other European cities, collective organization against platforms is relatively modest

and no worker cooperatives/ community-owned platforms exist. When taxi platforms entered

the taxi industry and were not regulated at all, both taxi drivers and taxi companies addressed

the unequal situation in the media, and with policymakers and politicians. The main heated

debates took place when the changes in the Public Transport Act were discussed (in 2016) at the

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

20

state level. Traditional taxi companies and taxi drivers (Estonian Taxi Companies Association,

Association of Car Companies, and Estonian Taxi Drivers Association) demanded the equal

treatment of platform taxi services and regular taxis. The Estonian Taxi Drivers Association (Eesti

Taksojuhtide Liit) argued that the activities of Uber and Taxify should not be legalized. There

were public debates in the mass media, but also a demonstration in front of the Parliament. On

the other hand, the platform companies (including Taxify) established the Estonian Sharing

Economy Association (Eesti Jagamismajanduse Liit) to promote this line of economic activity

(Elijas-Taal et al., 2016).

In January 2018, Taxify (now Bolt) drivers gathered in front of the Taxify office to demonstrate

against the platform’s new dynamic price policy – meaning that in the area where there are a lot

of clients the prices will be higher/doubled, but otherwise the price will be low. According to the

protesting drivers, that would have decreased their income by 30%. According to one Taxify

driver, those who rented the car from Taxify were in the worst position, as they had to pay rent

to Taxify and were obliged to accept mandatory rides. In order to earn enough to pay the rent,

they would have needed to accept 100 rides per week (70–80 hours) (Pärgma, 2018). However,

the protests did not lead to any tangible outcomes and in 2020 Bolt was still using dynamic

pricing.

There exist no specific unions for platform workers and traditional unions have also been rather

reluctant to mobilise them. In addition, in sectors like the taxi industry, unionisation is very low

in general. As platform workers are not employees (see section 1.3), there are also legal

challenges for unionisation. Representatives of the Estonian Taxi Companies Association (Eesti

Taksoettevõtete Liit) have been of the opinion that Uber and other taxi platforms should be

considered as regular taxi companies, and thus their drivers should have either employment

contracts or they should be entrepreneurs (not work as private individuals) (Kalev, 2018). It is

important to understand that in the Estonian taxi industry, employment contracts did not prevail

before platform taxis entered the market and for most taxi companies their drivers are own-

account workers.4 A recently established community organisation called Estonian Taxi

Association (Eesti Taksokoondis, representing the interests of taxi drivers) has expressed clear

dissatisfaction – mainly in social media – with the current situation and has also highlighted that

taxi platforms should be considered regular taxi companies (Eesti Taksokoondis, 2021).

1.3. Employment status and tax compliance

Regarding the changes (or lack thereof) in the employment situation for taxi drivers in Estonia, it

is crucial to understand that the taxi industry was lightly regulated (compared to other European

countries) already before platform taxis came onto the market (Tallinn taxi industry focus group

March 2021). For example, in Estonia even one-man companies can have taxi licences. Most

importantly, traditional taxi companies are not providing (and usually have not provided) the

4 Own-account workers are those who are "working on their own account or with one or more partners, hold the type

of job defined as self-employed, and have not engaged, on a continuous basis, any employees to work for them" (ILO,

2003).

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

21

possibility to be employed as a taxi driver. Rather, when hiring taxi drivers the sector has used a

model of (false) self-employment (own-account workers) for a long time.

Platforms in Estonia can sign different types of contracts with their workers. Few (Wolt,

GoWorkaBit) offer the possibility to sign a service contract (authorisation agreement) between

them and the workers based on the Law of Obligations Act (Võlaõigusseadus, käsundusleping),

not based on the Employment Contracts Act. Taxi platforms, however, do not provide

employment or even service contracts to their drivers. Similarly, platforms have no obligation to

provide work for the drivers who are paid only when they have provided a concrete service to a

customer, out of which a part goes to the platform. Therefore, in regard to platform taxi driving,

we really cannot talk about having any amount of job security. This became painfully evident

during the pandemic, when the taxi industry was hit quite hard.

Legally, platform workers operating via Uber, Bolt or other taxi platforms fall either under the

status of sole proprietors (self-employed persons), work through their own small company

(either paying social security tax for themselves or not) or provide services as natural persons

who declare taxes (if at all) as occasional income or through a newly established business account

for natural persons (see Table 2). In case of the first two options – sole proprietors (self-employed

persons) or working through their own small company – they can pay their own labour taxes and

also get social security coverage. However, this makes earning opportunities very low and as it is

clear from the interviews we conducted for this project, rarely happens. From occasional income,

only income tax is paid, which does not provide social security. In the sample of drivers we

interviewed for the PLUS project, the most dominant relational status with the platforms was that

of a natural person who declares taxes as occasional income (one did it also through a newly

established business account for natural persons). Few were self-employed and few provided

services via their own company.

Table 2: Variety of statuses, tax obligations and social security situations of platform taxi drivers Status Tax obligations Social security coverage Private individual declaring occasional income via tax return

Income tax No

Private individual using entrepreneur account

Tax rate is 20% if the annual receipt is up to 25,000 euros and 40% if the annual receipt exceeds 25,000 euros. The rate covers income tax, social tax and contributions to mandatory funded pension.

Yes, if the sum is continuous and above the mandatory threshold.

Sole proprietors (FIE) Have to pay income tax and social tax, and also contributions to mandatory funded pension in case they have joined the mandatory funded pension system from their profit.

Yes, if the sum is continuous and above the mandatory threshold.

Private limited company (OÜ)

Labour taxes need to be paid on income from employment: income tax, social tax,

Yes, if the sum is continuous and above the mandatory threshold.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

22

unemployment insurance premiums and in case the driver has joined the mandatory funded pension system then also contributions to mandatory funded pension.

Source: https://www.emta.ee/en/private-client/muud-tulu-liigid

This kind of legal situation, where platform workers are not considered by platforms as

employees or even someone who should have service contracts with platforms, is also accepted

by state authorities. The debate in public regarding the rights and obligations of platforms

towards their workers is a bit more mixed, yet modest. Some trade unions highlight that taxi

platforms can be considered as employers, and the representatives of the traditional industry

have suggested that platforms should operate under the same rules and regulations as they are

(E-mail interview with the representative of EAKL, 2021; Kalev, 2018).

Still, the fact that platforms like Uber do not provide employment contracts for their drivers has

not had as huge a disruptive impact on the taxi industry as it probably has in some other

countries. In Estonia, most “traditional” taxi services also do not employ taxi drivers, but drivers

are either self-employed or working through their own company. Therefore, the relationship

between drivers and traditional taxi companies versus ride hailing platforms like Uber does not

differ hugely. A representative of a taxi company explained that if a company would actually hire

taxi drivers, they would be in a clearly disadvantageous position in the market, as happened with

one of the biggest taxi companies:

They [the taxi company] initially hired all the taxi drivers, produced… I don’t know, like several million of kroons [previous currency] of losses every year until they realised that this model does not work. And they started using the same model as all the rest. This [the model] means that the taxi driver is a private entrepreneur, typically either renting or owning the car and offering the service as (s)he pleases. So in that sense there has not been any important change. (Tallinn taxi industry focus group, March 2021)

This means that the employment conditions for taxi drivers have been largely dependent on

themselves already for a long time and to some extent on specific agreements with the companies

they provide the service for. It was also highlighted during the taxi industry focus group that the

usual practice among taxi drivers seems to be to pay only the minimum amount as a salary to

receive social security coverage and not pay taxes on the rest of their income, so the grey economy

has long been part of Tallinn’s taxi industry.

What did change with the entrance of platforms, however, is that platforms normalised the

process of taxi driving as a source of occasional income without providing an employment contract

to them5 as opposed to registered self-employment used by traditional taxi companies. The grey

5 When natural persons work through platforms they might declare their income from driving as occasional income

and pay only income tax from it or pay the taxes through the new option of the Simplified Business Income Taxation

Act, which sets a lower level of tax obligations, but when fulfilling certain requirements, a person may also receive social

security through that.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

23

economy increased further as more drivers are not paying taxes at all. It is also extremely difficult

and energy inefficient to monitor the taxi drivers for tax avoidance:

I have spoken to a tax official once who tried to carry out tax inspections in the taxi industry. (S)he said they spent two kroons to find one. This means they cannot in any way monitor sole proprietors’ [type of registered self-employment in Estonia] when it comes to taxes, so actually no control was exercised. This means taxi drivers always declare the minimum wage…tax expense. This means they pay the necessary minimum social tax to get social insurance. (Tallinn taxi industry focus group, March 2021)

Due to the lack of tax compliance, the platform economy has forced the state tax authorities to

revise the tax collecting system. Interestingly, in Estonia, this has been done with the cooperation

of the platforms. In 2015, the Estonian Tax and Customs Board (ETCB) started cooperation with

Uber and Bolt to establish a system to simplify the declaration of taxes related to earnings through

the platforms: drivers have the option to declare their income through a prefilled form provided

by the ETCB through the platform, and platforms then send the information to the ETCB.

However, it is not mandatory for the drivers to share this information and generally very little is

known about platform workers (numbers, income, hours worked etc.), as the platforms are

rather non-transparent and do not share the data.

Furthermore, in 2018, the Simplified Business Income Taxation Act was passed with the aim of

simplifying tax responsibilities for natural persons. The act stipulated that the rate of business

income tax is: 1) 20% of the amount received on the business account if the amount does not

exceed 25,000 euros in a calendar year; 2) 40% of the amount received on the business account

exceeding 25,000 euros in a calendar year. The rates include both income and social taxes, which

are low compared to the regular Estonian employment tax rate of over 50%. The taxes can be

paid through the newly established entrepreneur account. Therefore, the platform economy has

introduced changes in the Estonian tax system.

1.4. Social protection

Between 2005 and 2018, Estonia was among the European countries with a relatively low level

of expenditure on social welfare in the areas of labour market policy (approximately half of the

EU28 average), social protection (67% of the EU28 average), social exclusion (25% of the EU28

average) and old age provision (67% of the EU28 average). Child and family policies have been

approximately 40% above the EU28 average (Unt et al., 2021). Estonia also has a minimum salary

which lags behind most of the other EU countries – only Latvia and Bulgaria have lower

purchasing power parities of their minimum salary (Hankewitz, 2020).

The access to social protection in Estonia is earnings related; therefore, contribution-based and

depends on whether a person has paid enough social security contributions (and for long

enough). For all those who have an employment contract there is insurance coverage as even in

the case of part-time work, the minimum amount of social tax is requested from employers. The

insurance cover will be suspended two months after the date of the suspension of work entered

in the employment register (e.g., an employee enters active service, goes on parental leave, etc.).

The insurance cover will not be suspended in the case of unpaid leave granted by agreement of

the parties, provided that social tax is paid for the person at least to the extent of the monthly rate

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

24

of social tax. In 2021, the monthly rate on which social tax liability of based is 584 euros and thus

the minimum amount of social tax is 192.72 euros per month.

There are other ways to be eligible for social protection covered by the state like being officially

registered as unemployed (for a full list of those eligible for social protection by the state, see

Eesti Haigekassa, 2021a). Unemployed can apply for unemployment insurance benefit (UIB),

which is more generous and related to the previous salary level or for unemployment allowance

(in the case of no sufficient prior work experience). From 1 September 2020, it is possible to work

up to 8 days a month, with the maximum sum of 233.60 euros (gig work or any other form of

work) and still be officially registered as unemployed with full rights, including social insurance

coverage. In addition, the eligibility for unemployment insurance benefits was relaxed to promote

seasonal work. Nevertheless, the employment has to be on the basis of an employment

contract/contract under the law of obligations or civil service (Eesti Töötukassa, 2020a). As taxi

platforms do not offer these options, the system is not applicable for working under these

platforms. Self-employed can register as unemployed if they stop their self-employed status.

The regulation of insurance cover differs if a person works on contract under the law of

obligations, which is the usual contract if the platform has a contract with an individual. In order

to qualify for health insurance coverage, social tax must be declared either in the tax form of one

employer or several employers in the monthly total amount of at least the minimum social tax

obligation. However, an important difference between people working under an employment

contract from those working under a law of obligations contract is that the insurance coverage of

an insured person is suspended after one month has passed from the due date where according

to the information received from the Tax and Customs Board social tax has not been declared to

the extent of the minimum social tax obligation in the latter case (not after two months as is the

case with an employment contract).

Considering sickness benefits, employees working under an employment contract are in an

advantaged situation, as in general individuals working as self-employed or under the law of

obligations will receive compensation only from the ninth day of incapacity:

“In case of illness, quarantine, non-work and traffic injury and the complications or illnesses caused by it, the benefit is paid by the employer from day 4 to day 8. Since day 9, the benefit is paid by the Health Insurance Fund. In other cases (occupational disease, occupational accident (incl. traffic-related occupational accident and complications or illness caused as a result of an occupational accident), injuries caused as a result of protecting national or public interests and preventing a criminal offence), the benefit is paid only by the Health Insurance Fund and from the second day of the exemption from work.

Furthermore, the benefit is paid by the Health Insurance Fund as of the second day if the reason for the certificate of incapacity of work is the provision of work corresponding to the person’s state of health and transferring the person to a less strenuous position, or if the employer does not have work suitable for a pregnant woman. If the employer does have less strenuous work that the employee could do during the period of incapacity for work caused by the above, then the Health Insurance Fund shall compensate the

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

25

difference in wages caused by the transfer to lighter work since the first day of the certificate of incapacity of work.” (Eesti Haigekassa, 2021b)

All benefits, such as maternity, parental, and unemployment benefits, are related to contributions

and calculated on the basis of social tax paid. For instance, maternity benefit is related to the

contributions during the previous year. It is paid fully by the Health Insurance Fund, and is

accessible to all insured pregnant women (Eesti Haigekassa, 2021c).

As was emphasised in the previous section, platforms like Uber do not take responsibility for

people providing services via their platform, as they refuse to define themselves as employers.

Therefore, all issues related to welfare and social protection are the responsibility of the person

providing the service via a platform, not the platforms themselves. In Estonia, a person has social

insurance only if she is able to pay the social tax contribution above a certain threshold. If the

payment of social tax is not regular or for a month does not reach the set threshold, a person is

not covered by social insurance. According to one social insurance expert (TA-EX-1), usually

platform work does not guarantee social insurance coverage, but in most cases it is not a problem

as most people also have a “main” job or are registered as self-employed or work under their own

company, and might cover it through that. It was commonly agreed by the experts interviewed

(despite scarce empirical evidence) that social tax is usually not paid from the income gained

from platform work. Arguably, in most cases it is declared as occasional income or avoided in

other ways as the reduction of labour taxes is the crucial competitive advantage of platform work

in most cases.

In order to tackle possible tax avoidance, lack of social security coverage and offer less costly

opportunity to declare taxes from (irregular) platform work, the aforementioned special

regulation passed in 2018, the Simplified Business Income Taxation Act, together with the

entrepreneur account system for natural persons was created. It provides possibilities to pay a

lower rate of social taxes and be covered by social insurance. This is meant to be a way for a

private person to operate as an entrepreneur in a bureaucracy free and affordable manner,

without worrying about financial statements, monthly tax declarations and the payment of taxes

– all of this takes place automatically under entrepreneur account entries. It is also possible to

obtain social insurance coverage based on income on an entrepreneur account, but payments

need to pass the minimum social tax threshold. Furthermore, no costs can be deducted from the

income, which might hinder the interest of those who have considerable work-related costs (like

fuel, insurance) to opt for an entrepreneur account.

To conclude, the main challenge related to platform work is that the social insurance coverage is

guaranteed in the case of continuous payments of social tax above the threshold. In the case of

strong income fluctuations from month to month, it might not be possible to obtain continuous

coverage. Furthermore, people working for platforms without an employment or service contract

do not collect/pay unemployment insurance (except for when they pay labour taxes for

themselves) and thus, do not qualify for unemployment insurance benefits in case they register

as unemployed.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

26

1.5. Regulations ‘in becoming’

In several countries massive uprisings against the platformisation of taxi and other services have

taken place, and companies like Uber have been sued on numerous occasions (Kelly, 2016; Sarah,

2021). Similar court cases have been absent in Estonia, some of the reasons being the more

techno-optimistic attitudes prevalent in society and the lack of collective mobilisation amongst

platforms by their workers. In general, the attitude in Estonia society is very liberal, and therefore

changes will more likely be driven by European Commission initiatives and regulations than by

local ones.

Importantly, in Estonia the image of platform work is probably not as negative as it is in some

other European countries. For example, Estonian men, who are generally in a more advantageous

position in the labour market than women or minority population (Täht, 2019; Aavik, 2015 a, b)

do not seem to be ashamed of this line of work. Similarly, they might not complain about the

working conditions or take (collective) action trying to improve them. Rather, they can even take

a lot of pride in the work, especially when framing their work as that of small-scale entrepreneurs.

The latter status is also legitimized by current legal regulations of employment and emphasised

by the taxi platforms, who call their workers “partners”.

The debates (at least public/semi-public) on employment regulations related to the situation of

platform workers have been very modest. In general, it seems to be taken for granted that

platform work is outside employment relationships and platforms are service intermediaries, not

employers. When platforms entered the market and were not legal/regulated, then some

discussions regarding the employment status of platform taxi drivers were raised. For example,

during the parliamentary debate in 2016 related to the possible changes in the Public Transport

Act that would legalise platform taxis, opposition Center Party representative raised the issue of

whether Uber and other platforms should be considered as employers:

“Looking at other spheres based on this analogy that you claim to be 21st-century regulation, does it make a participant in the sharing economy also to a certain extent an employee with rights and obligations resulting from this [including rights to health benefits, retirement benefits, insurance in the event that they become unemployed, and other social net features]? What might be the status of such an individual, as an individual entrepreneur (FIE) or something else—is it rather hobby-work?” (MP Andrei Novikov,

opposition Center Party, cited in Lanamäki & Tuvikene, 2021).

The topic was silenced by the liberal Reform Party representative by highlighting that even in the

traditional taxi industry most drivers are not employees. Furthermore, ride-hailing was depicted

as a steppingstone towards becoming an entrepreneur (see Lanamäki & Tuvikene, 2021).

In addition, even if some discussions and media coverage exists, these seem to be characterized

by a lack of concrete proposals on how to change the legal status of the platform workers in order

to fit the Estonian labour market (Tuvikene & Holts, 2021; Vallistu, 2018b). There have been

actors, namely researchers and labour representatives, who have highlighted the shortcomings

of the current situation, where platform workers remain in a grey zone: they are not (totally)

independent in regulating their working conditions, but the platforms do not take any

responsibility for their employment. While researchers from the project Cities, Work and Digital

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

27

Platforms (Tallinn University, 2021) have highlighted possible shortcomings of the current

situation – where platforms/algorithms dictate service prices and other working conditions, but

do not take any responsibility for their workers – they have not offered concrete solutions, but

rather highlighted the need for further dialogue. The latter is necessary also as platforms and

platform workers differ and so do their motivation and expectations towards platform work, so one

regulatory system might not be suitable for all (Tuvikene & Holts, 2021).

The Estonian Trade Union Confederation (Eesti Ametiühingute Keskliit, EAKL), who has

discussed the topic of platform work amongst its member unions, argues that they would prefer

the same/similar rules to apply to platform work as to regular work, as the aspect that digital

tools are used for managing the work relationship does not change the relationship itself.

Furthermore, only an employment relationship currently offers benefits like paid sick days and

holidays. If platform work is regulated separately from other kinds of work, it will create unfair

competition and potential for tax evasion. The EAKL has made a proposal to extend the concept

of employee, so that all workers would receive at least minimal rights (including the right to

collective representation) and social benefits (E-mail interview with the representative of EAKL,

2021).

Bolt (direct competitor of Uber who is dominating the platform taxi market in Tallinn) is of the

opinion that employment regulations should not apply to their drivers, but when there has been

allegations that the current situation is unfair, they have suggested looking at some alternatives

on how to regulate a status that stays between employee and independent contractor (Pärli,

2021). Recently (June 2021), one of the founders of Bolt argued that the taxi platform industry

needs to be more regulated indeed, but something alternative to the employment contract is

needed: one of the options could be a taxi insurance system through which drivers would get

social security. Another alternative could be an in-between worker status as in the United

Kingdom (Võsumets, 2021).

Although Estonian labour legislation is rather flexible; for example, it is possible to end an

employment contract with agreement between the parties, currently there are further plans to

make it even more so. Flexibility in the labour market seems to be something promoted in Estonia

rather widely and strongly. An example is the current agreement between labour and employer

representatives and the Ministry of Social Affairs to allow employment contracts with flexible

working hours in the retail sector. This would make it possible for employers to use employees

as much as needed on a specific period and pay less for over-time. The agreement is currently

being discussed in the parliament. If it passes, it will most probably enter into force for a test

period (Soobik, 2021). In a recent seminar by the research group Cities, Work and Digital

Platforms (Tallinn University, 2021), where this agreement was discussed related to its

applicability to platform work, some participants (researchers, state officials and labour

representatives) highlighted that it is probably not suitable for platform work, as it is still not

flexible enough (e.g. the employer still has to employ a person in a minimum of a 0.3 position, and

can change it only by 0.2 of a position), and 0-hour contracts might be more suitable with platform

business models.

When specifically regulating platform work, Estonian policymakers seem to be waiting for

instructions and directives from the European Union. For example, the minister of health and

labour argued that although the social guarantees of platform workers is a topic that is often

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

28

discussed, the regulation of this kind of work in Estonia will be dependent on the legislative acts

that the EU will pass (Võsumets, 2021). Therefore, it seems likely that platform labour will not be

regulated before the (binding) instructions from Europe arrive.

1.6. City-level regulations for the taxi industry in Tallinn

In Tallinn’s taxi industry, the local government is able to establish some small-scale local

regulations for traditional taxis, but not for the platform taxis. Indeed, Tallinn has established a

few local rules, including maximum prices for rides and the requirement to have a less-polluting

car (Riigi Teataja, 2021b). These distinctions still create quite a lot of inequality between

traditional versus platform taxis. Furthermore, the local government (in the case of Tallinn: the

Municipal Police Department (Munitsipaalpolitsei, MUPO)) is the authority issuing the necessary

licences for taxi drivers and monitoring the drivers at the local level (including platform taxi

drivers). As was evident from the taxi industry focus group, MUPO is not satisfied with their

current power over the platform taxi industry, as declining to give licences and monitoring

platform taxi drivers is difficult under the current Public Transport Act (Tallinn taxi industry

focus group, March 2021).

Monitoring platform taxi drivers is complicated for several reasons. First, it is difficult to

distinguish platform taxis from regular cars, as the former might not have any visual clues

indicating the car is used to provide taxi services. Second, the legislation does not provide an

effective means for monitoring platform taxies (ERR, 2021b). MUPO does not have the right to

stop “regular” cars for the purpose of checking documentation. They do conduct inspection

purchases by ordering platform taxis, but a recent decision by the Supreme Court of Estonia

(Riigikohus, 2021) stated that this strategy of targeting and inspecting platform taxi drivers is

illegal, as the Public Transport Act does not provide such an option for the authorities to order

taxis via mobile app in order to inspect the driver. The court concluded that the current legislation

does not provide the possibility to effectively inspect platform taxi drivers. MUPO officials

expressed the hope that maybe due to this recent decision finally some changes to the legislation

will be made to make such activities possible. However, the participants were also a bit sceptical

that such changes would occur, as the political will to regulate the sector more strictly might be

missing: “[for more than three years] we have made proposals to the ministry on how to better monitor [platform taxis].” Nothing has changed this far (Tallinn taxi industry focus group, March

2021).

Although the Public Transport Act does not explicitly set language requirements on drivers, the

Language Act states that drivers of public transport vehicles have to be proficient in Estonian at

least at B1 level. MUPO does not have the competence to check the language level of the taxi

drivers, but they have refused to issue the service provider card if applicants were not even able

to express themselves at an elementary level when turning to MUPO. According to the officials,

some potential taxi drivers do not speak Estonian, English or Russian. The Estonian Ministry of

Economic Affairs and Communication on the other hand argues that MUPO should issue the card

and only if the driver is caught during a drive, then their data can be sent to the Language

Inspectorate who should evaluate the situation (Pau, 2019). On the other hand, in May 2019, the

Estonian Sharing Economy Association – representing Uber and Bolt – sent a letter to the Ministry

of Economic Affairs and Communications arguing that Estonian language requirements for

drivers are threatening their economic freedom (Pau, 2019).

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

29

1.7. Impact of the 2020 Covid-19 pandemic

The Estonian government declared a Covid-19-related emergency on 12 March 2020 (first

recorded case was on 27 February) and it ended on 18 May (some restrictions still remaining).

Although remote work was promoted and sports clubs and other non-essential services were

closed, people were still allowed to move freely within the country (except for one island,

Saaremaa, with a high outbreak) with reasons such as work obligations, shopping for basic

necessities and exercising, following the 2+2 rule in public spaces. Neither taxis nor

accommodation services were prohibited but as tourism collapsed and people were staying at

home, these services diminished considerably.

The state implemented several measures to support competitiveness and the survival of

companies, such as postponing the repayment of loans, postponing taxes, providing salary

subsidies and other similar measures. During the pandemic, Estonia implemented two state-wide

temporary subsidy programmes (one for March to June 2020 and one for March–May 2021) for

employees whose employers’ business had been disrupted by the restrictions set by the state. To

be eligible in 2020 for the March-May subsidy (eligibility for the June one was a bit higher), the

employer had to comply with at least two of the following conditions: “The employer must have

suffered at least a 30% decline in turnover or revenue for the month they wish to be subsidized

for, as compared to the same month last year; The employer is not able to provide at least 30% of

their employees with work; The employer has cut the wages of at least 30% of its employees by

at least 30% or down to the minimum wage.” (Eesti Töötukassa, 2020b). The 2021 programme

stated that:

“An employer can apply for a temporary subsidy for employees, if: their turnover or

income for the calendar month for which temporary subsidy is claimed has fallen by at

least 50 per cent in comparison with the average turnover or income for the period

December 2019 through February 2020, or the average turnover or income for the period

July 2020 through December 2020; they can no longer provide employees with work to

the agreed volume (§ 35 of the Employment Contracts Act applies) or they have reduced

the remuneration of the employees (§ 37 of the Employment Contracts Act applies); they

are not the subject of compulsory dissolution, liquidation or bankruptcy proceedings and

at the moment of application they had no tax arrears or they have been deferred.” (Eesti

Töötukassa, 2021)

Furthermore, “Self-employed persons whose business income in 2020 was 50% lower than in

2019 can also apply for compensation. The amount of compensation for a FIE is EUR 584 per

month.” (ibid.). Also, starting from January 2021, only the first day of sick leave is not covered by

the employer, who has to cover days 2–5, and after that period sick leave is paid for by the

Estonian Health Insurance Fund (Eesti Haigekassa, 2021d). For people officially living in Tallinn,

the city was also reimbursing the first sick day between the period of March and May 2021

(Tallinna e-teenindus, 2021).

Uber continued providing services in Tallinn, but during Uber’s first wave of redundancies related

to Covid-19 pandemic, they made all the employees in their Tallinn office redundant (Härma,

2021). It appears that Uber drivers were not even notified about this, as in a Facebook forum of

Uber drivers at the end of May 2020, people were complaining that the Uber office is closed and

asked if anybody knows when it will be open. In response, people were sharing a newspaper

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

30

article about Uber redundancies. Apparently, now driver support for Tallinn drivers is delivered

through Uber’s Polish office. It is also clear that the number of clients had decreased considerably,

with some drivers claiming that they have not had any orders for days or even weeks (see part

2.4 for the Uber drivers’ views). This is understandable, as the Tallinn market is divided by

different taxi platforms, and the local market is ruled by Bolt. Therefore, most Uber clients are

tourists, and tourism has collapsed due to the pandemic and related restrictions.

The current pandemic situation has clearly decreased earnings for drivers and shrunk the taxi

market in general. It is not clear how much Uber’s turnover has changed during the period, but

Bolt announced an 85% decrease (Postimees, 2020). As explained above, the government

compensated salary losses for the employees of many companies whose business activity

decreased, as the main social policy measure used in Estonia in connection with Covid-19 has

been income replacement, implemented by the Estonian Unemployment Insurance Fund (EUIF).

From March to June 2020, the subsidy was paid to enterprises which were struggling and

followed certain criteria. One of the biggest support payments (5th place) was given to Bolt

Technology OÜ: 1.4 million euros. However, it is important to note that drivers were not eligible,

as they are not employed by the platforms. Support from the state only concerned actual

employees (EAS, 2021).

As highlighted by one of the experts, “Platform workers are not real employees nor real

employers and they are pushed into the grey zone” (Focus group discussion on platform

economy, November 2020). During the Uber and Airbnb focus group in November 2020, the

participants clearly emphasised that none of them received the subsidy meant for own-account

workers, as they did not fulfil the necessary employment/tax conditions. The experts that we

interviewed in November 2020 also agreed that due to their grey zone status, platform workers

can fall through the cracks of any state compensation mechanisms (Focus group discussion on

platform economy, November 2020).

Furthermore, platform taxi drivers encountered increasingly risky working conditions: “related

to the pandemic we are trying to lessen human contact, but when doing this kind of work it is

necessary to be in contact with different people, for example, if you offer a transport service; and

this crosses to the aspect that they [platform workers] do not have social security, but their risk

level, risk factors are amplified considering health, wellbeing” (Focus group discussion on

platform economy, November 2020). Regarding the safety of drivers, MUPO representatives

elaborated that traditional taxi companies usually have provided decent protection

(masks/shields/disinfectants), but with platforms it has mostly been the responsibility of the

driver. Uber did not provide their drivers any safety equipment during the pandemic (some

drivers were complaining that only in May 2020 did Uber order them to wear face masks). Some

taxi companies installed partition screens and started providing delivery services (PM Tarbija,

2020). Regarding social security, it depended on the concrete driver. If s(he) paid the minimum

amount to receive social security, then s(he) had access to it (e.g. when losing a job during the

pandemic), otherwise not.

A hot debate ensued at the beginning of April 2020 around the issue of Bolt asking to borrow 50

million euros from the state (as their revenues had dropped and commercial banks were not

willing to provide loans on suitable conditions) (ERR, 2020), but the state did not grant them this

support (Digigeenius, 2020).

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

31

PART II: PLATFORM WORKERS’ PERSPECTIVE

The following part of the report is devoted to the perspectives of platform workers working for

platform taxis. Although all of them worked for Uber, most also used other applications and were

able to compare their experiences of working for different apps. The topics are divided into three

broad categories – labour process, skills, social security – followed by a fourth section about how

some of these workers have experienced the 2020 pandemic in their working lives. The last

section of this chapter raises some alternative ways highlighted by the interviewees on how the

(platform) taxi industry could be organised.

2.1 Labour process

The reasons the interviewees started to work through Uber and other taxi apps are manifold. So

are the reasons they still continue to, and these reasons might be different to their initial ones.

These include the need/desire to earn extra income to their main employment, not finding a more

secure or higher paid or professional job, and using the option to earn income (during studies)

with a flexible schedule. For those who started out as traditional taxi drivers, using apps in

addition to or instead of their traditional taxi was because driving only traditional taxis became

more difficult after platforms entered the market. For new migrants, taxi apps can also be a way

to enter the labour market even without speaking Estonian. There are those who started

providing the services in the very beginning when Uber and the locally-owned Bolt (back then

Taxify) came to the market in 2015 and 2013, respectively, and this area was without any

regulations and was advertised as a way to share your car and earn something extra. Some

interviewees expressed nostalgia for the period of ‘actual ride-sharing’ where their service was

different from the traditional taxi service.

The flexibility of choosing your own working time was often highlighted as the main advantage of

this kind of work (some had other work/study obligations, so having a free schedule was seen as

an advantage that not many jobs provide). Some embraced the flexibility aspect totally, talking

about taxi-apps as client-providing platforms, not employers, and vouching for fewer regulations.

Others, although highlighting the positive aspect of flexibility, also saw working for taxi platforms

as a necessity that they would stop if there would be other (better) options (e.g. pay rise in the

main job or professional job available). In the Estonian context, flexibility seems to be often

internalized as something to be strived for in the labour market, and even if you are not satisfied

with your work situation, you might still speak up for loosely regulated platform work. For

example, one of the respondents who did not manage to find better employment highlighted: “/.../

the platform is like an ultimate expression of the free market” (U_M_Tln_6), and emphasised that

the less regulated, the better. Another interviewee also argued for the fact that platforms cannot

be considered as employers:

As I see the service: I choose when I work and for whom I work and what kind of clients I provide the ride to. Uber is just the communication channel that brings together me as the driver and the clients. And it does not give me work directly, but just informs me about the clients. (U_M_Tln_3)

There were also those interviewees who saw taxi platforms indeed as employers and doubted if

the current situation where they do not have any responsibility for their workers is fair. On the

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

32

other side of this flexibility, the drivers also expressed feeling insecure: you can never be certain

how much work there is, how much you have to wait. The interviewees argued that the taxi

business, which platforms are also part of as the rules and regulations generally overlap with the

ones for traditional taxis, is built on the lowest price offers (price dumping) and although clients

could be satisfied, taxi drivers are the ones taking the blow of ever decreasing rates. They

acknowledged that if you actually pay all the taxes and taxi insurance,6 there is no real possibility

to earn a decent income (not even a minimum hourly wage).

In Tallinn, it is a common strategy to use multiple apps (Uber, Bolt, Yandex, Taxigo) at the same

time and when talking about Uber, the interviewees generally compared it with other apps, whilst

Uber stood out as the most elitist. Uber is the app with the highest entry barriers. Mainly because,

compared to others, they check if the driver has all the necessary documents, including taxi

insurance – and if these are lacking or expired, you cannot provide the services. Second, Uber

restricts consecutive driving time to 12 hours, after which you have to take at least a 6-hour break.

Some interviewees highlighted that overworking is a serious issue that can cause accidents in this

line of business (so Uber is doing at least something to prevent it). Furthermore, there seems to

be a generally agreed hierarchy between apps, where some interviewees highlighted that using

Uber is their first choice – the reasons for that included better clients and rates (although rate

differences with Bolt are small), better functioning application (e.g. client can leave tip via the

app) and Uber’s decent communication with drivers. However, multi-apping seems to occur

because Uber lacks clients, so there is a lot of waiting time. The high season for Uber-driving is

when there are tourists and big events, as Uber is to a large extent used by tourists, and Estonians

seem to prefer Bolt. In addition, some negative aspects of Uber include not correctly functioning

price multiplier, no bonus for longer trips and no possibility to set a radius for your rides – in

Tallinn you have to service the whole city and surrounding Harju county, which is not densely

populated and longer trips are usually not beneficial for drivers:

To travel 15 kilometres and then get three-four euros and we have to pay … Uber will deduct its commission and then we will get only three or four euros, so … And we have to come back again, to the city, empty. (U_M_Tln_7)

Although some argued that refusing rides in Uber is not problematic, as Uber has enough drivers

and they do not block people as aggressively as Bolt, others felt that the difficulty of refusing rides

is really problematic: if you are your own boss (as Uber emphasises) you should be able to decide

your working methods yourself, including being able to choose your rides. This is especially

important in the Tallinn context, as some rides (e.g. longer ones outside Tallinn) are not really

beneficial for drivers. Those who had been blocked by Uber had gotten their accounts open quite

effortlessly. However, some highlighted that when they have asked Uber why exactly people get

blocked from the app, the Tallinn office (that was closed at the beginning of the pandemic of 2020)

seems not to know how the application works. This leads us to the algorithmic management and

surveillance that some interviewees felt rather strongly. For example, one of the drivers explained

6 It is important that in Tallinn all platform taxi drivers have to have taxi insurance (even if they do only a few rides per

week), which can be several (five) times higher than regular car insurance. Also, for leased cars, banks have to give

permission to use it for taxi-driving and they might say no.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

33

how he has been put onto a black list because he cancelled three clients in a row as they just did

not show up:

Well, the last time was just a couple of weeks ago. Completely by accident on a Friday night I had three clients via Uber: I got to the place, waited in the right place and they did not show up. What else can I do but to try to contact them. Client does not answer. So initially I cancelled the ride. It wasn’t a problem for me as there is this cancellation fee for a driver, but the thing is that there were three clients in a row like that! It was a bit suspicious even for me, but I guess the Uber system saw that it is suspicious and blocked my account. (U_M_Tln_8)

Another driver explained how he received a letter from Uber accusing him of cheating the system.

He got pretty upset as he argued that cheating is practically impossible. Furthermore, some

argued that the pricing policy is not transparent (e.g. the app says the price should be 3x higher

indicating a red colour on the map, but the rides that are offered are still at a lower price). A few

of the respondents contacted Uber asking about this, but the answer was something like “it

depends on demand”. The lack of transparency is also felt in relation to client feedback: there is

no way to know who gave a bad rating and why if they do not add comments (which, fortunately,

Uber enables). During the interviewing period, Uber had a bonus system (do an increasing

number of rides and you get a bonus), which one of the respondents summed up as a game that

makes you want to work more and more, which can be dangerous (e.g. sleepy drivers driving

around). Bonus systems seem to be one of the main management strategies that taxi platforms

use to get drivers working for them: both away from competing platforms but also to motivate

drivers working at specific times:

Every application is trying to give some incentive, so the driver might be in a hurry, to do those stuffs. This is the most difficult part that you have to follow the signs. Two days… Three days back I was going and the … I was in a hurry, I wanted to complete that ride and get back to the city, because my friends said that there’s a peak hour in the city. So, on 30 [km/h] speed-limit road I was going on 50. (U_M_TLL_12)

These systems are sometimes not transparent and can change rather quickly, increasing the

insecurity felt by the drivers. Furthermore, these incentives from the platforms can also motivate

drivers to take risks in traffic, as the above quote highlights. Bonus systems are something that

drivers generally highlight when asked about how taxi platforms direct their work. However, not

all drivers think of the algorithmic management and surveillance as problematic (or think about

it at all):

I do not think they do anything [to direct drivers]. Well, they only direct by putting back the bonuses, so people would get out more to drive, to switch their apps on. (U_F_Tln_2)

There are drivers who use several strategies to increase their livelihoods, while others seem not

to. It seems that tech-savvy people can strategise better (e.g. know how to check out how long the

ride might be). Bonus systems also change quickly. Although flexibility in terms of working time

was often highlighted by the interviewees as the main advantage of this line of work, when talking

about strategies that make the work more profitable, one of them actually chooses the “right time”

to work. However, the work schedule of the interviewees varied. While some drivers preferred

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

34

to work during evenings and/or nights and weekends, others worked during regular working

days (e.g. from 8am to 6pm). The main strategy for increasing income for many interviewees was

to take orders from the app that provided the best prices at that time (e.g. had peak time

coefficients or some bonuses); therefore, these strategies extended beyond a single app.

The Estonian ride-hailing sector stands out compared to the same sector in several other

countries for its lack of collective mobilisation against taxi-app companies. There have been some

collective actions against Bolt’s dynamic pricing policy, but these did not lead to any changes

(Pärgma, 2018). Taxi-app drivers, even those who see the need to regulate the market more and

demand better conditions from the apps seem to express the attitude that it is very difficult, if not

impossible, for workers to change anything, as Story-box 1 below exemplifies.

Story-box 1: Navigating Gig Work, Understanding Risks, Questioning, Keeping up hope

Indrek is a driver in his 40s. With secondary education and professional background in a variety

of different service positions, where he has always had direct contact with customers. He once

considered working in public transport, but felt this might be too much responsibility for

passengers, their being on time and safe. So, he took training as a taxi driver and started working

as one, and it is from that professional background as a taxi driver that he looks at platforms like

Uber.

Today, Indrek diversifies his income by operating in several ride-hailing platforms in parallel and

also doing other gigs. He has a one-man traditional taxi company, as this allows him to offer taxi

services the normal way – getting clients from the street. Of course, this way he sets higher tariffs

than what Uber and other apps provide. However, nowadays getting clients that way – from the

street – tends to be difficult. Nevertheless, he seems to take pride in the work; he regularly works

as a driver equivalent to a full-time job.

He tends to see the most important skills being those related to customer service, communication

and the management of his customers’ mood and behaviour. He mentions politeness,

communication skills, and knowledge of the city as easy, really basic things that are not even

worth mentioning. However, he still mentions the taxi driver training as useful and valuable –

meaning, this cannot be that easy, but he does not seem to notice. On the other hand, the way he

manages his work – combining regular taxi service with several platforms – indicates that he is

quite tech-savvy and feels strongly about optimizing his working times, routes, schedules, and

customers. So, he takes the work quite seriously, and likes the customer service part of it.

This does not prevent him seeing the wrongs in this line of work. He also regularly stands up for

his rights against taxi-platforms, especially through the drivers’ Facebook forums. Being

outspoken against wrongdoings by several platforms, he has been noticed by the offices and

sometimes gets punished. Therefore, his experience makes him see very little possibility for

workers to stand up against the platforms when they see injustice, or when they would like to try

to improve their working conditions, when the app can just block you and take away your

livelihood:

Well, those who stood against Bolt’s system, they were all blocked. So like, tried to make a trade union or something like that… /…/ The process stayed unfinished. That was two years ago. So we

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

35

do not have a trade union here in Estonia. We have the power of the app. Let’s say that the drivers do not have a right to say anything.

Being blocked by a platform seems to be a huge issue in Tallinn. Most platform taxi drivers do not

drive for only one platform, but diversifying their platforms is a necessity to earn a decent income.

In addition, when asked about the positive and negative aspects of working for Uber, drivers

frequently compare it to other platforms. For example, for Indrek, compared to other ride-hailing

platforms, Uber stands out as better (‘best of the worst’), as although it lacks clients, thus there is

constant uncertainty of earning with Uber, (at least before the pandemic when Uber still had an

office in Tallinn) their communication with drivers was rather decent and they are the only ride-

hailing platform that explicitly follows the state regulations. He sees one reason why Uber has

less clients is its very marginally higher rates, as he considers clients to be extremely price

sensitive.

When asked what would make him stop working for Uber, he only lists possible wrongdoings that

Uber would consider offensive – but nothing that he might himself feel as a good enough reason

to leave. This is regardless of him protesting the price arrangements, questioning the platform’s

business practices, thinking about his lack of health insurance and problems arising from

retirement age getting nearer. With all the quiet or louder protesting about working conditions

and unfair treatment by the platform, and while questioning the dignity of the line of business

because of this, he does not really see himself leaving the rideshare platforms.

These days, he feels that going back to work in other parts of the service sector would mean giving

up much of the comfort arising from the flexibility of being able to choose his own working hours

and deciding to change them whenever he needed to.

The problems behind why collective mobilisation seemed unfruitful to the interviewees relate to

the huge pool of available drivers, general (perceived) weakness of trade unions in Estonia, and

the ‘power of apps’ to block drivers who ‘act out’. Some drivers practice individual agency by

sending Uber feedback on how to improve the app, but it was also mentioned that the only

feedback Uber expects from drivers is the following:

Some kind of simple and fast survey that wants to know how satisfied you are with Uber as your cooperation partner. I have filled it a few times, but it is very general, rating some stuff on a scale of 1–5. (U_M_Tln_8).

Generally, it seems that both individual and collective agency is constrained by the power of the

app. The work can also be quite isolating, as some do not know any colleagues that use a taxi-app

and sometimes (Facebook) forums are the only place to discuss work related matters (In Tallinn

Uber-drivers’ forums are less active than Bolt-drivers’). Uber also expects all communication with

them to be done via the app, so there is no personal communication. The former can sometimes

be too slow and inconvenient, although there are drivers who expressed satisfaction with that as

well.

Those who have also worked as traditional taxi drivers highlighted how the working conditions

have deteriorated, although in Estonia traditional taxi companies also do not provide

employment contracts to their drivers. One of the differences that those who also drove or had

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

36

driven traditional taxis emphasised, is that with traditional taxis you can set your own price

tariffs. Taxi-apps also compete with each other, driving down the prices. One of the interviewees

with longer taxi driving experience stated:

Then when I came back [from a foreign country], I started to drive Uber. And also Taxify, which is now Bolt. It was rather good in the beginning, but the situation has become worse and worse. There are more drivers and now the situation is really bad. This is not a job any more. If we look it like that, a few years ago you could say it was a real job. When you did it full-time, then you could get a decent salary after all expenses, but now you cannot. You cannot earn the same income working the same hours. (U_M_Tln_14)

Some respondents who complained about low prices and high competition thought that the

situation would improve if other platforms would also start requesting some documentation from

drivers as Uber does, but generally the interviewees did not express strong attitudes towards the

need to regulate the market more.

Relationship to work-related resources

Starting ride-hailing requires investing some initial resources, including economical, such as

equipment (e.g. buying, leasing or renting the car, but also other technical devices to stay

connected), as well as investments in insurance and all the permits (including the vehicle card

that increases the insurance payment). This means that one needs to have access to crucial

information;7 in addition, one interviewee relied on the suggestion from their friends who knew

they liked driving, so information is important in more ways than one in deciding to start driving

for a platform. Apart from specific assets and information, other useful resources included

different aspects related to human capital – previous work experience in logistics, earlier

experience in the field, knowing the customers and the city, knowing the titular language, but

possibly also a level of formal education.

Some of the drivers we interviewed had access to their own car, a few were renting, others had

rented and decided to get their own car. It is difficult to say how this defines their experience of

ride-hailing: most rely on multiple apps at the same time, making use of the diversity of customer

profiles that different apps offer, and therefore also experiencing the different business practices

provided by the different brands. Renting a car for ride-hailing purposes was generally deemed

to be too costly, and therefore not economical, even pushing one into poverty, and owning the car

was sometimes mentioned as giving an additional sense of security. However, those who were

renting their car from a rental company (e.g. weekly) were happy with the fact that insurance and

other documents “are in order” and they do not have to worry about repairs, while some

highlighted special arrangements that the rental company offered in relation to ride-hailing apps.

One interviewee calculated that driving a rental car is actually more economical, so it seems that

the perception of the need to own a car depends on how one works and, perhaps, on good

accounting skills.

7 While Uber is good at providing a list of such prerequisites, some other platforms may not, and sometimes our

interviewees pointed out even assuming (wrongly) that in the other platforms this is not required, making Uber seem

too difficult to access.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

37

Some also drive a regular taxi in parallel; these workers usually have their own car and provide

taxi-services through their own company. We see this experience with regular taxis as defining

the experience with ride-hailing. Those in our sample who combine regular taxi services with

ride-hailing often make reference and direct comparisons between these, while those who only

use apps like Uber in their driving do not otherwise compare Uber with taxi companies but to

other platforms operating in Estonia (that they themselves may have been using now or in the

past).

It is interesting to note that a number of our interviewees also had entrepreneurial backgrounds,

either in the past (sometimes ending in a lack of success) or currently (e.g. one interviewee runs

a business abroad). As one of the interviewees put it, being an app-taxi driver equals being an

entrepreneur, except it is easier because there is less mental pressure and no employees:

Being a taxi driver is a cowboy-like job. You are alone really /…/ Everything depends on yourself. (U_M_TLL_15)

On the other hand, one freelancer with a one-man-company considered working for multiple

firms too exhausting, and they resorted to Uber when their main “customer” cut their orders.

2.2 Skills

Starting as an app taxi driver, as the interviewees confirm, requires minimal skill – sufficient

driving skills and knowledge of traffic rules to pass the tests to apply for a driver’s licence.

Therefore, it is generally perceived as a low-skilled occupation: if you can drive a car, you can do

the job. No actual formal education requirements are made beyond schooling to get a driver's

licence, even though some have also graduated from a special training programme for taxi

drivers.

The skills that are important for success as an Uber driver are likely learned on the job as informal

learning, while the basis for these is laid in prior learning and work experiences. Relevant skills

highlighted by the interviewees when asked included references to hard skills and soft skills. Hard

skills here mean anything that can be learned and objectively tested, included several types,

especially those related to: (1) having good driving skills and being a polite driver, (2) knowing

the city, and (3) having a good knowledge of the local language, but also (4) knowing the rules of

the country (e.g. traffic rules, but also informal rules) and (5) knowing regulations (e.g. related

to the car); (6) being familiar with accounting; (7) being tidy, and (8) having a clean car. Hard

skills are often reflected via negative examples, where their lack is apparent, as in those examples

discussing the relevance of good driving skills, knowing the city, and knowing the rules and

regulations:

Quite often you see those cars (with signs from another platform), which attempt rather obscure moves every now and then. And then you have a glimpse of the driver at the wheel, and you can tell that they are probably not local! (U_M_TLL_8)

The latter example also demonstrates how lack of some skills is related to the driver being a

foreigner.

In the next example, there is even some nostalgia about knowledge of the city being important:

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

38

In the old days we had decent taxi drivers: they drove a Volga [type of car], they knew the map of the city by heart. (U_M_TLL_11)

This reflects a critique of the applications that predefine the routes, and the pride locals have in

knowing their city.

Of soft skills, or those that may vary from situation to situation and are regarded mostly

subjectively in the given context, the interviewed drivers mentioned the following: (1)

elementary politeness, (2) communication skills, (3) excellent knowledge of the Estonian

language, (4) some knowledge of other languages, (5) being calm (in traffic as well as with

customers), (6) being patient (also, while waiting for the next ride), (7) having self-control, and

(8) not being afraid of instability.

Among the hard skills, good knowledge of the Estonian language was mentioned, which can be

measured rather easily by, say, tests. When listed among the soft skills, the same knowledge of

the Estonian language points to the fact that someone who prefers to speak in Estonian might

reward the driver based on their excellent knowledge of the Estonian language, and give less

regard to the one whose spoken Estonian is poorer, even if it is, strictly speaking, good enough to

understand and communicate everything that is necessary to provide a good service.

Among the soft skills, there is also, for example, knowledge of any additional, unexpected

language of some customer that enables the driver to provide something additional to the service:

Given that I know the Finnish language quite well, then they [Finnish tourists] are usually rather pleasantly surprised – and it is possible that I might earn additional tips for this. [U_M_TLL_11]

This citation refers to additional skill. The next also quote mentions knowledge of language as

important, but suggests that this is important insofar as it helps provide the customers the

attention they seek:

Well, this ridesharing customer, of course she seeks a lower price, but in fact she actually wants personal attention. This is almost impossible to offer if one does not know the language well! [U_M_TLL_1]

Here we see how language skill is seen as necessary as a soft skill: not just to discuss traffic, but

to be able to keep up small talk or provide a supportive environment and attention, when it is

sought. Offering personal attention, human attention, or even recognising that this is expected, is

to be seen among the soft skills that may be needed to secure success. Focus on communication

skills that include politeness but also go beyond politeness are reflected in the next quote:

One has to be kind of calm, patient, both in traffic and in communication with customers. There are different kinds of customers. And, beyond this, one also has to be flexible, to the extent that some situations require very quick reactions. In some cases one may be provoked, while some cases are such that one shouldn’t get involved…Well, sometimes, perhaps it is useful to be able to just hear a person out. [U_M_TLL_1]

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

39

So, it is kind of expected that a good driver is also a good psychologist, as hearing a person out

and knowing when not to get involved are definitely soft skills.

Beyond hard and soft skills, some additional skills emerged as important from the drivers’ own

descriptions of their work, even though these remained more salient, implicit skills, as the

interviewees did not consciously reflect on them. These involved: (1) having good technical skills;

(2) having strategic thinking (tech-savvy was mentioned to be able to strategize better given the

bonus systems); (3) having good long-term (e.g. monthly) strategies for planning the work time

(including rests) and earnings; (4) an entrepreneurial spirit and experience; (5) earlier

experience of driving a regular taxi (though this might increase the expectations); (6)

responsiveness to keep up with the rating (such as checking if somebody gives a bad score and

try to improve the situation, as in adding more fresheners, when a bad smell is mentioned); (7)

resilience (given the dangers of the job, survive experiences of racism and criticism, e.g. for not

speaking good Estonian); (8) ability to build social capital and rely on professional networks.

It seems even fair that good technical skills are not mentioned in a straightforward way because

they are taken for granted. Responsiveness and resilience are reactive strategies related to self-

control: either doing something or refraining from something, they are about answering to

someone else’s needs. On the other hand, aspects such as building social capital include groups of

drivers meeting regularly [U_M_TLL_14], relying on friends who suggested the app in the first

place [U_F_TLL_13], discuss taxi-app related issues on the social media group [U_F_TLL_12], and

these are all indeed strategies for coping better in the field. But specific skills can be recognised

within these practices, and thus it is important to be aware of these.

The skills relevant for success in driving a platform taxi are practiced in the daily work, and are

also likely developed over the course of one’s engagement with the platforms. The drivers

specifically mentioned possibilities for learning languages and getting better over time at

recognising people. However, if any of this learning is taking place, this is specifically through

informal, self-planned, independent learning, supported by the desire to do their work better (=

most efficiently) and in response to feedback received through the rating systems.

It is important to note that some skills are not mentioned explicitly. The drivers do mention being

calm and polite, safe driving, good knowledge of the streets and excellent communication skills

with a good knowledge of languages, being friendly, listening, and avoiding troubles escalating.

However, there is a myriad of other, more implicit skills that are necessary, and that are being

developed during the customer-oriented ride-sharing experience. It seems that formal education

may or may not have prepared the drivers for these challenges, and if those with a higher

education have some advantages, these are more along the lines of being able to better plan and

strategize – but not necessarily in all cases.

The skills are on the whole also well in line with what is often portrayed as traditional taxi driving

skills, with perhaps a few exceptions: extra possibilities for the tech-savvy (emerging from

platform-economy specifics); strategic thinking and entrepreneurial spirit (compared to regular

taxi companies, each driver has more responsibility); responsiveness (vis-a-vis ratings);

resilience (since there is no one to protect and mediate). Coping with puzzling ratings, for

example, seems to require both resilience and responsiveness, but mostly requires drivers to be

calm and patient. The experience with ride-hailing platforms has also shown (some)

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

40

customers/riders appreciate the skills that platforms could have made redundant by design:

verbal communication skills that require a common language (by design, the rides could also

occur without exchanging words) and knowing the city to navigate most efficiently (by design,

the apps provide the routes and maps to drive by). Relaxing those requirements may have

lowered the access barriers to bring new demographics and especially recent immigrants to the

ride-hailing scene, but as the riders are used to these perks, some seem to mourn their loss. While

new generations of digital native riders might get more used to the new standards, it seems that

top ratings may motivate many drivers (especially those from older generations or with earlier

experience from regular taxis) to keep up with these special features, thereby pushing up the

standards and, eventually, setting this as a new norm for the highest ratings.

Deskilling

The respondents had diverse educational backgrounds: there are those with only a secondary

level of education, those with higher education as well as those with master’s degrees or

equivalent, and some of the interviewed drivers were currently in their degree studies. Some had

a professional job, or have had it earlier. Some in our sample once took the taxi driver’s training

programme.

No one in the sample sees app driving as strongly contributing to their skills, or their CV in

general, and this is sometimes rather seen as eroding their human capital, as they do not make

use of all their skills and there is a lot of waiting time involved. The drivers do not see that they

make full use of their skill-set in this line of work, but they also do not feel that it intervenes in

the rest of their life:

I switch myself on exactly when I want, and also just like that I switch myself off exactly when I want. In this sense the job is rather easy, because then my brain is also switched off from the work and I do not have to think about it anymore [when I am switched off] (U_M_TLL_8)

Especially those with (previous) experience of driving for a regular taxi service or with a positive

image of this job from earlier, they see the de-professionalisation of the occupation. The following

quote shows how one interviewee discussed deskilling in the profession, discussing the loss of

requirements and necessary criteria for applying for a licence to be allowed to drive a taxi:

Four years ago, I applied for a taxi licence and there were tougher rules then, requiring training, taking exams, and… And I needed to prove my fluency in the Estonian language and some English… But now, nothing much is expected of you [to get the papers]! (U_M_TLL_12)

Some of the drivers admitted to breaking their own earlier stereotypical opinions of app-riders

only when joining the scene:

In the very beginning [of ridesharing starting in Estonia] people met in parks and talked, so I understood that the group of drivers is very diverse. I would say that in the beginning that stereotype of the full-time Uber driver I had was that he is like a loyal customer at a

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

41

pawnshop, and he is in fact acting, well, for a system that enables him to still earn his own money, not from the creditors. [U_M_TLL_1]

On the individual level, it seems that the expertise required and collected for successful Uber-

driving goes rather unnoticed even by those in whose case this could be expected to be most clear

– those who have been active with the apps already for a longer time.

We could see indications during the interviews that a certain upgrading is possible through a

tech-savvy, rational approach (but not through using one’s voice to make suggestions to the

central office or protesting online) that would not be available in the regular taxi business.

Therefore, it may be that the most entrepreneurial souls with greater initial resources might find

ways to make the most out of this profession. At the other extreme, there are claims that drivers

do not see this as an earning possibility but rather as a way of life, committing themselves to the

philosophy of ‘sharing rides with others’ – and it may happen that sometimes these are the same

group. In between the two extreme approaches – making the most of the app or seeing it as a

lifestyle extension – there is the discourse about Uber being just one option to get by, perhaps

until something better comes up, and a sub-discourse where ride-hailing is presented as a way to

earn a little extra on a regular basis. For the different types of arguments, different sets of skills

seem crucial (see Table 3).

Table 3: The key skill sets for drivers according to their main motives for driving

Type Economic arguments prevailing Non-economic arguments prevailing

Sub- type Important source of income but only temporarily

Regular source of extra income

Making the most of the app (main or extra income, temporary or regular)

Lifestyle- discourse

Main feature

Would rather quit and earn money elsewhere

Would rather not need to earn extra, but is quite ok with earning this way

Tech-savvy strategic planning to platform- generated income

Ride-sharing philosophy

Key skills Politeness; ability to stay calm and patient; good self-control; being tidy and having a clean car; resilience; responsiveness to ratings

Technological skills Communication

Strategic planning, entrepreneurial spirit, good knowledge of the city

Recognising opportunities, entrepreneurial spirit

Description Getting by, doing your best: resilience and responsiveness become key skills

Some passion in the game, can use the app for self-expression: strategising, enjoying communication, choosing customers

Source: authors’ own analysis

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

42

In fact, in some cases, the skills stepped in as a proxy for other dimensions: during the interviews

some of the drivers from the dominant socio-demographic group kept making reference to the

other socio-demographic groups, such as referring to visible minorities and recent immigrants

without any knowledge of the local language. However, rarely did those comments mention the

age or gender or even the ethnic or migration background of the others, and the criticism is rather

presented as a way of highlighting the lack of skills – most clearly, lacking knowledge of the

language and lacking the knowledge of the city (in terms of maps, routes, etc.).

2.3 Social protection

Social protection plays a central role in reducing and preventing poverty (e.g. via unemployment

insurance) and providing access to health care via health insurance. What are the strategies

platform taxi drivers use to access social security and which trade-offs do they face in their

choices? In this section we will look in detail at the grounds the drivers use for obtaining social

protection and how they make sense of their choices.

In our sample, we have a wide range of possible ways to obtain social protection. We have

interviewed drivers for whom the social tax contribution is paid outside of the platform work and

they are covered by health insurance and other aspects of social security from their main job or

because they are officially enrolled as students. We also interviewed drivers who make social

contributions and obtain coverage through platform labour using multiple apps, including Uber.

Finally, we also have one Uber driver who does not have health insurance in our sample. We

cannot make any conclusions about how many platform taxi drivers are without social security

coverage; however, we can open up the reasons behind their choices. One of the repeated

arguments was that “one can work for Uber if one has health insurance from somewhere else”.

These were people for whom platform work was a source of extra income alongside other

activities. For example, some were officially registered as students, while others had a main day

job and drove a platform taxi for extra income.

‘Students’

On the one hand, the students in our sample were from different ethnic backgrounds – foreign

students and local students with minority ethnic backgrounds. From a social status perspective,

however, they were a rather homogenous group. All were below 30 and were somewhat

privileged at least in our selection as being able to afford to own a car. It seemed that the

opportunity to have health insurance from the state, thus avoiding the necessity to pay social tax

contributions and having their own car made it possible to more easily find a balance between

the costs and benefits.

‘Extra income – patchwork’

Second, a very heterogeneous group of drivers combines this line of work with other income

sources: some lacked enough income from their main job and so drove a platform taxi out of

necessity, others were driving Uber for techno-optimistic reasons trying to keep up ‘the original

spirit of sharing the car while going in the same direction’. One common line in this group was

that they all drive around 10 hours per week and worry that the costs are too high and there is a

strong need to find a balance so as to earn something. One way for this is to avoid taxes or opting

to go without the mandatory car insurance for taxis (i.e. risking a fine) or other strategies used to

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

43

find a balance and make driving beneficial economically. They constantly brought up that paying

taxes is a grey area, costs are too high to follow all the regulations. They are covered by social

protection either from their main job if they are employees or from their own company, which is

often a one-person business created to optimise the taxes. One specific type, who invest

considerable time in analysing how to optimise their outcomes, also emerges: techno-savvy tax

optimizers – they have developed detailed logic and strategies to make the most of the benefit

systems offered by Uber, Bolt or other platforms, trying to pinpoint the logic behind the

algorithmic management by optimising their work hours and timing and also finding ways to

reduce the tax load via their own small company.

‘App taxis as a source of social protection?’

In Tallinn, it is not possible to survive driving only one app. Thus, if driving platform taxi the only

source of income, drivers are simultaneously using multiple apps, next to Uber also Bolt, Yandex

and sometimes others. There are not enough clients available through Uber and the Uber market

is characterised by strong seasonality which depends on tourists, since locals use the other apps

more.

Everyone in our sample who pays their social tax from their taxi driving income drives at least 40

hours per week and in one case 60–70 hours per week. Several interviewees outlined that there

are many taxi drivers who work enormously over the normal workload and that Uber is the only

app on the market which requires a break after 12 hours of work. However, drivers can then drive

using other taxi apps.

There are different strategies to negotiate an income – some have their own car, some rent.

However, all are own-account workers and pay themselves a minimum salary (or slightly more)

to get health insurance, even if they actually earn a considerably higher income. It is not possible

to cover the costs related to the car, pay all the taxes and earn an acceptable amount; the

respondents repeatedly outlined that this needs a lot of balancing and finding trade-offs.

‘Without any social protection’

When drivers have to cover a lot of car expenses (like lease payments for a car), the income earned

through taxi platforms is not enough to cover even the minimum amount of social tax. As the case

of Jane (female, 48) shows, social tax contributions might be the first to be sacrificed and she

drives without social insurance. She is not worried about it and does not look for solutions as it

does not seem possible for her under the current circumstances. She stresses that the flexibility

of such work is a great asset and she sees the freedom in it. She works for multiple apps, including

Uber, and does not want any stickers on the car to indicate a specific company. As we have only

one case without health insurance, this is a possible area for further attention (see more from

Story-box 2 below).

Story-box 2: Carefree attitude towards platform taxi driver’s life despite social insecurities

Jane (female, 48, with secondary education) has been a platform taxi driver for around four years.

She started with the local platform Taxify (now Bolt), then joined Uber 2.5 years ago and currently

also uses another local app TaxiGo. Jane started based on the recommendation from friends who

knew she liked driving. Currently, taxi apps are her only source of income. Whereas working in

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

44

the taxi industry was “like a flower” when she started four years ago, Jane emphasises that price

dumping due to competition between apps has been intense and made the life of drivers much

more difficult.

She likes the freedom to choose when and how much she works as well as communicating with

the clients and says she could not imagine having a nine-to-five job. Jane works mostly at nights,

about 16 to 20 hours a week. She starts from 10 or 11pm and drives until about 2 or 3am on

weekdays and until 6 or 7am on Friday and Saturday. Jane has chosen to drive at nights because

she feels that sitting in traffic jams does not pay off. She does not feel more threatened by the

clients because she is a woman. She feels she has sufficient experience to position herself in town

so that she has enough clients and also pays attention to large events and waits nearby to get

clients. To be a good taxi driver, Jane thinks that you need to know the city, speak different

languages and be polite. She stresses, you cannot really trust only the app to get you where you

need to go. She keeps an eye on the price coefficients to choose the app offering the best price but

when there is little work, she keeps all apps open in parallel. She rejects the attempts of different

apps to mark the cars with their logos and likes to leave herself the room to manoeuvre between

the apps.

Overall, Jane is satisfied with Uber, especially compared to the other apps: she thinks the Uber

pricing system works better; there is a working feedback system with the app administrator and

Uber keeps track of all documentation. The only thing Jane would like to change is to choose a

preferred area where she wants to work so she would not lose in activity score when refusing

clients far away. Jane also talks about low prices and fierce competition, but thinks this problem

would decrease if other platforms would also start requesting some documentation from drivers

as Uber does. Jane knows some other drivers and sometimes they discuss work. She also belongs

to the online forums of the apps she uses and once participated at a demonstration in front of the

office of Taxify/Bolt to protest against dynamic pricing. No one from the office came to talk to the

protesters; however, free pizzas were sent out to them. Jane has heard some talk about

establishing a union for the app-drivers, but until now no leaders for the movement have

appeared and she is doubtful if drivers are actually able to fight against taxi apps, as “/…/what damage could a small breeze do to the fence anyway /…/.

Jane drives a leased car, which will be hers in two years. Until then the payments for rent and

other expenses are quite high, so something has to give to make ends meet. A cost saving strategy

for Jane is to only pay income tax once a year (but not social security contributions). Although

this is a legal option in Estonia, paying only income tax is usually meant for those people who earn

extra (occasional) income this way and get social security coverage from alternative sources. As

taxi platforms are Jane’s only source of income (and she only pays income tax), she does not have

health insurance and she is not saving for a pension nor has she other social guarantees.

Nevertheless, she does not seem to be worried about that too much and is not looking for

alternative jobs. With her income, she also supports her mother who she lives with. She seems a

bit doubtful (maybe lacking enough information) about whether alternative ways of paying taxes

(e.g. through her own company or through the entrepreneur account for natural persons) might

be more reasonable options for her.

Therefore, registration in the public social security system varies among the interviewees. One

main pattern among the interviewees (albeit not generalizable to the whole population of Uber

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

45

or Bolt drivers) is that inclusion comes with the “main job”, a stable position in the labour market,

or as a result of a special condition granting access to the public health care system (e.g. health

problems, student status) or from their own small (usually one-person) company.

2.4 Perception of the impact of the pandemic

During the focus group interview on 6 November 2020, Uber drivers highlighted that the most

important development during the pandemic was the further diminishing role of Uber in Tallinn.

The drivers agreed that the number of clients has decreased on all platforms, but Uber has

become an especially marginal player in the field. This is firstly related to their clients being

overwhelmingly tourists: ‘the road to the airport is getting grassy’, as one of the Uber drivers put

it. Furthermore, the drivers contemplated whether Uber has ‘given up’ on Tallinn: they closed

their Tallinn office in May 2020. Communicating with Uber is now done via the Polish office, which

is rather difficult for the drivers – answers can take several weeks. What is more, as no office

worker is aware of the actual situation in Tallinn, they seem to be detached from the local market

realities and, as the drivers argued, both Uber and the drivers lose.

Another effect of the pandemic was decreased rates for rides by all of the taxi platforms: Uber was

the last to lower them, but they still did. One of the drivers emphasised that related to the

decreased number of clients (and lack of some client groups), drivers who want to earn a decent

income also have had to change their driving time strategies. It seems this has reduced the

working time flexibility even more: certain hours that previously enabled them to earn decently

(like times when flights and ships left) do not anymore. It also means that driving via taxi

platforms is increasingly difficult to do as a side-job, as your main job requires you to be present

at specific times.

Uber drivers also emphasised that Airbnb actually operates as a free market platform, as hosts

can set their own prices, but taxi platforms do not, as prices are set for them. Furthermore, the

drivers seemed to feel the increasing amount of control exercised over them that can end in

platforms blocking drivers without due process. As one of the participants explained:

I had a situation recently. The client made a complaint, gave me a rating of 1: the driver is not wearing a mask. I got a letter from Uber saying that one more complaint about not wearing a mask, and it will be the end [you will be blocked].

Although Uber informs the drivers regularly that wearing masks is mandatory (unlike other taxi

platforms), in Estonia wearing it during the time of the focus group was still rare, even in the

service sector. Another driver has had the opposite experience with clients: he wears a mask and

some associate this with him being ill. He also stated that the only thing the driver actually can

choose is whether to drive or not, everything else is controlled by the platform:

All the platforms try to wriggle out of being employers. If you look at the contract, what is written there: the platform is basically absent. Everything is the responsibility of the driver. They [the platform] do not know anything. In practice, if you make a couple of ‘wrong moves’, you will get [no orders, or even blocked from the platform].

One of the drivers in a focus group emphasised that Uber does offer some kind of health insurance

related to Covid if you have ‘diamond driver’ status and you get a doctor’s notice that you have

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

46

Covid-19 (and, possibly, also some other health issues), but the drivers all seemed to be sceptical

about whether it would actually be possible to get this. Luckily, none of the focus group

participants had fallen ill with Covid-19, or at least they did not mention it and when asked if a

lot of drivers have been ill, they did not think so.

To sum up, it seems that platforms were seen by the focus group participants as untrustworthy

partners, as they do not consider the driver’s feedback, and their systems, even if in place, do not

function properly (especially after Uber shut down the local office). This scepticism towards the

platforms might have increased during the pandemic, as generally the world can seem a bit more

insecure. Nevertheless, the drivers were not very demanding towards the platforms, but just

wanted them to ‘play fair’: at least to establish some kind of due process, as currently the fear of

just getting blocked was strong (two of the participants were actually already blocked by Bolt, as

they spoke out against some practices of that platform). In general, we can conclude that the

workers were undergoing difficult times because of the pandemic situation.

2.5 Alternatives to the current status quo

Although the interviewees did not highlight many possibilities that they saw to improve their

situation as (platform) taxi drivers, there were some ideas brought up that might deserve further

discussion. Although the collective mobilisation of platform workers as a way to improve their

working standards was one of the topics brought up by some drivers, we already discussed the

difficulties or particular barriers related to unionisation in Tallinn. As one interviewee

mentioned, she has heard of some talk about establishing a union for the app drivers, but until

now no trustworthy leaders for the movement have appeared. She does not feel very positive

about the benefits it could bring, as she states “what damage could a small breeze do to the fence!”

[U_F_TLL_13], likening potential unionising efforts to a breeze that could not possibly alter the

general course of social norms and practices. While she seems to at least want such a breeze to

turn into a more serious storm, the extreme capitalist viewpoint of another driver, who favours

minimal regulations and interference by the state, sees no possibility for any collective action

against the platforms. Instead, he sees the possible unionisation efforts as an opportunity for

those not involved to earn more, stating:

Being a taxi driver is a cowboy-like job. You are alone really /…/ Everything depends on yourself. [U_M_TLL_15]

Although the app drivers saw that many taxi clients are extremely price sensitive, so that even a

tiny price difference matters to them, it was still highlighted that a fixed price floor for rides might

be one option to end the downward spiral of prices:

…like price ceilings, price floors should be established, so that the driver could ask for fair fee. [U_M_TLL_5]

What else, beyond setting price floors, could help fairer practices to emerge? The idea of a city-

supported communal taxi-app was introduced:

Well, there could be a commonly administered application, whereby the city pays the drivers directly and then the drivers are driving as public transport. It would be free of charge for the customer, as I would not need this one euro from them. The city could pay

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

47

the wages and other costs. Now, of course, this will not happen, I am sure. But I wanted to simply mention it. [U_M_TLL_9]

This is to be understood partly in the context of free public transport, which is the case in Tallinn.

Therefore, the idea would be to turn the private entrepreneurs (or at least some of them) into

public service providers. Interestingly, of course, this also points to the idea of someone securing

the cost of living – and the dignity of the driver’s work would be restored. This also shows the

view of drivers who would like to drive, or would not mind driving, but find it unfairly paid.

Another solution to restore the dignity of the taxi driver’s profession involves the idea that given

driverless cars in the future, the job of drivers will become redundant altogether. Another

example drivers mentioned asked why shouldn’t rental cars be cheaper than ride-hailing because

“you only exploit the car, not the car with a driver” [U_M_TLL_5]. The need to share cars would be

there in both cases, whether with driverless cars or with ride sharing, as according to that

interviewee, more and more people understand that it is not feasible to own a car, or at least it is

not worthwhile to drive it in the city, for example, because of traffic jams and difficulties related

to parking. The interviewee also pointed out that according to his observations, ride-hailing has

not increased traffic in the city – there is no more demand for rides. To develop the idea further,

if giving up private cars could mean there are less cars in cities, this might also mean more

environmental sustainability at some point, as well as restoring the dignity of the drivers.

Perhaps more options should become available even in public discourse for other solutions to be

realised.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

48

References

Aavik, K. (2015a). The most important decisions are made in the sauna’: the role of social capital

in creating intersectional privilege in the career narratives of Estonian male

managers. NORMA: International Journal for Masculinity Studies, 10(1), 39-54.

Aavik, K. (2015b). Intersectional disadvantage and privilege in the Estonian labour market: An analysis of work narratives of Russian-speaking women and Estonian men. Doctoral

dissertation in Sociology. Tallinna Ülikool.

Altenried, M., Bojadžijev, M., Chicchi, F., Frapporti, M., Marrone, M., Niebler, V., & Pirone, M.

(2021). Final report on impact of technologies on workers and labour process in the platform economy. PLUS unpublished report.

Booth, R. (2020). Uber drivers to launch legal bid to uncover app’s algorithm. The Guardian web

page. Available at: https://www.theguardian.com/technology/2020/jul/20/uber-drivers-to-

launch-legal-bid-to-uncover-apps-algorithm (31.10.2021)

Digigeenius (2020). Bolt ootas üle kuu aja ministri vastust: laenu ei saa. Digigeenius web page.

Available at: https://digi.geenius.ee/rubriik/uudis/bolt-ootas-ule-kuu-aja-ministri-

vastust-laenu-ei-saa/ (15.10.2021)

EAS (2021). EAS web page. Available at: https://www.eas.ee/teenus/covid-ettevotlus/

(15.10.2021)

Eesti Haigekassa (2021a). Riigi poolt kindlustatud. Eesti Haigekassa web page. Available at:

https://www.haigekassa.ee/inimesele/kindlustus/riigi-poolt-kindlustatud

(15.10.2021)

Eesti Haigekassa (2021b). Benefits for temporary incapacity for work. Eesti Haigekassa web page.

Available at: https://haigekassa.ee/en/people/benefits/benefits-temporary-incapacity-

work (15.10.2021)

Eesti Haigekassa (2021c). Maternity benefit. Eesti Haigekassa web page. Available at:

https://haigekassa.ee/en/node/7/maternity-benefit (15.10.2021)

Eesti Haigekassa (2021d). Sickness benefit. Eesti Haigekassa web page. Available at:

https://haigekassa.ee/en/people/benefits/sickness-benefit (15.10.2021)

Eesti Liikluskindlustuse Fond (2020). Taksode ja muude sõidukite riskiohtlikkuse võrdlus.

Available at: https://www.lkf.ee/sites/default/files/20200304-

Taksod_tavasoidukid.pdf?750 (15.10.2021)

Eesti Taksokoondis (2021). Eesti Taksokoondis web page. Available at:

https://www.facebook.com/eestitaksokoondis/?__xts__=68.a (15.10.2021)

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

49

Eesti Töötukassa (2020a). Temporary employment during the period of registration as

unemployed. Eesti Töötukassa web page: Available at:

https://www.tootukassa.ee/eng/content/job-search/temporary-employment-during-

period-registration-unemployed (15.10.2021)

Eesti Töötukassa (2020b). Temporary subsidy program 2020. Eesti Töötukassa web page.

Available at:

https://www.tootukassa.ee/sites/tootukassa.ee/files/temporary_subsidy_program_202

0.pdf (15.10.2021)

Eesti Töötukassa (2021). Temporary subsidy program. Eesti Töötukassa web page. Available at:

https://www.tootukassa.ee/eng/content/subsidies-and-benefits/temporary-subsidy-

program-0 (15.10.2021)

Eurofound (2018). Employment and working conditions of selected types of platform work.

Publications Office of the European Union.

European citizens’ knowledge and attitudes towards science and technology (2021). Available at:

https://europa.eu/eurobarometer/surveys/detail/2237 (13.10.2021)

European Commission (2021a). European pillar of social rights. Available at:

https://ec.europa.eu/info/strategy/priorities-2019-2024/economy-works-people/jobs-

growth-and-investment/european-pillar-social-rights_en (13.10.2021)

European Commission (2021b). Protecting people working through platforms: Commission

Launches second-stage consultation of social partners. Available at:

https://ec.europa.eu/commission/presscorner/detail/en/IP_21_2944 (13.10.2021)

Eljas-Taal, K., Rõa, K., Lauren, A., Vallistu, J., & Müürisepp, K. (2016). Jagamismajanduse põhimõtete rakendamine Eesti majandus- ja õigusruumis. Technopolis |group| ja Ernst &

Young Baltic AS.

Erikson, M., & Rosin, A. (2018). Tuleviku töötegija õiguslik staatus. Töötaja või iseseisev teenusepakkuja? Arenguseire Keskus.

ERR (2020). Bolt asks for €50 million loan from state. ERR news web page. Available at:

https://news.err.ee/1074851/bolt-asks-for-50-million-loan-from-state (15.10.2021)

ERR (2021a). Kohus: Bolt vastutab ettevõttena oma juhtide tekitatud kahju eest. ERR news web

page. Available at: https://www.err.ee/1608147517/kohus-bolt-vastutab-ettevottena-

oma-juhtide-tekitatud-kahju-eest (13.10.2021)

ERR (2021b). Mupo on teinud sadu ebaseaduslikke kontrolloste. ERR Majandus web page.

Available at: https://www.err.ee/1608103294/mupo-on-teinud-sadu-ebaseaduslikke-

kontrolloste (15.10.2021)

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

50

Gubrium, J. F., & Holstein, J. A. (2008). Narrative ethnography. In S. N. Hesse-Biber, & P. Leavy,

Handbook of emergent methods (pp. 241-264). Guilford Press.

Hankewitz, S. (2020). Estonia’s minimum wage is some of the lowest in the EU. estonian world.

Available at: https://estonianworld.com/life/estonias-minimum-wage-is-some-of-the-

lowest-in-the-eu/ (15.10.2021)

Helemäe, J., & Saar, E. (2015). Estonia: visible inequalities, silenced class relations. East European Politics and Societies, 29(3), 565–576.

Holts, K. (2018). Understanding virtual work. Prospects for Estonia in the digital economy. Arenguseire Keskus.

Huws, U., Spencer, N. H., Coates, M., & Holts, K. (2019). The paltformisation of work in Europe. Results from research in 13 European countries. Foundation for European Progressive

Studies (FEPS).

Härma, K. (2020). Uberi suurkoondamine jättis firma Eestis töötajateta. foundME web page.

Available at: https://www.foundme.io/uudised/2020/05/26/uberi-suurkoondamine-

jattis-firma-eestis-tootajateta (15.10.2021)

ILO (2003). Own-account workers – ILO. Glossary of Statistical Terms. Available at:

https://stats.oecd.org/glossary/detail.asp?ID=1986 (29.10.2021)

Kalev, R. (2018). Taxify taksojuhid on ise valinud peksupoisi staatuse. Ärileht web page. Available

at: https://arileht.delfi.ee/artikkel/80951455/raul-kalev-taxify-taksojuhid-on-ise-

valinud-peksupoisi-staatuse (13.10.2021)

Kampouri, E., & Walsh, T. (2021). WP6: Report on Welfare in Platform Economies. PLUS

unpublished report.

Kantar Emor (2018). Kantar Emor web page. Available at:

https://www.kantaremor.ee/pressiteated/taksoappide-uuring-noored-pigem-toksivad-

vanemad-helistavad/ (13.10.2021)

Kelly, H. (2016). Uber’s never-ending stream of lawsuits. CNN Business web site. Available at:

https://money.cnn.com/2016/08/11/technology/uber-lawsuits/index.html

(15.10.2021)

Kranich, K. (2018). Graafik: Taxify jääb Eesti taksoturul alla üksnes ühele taksofirmale. Postimees

web page. Available at: https://majandus24.postimees.ee/4603401/graafik-taxify-jaab-

eesti-taksoturul-alla-uksnes-uhele-taksofirmale (13.10.2021)

Lanamäki, A., & Tuvikene, T. (2021). Framing digital future: Selective formalization and

legitimation of ridehailing platforms in Estonia. Geoforum.

https://doi.org/10.1016/j.geoforum.2021.01.016

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

51

Lauristin, M. (2003). Social contradictions shadowing Estonia's "Success Story".

Demokratizatsiya, 11(4), 601–616.

Maffie, M. D. (2020). Are we ‘sharing’ or ‘gig‐ing’? A classification system for online platforms.

Industrial Relations Journal, 51(6), 536-555.

Monahan, T. (2020). Monopolizing mobilities: The data politics of ride-hailing platforms in US

cities. Telematics and Informatics, 55, 101436.

https://doi.org/10.1016/j.tele.2020.101436

Pau, A. (2019). Haige olukord Tallinnas: umbkeelesete taksojuhtide pealetung. Pm Tehnika web

page. Availabe at: https://tehnika.postimees.ee/6681315/haige-olukord-tallinnas-

umbkeelsete-taksojuhtide-pealetung (15.10.2021)

PM Tarbija (2020). Tallink Takso hakkab kaupa vedama. PM Tarbija web page. Available at:

https://tarbija24.postimees.ee/6936257/Tallinnk-takso-hakkab-kaupa-

vedama?_ga=2.165490495.1688433571.1591473113-2133727612.1578647596

(15.10.2021)

Postimees (2020). Aivar Kokk: Bolt peaks enne riigilt raha küsimist peeglisse vaatama. Postimees

web page. Available at: https://www.postimees.ee/6946895/aivar-kokk-bolt-peaks-

enne-riigilt-raha-kusimist-peeglisse-vaatama (15.10.2021)

Pärgma, R. (2018). Sadakond Taxify juhti streigivad uue hinnapoliitika vastu: me oleme nagu

taksoahvid, kes töötavad end iga nädal miinusesse, et Taxify võidaks hinnasõja! Ärileht

web page. Available at: https://arileht.delfi.ee/news/uudised/fotod-ja-video-sadakond-

taxify-juhti-streigivad-uue-hinnapoliitika-vastu-me-oleme-nagu-taksoahvid-kes-

tootavad-end-iga-nadal-miinusesse-et-taxify-void?id=80920773

Pärli, M. (2021). Äpitaksojuhid ja -toidukullerid: elame peost suhu, mitte peost pittu. ERR web

page. Available at: https://www.err.ee/1608113332/apitaksojuhid-ja-toidukullerid-

elame-peost-suhu-mitte-peost-pittu (15.10.2021)

Riigikohus (2021). Riigikohus: Seadus ei luba mobiilirakenduse abil taksosid kontrollida.

Supreme Court of Estonia’s web page. Available at: https://www.riigikohus.ee/et/uudiste-

arhiiv/riigikohus-seadus-ei-luba-mobiilirakenduse-abil-taksosid-kontrollida (31.10.2021)

Riigi Teataja (2021a). Public Transport Act. Available at: in Estonian:

https://www.riigiteataja.ee/akt/122012018016?leiaKehtiv and in English:

https://www.riigiteataja.ee/en/eli/ee/513042021002/consolide/current (13.10.2021)

Riigi Teataja (2021b) Tallinna taksoveonõuded. Available at:

https://www.riigiteataja.ee/akt/429102015031 (13.10.2021)

Riigi Teataja (2021c). Occupational Health and Safety Act. Available at:

https://www.riigiteataja.ee/en/eli/ee/522042021002/consolide/current (13.10.2021)

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

52

Riispapp, J. (2021). Jurist: Bolti ja kliendi vaidlus võinuks olla esimene samm platvormitöötajate

kaitsmiseks. ÄrigeeniusPRO web page. Available at:

https://ari.geenius.ee/rubriik/riikjaettevotja/jurist-bolti-ja-kliendi-vaidlus-voinuks-

olla-esimene-samm-platvormitootajate-kaitsmiseks/ (27.10.2021)

Sarah, B. (2021). Uber drivers entitled to workers’ rights, UK supreme court rules. The Guardian

web page. Available at: https://www.theguardian.com/technology/2021/feb/19/uber-

drivers-workers-uk-supreme-court-rules-rights (15.10.2021)

Soobik, R. (2021). Hea tahte kokkuleppega tehakse ettepanek piloteerida jaekaubandussektoris

paindlikumaid töölepinguid. Sotsiaalministeerium home page. Available at:

https://www.sm.ee/et/uudised/hea-tahte-kokkuleppega-tehakse-ettepanek-

piloteerida-jaekaubandussektoris-paindlikumaid (15.10.2021)

Srnicek, N. (2017). Platform capitalism. John Wiley & Sons.

Statistics Estonia (2021). Statistics data base. Available at: http://pub.stat.ee/px-

web.2001/Database/Majandus/05Infotehnoloogia/04Infotehnoloogia_leibkonnas/04In

fotehnoloogia_leibkonnas.asp (13.10.2021)

Tallinna e-teenindus (2021). Esimese haiguspäeva hüvitise taotlemine. Tallinna e-teenindus.

Available at: https://taotlen.tallinn.ee/haiguspaev (15.10.2021)

Tallinn University (2021). Cities, work and digital platforms. Tallinn University’s web page.

Available at: https://www.tlu.ee/en/cities-work-and-digital-platforms (15.10.2021)

Tulika Takso (2021). Tulika Takso web page. Available at: https://tulika.ee/en/tule-taksojuhiks/

(13.10.2021)

Tuvikene, T., & Holts, K. (2021). Platvormitöö ümarlaud: vastutusega seonduv tuleks täpsemini

paika panna. Novaator web page. Available at:

https://novaator.err.ee/1608127870/platvormitoo-umarlaud-vastutusega-seonduv-

tuleks-tapsemini-paika-panna (15.10.2021)

Täht, K. (Toim.) (2019). Soolise palgalõhe kirjeldamine ja seletamine – tehniline ülevaade. RASI

toimetised nr 10. Tallinna Ülikool.

Unt, M., Taru, M., & Märtsin, M. (2021) Social Citizenship for young labour market entrants: Cross- country comparison of institutional settings. EUROSHIP Working Paper No. 3. Oslo

Metropolitan University. https://doi.org/10.6084/m9.figshare.14617635

Vallistu, J., Erikson, M., Eljas-Taal, K., Tampel, R., Ratnik, H., Nausedaite, R., Aksen, M., & Pruks, P.

(2017). Tuleviku töö – uued suunad ja lahendused. Technopolic Group.

Vallistu, J. (2018a). Tööturg 2035. Tööturu tulevikusuunad ja -stsenaariumid. Riigikogu

Toimetised.

PLATFORMISATION OF TALLINN’S TAXI INDUSTRY

53

Vallistu, J. (2018b). Platvormitöötaja – tööline või ettevõtja. Äripäev web page. Available at:

https://www.aripaev.ee/arvamused/2018/01/12/johanna-vallistu-mis-elukas-on-

platvormitootaja (15.10.2021)

Vallistu, J., & Piirits, M. (2021). Platvormitöö Eestis 2021. Küsitlusuuringu tulemused. Arenguseire Keskus.

Võsoberg, V. (2016). Mehelik amet: kuidas saada taksojuhiks. Õhtuleht web page. Available at:

https://mehele.ohtuleht.ee/774254/mehelik-amet-kuidas-saada-taksojuhiks

(13.10.2021)

Võsumets, M. (2021). Tööampsudega vaesusesse – kus on 21. sajandi heaoluriik? Edasi.org web

page. Available at: https://edasi.org/84657/marian-vosumets-tooampsudega-

vaesusesse-kus-on-21-sajandi-heaoluriik/ (15.10.2021)

Zhao, Q., Zuo, C., Pellegrino, G., & Zhiqiang, L. (2019). Geo-locating drivers: A study of sensitive data leakage in ride-hailing services. In Annual Network and Distributed System Security

symposium, February 2019 (NDSS 2019).

December 2019

Social Europe

Study to gather evidence on the working conditions of

platform workers VT/2018/032

Final Report

13 December 2019

CEPS, EFTHEIA, and HIVA-KU Leuven

Authors:

Zachary Kilhoffer, Willem Pieter De Groen (CEPS)

Karolien Lenaerts, Ine Smits (HIVA KU-Leuven)

Harald Hauben, Willem Waeyaert and Elisa Giacumacatos (EFTHEIA)

Jean-Philippe Lhernould (University of Poitiers)

Sophie Robin-Olivier (Sorbonne University)

EUROPEAN COMMISSION

Directorate-General for Employment, Social Affairs and Inclusion Directorate B - Employment Unit B2 – Working conditions

Contact: [email protected]

European Commission B-1049 Brussels

European Commission Directorate-General for Employment Social Affairs and Inclusion

Directorate B

2019

Study to gather evidence on the working conditions of

platform workers VT/2018/032

Final Report

10 November 2019

CEPS, EFTHEIA, and HIVA-KU Leuven

Authors:

Zachary Kilhoffer, Willem Pieter De Groen (CEPS)

Karolien Lenaerts, Ine Smits (HIVA KU-Leuven)

Harald Hauben, Willem Waeyaert and Elisa Giacumacatos (EFTHEIA)

Jean-Philippe Lhernould (University of Poitiers)

Sophie Robin-Olivier (Sorbonne University)

This Final Report presents a deliverable for the “Study to gather evidence on the working conditions of platform workers”, funded by the Directorate-General for Employment, Social Affairs and Inclusion of the European Commission (VT/2018/032). This project was awarded to the consortium composed of CEPS, Eftheia, and HIVA-KU Leuven. The consortium gratefully acknowledges the funding of the European Commission in carrying out this research.

The authors would like to thank Ivana Vukorepa (University of Zagreb), Martin Risak (University of Vienna), Six Silberman (IG Metall), Antonio Aloïsi (Bocconi University), Jovana Karanovic (Vrije Universiteit Amsterdam),

Leonie Westhoff (CEPS), Bartłomiej Bednarowicz (University of Antwerp) and the members of the EC DG EMPL

Steering Group for this study for their valuable contributions to and review of this report. Furthermore, the authors are grateful to the legal and socio-economic experts who prepared the national contributions.

LEGAL NOTICE

Manuscript completed in December 2019

Neither the European Commission nor any person acting on behalf of the European Commission is responsible for the use that might be made of the following information. More information on the European Union is available on the Internet (http://www.europa.eu).

Luxembourg: Publications Office of the European Union, 2020

PDF ISBN 978-92-76-10304-2 doi: 10.2767/26582 KE-01-20-054-EN-N

© European Union, 2020

Reuse is authorised provided the source is acknowledged. The reuse policy of European Commission documents is regulated by Decision 2011/833/EU (OJ L 330, 14.12.2011, p. 39). For any use or reproduction of photos or other material that is not under the EU copyright, permission must be sought directly from the copyright holders.

Study to gather evidence on the working conditions of platform workers

5

ABSTRACT

Platform work is a type of work using an online platform to intermediate between

platform workers, who provide services, and paying clients. Platform work seems to be

growing in size and importance. This study explores platform work in the EU28, Norway

and Iceland, with a focus on the challenges it presents to working conditions and social

protection, and how countries have responded through top-down (e.g. legislation and

case law) and bottom-up actions (e.g. collective agreements, actions by platform

workers or platforms). This national mapping is accompanied by a comparative

assessment of selected EU legal instruments, mostly in the social area. Each instrument

is assessed for personal and material scope to determine how it might impact such

challenges. Four broad legal domains with relevance to platform work challenges are

examined in stand-alone reflection papers. Together, the national mapping and legal

analysis support a gap analysis, which aims to indicate where further action on platform

work would be useful, and what form such action might take.

Study to gather evidence on the working conditions of platform workers

6

ABBREVIATIONS

AMT Amazon Mechanical Turk

AT Austria

BE Belgium

BG Bulgaria

CJEU Court of Justice of the European Union

CY Cyprus

CZ Czechia

DE Germany

DK Denmark

e.g. for example

EC European Commission

EE Estonia

EL Greece

EP European Parliament

EPSR European Pillar of Social Rights

ES Spain

ESIP European Social Insurance Platform

ETUI European Trade Union Institute

EU European Union

EU-OSHA European Union Occupational Safety and Health Association

FAU Freie Arbeiterinnen- und Arbeiter-Union [Free Workers Union]

FI Finland

FNV Federatie Nederlandse Vakbeweging [Federation of Dutch Trade Unions]

FR France

GDPR

General Data Protection Regulation: Regulation of the European Parliament and of

the Council n. 2016/679 of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC

HR Croatia

HU Hungary

i.e. in other words

ICT Information and communications technology

IE Ireland

IG BAU Industrielle Gewerkschaft Bauen-Agrar-Umwelt [Trade Union of Construction, Agriculture and Environment]

IG Metall Industriegewerkschaft Metall [German Metalworkers' Union]

IGAS Inspection générale des affaires sociales [General Inspectorate of Social Affairs]

ILO International Labour Organization

IRES Institute of Economic and Social Research

Study to gather evidence on the working conditions of platform workers

7

IT Italy

JRC Joint Research Centre of the European Commission

LT Lithuania

LU Luxembourg

LV Latvia

MS Member State (of the European Union)

MT Malta

NGG Gewerkschaft Nahrung-Genuss-Gaststätten [Food, Beverages and Catering Union]

NL Netherlands

OECD Organisation for Economic Co-operation and Development

op.cit the work cited

OSH Occupational safety and health

P2B

Platform to Business Regulation: Regulation (EU) 2019/1150 of the European Parliament and of the Council of 20 June 2019 on promoting fairness and transparency for business users of online intermediation services (OJ L186/57

11.07.2019)

PL Poland

PT Portugal

RO Romania

SE Sweden

SEK Swedish krona

SI Slovenia

SK Slovakia

STEM Science, technology, engineering and mathematics

TFEU Treaty on the Functioning of the European Union

UK United Kingdom

UNECE United Nations Economic Commission for Europe

WES Work, Employment and Social Relations model

Study to gather evidence on the working conditions of platform workers

8

GLOSSARY

Term Definition

atypical contracts

Contracts that do not conform to those in standard work. Examples include part-time, fixed-term, temporary, casual and seasonal work contracts

atypical employment/work See non-standard employment/work

bogus self-employment

Business activities that do not include any managerial or proprietary tasks and which possess the attributes of an employment relationship

but without entitlement to the corresponding labour law protections 1

clickworkers Platform workers who engage in small-scale and repetitive online work, such as tagging images

collaborative economy A term sometimes used instead of platform economy, sometimes with normative or non-economic connotations

crowdwork(ing) A type of platform work where clients place an open call for services to a large, global body of potential crowdworkers

employee A worker that is in a contractual employment relationship with another person (an employer) in return for remuneration

employer A natural or legal person that is bound by one or more natural persons (employees) via an employment contract

employment contract

The contract that formalises the employment relationship between an employee and employer: characterised by the subordination of an

employee to an employer

employment relationship

The relationship characterised by an employment contract or service

contract between the employer and employee

food delivery riders A type of platform worker providing services intermediated by a food delivery platform such as Deliveroo, Foodora, or UberEats

genuine self-employment

When the contractual status and factual characteristics of a natural person's work are both self-employment: distinguished from bogus self-employment

homeworker One who performs work from home or other premises of their choosing, creating a product or service specified by an employer

independent contractors

A person contracted to provide services to another entity as a non-

employee

intermediary A person, business, or function that serves to connect one or more entities to facilitate exchange of information, goods or services

job quality

A multi-disciplinary, multidimensional concept that is generally understood as ‘the extent to which a job has work and employment- related factors that foster beneficial outcomes for the employee, particularly psychological well-being, physical well-being and positive

attitudes such as job satisfaction’ 2

non-standard employment/work Work that diverges in one or more aspects from standard work

online labour platform A term sometimes used to distinguish a platform (as in platform work)

from other sorts of platforms. For example, Facebook and Uber are

1 See Eurofound (2008)

2 Holman (2013)

Study to gather evidence on the working conditions of platform workers

9

both online platforms, but only the latter is an online labour platform

online platform An online service that facilitates communication between one or more parties, especially to exchange services for payment

platform An online platform on which services are exchanged for payment (unless otherwise specified)

platform economy Economic and social activity facilitated through online platforms

platform work

All labour provided through, on, or mediated by online platforms in a wide range of sectors, where work can be of varied forms, and is provided in exchange of payment

platform worker A natural person providing platform work

self-employment

A broad set of labour practices wherein a natural person earns income

without an employment relationship with an employer; both bogus

self-employment and genuine self-employment are types of self- employment.

sharing economy

A term sometimes used instead of platform economy, sometimes with normative or non-economic connotations, and sometimes referring specifically to 'sharing' goods or services through an online platform

standard employment/work

Employment relationships between a natural person (employee) and a natural or legal person (employer) that are indefinite (open-ended) and full-time

traditional employment/work

Employment relationships between a natural person (employee) and a

natural or legal person (employer) that are indefinite (open-ended) and full-time, especially prior to the platform economy

worker

Anyone who performs work for pay, regardless of employment status, and without legal connotations (unless otherwise specified, especially in discussions of EU law)

Study to gather evidence on the working conditions of platform workers

10

CONTENTS

Abstract ............................................................................................................... 5

Abbreviations ....................................................................................................... 6

Glossary............................................................................................................... 8

Table of Boxes, Figures and Tables ........................................................................ 13

Executive Summary ............................................................................................. 15

1. Introduction .............................................................................................. 25

1.1 Previous key research and main findings .................................................... 25

1.2 Knowledge gaps and research objectives ................................................... 26

1.3 A need for EU action? .............................................................................. 27

1.4 Structure of the study ............................................................................. 28

2. Methodology .............................................................................................. 29

2.1 Research questions and data collection .......................................................... 29

2.1 Data analysis .......................................................................................... 32

2.2 Quality control ........................................................................................ 33

2.3 Limitations and constraints ....................................................................... 33

3. Conceptualisation of platform work .............................................................. 35

3.1 Standard and non-standard employment ................................................... 35

3.2 Conceptualisation of ‘worker’ .................................................................... 37

3.3 Conceptualisation of platform work(er) ...................................................... 39

Typology – distinguishing types of platform work ..................................... 40 3.3.1

Size, prevalence, and expected evolution of platform work in the EU .......... 42 3.3.2

4. Challenges for platform workers ................................................................... 48

4.1 Framework ............................................................................................. 48

4.2 The work dimension ................................................................................ 52

Autonomy in the allocation of tasks ........................................................ 53 4.2.1

Autonomy in work organisation .............................................................. 54 4.2.2

Surveillance, direction and performance appraisal .................................... 55 4.2.3

Task complexity ................................................................................... 57 4.2.4

Work intensity and speed pressure ......................................................... 58 4.2.5

Emotional demands .............................................................................. 59 4.2.6

Physical environment ........................................................................... 60 4.2.7

In short: challenges related to the work dimension .................................. 62 4.2.8

4.3 The employment dimension ...................................................................... 65

Employment status .............................................................................. 65 4.3.1

Determination of the employer in platform work ...................................... 67 4.3.2

Contracts: type of employment contract, contractual information provision 4.3.3 and termination of the employment contract ..................................................... 68

Social protection .................................................................................. 69 4.3.4

Earnings: price-setting, wages and additional fees ................................... 70 4.3.5

Study to gather evidence on the working conditions of platform workers

11

Working time ....................................................................................... 73 4.3.6

Career opportunities ............................................................................. 75 4.3.7

Training and skills ................................................................................ 77 4.3.8

In short: challenges related to the employment dimension ........................ 78 4.3.9

4.4 The social relations dimension .................................................................. 82

Representation .................................................................................... 82 4.4.1

Participation in decision-making ............................................................. 84 4.4.2

Supportive management and social support ............................................ 86 4.4.3

Adverse social behaviour and equal treatment ......................................... 86 4.4.4

In short: challenges related to social relations ......................................... 88 4.4.5

4.5 Other challenges ..................................................................................... 91

Undeclared work .................................................................................. 91 4.5.1

Cross-border work ............................................................................... 92 4.5.2

Data protection .................................................................................... 93 4.5.3

In short: challenges related to the other indicators .................................. 95 4.5.4

4.6 Summary of the challenges for platform work ............................................. 95

5. National tools and responses to platform work challenges ............................... 97

5.1 Conceptual/theoretical framework ............................................................. 97

5.2 Mapping of national tools and responses .................................................... 98

5.3 Top-down responses and tools .................................................................. 99

Legislation .......................................................................................... 99 5.3.1

Case Law .......................................................................................... 108 5.3.2

Administrators and inspectorates ......................................................... 113 5.3.3

5.4 Bottom-up tools and responses ............................................................... 115

Collective agreements ........................................................................ 115 5.4.1

Platform worker organisation ............................................................... 117 5.4.2

Platform actions ................................................................................. 118 5.4.3

5.5 Other tools and responses ...................................................................... 120

6. Instruments and actions at EU level ........................................................... 122

6.1 Analysis of relevant EU legislation ........................................................... 123

EU labour and social protection legislation ............................................. 123 6.1.1

Working conditions ............................................................................. 141 6.1.2

Collective labour rights ....................................................................... 152 6.1.3

Social protection ................................................................................ 157 6.1.4

The P2B Regulation ............................................................................ 166 6.1.5

Platform work and the General Data Protection Regulation ...................... 172 6.1.6

Preliminary conclusions ....................................................................... 176 6.1.7

6.2 Other EU actions ................................................................................... 194

Communications ................................................................................ 194 6.2.1

Information gathering and dissemination .............................................. 194 6.2.2

Study to gather evidence on the working conditions of platform workers

12

European Pillar of Social Rights ............................................................ 196 6.2.3

European Labour Authority .................................................................. 196 6.2.4

Case Law .......................................................................................... 198 6.2.5

Preliminary conclusions ....................................................................... 198 6.2.6

7. Gap analysis: which challenges remain to be addressed? .............................. 199

7.1 Conceptual framework ........................................................................... 199

7.2 Responses to platform-specific challenges ................................................ 202

Autonomy in the allocation of tasks ...................................................... 202 7.2.1

Physical environment ......................................................................... 203 7.2.2

Surveillance, direction and performance appraisal .................................. 204 7.2.3

Contracts .......................................................................................... 205 7.2.4

Determination of employer .................................................................. 206 7.2.5

Data protection .................................................................................. 207 7.2.6

7.3 Responses to challenges common in other forms of non-standard work ....... 208

Employment status ............................................................................ 208 7.3.1

Representation .................................................................................. 209 7.3.2

Social protection ................................................................................ 210 7.3.3

Earnings ........................................................................................... 212 7.3.4

Working time ..................................................................................... 213 7.3.5

Undeclared work ................................................................................ 214 7.3.6

7.4 Responses to challenges found in the general labour market ...................... 214

Autonomy in work organisation ............................................................ 214 7.4.1

Adverse social behaviour and equal treatment ....................................... 215 7.4.2

Cross-border work ............................................................................. 216 7.4.3

7.5 Summary: challenges that are not fully addressed at national/EU level ........ 217

8. Conclusions and policy pointers ................................................................. 219

8.1 Takeaways from the gap analysis ............................................................ 219

8.2 General conclusions and policy implications .............................................. 220

Bibliography ..................................................................................................... 225

Reflection Paper 1 .......................................................................................... 237

1. Introduction ......................................................................................... 238

2. EU competition law and the right to collective bargaining ........................... 239

2.1 The Albany case: The right to collective bargaining for employees as an ‘exception-to-the-rule’ approach ................................................................... 239

2.2 The FNV Kunsten Informatie case: the right to collective bargaining for

‘false self-employed’ persons ........................................................................ 240

3. Collective rights for platform workers in Member States .................................. 243

3.1 Statutory law ......................................................................................... 244

3.2 Collective agreements............................................................................. 247

3.3 Other trade union initiatives .................................................................... 248

3.4 Grassroot organisations’ initiatives ........................................................... 249

Study to gather evidence on the working conditions of platform workers

13

3.5 Action by platforms ................................................................................ 250

Some preliminary conclusions .......................................................................... 250

Reflection Paper 2 .......................................................................................... 253

1. Algorithmic management in the platform economy .................................... 254

1.1 General considerations ....................................................................... 254

1.2 Impact on employment or labour market status classification .................. 255

2. The GDPR and its implications for platform workers ................................... 255

2.1 The right to access personal data (Article 15 GDPR) ............................... 256

2.2 The right not to be subject to a decision based solely on automated processing (Article 22 GDPR) ........................................................................ 258

2.3 The right to data portability ................................................................. 261

2.4 Discriminatory biases underlying the algorithm ...................................... 264

Annex I: Synopsis Report of consultations ............................................................ 266

Overview of fieldwork activities ........................................................................ 266

Summary of findings from fieldwork activities .................................................... 267

Annex II: A note on what is not platform work ...................................................... 276

TABLE OF BOXES, FIGURES AND TABLES

Box 1: The Fairwork Framework: a job quality framework tailored to platform work .... 50 Box 2: Price setting in platform work ..................................................................... 72 Box 3: Platform work, collective bargaining and EU competition law .......................... 83 Box 4: Platform cooperatives: comparing TaskRabbit with Loconomics ....................... 85 Box 5: Licensing requirements for drivers ............................................................ 100 Box 6: The Lazio region's platform work law ......................................................... 101 Box 7: Belgium's special tax categories ................................................................ 105 Box 8: Denmark's collective agreements for platform work ..................................... 114

Figure 1: Challenges summary .............................................................................. 16 Figure 2: Conceptualisation of standard employment ............................................... 35 Figure 3: Conceptualisation of platform work .......................................................... 39 Figure 4: Typology of platform work ...................................................................... 41 Figure 5: Revenues and transaction values facilitated by collaborative

economy platforms in Europe ............................................................................... 46 Figure 6: Example of AMT 'Mechanical Turk Masters' ................................................ 56 Figure 7: Stylised representation of work challenges for different types of

platform work ..................................................................................................... 64 Figure 8: Stylised representation of work challenges for different types of

platform work ..................................................................................................... 64 Figure 9: Stylised representation of employment challenges for different types of

platform work ..................................................................................................... 81 Figure 10: Stylised representation of employment challenges for different types of

platform work ..................................................................................................... 81 Figure 11: Stylised representation of social relations challenges for different types of

platform work ..................................................................................................... 90 Figure 12: Stylised representation of social relations challenges for different types of

platform work ..................................................................................................... 90 Figure 13: Summary chart of the challenges facing platform work (by level of

specificity).......................................................................................................... 96

Study to gather evidence on the working conditions of platform workers

14

Figure 14: National responses summary ................................................................. 99 Figure 15: Graphic expression of conceptual framework gap analysis ....................... 199

Table 1: Summary of gap analysis ......................................................................... 23 Table 2: Addressing research questions ................................................................. 30 Table 3: Platform worker prevalence estimate as percentage of population

16-74 years ........................................................................................................ 44 Table 4: Self-reported employment status by platform work type .............................. 45 Table 5: Comparison of the WES model with other job quality frameworks and their

definitions and indicators ..................................................................................... 49 Table 6: Overview of job quality models developed in different fields ......................... 50 Table 7: Adjusted WES model ............................................................................... 51 Table 8: Summary table of challenges related to the work dimension ........................ 63 Table 9: Summary table of challenges related to the employment dimension .............. 79 Table 10: Summary table of challenges related to the social relations dimension ......... 89 Table 11: Summary table of challenges related to undeclared work, cross-border

platform work and data protection ......................................................................... 95 Table 12: Summary table of the challenges facing platform work

(by level of importance) ....................................................................................... 96 Table 13: Overview of factors indicating whether a platform worker is employee

or self-employed ............................................................................................... 104 Table 14: List of relevant EU legislation in the area of working conditions and

social protection ................................................................................................ 123 Table 15: Personal Scope as defined in EU legislation ............................................ 126 Table 16: Comparison between the Directive on transparent and predictable

working conditions and the P2B Regulation ........................................................... 169 Table 17: Overview of working conditions and social protection of platform

workers as protected by EU legislation ................................................................. 188 Table 18: Most important challenges for platform workers ...................................... 199 Table 19: Frequency of national responses to most important challenges ................. 200 Table 20: Gap analysis assessment key ............................................................... 202 Table 21: Summary of gap analysis ..................................................................... 218 Table 22: Participants in semi-structured interviews .............................................. 269 Table 23: Participants in country focus groups ...................................................... 272 Table 24: Participants in the validation workshop .................................................. 275

Study to gather evidence on the working conditions of platform workers

15

EXECUTIVE SUMMARY

Introduction

Platform work is a small but diverse and seemingly growing form of labour.

Platform work differs from many traditional patterns of work, and while it may increase

labour market access and lead to innovation and entrepreneurship, and in so doing help

achieve EU policy objectives, it also challenges existing labour and social law

frameworks. European policymakers and stakeholders have highlighted the need for

better understanding of platform work so its potential can be maximised and its harms

minimised.

Against this background, this study combines fieldwork and desk research, and both

legal and socioeconomic perspectives, for each of the 28 EU Member States, plus Norway

and Iceland, as well as for the EU level. The primary goal is to assess, in view of the

particular challenges faced by platform workers, whether EU action is required

to improve their working conditions and social protection, and if so, what form

such action could take.

Conceptualisations

Platform work is understood as all labour provided through, on, or mediated by

online platforms in a wide range of sectors, where work can be of varied forms,

and is provided in exchange for payment. It features a triangular relationship

between platform, platform worker and client, using online intermediation. The

intermediation largely uses technology and algorithms, is often intransparent, and may

significantly affect working conditions, for example, by its impact on the allocation and

organisation of work, and the evaluation of platform workers. This ‘black box of

intermediation’ is a distinguishing feature of platform work.

Platform work refers to very heterogeneous forms of work. We can distinguish between

platform work types based on three primary factors:

 Location: whether a task is performed online (from anywhere with an internet connection) or on-location;

 Complexity: higher- or lower-skill requirements;

 Allocation of work: primarily determined by platform, platform worker, or

client.

Platform work can be grouped with other types of non-standard work or self-

employment. Non-standard work refers to arrangements that diverge from a full-time,

open-ended employment contract with one employer. Platform workers rarely have an

employment contract with the platform and are mainly considered self-employed in

practice. This self-employment can be bogus or genuine, but it is often difficult to

distinguish between them, and there may also be differences across countries, types of

platform work, and even individuals using the same platform in the same city.

The size or prevalence of platform work is much debated and estimates vary widely.

According to the COLLEEM II survey data3 for 16 EU countries, an average of 11% of the

adult population has performed platform work at least once. Those who rely on platform

work for their main income are far fewer, averaging 1.4% of adults. The most common

platform work tasks include online clerical and data entry. Men are much more likely to

perform transportation and delivery services, while women perform more translation and

certain on-location services (e.g. housekeeping or beauty services). The survey also

3 JRC’s COLLEEM II survey

Study to gather evidence on the working conditions of platform workers

16

finds that those performing platform work as a primary or secondary occupation

(relying on it for a large proportion of their income) are less likely to be employees

than those performing platform work sporadically. Platform workers’ employment status

is often unclear even to the platform workers themselves, which means that self-

reporting of status can affect the reliability of aggregate data.

The future trajectory of platform work is also much debated. Available data and expert

opinion suggests the total number of platform workers is growing and likely to continue

to do so, with demand for certain services (e.g. training AI, care for the elderly or

childcare) likely to grow more than for others (e.g. tasks more prone to automation, and

food delivery or personal transport services). Moreover, certain characteristics of

platform work are becoming more common in the overall labour market (e.g. algorithmic

management).

Challenges for platform workers

The challenges related to the working conditions and social protection of platform

workers are mapped in accordance with a job-quality framework based on the Work,

Employment and Social Relations (WES) model.4 This model consists of three

dimensions: work, employment, and social relations. We also consider ‘other’

challenges relevant to platform work.

Potential challenges facing platform workers were assessed and summarised by their

significance (high, medium, low, or none), what types of platform work seem most

affected, and their specificity (specific to platform work, common for non-standard

work, or present in the general labour market). Figure 1 shows how many of these

challenges are not specific to platform work.

Figure 1: Challenges summary

Challenges were found to vary a great deal across different types of platform work. For

example, the physical environment differs for online and on-location forms of platform

work, and autonomy in work organisation changes depending on whether the client,

platform, or platform worker determines which tasks are performed, when and how. The

4 Lamberts, M. et al. (2016), Jobkwaliteit in België in 2015. Analyse aan de hand van de European Working

Conditions Survey EWCS 2015 (Eurofound), HIVA Onderzoeksinstituut voor Arbeid en Samenleving; Leuven

Study to gather evidence on the working conditions of platform workers

17

importance given to particular challenges also varies across countries, depending on

their policy and legal frameworks. For example, social protection may be a greater

challenge in countries where the self-employed have significantly less statutory coverage

than employees. A summary of the WES dimensions and significant platform work

challenges (of high or medium importance) follows.

The work dimension primarily concerns job content, working conditions and work

organisation, which impact physical and psychological risks for the platform worker. The

use of technology, apps and algorithms particularly affects this dimension, which

therefore contains certain challenges more specific to platform work.

Of highest concern for platform workers are the challenges of autonomy in the allocation

of tasks; of surveillance, direction, and performance appraisal; and of physical

environment. Autonomy in work organisation is assessed to be of medium concern.

The employment dimension relates to the formal context in which a platform worker

performs tasks, such as their employment status, the nature and content of their

contract with the platform, the level of social protection, and the composition of their

earnings. It also entails issues that directly affect a platform worker’s personal life, such

as working time, training and career opportunities. This dimension contains some of the

most discussed challenges of platform work, including employment status, social

protection and earnings, which are also significant issues in non-standard work.

The challenges of highest concern are employment status, determination of employer,

and contracts (including type, termination, and provision of contractual information). The

medium concern challenges are social protection, earnings (including wages, fees and

price setting), and working time.

The social relations dimension concerns social relations and interactions, social

dialogue and representation at work, both formally and informally. Social support is an

important resource for the well-being of platform workers that can help achieve work-

related goals, encourage personal growth and compensate for job demands. These

challenges are mostly common to non-standard work.

Two challenges in this category are assessed as significant for platform workers:

representation (high concern), and adverse behaviour and social treatment (medium

concern). Notably, the latter is also a concern in the general labour market.

The ‘other’ dimension covers challenges that do not fit within the previous categories

but are nevertheless important for platform workers: undeclared work, cross-border

issues, and data protection. Each is assessed as a medium challenge, but these are

some of the least understood issues in platform work.

National tools and responses to platform work challenges

Mapping national responses to platform work helped us understand which strategies

exist and how effective they are. Responses were listed by category, with legislation,

case law and administrator/inspectorate action constituting top-down responses, and

collective agreements, platform actions, and platform worker actions considered

bottom-up responses. Descriptive data for each response was also provided, such as

who initiated the response, the degree of implementation, and so on.

There were 177 responses from the 30 countries, including regional and local

levels. The number of responses varied widely from country to country, from zero to

nineteen. This variation may reflect how some countries consider platform work to be

less of a challenge, or have adopted a ‘wait and see’ approach, or that some gave fewer

but more wide-reaching responses. In some countries (e.g. Iceland, Malta, and

Bulgaria), platform work has hardly registered as a topic of concern, whereas in others

(e.g. Spain, Germany, Italy, and France) numerous stakeholders have taken concerted

action. Most responses concerned employment status (54), representation (46),

Study to gather evidence on the working conditions of platform workers

18

earnings (32), and social protection (31), indicating that policymakers and social

partners are likely to be especially aware of challenges in these areas.

Top-down responses

National legislation specific to platform work is very rare in the EU28, Iceland and

Norway. Working conditions and social protection of platform workers do not generally

constitute the direct material scope of national statutory legislation. France is the only

country that has enacted national legislation with a view to improving the labour and

social rights of platform workers.5

Other recent national legislation has indirectly tried to regulate working conditions and

social protection of platform workers, either through defining the employment status of

the platform workers, by regulating the working conditions and social protection for

persons in non-standard employment, or by strengthening rights and protection of the

self-employed. Such legislation mostly concerns specific business sectors, such as

personal transportation services (provided by platforms such as Uber) and food delivery

services (from platforms such as Deliveroo).

Rather than focusing on working conditions or social protection, national legislation

has, especially initially, primarily aimed to ensure fair competition and effective

market functioning in specific market segments such as personal transportation. Two

main approaches entail deregulating the traditional business sectors, and explicitly

applying existing standards and requirements to the new (platform) entrants. National

legislation may also focus on proper registration of platforms and taxation for

platform work alongside other sources of income.

Case law on platform work was reported in 16 of the 30 surveyed countries.

Many cases are ongoing, and many rulings are being appealed, so it is difficult to draw

clear and firm conclusions. Initially, numerous cases in national courts concerned

competition law issues and the personal transport sector. Most court cases

considered whether the services provided by Uber or similar platforms amount to

taxi/transport services, or those with different standards and requirements, for example

limousine or information society services.

National labour courts play a key role in defining the employment status of

platform workers, with many cases ruling on this, especially those providing personal

transport and food delivery services. However, courts have reached different conclusions

from similar evidence. These contradictory rulings reflect the different facts and

arguments raised in court, and the discretionary power of labour judges, who assess

facts on a case-by-case basis. Overall, national case law has thus far only modestly

improved clarity on employment status of platform workers.

Inspectorates and administrators have targeted undeclared work, social

protection contributions and coverage, safe working conditions and even

employment status, with authorities in Belgium, Denmark, France, Sweden and the UK

particularly active. However, inspectorates often struggle to address platform work,

which typically occurs in private spaces (homes) or dispersed public spaces (city

streets), rather than a ‘normal’ static workplace.

Bottom-up responses

Bottom-up responses focus more on the challenges of representation, earnings,

physical environment and working time, and especially involve food delivery.

Eight formal collective agreements between platforms and platform workers

were identified, with more pending. In several cases trade unions (e.g. Germany’s

5 Given the timeframe of the study, some of the most recent developments at national level (e.g. in some

sectors late 2019 in Italy) could not be included and analysed – demonstrating the fast-changing nature of the policy and regulatory framework of platform work.

Study to gather evidence on the working conditions of platform workers

19

NGG and IG BAU, and the Norwegian Transport Union) assisted platform workers in

organisation and negotiations, while in other cases platform workers organised

independently. These agreements cover a single platform, groups of similar platforms in

a country, or even national sectoral level (e.g. logistics).

Platform workers have organised strikes and demonstrations to improve

working conditions. Some have also created or joined cooperatives and collectives

(e.g. the Koeriers Kollektief [Courier’s Collective] in Belgium and cooperative SMart in

multiple countries). These aim to improve the collective voice and social protections,

among other goals.

Some platforms have taken action to address working condition challenges

faced by platform workers, or manage criticism of their practices. Several platforms have

modified their terms and conditions or specific practices to avoid legal challenges such as

lawsuits on employment status. Others have registered with national authorities or trade

associations, either voluntarily or through legal necessity, thereby formalising their

participation in labour markets.

Other forms of self-regulation have emerged, such as platforms creating

partnerships to provide platform workers with insurance and training. Some platforms

have signed on to charters or codes of conduct, agreeing to abide by certain principles

and decent work standards. Some of these initiatives appear to be innovative and

promising in addressing working conditions and social protection. The Frankfurt

Declaration,6 Crowdsourcing Code of Conduct,7 and Carta dei diritti fondamentali del

lavoro digitale nel contesto urbano8 [Charter of fundamental rights of digital work in the

urban context] are examples covering various forms of platform work.

In a few cases, platforms have taken punitive or ‘union-busting’ actions against

platform worker organisation. But others have encouraged platform workers to organise

and engage in social dialogue.

Instruments and actions at EU level

European institutions have released communications and initiated research specific to

platform work. Recent EU labour legislation explicitly refers to platform work as a type of

non-standard work and introduces material provisions with specific relevance for

platform workers who have an employment relationship.

This study assesses the relevance of EU law to platform work challenges. Twenty-one EU

instruments were selected for in-depth analysis, based on their probability of impacting

the working conditions and social protection of platform workers, and grouped as

follows:

 Non-standard work: Part-time Work Directive, Fixed-term Work Directive and

Temporary Agency Work Directive;9

 Health and safety: Health and safety for fixed-term work Directive and

Pregnant Workers Directive;10

6 Fair Crowd Work (2016), “Frankfurter Erklärung zu plattformbasierter Arbeit | Vorschläge für

Plattformbetreiber, Kunden, politische Entscheidungsträger, Beschäftigte und Arbeitnehmerorganisationen”, Declaration, Frankfurt, December 7 IG Metall (2019), “Report of the activities of the Ombuds Office of the Code of Conduct for Paid

Crowdsourcing, 2017-2018”, IG Metall, January 8 Comune di Bologna (2018), “Carta dei diritti fondamentali del lavoro digitale nel contesto urbano”, May

9 Council Directive 97/81/EC of 15 December 1997 concerning the Framework Agreement on part-time work

(OJ L 14, 20.1.1998); Council Directive 1999/70/EC of 28 June 1999 concerning the framework agreement on fixed-term work (OJ L 175, 10.7.1999); Directive 2008/104/EC of the European Parliament and of the Council of 19 November 2008 on temporary agency work (OJ L 327, 5.12.2008, p. 9–14)

Study to gather evidence on the working conditions of platform workers

20

 Individual labour rights: Written Statement Directive, Transparent and

predictable working conditions Directive (TPWC) and Working Time Directive;11

 Collective labour rights: Information and consultation Directive, Insolvency

Directive, Collective Redundancies Directive and European Works Council

Directive;12

 Work-life balance: Parental Leave Directive and Work-life Balance Directive;13

 Social protection: Recommendation on access to social protection;14

 Various aspects of non-discrimination: Employment Directive, Race Directive,

Gender equality in employment Directive, Gender equality in access to goods and

services Directive, Gender equality of self-employed Directive and Gender

equality in social security Directive.15

We further consider the General Data Protection Regulation16 (GDPR) and Regulation

2019/1150 on promoting fairness and transparency for business users of online

intermediation services17 (P2B). Competition law and collective bargaining as well as

data protection are handled separately in two reflection papers annexed to the study.

10

Council Directive 91/383/EEC of 25.06.1991 supplementing the measures to encourage improvements in the

safety and health at work of workers with a fixed-duration employment relationship or a temporary employment relationship (OJ L 206/19 of 29.07.1991), Council Directive 92/85/EEC of 19.10.1992 (OJ L 348/1 of 28.11.1992) on the introduction of measures to encourage improvements in the safety and health at work of pregnant workers and workers who have recently given birth or are breastfeeding 11

Council Directive 91/533/EEC of 14 October 1991 on an employer's obligation to inform employees of the

conditions applicable to the contract or employment relationship, Directive EU 2019/1152 of the European Parliament and of the Council of 20 June 2019 on transparent and predictable working conditions in the European Union (OJ L186/105 11.07.2019), Directive 2003/88/EC of the European Parliament and of the Council of 4 November 2003 concerning certain aspects of the organisation of working time (OJ L 299, 18.11.2003, p. 9–19) 12

Directive 2002/14/EC of the European Parliament and of the Council of 11 March 2002 establishing a general

framework for informing and consulting employees in the European Community (OJ L 080, 23/03/2002 p. 29 – 34), Directive 2008/94/EC of the European Parliament and of the Council of 22 October 2008 on the protection of employees in the event of the insolvency of their employer (Codified version) (OJ L 283, 28.10.2008, p. 36– 42), Council Directive 98/59/EC of 20 July 1998 on the approximation of the laws of the Member States relating to collective redundancies (OJ L 225, 12.8.1998 p.16-21), Directive 2009/38/EC of the European Parliament and of the Council of 6 May 2009 on the establishment of a European Works Council or a procedure in Community-scale undertakings and Community-scale groups of undertakings for the purposes of informing and consulting employees 13

Council Directive 2010/18/EU of 8 March 2010 implementing the revised Framework Agreement on parental

leave concluded by BUSINESSEUROPE, UEAPME, CEEP and ETUC and repealing Directive 96/34/EC (OJ L 68, 18.3.2010, p. 13–20), Directive (EU) 2019/1158 of the European Parliament and of the Council of 20 June 2019 on work-life balance for parents and carers and repealing Council Directive 2010/18/EU (OJ L 188, 12.7.2019, p. 79–93) 14

Council Recommendation of 8 November 2019 on access to social protection for workers and the self-

employed (OJ C 387, 15.11.2019, p. 1–8) 15

Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in

employment and occupation (OJ L 303, 2.12.2000, p. 16–22), Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin (OJ L 180, 19.7.2000, p. 22–26); Directive 2006/54/EC of the European Parliament and of the Council of 5 July 2006 on the implementation of the principle of equal opportunities and equal treatment of men and women in matters of employment and occupation (recast) (OJ L 204, 26.7.2006, p. 23–36), Council Directive 2004/113/EC of 13 December 2004 implementing the principle of equal treatment between men and women in the access to and supply of goods and services (OJ L 373, 21.12.2004, p. 37–43), Directive 2010/41/EU on gender equality of self-employed (OJ L 180, 15.7.2010, p. 1–6), Directive 79/7/EEC on the progressive implementation of the principle of equal treatment for men and women in matters of social security (OJ L 6, 10.1.1979, p. 24–25) 16

Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection

of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) 17

Regulation (EU) 2019/1150 of the European Parliament and of the Council of 20 June 2019 on promoting

fairness and transparency for business users of online intermediation services (OJ L186/57 11.07.2019)

Study to gather evidence on the working conditions of platform workers

21

Scope of EU action

The EU has the ability, where justified, and in accordance with the principles of

subsidiarity and proportionality, to set minimum requirements in the area of social

policy. However, and especially regarding working conditions, social legislation and policy

for platform workers remains mostly under Member States’ competences. In this

context, the EU has gradually taken legislative action, mostly in the areas of

employment and working conditions.

In terms of personal scope, the assessed EU legislation that regulates working conditions

under Article 153 TFEU applies to ‘workers’, referring to people with an employment

relationship or contract. The personal scope of the assessed EU directives hinges on

national legislation defining the concepts of ‘worker’, ‘employee’, ‘employment contract’,

or ‘employment relationship’. Through its extensive case law, however, the Court of

Justice of the European Union (CJEU) has gradually developed an EU-wide concept of

‘worker’ determined by the criteria of ‘subordination’ or ‘direction’. This helps ensure the

effectiveness of some directives that rely on national definitions of ‘worker’. CJEU case

law also ruled that bogus self-employed platform workers are to be reclassified as

workers (under the EU meaning of the concept) irrespective of the status agreed upon

by the contractual parties or defined in national legislation.

In this context, the following considerations are central:

 Platform workers who are classified as workers (including bogus self-employed

platform workers) fall within the remit of EU labour legislation.

 Platform workers who are self-employed fall outside the scope of EU labour

legislation.

Still, platform work profoundly challenges the binary divide of ‘workers’ and

self-employed that has been the cornerstone of labour legislation at national and

international level for decades. Evidence suggests varying approaches and

interpretations between EU Member States, and even within the same Member State for

labour and social law. Overall, the determination of self-employed versus ‘worker’ has a

crucial impact on applicable legislation and hence on the working conditions and

protection against social risks of individual platform workers.

In assessing the material scope of the legislation, two instruments were found to be

particularly relevant and adequate for one or more significant challenges in the area

of working and employment conditions for platform workers: the Transparent and

predictable working conditions Directive (TPWC) and GDPR. Neither instrument,

however, addresses all (significant) challenges of platform workers and may require

further adjustments to content and/or complementary action regarding enforcement,

especially in the case of GDPR. The TPWC Directive’s legal base in Article 153 TFEU

means it is limited in personal scope to workers and hence does not apply to the genuine

self-employed. The TPWC Directive, however, is relevant to challenges including

protection against abusive practices, obligatory information provision, the right to

parallel employment, protection against 'dismissal' (suspension, termination and other

restrictions), the right to effective legal redress and several rights relating to working

time. GDPR applies to all natural persons, and is relevant for establishing the right to

access personal data, including data concerning work allocation, work performance and

evaluation, and the right to data portability. As both are fairly new, our assessment is

preliminary.

Platform work generally qualifies as non-standard employment, and several of its

challenges are common to all types of non-standard work. EU legislation has tackled

these challenges, particularly through the non-standard work directives, and the new

TPWC Directive and Work-life Balance Directive. The Council Recommendation on access

to social protection is also of particular interest to the protection of rights of persons in

Study to gather evidence on the working conditions of platform workers

22

non-standard forms of employment, the genuinely self-employed, and persons

transitioning between labour market statuses. Enforcement of the relevant EU

legislation, and the issue of the minimum qualifying periods applicable to social

protection and parental leave schemes, are of particular relevance for platform workers

who work digitally with limited human supervision, at varying locations and often on

fragmented and small-scale tasks.

All other assessed EU legislation (collective labour rights, health and safety, and working

time) has varying relevance for platform workers, but only applies when there is an

employment relationship. Furthermore, these instruments are not adapted to the

specific working environment of the platform workers, and/or use concepts that

are not entirely fit for the purpose of regulating platform workers’ working conditions.

While somewhat broader in scope than platform work, P2B is an important step: an EU

legislative action addressing challenges specific to platform practices. P2B has great

relevance for platform workers, particularly regarding fair and transparent

intermediation when they are classed as ‘business users’ of information society

intermediation services. P2B is to some extent comparable with the TPWC, covering as it

does issues such as obligatory information provision, the right to notice in case of

contract revision, and restrictions on dismissal or equivalent measures. In personal

scope P2B applies to ‘business users’ – self-employed natural persons who exchange

services or goods with clients (legal persons) via an online intermediary service – which

includes a portion of self-employed platform workers. However, it is unclear at present

how many platform workers will be affected. Further limitations to personal scope are

also evident. P2B, for example, defines online intermediation services as ‘information

society services’, which does not include all types of platforms through which ‘work’ is

allocated and organised.

Assessment of the EU instruments finds that:

 Since the adoption of the TPWC Directive, platform workers who have an

employment relationship, and bogus self-employed platform workers that have

been reclassified as such by national judiciaries, have access to a wider scope of

protected labour rights, specifically in relation to working conditions. However,

there still needs to be better enforcement of their collective labour rights as

enshrined in current EU legislation.

 Self-employed platform workers who are economically dependent on a

single platform, and who work solo and in precarious situations, appear

to be the most vulnerable and least protected by individual and collective

labour rights, or by social protection legislation at both national and EU level. For

the small amount of self-employed who qualify as business users of information

society intermediation services, P2B should soon ensure access to similar (and in

some respects better than) protections of working conditions to those provided

for by EU labour legislation.

 The TPWC Directive and P2B have only very recently been adopted and it is

necessary to ensure their effective implementation.

Gap analysis

We analysed the extent to which national responses and selected EU instruments

address the most significant challenges of working and employment conditions facing

platform workers. The national and EU-level responses were rated by the extent to which

they addressed the individual challenges. For national responses we considered the

percentage of countries with one or more relevant responses, whereas for EU legislation,

we considered its personal scope, relevance, and adequacy.

Study to gather evidence on the working conditions of platform workers

23

Table 1: Summary of gap analysis

Note: In ‘countries w/responses’, the ‘total’ column refers to the percentage of countries with any relevant response, not the sum of on-location and online. P2B and GDPR could significantly influence these assessments, especially those deemed indeterminate. N/A: Assessed tools are not applicable to the challenge, NP: natural persons, W: workers, SE: self-employed. *Workers and self-employed are both in the personal scope of legislation, but certain additional limitations may effectively limit which platform workers are covered (See 7 of the study for a full key).

While very few responses target working conditions and social protection for all

platform workers, over half of all national responses specifically concerned personal

transportation and (food) delivery platforms. This may reflect a lack of awareness of

online platform workers, or the difficulty of addressing challenges associated with online

platform work at national level.

Overall the gap analysis suggests the following:

 Virtually no significant challenges are entirely resolved by national or EU

responses and instruments. The sole exception is data protection, but even for

this challenge, proper enforcement must be ensured.

 National responses at least partly address the most significant challenges for on-

location platform workers, but do very little for online platform workers.

 Platform workers meeting the criteria for worker status are generally better

protected, yet even here EU tools are not always fit for purpose because of the

differences between traditional and platform work.

 In spite of recent legislative initiatives, the impact of assessed EU instruments is

still limited in addressing the working conditions and social protection challenges

of platform workers, in particular when they are self-employed.

Conclusions and policy pointers

All platform workers, irrespective of their employment status, could benefit from

measures that aim for better (or more suited to platform work) protections in terms of:

(i) obligatory and timely provision of information about the terms and conditions

of collaboration, including on work allocation, organisation and evaluation, as well

as on a series of other dimensions specifically related to platform work

businesses;

Study to gather evidence on the working conditions of platform workers

24

(ii) advance notification, and for the right to an explanation in cases of refusal to

open an account, and of both temporary and more permanent termination of the

collaboration;

(iii) access to effective and timely out-of-court dispute-resolution mechanisms;

(iv) appropriate and transparent data protection when collecting and processing

personal and behavioural data;

(v) ‘collective’ rights, including rights to be represented, informed and consulted,

and the right to conclude agreements; and

(vi) effective application of the non-discrimination principle.

The most vulnerable forms of self-employment, non-standard work and indeed

all forms of work share many significant challenges with platform work.

Policymakers should be aware of these commonalities and consider broader approaches

rather than specific measures.

Employment status remains a core issue when addressing working conditions

and social protection challenges for platform workers at national and EU levels.

Most platform workers are self-employed, which excludes them from the personal scope

of much labour and social legislation at national and EU levels. In the types of platform

work that are more prone to bogus self-employment, determining specific platform

workers’ employment status is frequently challenging; case law moves very slowly on

this issue and has not offered much clarity. Policymakers may therefore consider actions

that make it easier to identify and reclassify bogus self-employed platform workers, and

clarify which platform practices are incompatible with self-employment.

Challenges related to intermediation, including algorithmic management, are

largely unaddressed, especially at national level. Intermediation in platform work

can entail surveillance, performance appraisal, and intransparent contracts. These

features seem to be growing beyond platform work as well. GDPR, P2B, and the TPWC

Directive are extremely important for increasing transparency and addressing such

issues. However, P2B only impacts a portion of self-employed platform workers, and the

TPWC Directive can only help platform workers with an employment contract. While the

actual impact of these tools on platform work is not yet fully clear, EU institutions may

consider further modifications or clarifications to ensure more platform workers fall

within their personal scope, and enforcement is effective.

Online platform workers are less visible and receive little attention, despite them

being probably the most numerous, and doing a form of work that is often inherently

cross-border. This implies that national authorities will find it difficult to address the

challenges of this type of work, while it is a more natural fit for EU action.

Besides regulatory options, the EU and Member States could consider

promoting voluntary codes of conduct or charters for platforms to commit to

upholding fair working conditions, for example to ensure dispute resolution mechanisms

are available to all platform workers, in a similar way to P2B requirements.

Finally, lack of data and the very recent adoption of relevant responses and

instruments limit our understanding of platform work. Coordinated action by

policymakers would ensure that high-quality data contribute to evidence-based actions,

social security coordination, and prevention of abuses and undeclared work. Moreover,

new national and EU legislation discussed in this study should be closely monitored to

assess whether it is sufficient, or if amendments or entirely new instruments are

required.

Study to gather evidence on the working conditions of platform workers

25

Europe stands on the brink of a new industrial revolution,

driven by new-generation information technologies […].

European Parliament (2019)

1. INTRODUCTION

Digitalisation, or the transformation of processes through new digital technologies,

continues to drive social and economic change. With this backdrop, the platform

economy has rapidly expanded in size and relevance. Online platforms such as Uber and

Deliveroo have quickly grown, spawned imitators, and disrupted economies and labour

markets around the world. Other types of online platforms such as Amazon Mechanical

Turk (AMT) and Jovoto have facilitated new types of outsourcing and innovation.

One function of such digital platforms is to leverage technology to intermediate supply

and demand for services, typically using internet-connected computers and

smartphones. When services are exchanged for payment, this phenomenon is known as

platform work.18

Researchers, policymakers, social partners and others have engaged in dialogue and

debate about platform work as it has grown in relevance across Europe. But a conflicted

narrative has emerged. Most of the literature recognises that platform work brings the

potential for both opportunities and challenges. Benefits include greater labour market

participation, economic growth, and social and technical innovation. However, platform

work also creates challenges for regulatory regimes, competing industries, and the well-

being of platform workers (Eurofound, 2018).

The focus of this study is the latter: understanding the working conditions and

social protection challenges of platform workers.

1.1 Previous key research and main findings

Policymakers have benefited from the existing research on platform work (Aloisi, 2015;

Eurofound, 2018; De Stefano, 2016; Lenaerts et al., 2017; Risak and Lutz, 2017), and

more recent publications, including from EU-OSHA (2017), Eurofound (Eurofound,

2018), the ILO (Berg et al., 2018), and the Joint Research Centre (JRC) of the

European Commission (Brancati et al., 2019; De Stefano and Aloisi, 2018; Pesole et

al., 2018), and other studies, have enabled preliminary conclusions to be drawn on the

working conditions and social protection of platform workers in the EU.

First, ‘platform work’ refers to many different sorts of work arrangements. This

heterogeneity can make it more difficult to consistently determine liability and assign

responsibility when things go wrong, such as work accidents or disputes between

platform workers and clients, and therefore properly apply labour and social regulations.

18

The definition and conceptualisation are developed in Section 3.

Platform work is all labour provided through, on, or mediated by online platforms in

a wide range of sectors, where work can be of varied forms, and is provided in

exchange for payment.

Platform worker refers to a natural person providing platform work.

Study to gather evidence on the working conditions of platform workers

26

Second, platform work presents difficulties for existing regulatory frameworks.

This is partially because the parties19 involved in platform work, and their legal

relationships, differ from those in standard work (Eurofound, 2018). Rather than

traditional management structures, platform work tends to rely on algorithmic

management. Moreover, platform work is considered a more ‘flexible’ form of work,20

whereas labour and social law frameworks remain better suited for more traditional

forms of work (ETUC, 2017).

Third, the employment status of platform workers is often unclear. For example,

employment status can be different even for platform workers using the same platform

in the same country (De Groen et al., 2018a). Most national legal frameworks would not

hold that platforms act as employers,21 though this assumption has been challenged by

legal scholars (Prassl and Risak, 2016). Arguments have also been made both for

(Drahokoupil and Fabo, 2016) and against (Aloisi, 2015) applying an intermediate

employment status, between employee and self-employed, to certain platform workers.

Fourth, platform workers face substantial practical and legal difficulties in

organisation and collective bargaining. To help mitigate power disparity and prevent

abusive situations, employees generally have the right to collective representation, with

the employer as counter party. However, factors such as their dispersed working

locations, lack of recognition that platform work qualifies as ‘work’, frequent turnover,

unclear employment status, and competition and cartel laws (Kilhoffer et al., 2017),

make this extremely difficult. Depending on the type of tasks performed and particular

platform used, platform workers can be cooperating or competing, and this further

challenges the suitability and practicality of more traditional collective bargaining

arrangements (Eurofound, 2018). Although new forms of organisation and

representation have been explored, the evidence indicates that most platform workers

have little bargaining power.

As a result, platform work can create challenges for decent working conditions.

These include a lack of career advancement opportunities, income insecurity,

psychosocial stress, and unsafe working environments.22 Platform workers are also at

risk of reduced access to social protection; they often forego social protections, pay

for more expensive options, or rely on social protection from a main occupation where

they have an employment contract (ESIP, 2019; Eurofound, 2018).23

1.2 Knowledge gaps and research objectives

Previous research has found that platform workers can face significant challenges in their

working conditions and social protection. If existing regulatory frameworks insufficiently

address these challenges, additional action may be appropriate at EU or national level.

However, significant gaps in our understanding of both challenges and regulatory

frameworks remain.

First, no existing research covers the entirety of the EU, Norway and Iceland,

even as the challenges of platform work and corresponding policy or regulatory

responses vary a great deal between countries (Lenaerts et al., 2017). We lack a

19

Platform, platform worker, and client – see Section 3.

20 See discussion of Challenges for platform workers in Section 4.

21 This assumption is being challenged, especially by case law and legislation. See National tools and

responses to platform work challenges.

22 See Challenges for platform workers.

23 This is because social protections like pension, paid sick days, and accident insurance are often interlaced

with employment contracts (and funded at least in part by employers). Note, however, that the statutory and effective access to social protections varies a great deal across Member States. See especially Eurofound (2018a) and EU-OSHA (2017).

Study to gather evidence on the working conditions of platform workers

27

complete picture of relevant policy and regulatory developments that accounts for both

EU and national initiatives.

Second, available information focuses on a variety of issues, platforms, and

countries, all assuming slightly different understandings of what constitutes platform

work. This creates a patchwork body of evidence that makes it difficult to

comprehensively assess working conditions and social protection for platform workers in

Europe.

Third, much of the research lacks a multidisciplinary approach. Understanding

platform work in all its complexity requires an approach combining policy, social, and

legal perspectives.

This study, therefore, seeks to contribute to a greater understanding of platform work

and the challenges facing platform workers with a comprehensive mapping of the

challenges and responses relating to their working conditions and social protection in the

EU28, Norway and Iceland. Up-to-date24 information on the regulatory environments and

relevant developments affecting all types of platform work has been analysed by:

1) identifying the relevance, diffusion and nature of these challenges for working and

social conditions;

2) mapping national developments;

3) examining select EU legislation that may affect working conditions and social

protections;

4) assessing the extent to which national and EU measures address the challenges.

The study thereby provides evidence on whether additional national or EU action, and

provides pointers as to what form such action might take.

1.3 A need for EU action?

While the European Commission (2016a) has shown awareness of the challenges

associated with platform work and is considering further action, the debate continues as

to whether and to what extent the EU could and should also act. In the meantime, the

European Parliament25 and EU-level social partners26 have advanced EU-level proposals

to address social challenges facing workers in the platform economy (EU-OSHA, 2017).

First, because platform work impacts the EU principle of proportionality, a clear

understanding of its size and prevalence is needed, which impacts the EU

principle of proportionality.27 Its dynamic and fast-changing nature keeps the

evidence scattered and quickly outdated (De Stefano and Aloisi, 2018), and some argue

that data collection and other action at EU level could lead to a better understanding of

platform work (Eurofound, 2018).

Second, any EU action must respect the principle of competency.28 While the EU

has the ability29 to set minimum requirements in the area of social policy, notably with

24

See Methodology and Limitations and constraints on timeframe.

25 See Report on a European Agenda for the collaborative economy, available at

http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//TEXT+REPORT+A8-2017- 0195+0+DOC+XML+V0//EN 26

See “ETUC resolution on digitalisation: ‘towards fair digital work’”, available at

https://www.etuc.org/en/document/etuc-resolution-digitalisation-towards-fair-digital-work#.WeYb77jg18E

27 Treaty on European Union, Article 5, states ‘the content and form of Union action shall not exceed what is

necessary to achieve the objectives of the Treaties’. The size and prevalence of platform work are further discussed in Size, prevalence, and expected evolution of platform work in the EU. 28

The EU has exclusive competences in certain areas, shared competences with Member States, and

supporting competences. See Title 1 of Part 1 on the Treaty on the Functioning of the European Union.

Study to gather evidence on the working conditions of platform workers

28

regard to working conditions, social and labour law remains mostly under Member

States’ competence. Several studies have pointed out that the transnational

character of the platform economy would justify an EU regulatory framework,

where the principle of subsidiarity in the field of social and labour law could be applied

(Eurofound, 2018; EU-OSHA, 2017). Such a framework may be appropriate in the

context of the EU’s interest in maintaining a well-functioning Digital Single Market30 and

ensuring proper competition (Dittrich, 2018). At present, regulatory and policy responses

vary greatly across Member States, creating a mosaic of markets and leading to calls for

a common policy response at European level.

Third, policymakers are cognisant of the potential trade-off between innovation and

new regulation (Ranchordás, 2015). Even if granted that additional regulation could be

useful, new regulatory constraints may hinder innovation and the economic potential of

platforms. A frequent discussion point concerns whether existing national and EU

regulations may suffice, if properly applied and enforced, and the extent to which new

legislation or other actions may be required (De Groen et al., 2018a).

Thus, the debate continues over the possible regulatory framework for platform work at

EU and national level. Recent developments, such as the European Pillar of Social Rights

(EPSR),31 could suggest a timely opportunity to better monitor new forms of employment

at a supranational level, and coordinate Member State initiatives (Pesole et al., 2018).

1.4 Structure of the study

The remainder of the study is as follows:

Section 2 Methodology explains how the research was carried out.

Section 3 Conceptualisation of platform work forms the conceptual basis for

platform work, particularly in what distinguishes platform workers from other types of

workers. It also briefly covers the types of platform work, the size of platform work in

the EU, and the expected evolution of the platform economy.

Section 4 Challenges for platform workers gives a framework for assessing job

quality and then applies it to information collected from an extensive literature review

and several types of fieldwork. This Section seeks to identify the kinds of working

conditions and social protection challenges platform workers face, for which types of

platform work, and in which countries.

Section 5 National tools and responses to platform work challenges gives a

framework for classifying and comparing tools and responses and then applies it to a

mapping of tools and responses at national level for the EU28, Norway and Iceland.

Section 6 Instruments and actions at EU level horizontally assesses selected EU

legislation for its impact on the working conditions and social protection of platform

workers. It further considers non-legislative EU actions.

Section 7 Gap analysis: which challenges remain to be addressed? builds on the

previous content to assess the extent to which challenges have been addressed by

national and EU tools and responses.

29

Where justified, and according to the principles of subsidiarity and proportionality.

30 The Digital Single Market strategy discusses platform work and the platform economy more broadly, though

most attention is paid to technological innovation as a means for economic growth and competitiveness, and reducing intra-European barriers to doing business. Less focus is paid to the social dimension and the challenges that platform workers face (European Commission, 2017). However, the European Parliament Resolution reacting to the Digital Single Market Act calls on Member States to ensure employment and social policies are fit for digital innovation, entrepreneurship, and the growth of the platform economy and its potential for more flexible forms of employment. See European Parliament resolution of 19 January 2016 on Towards a Digital Single Market Act (2015/2147(INI)). 31

See discussion in 6.2.3.

Study to gather evidence on the working conditions of platform workers

29

Section 8 Conclusions and policy pointers brings together all findings and offers

pointers for future policy consideration and research.

After the body of the study, we include additional inputs to the research.

Study to gather evidence on the working conditions of platform workers

30

2. METHODOLOGY

The study relied on a mixed-methods approach with desk research and several types of

fieldwork. The scope of the research was the EU28, Norway and Iceland from 2013 to 30

June 2019.32 The study covers all types of platform work meeting the definition in Figure

3.

The overarching objective of this study is to provide an evidence-based analysis of the

challenges faced by platform workers with regard to their working conditions and social

protection. It reviews policy and legal responses to those challenges at national and EU

level, and based on this analysis, examines the potential need for further EU action.

The research team undertook this research in four main steps. Step 1 conceptualised

platform work (see Section 3) and mapped the challenges it presents to working

conditions and social protection (see Section 4). Step 2 gathered data on tools and

responses to these challenges (see Section 5). Step 3 analysed a selection of EU

instruments for their relevance to challenges platform workers face (see Section 6). Step

4 synthesised all this evidence with a gap analysis (see Section 7), and conclusions and

policy pointers (see Section 8).

This Section outlines the research questions used in the study, methods of data

collection and analysis, quality control, and limitations and constraints.

Throughout, the research was carried out in full compliance with the European

Commission’s Better Regulation Guidelines. Special attention has been paid to vulnerable

groups and issues related to discrimination and inequality.

2.1 Research questions and data collection

The research questions were designed to allow the research team to meet the

overarching goal of the study: to assess whether further EU action on platform work is

merited. Table 2 lists the research questions, which step or steps of the research process

addressed them, and which research methods were utilised.

32

In a few exceptional cases, tools and responses were updated in early December 2019.

Study to gather evidence on the working conditions of platform workers

31

Table 2: Addressing research questions

Research questions

Step(s) addressing research question

Method(s) addressing research question

1 2 3 4 d e s k r

e s e a rc

h

S e m

i- s tr

u c tu

re d

in te

rv ie

w s

q u e s ti o n n a ir

e s

n a ti o n a l fo

c u s g

ro u p s

v a li d a ti o n w

o rk

s h o p

Challenges

conceptualisation How do we conceptualise platform work and its challenges? X X X

mapping Which challenges are present, and where? X X X X X

How prevalent are the challenges? X X X X X X

analysis How significant are the challenges? X X X X X X X X

How are the challenges likely to develop in the future? X X X X X X X X

Responses

mapping What are the national responses to the challenges? X X X X X X

What are the EU responses to the challenges? X X X X

analysis

Are existing national responses adequate now, and in the foreseeable future? X X X X X X X

Are existing EU responses adequate now, and in the foreseeable future? X X X X X X

Overarching objective

analysis Is further EU action needed? X X X X X

What particular EU action/s is/are suggested? X X X X X Note: Step 4 covers each research question insofar as it summarises all results.

Study to gather evidence on the working conditions of platform workers

32

Two main phases of desk research were conducted. An initial phase established the

current knowledge base to avoid duplicating existing research. A second round of more

targeted desk research was conducted at the beginning of each step, and throughout as

needed.

This initial desk research was broad in scope and supported all tasks. It was based on

keyword searches and a review of relevant academic and other providers of analysis. It

covered academic and grey literature on platform work in general and in relation to

working conditions and social protection, developments at national and EU level with

regard to policy, legislation, regulation, and collective agreements, media and public

debates on platform work, existing data and efforts to quantify the platform economy,

and similar topics. Legal documents, publications of policymakers and social partners,

existing databases, articles published on traditional and social media channels, opinion

pieces, and communication from platforms were also considered. Literature from

different scientific disciplines, such as law, economics, and sociology, was consulted to

form a complete perspective on platform work.

A second round of desk research was carried out at the start of each step to ensure that

any knowledge or data gaps were addressed. This stage focused on the specific topic

and objectives of each step. For Step 1, the initial desk research was complemented with

literature on (the working conditions and social protection of) other employment types

that have characteristics in common with platform work, for example temporary agency

work. For Step 2, additional efforts were made to gather literature in the national

language of each country of the EU28, Norway and Iceland, with particular attention paid

to the six Member States for which a focus group was organised. For Steps 2, 3 and 4,

further desk research was performed to collect relevant policy and legal documents that

support policy and legal analyses. Finally, the second round of desk research also helped

ensure that the latest literature and evidence were absorbed.

The research team also undertook extensive fieldwork to gather new data and

complement the desk research. Stakeholders, including policymakers, academic and

legal experts, platform representatives and social partners representing employers and

workers, national administrators, labour inspectorates and occupational safety and

health (OSH) authorities more broadly, and business sector associations were all

consulted.

The research team was responsible for preparing the list of stakeholders (including back-

up options), which was refined with input from the European Commission. Each

stakeholder was sent an invitation to participate, a letter on behalf of the commission

addressed to potential participants, a data protection notice and a privacy statement.

Semi-structured interviews were held with stakeholders and different interview

templates were developed for each interview partner for the respective step of the

research. The results of the interviews were used throughout the analysis to gain

conceptual clarity, provide the background to the research and verify the results of the

analysis.

Country expert surveys were carried out with one socio-economic and one legal expert

for each country, making a total of 60 survey responses. The experts were identified

through the existing expert network, the research team and informal networks. The

surveys included both unique and shared sections to ensure comparability and

triangulate information, while making use of the expert knowledge of respondents.

The experts received detailed instructions on how to complete the surveys, which asked

them to catalogue policy responses according to the prescribed typology and address all

stages of implementation, including initiated but abandoned, pending or under

discussion, or partially or fully implemented. While the focus was on responses, experts

were also invited to catalogue particularly relevant tools. They were asked to describe

responses in detail, including aspects such as the initiator of the development (e.g.

government, social partners or other stakeholders), the scope of the development (e.g.

Study to gather evidence on the working conditions of platform workers

33

the types of platform work or platform workers covered), the focus of the development

(e.g. direct or indirect impact on platform work), the degree of implementation, the

timing of the development (e.g. proactive or reactive, response to specific incident), the

originality of the development (e.g. new law or modification of existing) and other key

characteristics. In addition to their mapping of national developments, experts were

asked to include relevant literature to obtain further information and cross-validate it

with the desk research. Country expert surveys were thus used to provide an

assessment of the severity of individual country challenges, a mapping of the most

relevant national responses and tools, and a review of the most pertinent literature.

Focus groups were held for six countries,33 to learn more about especially interesting

policy developments and trends with potential relevance at EU level. Countries were

selected on whether particularly original policy developments or experimental

approaches to platform work had occurred there. Each focus group had between six and

twelve participants consisting of at least one policymaker, academic or legal expert,

social partner, platform representative, and platform worker.34

Finally, a validation workshop was held in Brussels where stakeholders, including

policymakers, academic and legal experts, social partners, platform representatives and

platform workers, were consulted on the findings. This workshop focused on confirming

findings of the gap analysis, addressing any potential shortcomings of the conclusions,

and discussing the merits of policy pointers.

2.1 Data analysis

Turning to the analysis, the research team explored both legislative and non-legislative

aspects of challenges arising from platform work, and potential national and EU-level

responses. This allowed for a broad, multi-disciplinary analysis of platform work

across the EU28, Norway and Iceland. The question about the potential need for EU

action in the field of platform work was addressed in a structured way, relying on the

information gathered throughout all project steps and delivering the final conclusions

and policy implications of the study.

In Step 2, special attention was paid to any national developments that referred to

existing EU framework, both legislative and non-legislative. Examples of these

are the EPSR, or the European Commission recommendations and directives, especially

those on non-standard work, on individual rights, for anti-discrimination measures and

relating to work-life balance, as well as law relating to areas such as competition law and

labour mobility.

These are analysed in Step 3, and findings were also used to draw conclusions about

what kind of EU action would be particularly appropriate, based on a detailed and

systematic analysis of the most relevant selected areas of EU social law and related case

law. Limits with respect to their relevance and application to platform workers will be

identified and solutions to address these limits are proposed.

Step 4 first condensed and systematised the information gathered through the country

experts survey and national focus groups. The aim was to perform a comparative

analysis and in particular look for patterns of national developments across different

countries (i.e. similar developments, either in content or form, which emerge in

countries that share the same characteristics). Step 4 establishes whether challenges

are recurring in several countries and examines whether challenges appear to have the

same roots or consequences within the socio-economic system. In addition, special

attention is paid to challenges that emerge across the EU and appear to show a

33

Denmark, Estonia, Spain, France, the Netherlands and Slovenia.

34 In one country, however, the platform worker declined to participate at the last minute.

Study to gather evidence on the working conditions of platform workers

34

transnational character or are a result of the transnational nature of specific types of

platform work, meaning EU action could play a significant role.

The analysis also assesses the extent to which existing national developments or

existing EU legislation potentially relating to national legislation address the

challenges, to understand what - if any - role the EU may have in complementing such

national developments. The purpose of this part of the analysis is to show what role EU

initiatives are already playing in addressing the challenges of platform work through

national developments, which will in turn inform their appropriateness and possible

scope in the future.

In sum, the combined analysis of findings seeks to establish whether challenges relevant

at EU level are addressed by existing EU initiatives and national developments,

highlighting possible gaps. In addition, it assesses whether and how, based on existing

linkage, national developments and EU initiatives can complement each other.

2.2 Quality control

Several methods were employed to ensure quality control was maintained throughout

the research.

First, a team of quality control managers provided input at critical junctures in the

research. This multidisciplinary team comprised leading European authorities on platform

work, who brought expertise from the legal, socio-economic, and academic spheres.

Second, findings were confirmed through triangulation.35 Findings derived from different

sources or research methods were thus validated against one another, and their

robustness across different national contexts, types of platform work, types of platforms

or workers, both within and across tasks, could also be verified. Information that could

not be validated through triangulation has either not been included in the final study or a

note included stating that the information in question was derived from just a single

source. Preference has been given in this study to information derived from academic

(peer-reviewed) publications, legal texts, policy documents and other validated sources.

Information primarily derived from stakeholder consultations, or anecdotal evidence

more generally, has been clearly indicated as such.

Third, this research has benefited from frequent consultations with the European

Commission. This has helped to ensure that the most suitable experts were identified,

the most important upcoming events and legislation were accounted for, and that the

methodology and content of the research were subject to rigorous review.

2.3 Limitations and constraints

A few constraints on the study should be noted. First, the quality of existing data on the

size and prevalence of platform work in the EU is suboptimal, and predicting future

developments is very difficult.36 Moreover, many relevant responses (e.g. court cases

and legislation) are new or pending. This makes assessing the gaps in working

conditions and social protection a more difficult task. For practical reasons, a cut-off date

of 30 June 2019 was implemented, with changes in legislation or other significant

developments after this date not considered, except for a few exceptional

circumstances.37

35

Findings derived from a single source were either not included in the final report, or noted (e.g. with a

footnote or brief statement in text) as less certain.

36 Data limitations are discussed in more detail in Section 3.3.2.

37 For example, the Recommendation on Access to social protection for workers and the self-employed, a key

initiative as part of the rollout of the European Pillar of Social Rights, was only adopted 7 November 2019. A few pieces of particularly relevant national legislation have also been updated past the cut-off date.

Study to gather evidence on the working conditions of platform workers

35

Second, while the analysis aims to establish a comprehensive mapping of responses at

the national level, time and logistical constraints meant that only a selection of EU law

could be analysed. The specific legislation38 was agreed upon between the research team

and the European Commission based on its likelihood of impacting working conditions

and social protection for platform workers. However, this is a non-comprehensive

analysis of EU law. Furthermore, the P2B Regulation39 was not originally a part of the

legal analysis but added partway through the research. Therefore, it is not assessed to

the same depth as other EU instruments.

38

Covered in Section 6.

39 Regulation (EU) 2019/1150 of the European Parliament and the Council on promoting fairness and

transparency for business users of online intermediation services (OJ L 186, 11.7.2019, p. 57–79).

Study to gather evidence on the working conditions of platform workers

36

3. CONCEPTUALISATION OF PLATFORM WORK

The conceptualisation and definition of platform work has been a topic of much scholarly

discussion. Similar and overlapping terms include crowdwork, gig work, and a plethora

of others, as explored by a number of high-quality research papers (Coyle, 2016;

Codagnone et al., 2016; Drahokoupil and Fabo, 2016; Martin, 2016; Maselli et al., 2016;

Frenken and Schor, 2017).

The European Parliament’s Directorate-General for Internal Policies expressed its

preference for the more neutral terms ‘platform economy’ and ‘platform work’, which

avoid potentially misleading connotations (Forde et al., 2017). For this reason, ‘platform

work’ is the preferred term for this study.

The remainder of this section further explains the notion of ‘platform work’, ‘worker’, and

other key conceptualisations.

3.1 Standard and non-standard employment

Before addressing platform work, it is useful to think of the traditional or ‘standard’

employment relationship as a reference point. This is illustrated in Figure 2.

Figure 2: Conceptualisation of standard employment

Source: adapted from Hasle (2007).

Standard employment or ‘standard work’ refers to a full-time, open-ended

employment contract with one employer. As shown, employers and employees are

traditionally bound by an employment contract or employment relationship. The

employment contract typically stipulates the rights and obligations of both parties, such

as the scope of the services and tasks to be performed by the employee, working hours,

the remuneration they will receive, and so on. Employees perform their services under

the ‘direction’ or ‘subordination’ of the employer. However, employees would have no

direct contractual relationship with a client. At most, employees render a service on

behalf of their employer (ILO, 2016).

Non-standard employment or ‘non-standard work’ refers to all forms of paid labour

that deviate from the standard reference or ‘standard work’ such as part-time work,

fixed-term work, temporary agency work, casual work, student jobs, zero-hour

Study to gather evidence on the working conditions of platform workers

37

contracts, and on-call work. This typology (Eurofound, 2017a) also includes self-

employment, although unlike the other non-standard forms, it does not correspond to an

employment relationship in any legal sense.

Non-standard employment does not necessarily entail precarious working conditions. It

can also refer simultaneously to situations with an employment relationship, as is the

case for temporary agency workers in the EU,40 and part-time or fixed-term contract

employees. However, many of the safeguards built into standard employment do not

always apply to non-standard employment (e.g. dismissal protection or paid sick leave),

and thus non-standard employment can create challenges for working conditions

(Eurofound, 2017b).41

Platform work, when performed without an employment relationship, is an example of

non-standard employment. Platform work challenges the fundamentals of the standard

reference; for example, (but with exceptions) platform work usually has no employer

giving the platform worker instructions in the traditional sense, remuneration is typically

paid per task rather than as a wage or salary, and work can take place on more than

one platform.

Although this study investigates job quality challenges in platform work, there are other

forms of non-standard work that could have similar challenges. As a starting point for

this analysis, casual work, interim work and portfolio work is briefly discussed to

illustrate the overlap and differences.

Casual work is irregular, on-call or intermittent work characterised by instability and

discontinuity, facilitated by the use of ICT, and without the prospect of permanent work

(European Parliament, 2000; Eurofound, 2018a). Like platform workers, casual workers

experience a high degree of flexibility, but also face irregular employment relationships,

which lead to unclear legal coverage and a lack of social protection and representation.

Lower wages (due to very low guaranteed hours and only completed work being paid),

insecurity, changing workplaces and limited guidance cause higher precariousness and

occupational risks (De Stefano and Aloisi, 2018; Vereycken and Lamberts, 2018;

Eurofound, 2018a; ILO, 2019a). Platform workers seem to have higher levels of

autonomy in choosing their working hours than casual workers in general, and on-call

workers in particular.

Interim or temporary agency workers enter a company to answer a periodic or

suddenly increased need for certain skills (ILO, 2019b). Similarly to platform work,

temporary agency workers have various skills and education levels, and operate in

diverse sectors (Vereycken and Lamberts, 2018; Eurofound, 2018b). Although both

types of employment are set within triangular relationships, the role of temporary work

agencies is more extensive than the role of platforms. Additionally, while there is some

diversity in the legal arrangements governing temporary agency work, the basic

functions and legal relationships of the temporary work agency, worker and user firm are

more or less identical across Europe.42 This is in stark contrast to the diverse

arrangements in platform work (WEC-Europe and UNI Europa, 2018). Whereas

temporary agencies are de jure, wherein the employer and temporary agency workers

are classified as employees, platforms typically claim they provide only a matching

service between labour supply and demand,43 and platform workers are usually

considered self-employed (ibid).

40

Under Directive 2008/104/EC, all temporary agency workers must have an employment contract.

41 See also Lamberts et al. (2016) for discussion of the European Working Conditions Survey’s findings on non-

standard working conditions.

42 Directive 2008/104/EC of the European Parliament and of the Council of 19 November 2008 defines each

party in temporary agency work and regulates their legal relationships.

43 For example, in C-434/15 (Asociación Profesional Elite Taxi v Uber Systems Spain, SL), Uber has argued that

it provides information services, not transportation services. For discussion on the role of platforms as

Study to gather evidence on the working conditions of platform workers

38

Portfolio work refers to ‘small-scale contracting by freelancers, self-employed or micro

enterprises, conducting work for a large number of clients’ (Eurofound, 2018b: p. 14).

Portfolio work is common in the creative sector, where it could support the career

development of professionals (Vereycken & Lamberts, 2018). As with some types of

platform work, portfolio work in principle puts the worker in charge of the choice of

tasks, working hours and workplace (Vereycken and Lamberts, 2018; Eurofound,

2018b). Portfolio workers risk health problems similar to those encountered by platform

workers, as both are prone to working during illness, disturbed work-life balance and

limited social interactions (Eurofound, 2018b).

In spite of the heterogeneity in platform work and non-standard forms of work, there are

shared characteristics such as short-term contracts, unstable and unpredictable work

schedules, and non-conventional workplaces. But in stark contrast with platform

work, and despite some differences across Europe, standard employment, and even

some forms of non-standard employment,44 has rather consistent and clear

legal relationships. (WEC-Europe and UNI Europa, 2018).

Therefore, some of the challenges platform workers face in the area of working

conditions and social protection are linked to factors also found in other (non-

standard) forms of work. Some challenges, such as discrimination, may also arise in

the broader labour market, but potentially have a bigger impact on platform workers

given that some are particularly vulnerable, for example those who are younger, less

experienced in the labour market, or recent migrants (Eurofound, 2018).

Other challenges are specific to platform work. For example, many platforms use

intransparent and disadvantageous terms and conditions, and may alter these without

notifying platform workers (Graham et al., 2019). Clickworkers45 risk unique forms of

psychosocial stress and non-payment, and this type of work only became possible with

the advent of online platforms. This study will attempt to clearly distinguish between

challenges particular to (a given type of) platform work, and non-standard work

arrangements.

A final but particularly important type of non-standard employment is self-

employment, which refers to many different types of work. Eurofound finds that

roughly half of self-employed people have high levels of job quality, whereas a quarter

are characterised by economic dependence, low levels of autonomy, and financial

vulnerability (2017c).

In this study, the term ‘self-employed’ is used to refer to anyone who works and earns

remuneration outside the context of an employment relationship and is thus

contractually self-employed.

An important distinction to make is between false or ‘bogus’ self-employment and

‘genuine’ self-employment. Bogus self-employment refers to individuals who are

factually employees (subordinate to an employer), but for reasons connected to the

evasion of regulatory legislation, are contractually self-employed (Eurofound, 2017c).

The opposite status is sometimes called genuine self-employed, which refers to

individuals whose factual circumstances and contractual status are both aligned with

self-employment.

intermediaries versus simple online marketplaces, and how this can impact platform workers, see De Groen et al. (2018).

44 For example, under Directive 2008/104/EC, temporary agency work is defined and many contractual

obligations established for all parties. Most relevant, all temporary agency workers in the EU must have an employment contract.

45 Platform workers using platforms such as Amazon Mechanical Turk, CrowdFlower, or Clickworker.

Study to gather evidence on the working conditions of platform workers

39

3.2 Conceptualisation of ‘worker’

The term ‘worker’ (NL: werknemer, FR: travailleur, DE: Arbeitnehmer, ES: trabajador)

may have different meanings when used in different EU languages and/or in different

contexts. It may even have different legal interpretations.

In English, ‘worker’ commonly refers to all who work (travailleur, trabajador, and so on),

irrespective of their legal employment status or classification in most national labour law

frameworks. By this understanding worker would include travailleurs salariés and

travailleurs non-salariés, werknemers and zelfstandigen, employees and the self-

employed.

The terms ‘worker’ and ‘employee’ have both been used in EU labour legislation, the

latter word most often being used in older directives and specifically those that concern

collective labour rights. Most recent EU labour legislation, however, uses the concept of

‘worker’. Both terms have the same legal connotation in EU labour legislation and refer

to an employment relationship.

Settled case law of the Court of Justice of the European Union (CJEU) has given an EU-

autonomous interpretation to the concept of ‘worker’. In the Lawrie Blum case,46 decided

under the free movement of workers framework, the CJEU defined a ‘worker’ as ‘a

natural person who for a certain period of time performs services for and under the

direction of another person in return for remuneration’. This definition of ‘worker’ has

also been applied by the Court to the concept of ‘employee’ in older EU directives.

However, there is no single definition of worker in EU law; it varies according to the area

in which the definition is to be applied.47

One should remain very careful and not lose the nuances, as the terms ‘worker’ and

‘employee’ may still have a slightly different meaning across the different directives,

depending on whether they refer to the national legislative definition of each Member

State, or to the autonomous EU meaning of ‘worker’ as interpreted by the CJEU. Member

States’ national legislation differs with regard to the concept of worker or employee, and

differences may even exist between labour and social protection legislation of a

particular country.

Under EU law, the key criterion for establishing the existence of an employment

contract, and hence classification of ‘employee’ or ‘worker’, is that of subordination. The

CJEU has given a broad interpretation of this criterion, which may go much further than

that determined by national legislation. In other words, the concept of subordination

under EU law may be broader than the one provided under national law with the

consequence that a person may be considered as self-employed under national law and

as ‘worker’ or ‘employee’ under EU law. In such cases, the person should be covered by

the protection of the relevant EU directives. There are, however, countries where the

subordination requirement has been interpreted more broadly by national legislation or

national courts than the current interpretation of the CJEU and/or where the

subordination requirement has been complemented by other and newer criteria in order

to determine the status of a worker.48

46

Judgment of the Court of 3 July 1986. - Deborah Lawrie-Blum v Land Baden-Württemberg. - Reference for a

preliminary ruling: Bundesverwaltungsgericht - Germany. - Worder - Trainee teacher. - Case 66/85.

47 cf. Judgment of the Court of Justice of 1 March 2012, O’Brien, C-393/10, ECLI:EU:C:2012:110, paragraph

30.

48 In Austria, economic dependency is taken into account for determining the status of an employee in social

security law apart from the personal dependency criterion which is similar to the ‘subordination requirement’ and the key determinant under labour law. In several countries there is no statutory definition of ‘employee’ and reference is made to the existence of an employment contract (examples are – non-exhaustive - AT, BG, BE, DK, FR, IE, IS, LU). In these countries but also in countries which have a statutory definition, the concept of employee is mainly developed through case law. Whereas the ‘subordination’ dimension is always one of the key criteria to determine the existence of an employment relationship, other aspects are also considered. In the UK, the judiciary has over the years developed a number of tests for helping to define those who are

Study to gather evidence on the working conditions of platform workers

40

In addition to the boundary given by the degree of subordination, which differentiates

between a worker/employee and a self-employed person, another boundary is relevant

to platform work: that of the marginality of the economic activity. Under a certain

threshold, an activity can be considered (by national or EU law) as too marginal or

ancillary to be an economic activity (be it performed by a worker or self-employed

person). The CJEU has, however, interpreted the criterion of ‘genuine and effective

activity’ very broadly, also identifying as workers those doing a limited number of hours

per week (e.g. five and a half), working students, au pairs,49 or students working for

four days during the holiday period. In some cases, national legislation seems to set

stricter thresholds.50

In this study, we use the term ‘worker’, as in ‘platform worker’, without any

legal connotation, unless specifically noted as otherwise. In these exceptional

cases, we state that we mean ‘worker’ as defined by national law, specific EU legislation,

or by the autonomous EU meaning of the word.

3.3 Conceptualisation of platform work(er)

The conceptualisation of platform work consists of two parts: identifying what platform

work is compared with other types of work; and distinguishing between different types

of platform work.

In this study, platform work is understood as all labour provided through, on or

mediated by online platforms in a wide range of sectors, where work can be of

varied forms and is provided in exchange for payment. This definition is derived

from Eurofound (2018) and visualised in Figure 3.

employees (which corresponds to the EU concept of ‘worker’ as the UK concept of ‘worker’ refers to a third category who is neither an employee nor a self-employed) and those who are not. The tests have included: 1) the control test – i.e. the level of control exercised by the employer; 2) the integration test – i.e. the extent to which a person is integrated into an organisation; 3) the economic reality test – i.e. looking at the contract as a whole and deciding whether the individual was in business on their own account; 4) the multiple factor test – i.e. that there are a variety factors that could apply depending upon the circumstances; and 5) mutuality of obligation – i.e. whether there is an obligation upon the employer to provide work and whether there is an obligation upon the employee to carry it out. In Ireland, courts assess the mutuality of obligations, right to use substitutes, the degree of integration in the workplace, the economic realities as criteria next to the right to control and the degree of control exercised by employers. In Poland, the statutory labour law definition of worker includes the criterion of economic, social and personal risk sharing by the employer apart from the subordination and supervision dimension, the personal provision of the services and the remuneration requirement.

49 The term ‘au pair’ refers to a young person who is temporarily hosted by a family, and provided room and

board in return for light everyday family tasks.

50 See Section 5.

Study to gather evidence on the working conditions of platform workers

41

Figure 3: Conceptualisation of platform work

Source: authors’ own elaboration.

The core features of platform work are a triangular relationship between platform,

platform worker and client, and online intermediation.51 Technology plays an

important role in work organisation, by using algorithms to pair clients and platform

workers, for example. Many forms of platform work involve smaller tasks, which

are part of a larger process (i.e. fragmentation and micro-tasks). Services are provided

on demand and the work is usually provided on a temporary or piecemeal basis.

Typically, platform workers do not have a long-term or stable relationship with the client

they are providing the services for, but there are some exceptions where platforms allow

their platform workers to have repeat clients or build up their own clientele. Platform

workers typically have no long-term relationship with the platform either.

However, it is important to note that the model in Figure 3 is very basic: a

simplification to help understand the labour element of platform work. In practice, many

variants and exceptions are possible, and more actors and intermediaries may be

involved. Furthermore - and critical for working conditions, social protection, and the

notion of subordination - the platform’s intermediation can entail from minimal to

very significant control over platform workers.

A few examples help illustrate the limits of this conceptualisation, when considering the

broader value chain of platform work. With personal transportation platforms, there are

frequently other actors (in addition to the basic three). For example, platform workers

may rent a car from a person or business. In Portugal, a platform worker cannot work

directly for Uber; rather, a transport company must have an employment contract with a

driver, and the company in turn has a commercial contract with Uber. In food delivery

platforms, the platform typically has two separate clients per transaction – the natural

person receiving the delivery, and the restaurant preparing it. A number of food delivery

platforms, such as Foodora in Austria, also formally employ some of their platform

51

In some exceptional cases, intermediation can take place through other means, such as through phone calls

or in person.

Study to gather evidence on the working conditions of platform workers

42

workers, making the legal relationship between two parties much more clear-cut. Some

platforms (e.g. cleaning platform Helpling) allow clients to select individual platform

workers. In the case of Uber and food delivery platforms, the idea is to make platform

workers as fungible (replaceable) as possible, homogenising provided services (Schmidt,

2019).

These and other exceptions demonstrate that the wide heterogeneity of platform work

(e.g. type of services, party setting price, and level of platform worker autonomy) makes

it very difficult to settle on a universally accurate model, and to disentangle platform

work from other types of non-standard work (Fabo et al., 2017; Eurofound, 2018; Riso,

2019a). Nevertheless, the essence of platform work remains the exchange of

services and payments, with terms and conditions, and intermediation provided

by the platform.

The legal relationship between platform and platform worker, and between platform

worker and client, are exceptionally difficult to pin down. In fact, platform workers in the

same city working through the same platform may have different employment statuses

(Eurofound, 2018). Compared with other forms of work, platform work’s most

distinguishing feature is the ‘black box’ of intermediation.

Typology – distinguishing types of platform work 3.3.1

Platform work is very heterogeneous, and it is important not to overgeneralise when

considering all its forms (Eurofound, 2018). Different types of platform work have

specific and shared working condition challenges.52 Moreover, policy or legislative actions

often apply to a single platform or platform type, rather than all platforms meeting the

definition provided above.53

To help group similar platforms, the research team drew from existing typology of

platform work to identify characteristics especially important for working conditions.

These are as follows and as shown in Figure 4.

Skill requirement for tasks: higher- or lower-skilled (De Groen et al., 2016)

Location of tasks: online or on-location (De Groen et al., 2016)

Selection process: decision made by platform, platform worker, or client (Eurofound,

2018)

52

See discussion in Section 4.

53 For example, work performed on-location has been the focus of more collective action (Kilhoffer et al.,

2017).

Study to gather evidence on the working conditions of platform workers

43

Figure 4: Typology of platform work

Source: authors’ own elaboration adapted from De Groen et al (2016); AMT refers to Amazon Mechanical Turk; on-location can refer to at the client’s premises, or performed elsewhere (such as a public space, e.g. city streets).

Note: though platforms usually tend towards one of the four quadrants, the figure is a spectrum rather than a typology. This figure is a simplification of very diverse forms of work and intermediation structures.

Online platform work refers to tasks that platform workers perform from any suitable

location on their electronic devices. In the literature, it may also be referred to as

‘crowdwork’, ‘location-independent’, ‘web-based’, or ‘online freelancing’ (e.g. Schmidt,

2017). In most cases, platform workers perform this type of work in their own workplace

(e.g. home) using a computer.

On-location platform work must take place in a specific physical location. However, the

matching still takes place online as in the case of online platform work. What differs is

the final step: whether the execution of the task takes place online or requires physical

proximity. In the literature, on-location may instead be referred to as ‘work on demand

via app’, ‘location-specific’, ‘location-based’, or ‘physical’ (Schmidt, 2017). This type of

work is often at the client’s premises but may be performed elsewhere. For example,

retail intelligence tasks take place in a private store, home renovation tasks might

necessitate fabricating wood or metal pieces in a private workshop, and personal

transportation and delivery take place in public spaces such as city streets.

On-location platform work seems to be better known and more discussed and debated

than online platform work. The former is certainly more visible; these platform workers

are easier to identify and organise, which has contributed to greater social partner

involvement (Kilhoffer et al., 2017).

The distinction between lower- and higher-skilled tasks refers to the service for which

payment is exchanged, and not the skill level of the platform worker, or the overall skill

level required. For example, clickwork (via AMT or similar) may include image

recognition and data entry – small-scale, simple tasks that generally require no

specialised training. However, the organisation of such work may entail highly complex

strategies to find the best-paid tasks, such as writing and leveraging algorithms, as well

Study to gather evidence on the working conditions of platform workers

44

as subcontracting tasks to a larger network.54 Food delivery (via Deliveroo or similar),

usually requires little or no training. Still, some platform workers doing these tasks use

management and interpersonal skills to oversee others and organise and conduct

negotiations between workers and management.55 Thus, lower- and higher-skilled are

simplifications of the complexity, scale, and specialisation required for tasks, and should

not be interpreted as absolute or normative judgements.

When helpful, the research team also considers other elements that differentiate various

sorts of platform work, such as task allocation and autonomy levels for platform

workers.

Note also that a great variety of online platforms, including social networks such as

Facebook, e-commerce websites such as Amazon and Etsy, and sharing services such as

CouchSurfing, are discussed under the umbrella of the platform economy (Fabo et al.,

2017). By this study’s understanding, these are a part of the platform economy in the

broadest sense but are not examples of platform work.56

Size, prevalence, and expected evolution of platform work in the EU 3.3.2

To address the main question of this study, some discussion of the size and expected

evolution of platform work is required. The more people take part in platform work, and

the more it is expected to grow, the more significant its associated challenges and the

stronger the case for intervention.57 This section provides a brief overview of the topic.

How prevalent is platform work?

Measuring platform work is fraught with conceptual and technical challenges,

such as considering the number of platform workers, the amount and frequency of work

performed, and the revenue generated. As of autumn 2019, the most complete overview

of data on platform work seems to be Riso (2019b), which discusses and attempts to

mitigate these challenges in great detail. Conceptual challenges include definitional

complexity and a lack of standardised terminology, as briefly discussed above. As

an example of technical challenges, data on platform work is not or cannot be

gathered in most of the ways that data on other forms of work can be, for

example national labour surveys and administrative reporting.58

Because of these challenges, estimates on the size and prevalence of platform work have

used a variety of conceptualisations of platform work and methodological strategies, for

example, online, offline, or mixed surveys, administrative data, and big data. As a result,

estimates on the size and prevalence of platform work are often vastly

different, and little consensus exists (Fabo et al., 2017; Kilhoffer et al., 2017; Riso,

2019b).

Two recent studies on platform work illustrate these differences in estimations. Huws et

al. (2019), using survey data, found that 17% of the working-age population in Spain

earn money from platform work at least weekly, and 10.5% less than weekly.

Meanwhile, 20.4% in Spain were found to be seeking, but not undertaking, platform

work. This would imply that 48.1% of adults in Spain are active or one-time platform

workers or trying to become one. Brancati et al. (2019), using COLLEEM II data (also

from surveys), finds that 18% of Spanish internet users between the ages of 16 and 74

years have at some point performed platform work, 4.7% at least monthly, and 4.1%

sporadically. Clearly the latter estimate is much more modest than the former, and this

54

See Annex I: Synopsis Report of consultations.

55 See Annex I: Synopsis Report of consultations.

56 For further discussion, see Annex II: A note on what is not platform work.

57 With respect to the proportionality principle of EU action.

58 The UK has attempted to address this with specific questions in labour force surveys.

Study to gather evidence on the working conditions of platform workers

45

reflects conceptual and technical differences in the research methodologies.

Conceptually, for example, survey respondents may not fully understand what is meant

by platform worker.59 And one clear technical difficulty is that individuals responding to

online surveys (especially for a payment) are more likely to be platform workers than

the general population (De Groen et al., 2017a).

To better understand the size and prevalence of platform work in Europe, the research

team considered available data (Eurobarometer, Eurostat, COLLEEM, etc.) covering the

size of platform work in (all or many) EU Member States. The research team assembled

these datasets and performed a simple correlation test to understand how similar or

dissimilar these estimations are. Strikingly, existing estimates of the size of platform

work in Europe are almost completely uncorrelated. This provides strong evidence

that current measurements of platform work in the EU cannot be compared and

are of questionable reliability.

Nevertheless, some sense of the size of platform work in the EU is necessary given the

range of available estimates, even if we cannot be completely certain. Overall, the

research team is best persuaded by the more modest estimates for the prevalence of

platform work. One high-quality study from the Organization for Economic Co-operation

and Development (OECD) (2019) notes that the most reliable estimates suggest

employment from platform work60 ranges from 1% to 3% of total employment.

In particular, the research team considers the most recent and reliable data on platform

work in Europe to be that of the COLLEEM II project (Brancati et al., 2019). This study

looked at 16 European countries, and found that 1.4% of the population aged between

16 and 74 years are platform workers as a main job. The percentage of the population

who had ever performed platform work ranged from under 6% in Czechia, to 18% in

Spain, and averaged 11% for all surveyed countries.

As the study points out, however, this measure is far too broad. Those who simply tried

out platform work, or do so very rarely, are less relevant from a policy perspective.

Thus, excluding those who do platform work less than once a month, 9% of the adult

population are platform workers. COLLEEM II data sub-divides this group into ‘sporadic’,

‘marginal’, ‘secondary’, and ‘main’ based on a combination of how many hours worked

weekly (less than 10 hours, 10 to 19 hours, or more than 20 hours) and contribution to

income (less than 25%, 25 to 50%, over 50%). Those performing platform work as a

main job ranges from 0.6% in Finland to 2.7% in the Netherlands and the average for all

countries is 1.4%.

Table 3: Platform worker prevalence estimate as percentage of population 16-74 years

Country Sporadic

(%)

Marginal

(%)

Secondary

(%)

Main

(%)

Netherlands 2.8 3.4 5.1 2.7

Spain 4.1 4.7 6.7 2.6

Ireland 2.6 3.2 5.2 2.0

UK 2.0 3.5 5.7 1.6

Portugal 4.2 3.7 3.9 1.5

59

This should be unsurprising given the definitional complexity of platform work, and the need for surveys to

stay brief. Huws et al. (2019) also used a rather broad definition of platform work, including other means of gaining income through ‘online labour’ (such as selling goods rather than services).

60 The OECD (2019) used the term ‘gig economy platform’, but its definition of this is essentially identical to

that of this study. It also covered countries beyond the EU, but regardless, the same range applied to EU and other OECD countries.

Study to gather evidence on the working conditions of platform workers

46

Germany 3.2 3.4 4.2 1.5

Romania 2.2 3.4 3.5 1.4

Hungary 1.7 1.4 2.2 1.4

Lithuania 3.8 3.6 2.7 1.2

Croatia 3.3 2.8 3.5 1.1

Sweden 3.0 2.6 3.7 0.9

Italy 1.5 2.5 3.9 0.9

France 1.5 2.6 2.8 0.9

Slovakia 1.2 2.2 1.8 0.9

Czechia 1.5 1.6 1.9 0.9

Finland 3.1 1.4 1.8 0.6

Total 2.4 3.1 4.1 1.4 Source: COLLEEM II survey data (Brancati et al., 2019)

Note: the total is for the available 16 countries. While no comparable estimates are available for the remaining Member States, Norway and Iceland, the data on available countries is assumed to be fairly representative for the EU, as a variety of regions, industrial relations systems, and government traditions are covered.

The employment status of platform workers is a highly relevant additional consideration

for policymakers, and COLLEEM survey data gives some indication of this. The COLLEEM

survey identified 10 types of tasks commonly associated with platform work, and

respondents were asked to indicate which they perform, as well as to select one of four

options for employment status: employee, self-employed, side gig as self-employed, or

not employed. However, the employment status was not specifically asked about in

relation to platform work, but generally. Previous research has found that many platform

workers are employees in standard employment, but self-employed when conducting

platform work, so this distinction is relevant (Eurofound, 2018).

Overall, 75.7% of platform workers claim to be employees while 7.6% claim to be self-

employed, but this includes very occasional platform workers. The self-reporting of

employment status is generally to be taken with great caution because survey definitions

and the manner of posing questions can differ. Table 4 shows self-described employment

status only for those performing platform work as a main or secondary occupation (not

sporadic), showing the type of task by COLLEEM’s 10 possibilities (far left column).

Looking at these results, those doing online micro-tasks (e.g. clickwork) and on-location

services (e.g. home repair or cleaning) were most likely to identify as self-employed.

Those providing online professional services (e.g. accounting, legal, or project

management) and transportation and delivery services, were most likely to identify as

employees. On-location services are most strongly associated with having a side gig as

self-employed, while platform workers providing online clerical and data-entry tasks, and

transportation and delivery services, are most likely to claim being not employed.

Table 4: Self-reported employment status by platform work type

Type of platform

work tasks

Employees

(%)

Self- employed (%)

Side gig as self-

employed (%)

Not employed (%)

Observ- ations (#)

Online clerical and data-entry tasks 37.5 13.5 37.2 11.8 862

Online creative and multimedia work 38.7 13.6 40.5 7.3 640

Online professional services 42.3 9.9 42.2 5.7 571

Study to gather evidence on the working conditions of platform workers

47

Online sales and marketing support work 38.4 10.2 41.6 9.8 559

Online writing and

translation work 37.8 10.9 40.9 10.4 546

Online micro tasks 38 15.5 36 10.5 465

Online software development and technology work 39.5 11.8 41.4 7.9 385

Interactive services 40.2 12 40.1 7.7 327

Transportation and delivery services 39.7 9.6 39.5 11.2 319

On-location services 31.6 14.5 45.6 8.3 271

Source: COLLEEM (Pesole et al., 2018)

Note: this Table only shows self-described employment status of platform workers, excluding ‘very occasional’ platform workers.

Furthermore, COLLEEM finds that most platform workers do multiple forms of platform

work and use more than one platform. The most common tasks platform workers

perform are online clerical and data-entry tasks, while the largest proportion of platform

workers (including sporadic activity) provide online professional services. This contrasts

with some earlier findings. For example, De Groen and Maseli (2016) estimated that in

2015, two-thirds of EU platform workers were Uber drivers. Possible explanations are

that the methodologies are not comparable, and that online forms of platform work have

grown in relative prevalence.

Online software development and technology work, and transportation and delivery

services, are much more likely to be performed by men, while translation and on-

location services are more likely to be performed by women. The prevalence of most

services does not vary significantly between countries, though some differentiation is

notable. Slovakia and Croatia are associated with platform work requiring a low or

medium level of education, while the Netherlands is associated with tasks requiring

specialised training (e.g. software development or interactive services such as teaching

or consultation).

Overall, COLLEEM II data suggests the following about the prevalence of

platform work in Europe:

1. 11% of adults have gained income from platform work at some time;

2. 3.1% of adults do platform work at least 10 to 19 hours/week or receive

between 25% and 50% of their income from platform work;

3. 1.4% of adults do platform work at least 20 hours/week or receive at least

50% of their income from platform work;

4. for individual Member States, this figure ranges from 0.9% to 2.7%;

5. the most common forms of platform work are online;

6. the prevalence of different types of platform work across different countries is

mostly the same;

7. those performing platform work as a primary or secondary occupation are less

likely to be employees than those performing platform work sporadically;

8. employment status of platform workers remains very ambivalent from existing

data, and/or to the platform workers themselves.

What is the trajectory of platform work?

Even less data exist on the outlook for platform work, namely whether it will grow or

contract. Aloisi notes, ‘[…] many online platforms are still in their business “infancy”, and

Study to gather evidence on the working conditions of platform workers

48

experts genuinely do not know how they will develop.’ (2018). Overall it remains

to be seen whether certain forms of platform work will continue to grow, stagnate, or

disappear as technology advances.

However, some quantitative research points to continued growth. Huws et al. (2019)

found that between 2016 and 2019, platform work effectively doubled in the UK,

based on proportion of adults carrying out platform work at least weekly.61 For the five

largest English language online labour platforms, representing at least 70% of the

market by traffic, the Online Labour Index shows roughly a 30% increase in tasks posted

between July 2016 and July 2019 (Kässi and Lehdonvirta, 2018). Based on a PwC

analysis of collaborative economy62 platforms, even starker growth (by revenue) is

visible from 2013 to 2016, as shown in Figure 5.

Figure 5: Revenues and transaction values facilitated by collaborative economy platforms

in Europe

Source: Vaughan and Daverio (2016: p. 14).

Most experts consulted over the course of this study agree that platform work is here to

stay in one form or another, and will probably grow.63 One trend with more clear growth

potential is algorithmic management, contained in the ‘black box of intermediation’,

which platforms have continued to develop and refine. Algorithmic management seems

to be spreading beyond the platform economy, which means that even if platform work

(as it currently exists) does not continue to grow, the lessons learned may continue to

be relevant.

Most experts consulted also indicated that online forms of platform work have much

more room for growth. For example, demand for micro-tasks boomed when companies

started using human input to train artificial intelligence (AI) (Schmidt, 2019). Car

companies are increasingly relying on online platform work to train their software and

creatively solve problems. This trend may not continue indefinitely, however. The need

for humans performing simple and repetitive micro-tasks may become increasingly

obsolete as AI, and machine learning in particular, continues to improve.

Some experts perceive that personal transportation and food delivery services are

beginning to plateau. Many cities seem to have an oversupply of these platform workers

61

The UK is the only country for which trend data are available. Note that Huws et al. (2016) would also

include individuals selling products on their own websites and selling self-made products. Selling goods is outside the scope of this study’s understanding of platform work.

62 The term ‘collaborative economy’ is still sometimes used in place of platform economy, even by EU

institutions. However, the terms ‘collaborative’ and ‘sharing’ have generally fallen out of favour in this context due to their normative connotations. See the European Parliament Opinion 2017/2003(INI), p. 4 at http://www.europarl.europa.eu/doceo/document/EMPL-AD-592420_EN.pdf?redirect.

63 For more detail, see Annex I: Synopsis Report of consultations.

Study to gather evidence on the working conditions of platform workers

49

relative to demand, as illustrated by protests over lack of work and reduced pay per

task. However, other forms of on-location work probably have more room to grow. For

example, caretaking and cleaning may grow in prominence, especially as the population

ages.

Another question concerns the economic sustainability of certain types of platform work.

After Uber’s highly-anticipated initial public offering in May 2019, many publications

question whether it will ever become profitable (Horan, 2017; The Economist, 2019).

One day in the indeterminate future, autonomous vehicles may well displace (platform)

work in the personal transport sector. Individual platform companies will certainly rise,

fall, merge and change, but charting the future course of platform work remains mostly

speculation.

Overall, available data and expert opinion suggest the following outlook for

platform work in Europe:

 platform workers are growing in number and will continue to do so for the

foreseeable future;

o demand for many services offered via platform work (e.g. training AI and

certain on-location services) is growing;

o certain services offered via platform work may not have as much growth

potential or may be more vulnerable to technological changes (e.g.

automation);

 certain characteristics of platform work are becoming more common in the

economy at large (e.g. non-standard employment, algorithmic management,

rating systems for people).

Study to gather evidence on the working conditions of platform workers

50

4. CHALLENGES FOR PLATFORM WORKERS

This section maps the challenges related to the working and employment

conditions of platform work. It builds on an extensive review of the academic and grey

literature and on the fieldwork conducted for the study (stakeholder interviews, focus

groups and a survey completed by national experts). National experts were asked to

indicate what challenges emerge in their country, specifying how prevalent these

challenges are, how they are expected to develop over time, and what types of platform

work are affected. Caution is needed when interpreting the responses of the national

experts on the expected evolution of the challenges because, in the absence of other

sources, these are mainly based on their personal views. National experts also

accounted for national developments, which are dealt with in section 5. Other

stakeholders were consulted to provide insights from an EU-level perspective or to

represent the views of an interest group. The inputs provided by the experts and

stakeholders are confronted with the literature, which serves as the baseline for this

Section. Relevant results from the fieldwork are highlighted where appropriate.

4.1 Framework

To perform this mapping of challenges in a structured yet comprehensive way, we use a

job quality framework based on the Work, Employment and Social Relations

(WES) model put forward by Lamberts et al. (2016). This gathers a wide range of

indicators capturing working and employment conditions and helps structure the

discussion on the working and employment conditions facing platform workers.

Job quality is a multidisciplinary, multidimensional concept that is generally understood

as ‘the extent to which a job has work and employment-related factors that foster

beneficial outcomes for the employee, particularly psychological well-being, physical

well-being and positive attitudes such as job satisfaction’ (Holman, 2013). Because of

the broadness of the concept, there is no single definition of job quality. Instead,

multiple indicators have been identified that contribute to its measurement. The job

quality concept used by Eurofound considers seven indices derived from the European

Working Conditions Survey: physical environment, social environment, work intensity,

working time quality, skills and discretion, prospects, and earnings (Eurofound, 2017b).

The OECD’s framework considers the worker’s situation as well as the labour market. It

includes three dimensions: earnings quality, working environment quality, and labour

market security (e.g. risk of job loss, benefits generosity and coverage) (Cazes et al.,

2015). The International Labour Organization (ILO), the United Nations Economic

Commission for Europe (UNECE) and Eurostat jointly developed a quality of employment

framework with seven dimensions: safety and ethics, income and benefits, working time

and work-life balance, security and social protection, social dialogue, skills development

and training, motivation and employment-related relationships. Finally, the European

Trade Union Institute (ETUI) created the widely used ETUI Job Quality Index using six

job features to compile one job quality score: wages, non-standard forms of work,

working time, working conditions and job security, skill and career development, and

collective interest representation (Leschke and Finn, 2016; Leschke et al., 2012).

The WES model organises the multitude of possible indicators describing the job quality

of platform work. Twenty-one indicators from the selection suggested by Eurofound

(2017b) are categorised into three broad dimensions that encompass all relevant,

objective job features needed to measure job quality at the worker level (Lamberts et

al., 2016). These three dimensions capture the bulk of the job quality literature: work,

employment, and social relations (Lamberts et al., 2016). The work dimension reflects

the organisation of specific tasks and the environment where a worker performs labour.

The employment dimension relates to those job characteristics that are mostly fixed

within formal employment agreements and interfere directly with workers’ private lives,

for example, wages, working time or training. The social relations dimension links

social relations and interactions, social dialogue and representation at work through

Study to gather evidence on the working conditions of platform workers

51

formal and informal channels. The WES model allows us to consider tasks as well as

jobs.

Table 5 shows the WES model and its main indicators and demonstrates how the

conceptualisation of different definitions overlap with those in other commonly used

frameworks.

Table 5: Comparison of the WES model with other job quality frameworks and their

definitions and indicators

WES model

(Lamberts et al., 2016)

Work organisation

Task autonomy

Autonomous teamwork

Task complexity

Speed pressure

Emotional demands Repetitive tasks

Risks (ambient, bio- chemical, musculoskeletal)

Employment conditions

Permanent contract

Full-time work

Wage, additional fees

Atypical working hours

Working time flexibility

Planning autonomy

Career opportunities

Training

Social relations

Participation

Representation

Supportive

management

Social support

Adverse social

behaviour

Eurofound

(2017b)

Physical environment

Work intensity

Skills and discretion

Working time quality

Prospects

Earnings

Social environment

Holman

(2013)

Work organisation Wages and payment

systems

Security and flexibility

Skills and development

Engagement

Green

(2006)

Intrinsic job quality Earnings

Prospects

Working time quality

-

ETUI Job Quality Index

Working conditions

Job security

Wages

Non-standard forms of work

Working time

Skill and career development

Collective interest

representation

OECD Job Quality Framework

Working environment

quality (level of job strain)

Earnings quality (level

and distribution across workforce)

-

ILO, UNECE and

Eurostat Quality of Employment Framework

- Safety and ethics

Income and benefits

Working time and work- life balance

Skills development and training

Security of

employment and

social protection

Social dialogue

Employment-

related

relationships and

work motivation

Source: authors’ own elaboration.

Study to gather evidence on the working conditions of platform workers

52

Box 1: The Fairwork Framework: a job quality framework tailored to platform work

Graham et al. (2019) have developed a job quality framework specific to platform work called the

Fairwork Framework. This framework helps to assess decent work standards in platform work by means of a scoring system with basis and advanced points, leading to a total score on ten per platform. Academics, policymakers and trade unions jointly defined five principles for fair platform work:

 Fair pay: the platform pays the local minimum wage (basis); pay is high enough to also cover costs (advanced)

 Fair conditions: task-specific risks are mitigated (basis); the platform is actively improving working conditions and workers’ health and safety (advanced)

 Fair contracts: the platform’s terms and conditions are transparent, concise and available

for workers (basis); and preferably also genuinely reflect the nature of the employment

relationship(advanced)  Fair management: the platform has a documented process for decisions affecting its

workers through which workers can be heard, consulted and informed (basis); the platform additionally provides evidence of equity and/or informed consent for data collection (advanced)

 Fair representation: freedom of association and worker voice mechanisms (basis); which is

completed with a recognised collective body that can undertake collective representation and bargaining (advanced)

The five principles of the Fairwork Framework are covered by the WES model: fair conditions are included in the work dimension; fair pay and fair contracts are part of the employment dimension; fair management and fair representation are taken up in the social relations dimension.

Besides these broad frameworks, academic studies have put forward job quality models

that focus on a subset of indicators and define the relationships and interactions between

them. The job demands – job control model developed by Karasek (1979) distinguishes

two types of job quality determinants, and focuses on the interaction between the

‘demands’ a worker faces and the available mechanisms to cope with them, or ‘controls’.

An example of a ‘job demand’ is working with tight deadlines; an example of a ‘job

control’ is having a high level of autonomy. The Karasek (1979) model only accounts for

the ‘work’ dimension. The extended job demands-control-support model and the

validated job demands-resources model were later introduced to account for a wider

range of indicators. These revised models also include indicators for employment

conditions and social relations. Other models that have been developed in the discipline

of psychology include the effort-reward imbalance model and the high-performance

works systems. Sociologists have introduced the ‘socio-technique’ that focuses on the

organisational level and suggests self-organising teams as a way to foster job quality. In

the medical literature, the focus is on occupational safety and health. Other theories,

such as the dual labour market theory, consider larger converging or diverging

tendencies of groups of workers, or take a perspective of management (human

relations) or consider it from a collective bargaining perspective (industrial democracy).

Table 6: Overview of job quality models developed in different fields

Work Employment Social relations

Job demands-control (JD-C) X

Job demands-control-support (JD-C-S) X X

Job demands-resources (JD-R) X X X

Effort-reward imbalance X (X)

High Performance Works Systems

(HPWS)

X X (X)

Study to gather evidence on the working conditions of platform workers

53

Socio-technique X (X)

Occupational safety and health (OSH) (X)

Dual labour market X

Human relations X X

Industrial democracy (X) (X) X

Source: authors’ own elaboration.

Because the WES model captures most of the aspects put forward in traditional job

quality models, as well as in frameworks developed with specific focus on platform work,

this model seems the best fit to provide a complete overview of all the relevant aspects

of work in the platform economy. It offers an elaborated overview of objective indicators

composing job quality, without predefining the relations between indicators. This allows

us to address indicators individually, while combining and adding other aspects specific

to platform work. Table 7 shows the adjusted WES model that guides our research. The

use of technology and algorithms is covered under each dimension.

Table 7: Adjusted WES model

Work dimension Employment dimension

Social relations dimension

Other indicators

Autonomy in the allocation of tasks

Employment status Representation Undeclared work

Autonomy in work

organisation

Determination of the

employer

Participation in

decision-making

Cross-border work

Surveillance, direction and performance

appraisal

Contracts Supportive management and

social support

Data protection

Task complexity Social protection Adverse behaviour and equal treatment

Work intensity and speed pressure

Earnings

Emotional demands Working time

Physical environment Career opportunities

Training and skills

Source: authors’ own elaboration, based on Lamberts et al. (2016).

Where possible, the study highlights whether a challenge is specific to platform work, or

to the use of algorithms, or whether it is part of a larger issue. In the WES model,

challenges specific to platform work are likely to be found under the work dimension and

to a lesser extent the social relations dimension, while challenges identified under the

employment dimension are typically broader and go beyond platform work. However, it

appears that the employment dimension has received most attention in the debate and

literature, precisely because it captures those aspects that interfere directly with

workers’ private lives. Given that platform work is mainly performed as a secondary

activity that comes on top of another job or activity (see Pesole et al., 2018), it likely

has a significant impact on work-life balance. This issue is also a major point of

discussion with other non-standard forms of work (Eurofound, 2017a). In addition,

because platform workers generally use their own equipment to perform the tasks at

hand (for example, riders use their own bicycle and smartphone), platform work is

almost automatically set in the private sphere.

Study to gather evidence on the working conditions of platform workers

54

Note, however, that any examination of the challenges related to platform work is

complicated by the high level of heterogeneity in the types of platform work, platforms,

and platform workers. In addition, because the WES model considers a wide variety of

job quality indicators across these types of platform work, it is very difficult to formulate

clear-cut conclusions for some of the challenges described. For that reason, the analysis

below is highly nuanced (although these nuances are necessarily abstracted in the

summary tables at the end of each part). Three points are considered in these summary

tables: the importance of the challenge (high, medium, low, or none), the

specificity of the challenge (challenge specific to platform work, common for non-

standard work, or found in the wider labour market), and the affected types of

platform work (all, on-location versus online, low- versus high-skilled, client-,

platform- or worker-determined). The section for each indicator gives examples of which

platform work types are affected. Similarly, comparisons between platform work and

other forms of non-standard work are drawn in the text.

How the importance of the challenge is categorised is much more complicated: it

accounts for the literature and fieldwork (notably the expert questionnaires and

interviews), the legal and socio-economic interpretation64 of a challenge, the extent to

which an issue is discussed and by which actors, the extent to which a challenge is

specific to platform work or not, the extent to which a challenge affects all or some types

of platform work, and so on. While such details and nuances are described in the

analysis, the summary tables combine all this information into a single ‘score’. To label a

challenge ‘high’, we considered whether it was identified in the literature and the

fieldwork as a ‘major’ challenge that moreover is specific to platform work and affects all

or most platform work types. The only exceptions are employment status and

representation, which, while highly relevant, are not specific to platform work.

Challenges that emerge from the literature and fieldwork as ‘major’ but that are not

specific to platform work, or only affect a few platform work types, are labelled

‘medium’. Classified as ‘low’ are those ‘minor’ challenges that arise among other non-

standard forms of work or in the labour market more generally, and that have been

raised in the literature or by the experts consulted for this study. These challenges are

typically especially relevant for one or a few types of platform work. In addition, some of

these challenges do arise in real life but are not seen as problematic by platform

workers, social partners, policymakers or other actors, atypical working times, for

example. Challenges with the label ‘none’ refer to those issue that may arise, though

only in a very few, specific cases affecting some individuals. For example, having to deal

with difficult clients is more likely to be a challenge with on-location work, but it does not

mean all platform workers engaged in this type of work experience emotional demands.

For some individuals, however, this may be the case. This issue has not been raised by

the consulted experts and there is little evidence beyond anecdotal reports available in

the literature. Although the available evidence is discussed, it is deemed insufficient to

justify a different categorisation.

The following sections describe the work, employment and social relations dimensions of

platform work and their related challenges. These analyses are primarily based on a

literature review and completed with data gathered through fieldwork. The identified

challenges have also been validated in an expert workshop.

4.2 The work dimension

The work dimension of the WES model combines the elements that are usually

distinguished by other job quality models: job content, working conditions and work

organisation. Combining these aspects into a single dimension is useful because these

64

This is an important point: when comparing the questionnaires for each country, for example, there were

several instance where socio-economic experts flagged as issue as ‘major’ whereas the legal experts did not consider it a challenge at all. This is taken into account in the analysis. This level of detail, however, is difficult to provide in the summary tables.

Study to gather evidence on the working conditions of platform workers

55

characteristics are often interrelated. The physical and psychological risks (traditionally

seen as working conditions) a worker encounters are not independent of the performed

task (job content) and the work organisation.

This section presents the platform work state of affairs on task allocation, autonomy in

work organisation, surveillance, direction and performance appraisal, task complexity,

work intensity, speed pressure, emotional demands and physical environment. The use

of technology, apps and algorithms has an impact on all elements of the work

dimension. Within the WES model, the work dimension is the most affected dimension in

this regard.

Autonomy in the allocation of tasks 4.2.1

The allocation of tasks, or the division of work, among workers, is one of the most

discussed issues related to working conditions. Along with the idea of ‘being your own

boss’, having the autonomy and flexibility to determine when to work, what tasks to do

and how to execute them has been identified as a key factor that motivates workers to

engage in platform work (Ivanova et al., 2018; Pesole et al., 2018). Freedom in the

selection and execution of tasks is very important to platform workers (Berg, 2016; De

Groen and Maselli, 2016). Having autonomy in the allocation of tasks is shown to be

beneficial for job quality (Karasek, 1979).

Compared to other forms of work, platform workers have a rather high level of

autonomy in the allocation of tasks (Eurofound, 2018; Pesole et al., 2018). However,

the allocation strategy used to divide tasks among the platform workers depends on the

type of work and the organisation of the platform (ibid.). Three task allocation strategies

can be distinguished, depending on who makes the selection (Eurofound, 2018): (i)

tasks are allocated by the platform (i.e. the platform matches client and worker -

‘platform-determined’ work;), (ii) tasks are allocated by the client (i.e. the client

selects a worker - ‘client-determined’ work), and (iii) tasks are allocated by the

workers (i.e. the worker chooses their tasks - ‘worker-determined’ work). The allocation

of tasks is linked to the matching process, which is usually based on an offer or a

contest, though the latter is much less prevalent (Eurofound, 2018). In addition,

platforms can move between allocation strategies over the course of their development.

The change can be to the advantage or detriment of worker autonomy. Cleaning

platform Helpling, for example, allowed its workers to have a fixed group of clients;

hence these platform workers have more control over task allocation.

In principle, all three allocation strategies can be found across all types of platform work.

In practice, however, platform-determined work seems to be more prevalent in low-

skilled on-location or online work, such as food delivery and click-work. Food delivery

riders and ‘taxi’ drivers typically receive task offers allocated by an algorithm, which they

can accept or decline. Although the option to decline tasks suggests relatively high

worker autonomy, it is important to take the precise platform mechanics into account.

For example, platforms differ as to which information is made available about the actual

task, and when. On most transportation platforms, riders only receive information about

the location of the client after accepting a task, limiting the capacity of the worker to

make an informed decision about the task acceptance (information asymmetry) (Lee et

al., 2015; Rosenblat and Stark, 2015; Rosenblat and Stark, 2016). In addition, the

platform worker’s autonomy in task allocation is harmed by systems imposing penalties

for declining tasks or stipulating a minimum number of obliged acceptances. Platforms

such as Uber and Lyft suspend drivers from the system if they cancel (possibly

unprofitable) trips too often (Rosenblat and Stark, 2016). The algorithm, ratings and

rewards systems thus undermine the formal autonomy of the platform worker regarding

the decision to work or not.

In client-determined work, the client selects a platform worker to do the task. This

can be based on a contest or competition between the workers, or a review of the offers

or profiles of the platform workers. In the latter case, platform workers are strongly

Study to gather evidence on the working conditions of platform workers

56

dependent on the ratings or reviews received by previous clients. Platform workers

without any ratings typically find it difficult to get work. Platforms are aware of this issue

and some have developed strategies to assist their workers. Examples include

background checks, a rating based on the completeness of their profile, a verification

that the worker is a real person who has had an intake interview with the platform, and

links to the worker’s profile on social media. These methods are used to create trust

(Schreieck et al., 2018). There are also differences between platforms as to whether

they impose penalties on platform workers who deny a task. Platforms such as ListMinut

or Hilfr allow workers a lot of autonomy to decide what tasks they accept. They

encourage platform workers to accept the allocated task, but do not penalise them for

declining (interviews with platform representatives). In other cases, platform workers

may feel pressured to accept a task to avoid a penalty or bad review. Finally, even in

client-determined platform work, platforms have significant power to control market

entry and to guide their clients’ choices.

Worker-determined platform work is associated with the highest allocation

autonomy. Here, clients post a task on the platform and the workers can select the ones

they prefer. In such cases, however, the platform workers’ profile and online reputation,

as well as the availability of tasks, also determine which tasks a worker can actually

access (Martin et al., 2016). Factors such as income insecurity may also lead platform

workers to accept any task posted, thereby restricting their autonomy in practice.

Although allocation autonomy in platform work can be higher than in regular jobs, there

are several explanations as to why it could be more limited in practice.

Autonomy in work organisation 4.2.2

Work organisation is defined as the organisation of the work to be performed with regard

to the order, method and tempo of the tasks (Szekér et al., 2017).65 This definition

largely coincides with the concept of discretion or ‘decision latitude’ put forward by

Eurofound (2017d). The concept of autonomy in work organisation correlates strongly

with the legal subordination ‘test’, which is crucial in determining the employment status

of a platform worker. A higher level of autonomy at the task level is associated with

higher levels of job quality because this can compensate for a high workload (job

demands) (Karasek, 1979). According to the job demands – job control model, the

amount of control and discretion a worker has moderates the effect of job demands on

psychological pressure. As a result, the psychosocial well-being of employees can be

improved by offering them a higher level of control and discretion without changing job

demands (Clays et al., 2007; Karasek Jr, 1979).

Autonomy regarding the work organisation, next to the flexibility to choose specific tasks

and when to work, are among the main motivations for workers to engage in the

platform economy (Yordanova, 2015; Berg, 2016; De Groen and Maselli, 2016;

Temper, 2017). While employees in regular companies usually receive tasks, instructions

and feedback from a supervisor, and may not have any contact with the company’s

clients, platform workers receive tasks via the platform, but instructions and feedback

stem directly from the client. This may cause confusion and reduce workers’ autonomy.

Once a task is allocated, platform workers’ autonomy regarding the work organisation

varies according to the type of work, and the design and organisation of the

platform. In relation to the type of work, the highest autonomy in work organisation

comes with online platform work based on differentiated skill use, similar to employees

in such jobs, examples of which include graphic designers and IT specialists. (Leimeister

et al., 2016; Pesole et al., 2018). Leimeister et al. (2016) describe how tasks requiring

specialised skills performed online feature more intensive interactions between the

platform worker and client. Whether an IT specialist works as an employee in a company

or uses a platform, the process leading from the task description to the final delivery is

65

Autonomy regarding working hours is discussed in a separate section under the employment dimension.

Study to gather evidence on the working conditions of platform workers

57

similar in both cases. To be able to perform the task, the worker needs to be entrusted

with a reasonable degree of task autonomy. A similar logic applies to on-location tasks

using differentiated skills, which tend to be complex and require the worker to have

some degree of autonomy in determining how to approach the task. The opposite applies

to low-skilled online and on-location work. Platform workers engaged in online

micro-tasks, such as household services working at the client’s location, and

transportation services involving passenger transportation or food delivery, have a lower

degree of autonomy in the work organisation (Eurofound, 2018; Pesole et al., 2018).

Sundararajan (2016) explains that organisational autonomy is lower in these cases

because the structure of the working relationships is more hierarchical. This implies that

instructions do not allow for changes in the completion, and that workers are closely

supervised.

Note, however, that there can be large differences between platforms intermediating the

same type of work, depending on how platforms are organised. For example, are tasks

fully predefined by the platform or can clients make their own proposition?

Nothwithstanding this, it does particularly appear that those platforms intermediating

work based on undifferentiated skills, regardless of whether they are online or on-

location, tend towards largely predefined tasks. This also reduces the platform worker’s

autonomy in determining the method, order and speed.

Surveillance, direction and performance appraisal 4.2.3

During the execution of the work, the platform and/or client monitors the platform

worker. The level of control the platform or client can impose has a strong impact on a

worker’s autonomy and well-being. In platform work, surveillance, direction and

performance appraisal techniques greatly rely on the use of technology, apps and

algorithms. Surveillance implies the monitoring of platform workers, allowing the

platform to interfere in the work process by providing direction and guidelines.

Performance appraisal is understood as an evaluation of the provided work, feedback

and ratings. The use of technology in such processes is a particular feature of

platform work, though in ICT-based mobile work, technology-based monitoring

instruments may be used to compensate for the employer’s loss of control (Eurofound,

2018b). Given that very little is known about how algorithms work (the ‘black box’ in the

conceptualisation), these processes are poorly understood, but nevertheless have a huge

impact on platform workers. For that reason, this challenge is deemed of high

importance.

The surveillance, direction and performance appraisal that take place in regular

employment relationships between an employer and employee are more scattered in

platform work. The triangular relationship between platform worker, platform and client,

which in some cases can include additional parties, implies that others may exercise

surveillance and direction vis-à-vis the worker. Platform workers might be monitored,

receive direction, evaluations or even penalties from the platform (or its algorithm), the

client or both (Eurofound, 2018; Rosenblat and Stark, 2015).

The literature flags up a number of challenges from identifying at least five managerial

control systems (Waters and Woodcock, 2017): surveillance; automatic evaluations;

automated decisions; automated messaging systems; and digital choice architecture.

These mechanisms are often related. Surveillance of the labour process, for example by

tracking, is often done by platforms intermediating transport services. It provides

platforms with detailed information on the location of the driver, their speed and general

work pace. Rosenblat and Stark (2016), for example, explain how Uber monitors the

trips of its workers based on GPS data (a geotracking system). Ivanova et al. (2018) find

similar cases for Deliveroo and Foodora. Tracking is used by platforms intermediating

online work as well. AMT allows clients to specify that platform workers must be

‘Mechanical Turk Masters’, which implies that platform workers allow online monitoring

of their work (see Figure 6). Some platform workers are in favour of this type of

tracking, as it provides them with proof in case of conflicts (Wood et al., 2019). The

Study to gather evidence on the working conditions of platform workers

58

existence and nature of such tracking mechanisms is indicative of the amount of control

a platform can exercise over its platform workers. It has, in fact, been used in court

cases to assess whether there is a link of subordination between a platform worker and a

platform (see infra).

Figure 6: Example of AMT 'Mechanical Turk Masters'

Source: screenshot provided by German expert consulted during fieldwork. Surveillance creates large amounts of data that platforms can use for automatic

evaluations of platform workers (Mohlmann and Zalmanson, 2017). If drivers or riders

fail to complete a delivery during their shift, they typically face penalties, such as the

strike system implemented by Uber. These evaluations might be at the basis of

automated decisions, such as task allocation systems, and data-driven automated

messaging systems influencing workers’ behaviour through reminders and pop-ups

(Gillespie, 2018; Van Doorn, 2017). Choices and behaviour of workers and clients are

led by the general design of the platform, creating a digital choice architecture of which,

for example, unwanted alternatives are filtered out (Sunstein, 2015). Ivanova et al.

(2018) present examples of direction mechanisms used by Foodora and Deliveroo in

Germany.

First, these platforms exercise control via automated messages based on what is

estimated normal behaviour for a rider (Mohlmann and Zalmanson, 2017). Using GPS

data and data received through the app when the rider indicates what step of the

delivery they are on (i.e. accept task, arrived at restaurant, collected meal, arrived at

client), the platform generates personalised estimates as to how long it will take to

complete a task. Riders receive an automated message such as ‘contact the dispatcher’

or ‘log in again’ whenever irregularities are observed. Although riders are closely

monitored, these platform workers ‘perceive the automated notifications as less

controlling than a human supervisor’ (Ivanova et al., 2018: p. 13).

Second, food delivery riders are incentivised through financial rewards to guarantee

that there are sufficient riders during busy shifts or to encourage them to complete

orders more quickly.

Third, deriving data from their personal statistics, the platforms adapt the selection of

possible shifts riders are able to access when choosing their working times. There is

high internal competition for popular shifts, so underperforming platform workers have

fewer chances to book shifts on their preferred hours. The financial rewards and the way

platforms shape workers’ weekly schedules are ways to nudge workers towards

behaviour that is considered desirable. Such mechanisms undermine the autonomy and

flexibility platforms claim to offer, especially for platform workers who depend on the

income gained through platform work (Ivanova et al., 2018; Lehdonvirta, 2018).

Whether a platform worker can refuse shifts or disregard previously agreed to shifts

Study to gather evidence on the working conditions of platform workers

59

without facing penalties is an indication of the degree of subordination and managerial

control.

Fourth, the platforms’ control systems create information asymmetries between the

platform worker, client and platform, limiting the actual decision latitude of these

workers (see supra). The lack of transparency regarding the algorithms used for

allocation and monitoring raises questions about the fairness of these mechanisms

(Ivanova et al., 2018; Lee et al., 2015).

Traditionally, performance appraisal forms part of the human resources management

activities organised by the employer in regular employment relationships. Performance

appraisal serves to justify workers’ rewards, as well as different levels of rewards within

and between companies, and to define required developments for a worker to perform in

the future (Dransfield, 2000). An appraisal usually starts with a clarification on the

evaluation criteria considered to accurately measure past performances. In the context

of platform work, however, rating systems are typically used to evaluate workers’

performance. Platform workers receive feedback from clients, who can either rate the

entire transaction or part(s) of it (Van Doorn, 2017). Ratings can take different formats,

depending on the platform (e.g. an option to provide text or only a score). These ratings

or evaluations feed into a platform worker’s online profile and determine their

reputation. Rating systems are often one-sided: clients can rate platform workers but

not vice versa. In the case of Uber, however, drivers and riders can rate each other

(two-sided reviews).

Rating, review or reputation systems are poorly understood and often intransparent. But

they have a major impact on platform workers because they affect their future work

opportunities on the platform (Aloisi, 2015; Lee et al., 2015; Rosenblat and Stark,

2015; De Groen et al., 2018a; Wood et al., 2019). And they are common in platform

work. Wood et al. (2019) describe rating systems as the most effective tool for

algorithmic control. On some platforms only the profiles of platform workers with the

highest scores are presented, making it difficult for those with low or no ratings to find

work. On others, for example Fiverr, all profiles are shown. Many platforms, including

Hilfr and Helpling, allow clients to search for platform workers based on ratings. Other

platforms, such as Uber, prohibit workers with a rating below a certain threshold to

work. These dynamics put significant pressure on platform workers to get good ratings

(Lee et al., 2015; Rosenblat and Stark, 2015; De Stefano, 2017; Wood et al., 2019),

which causes stress and raises the emotional demands of the work (Aloisi, 2015).

Platform workers have indicated that ratings are sometimes arbitrary, unfair or biased,

yet there are few options for recourse (ibid.).

Ratings are linked to a specific platform and workers cannot usually take their online

reputation with them when transferring between platforms. However, research on

transferable online reputation systems is growing, and arguments both for and

against the transferability of ratings have been put forward. Arguments against

transferability include those that platform workers should have the right to be forgotten

or that they may struggle to get rid of ratings they consider unfair or incorrect, and that

some types of platform work have little compatibility – for instance, what value does an

excellent profile as an Uber driver signify for a platform worker who wants to work as a

cleaner? (Lenaerts, 2018). On the other hand, the transferability of ratings may help

overcome the monopoly power of platforms, make it easier to work with multiple

platforms, and gives workers more control (ibid.). To this end, some new initiatives

support ratings transferability. Deemly, for example, is a platform that allows platform

workers to create an online profile combining reviews from different platforms that can

be shared via a digital link. Such initiatives could be further explored.

Task complexity 4.2.4

Task complexity is defined as the degree to which a worker is required to meet high

quality standards, solve problems independently, and learn new things in order to fulfil a

Study to gather evidence on the working conditions of platform workers

60

task. In the job demands – job control model of Karasek (1979), task complexity is

understood as a job demand. How task complexity contributes to job quality depends on

the interactions with other job characteristics, such as autonomy. More job demands

lead to more stressful situations and negative job outcomes, but high levels of job

demands combined with an increase in job controls, such as autonomy, can lift these

jobs into active jobs with more positive outcomes (Karasek, 1979).

In the context of platform work, task complexity varies according to the nature of the

work: is a task a routine task, a complex task, or a creative task (Rouse, 2010)?

Note that all three of these types are found online and offline. Routine tasks can be

found in transportation and household services, for example food delivery, and in low-

skilled online work, for example clickwork where a worker tags images. Routine tasks

require limited prior knowledge (Alkhatib et al., 2017). Pesole et al. (2018) report that

platform workers engaged in routine tasks are the least likely to learn new things (Pesole

et al., 2018). However, online micro-task workers report learning new things more often

than workers using an undifferentiated skill set in on-location platform work (Pesole et

al., 2018). The complexity of routine tasks is low.

Complex tasks, for example online content creation, reviewing and testing products or

applications, and participating in user surveys, require more effort. Creative tasks are

considered to have the highest complexity level, as the work is profession-based and

implies that workers have differentiated skills or are specialised in the area, and that

such tasks require workers to learn new things regularly (Pesole et al., 2018). Both

types of tasks include professional online or on-location work (e.g. IT specialists). These

platform workers find themselves significantly more often in stressful situations related

to their tasks (Pesole et al., 2018).

As the complexity of the task itself does not appear to be much affected by the use of

technology or algorithms, task complexity is not discussed as a challenge in the

literature on platform work, nor has it been identified as such by any of the experts and

stakeholders consulted for this study.

Work intensity and speed pressure 4.2.5

Work intensity is an indicator combining the amount of work and the work pace. This

indicator refers to ‘the effort and strain associated with carrying out the work’

(Eurofound, 2018). High levels of work intensity are associated with negative job quality

outcomes, such as burnout, stress and sleeping problems (Boekhorst et al., 2017;

Cottini and Lucifora, 2010; Maslach et al., 2009). Platform workers engaged in on-

location and online work based on undifferentiated skills report higher levels of work

intensity due to the competition between workers and their replaceability. Examples

include food delivery and passenger transportation, which are low-skilled tasks. A

platform worker participating in the French focus group emphasised how he felt that he

could easily be replaced.

Speed pressure is a related concept that gives an indication of the degree to which a

worker has to cope with tight deadlines, automatic speed, direct control or quantitative

production norms, whether they are dependent on external factors or colleagues to be

able to finish a task, and whether sufficient time is foreseen to complete a task (Szekér

et al., 2017). Speed pressure, as well as repetitive work, increases the probability of

suffering from mental health issues, such as stress, anxiety and sleeping problems, and

from physical health issues such as trembling hands (Cottini and Lucifora, 2010; EU-

OSHA, 2017). In platform work, speed pressure is determined by two factors: deadlines

posed by the client or platform, and having to combine multiple tasks and working on

different platforms at the same time and sometimes in combination with regular

employment (EU-OSHA, 2017; Pesole et al., 2018). The highly competitive environment

of platform work and its income insecurity have major consequences: platform workers

are likely to take on as many tasks as possible and work with tight deadlines, which

leads to an increased work pace or skipping breaks which in turn augments the risk of

Study to gather evidence on the working conditions of platform workers

61

injuries (Drahokoupil and Piasna, 2017). This is particularly the case with micro-tasks,

which are known for their repetitive nature and low remuneration (EU-OSHA, 2017).

Platform workers in food delivery have mentioned that they are entitled to only very

short breaks and that some of them do not manage to take breaks at all while working

(Eurofound, 2018).

The threat of replaceability causes stress among platform workers engaged in low-skilled

on-location or online work. Huws et al. (Huws et al., 2017) document how the threat of

replaceability leads to an increased work intensity among Uber drivers, with drivers

working at a rapid pace without taking breaks. This is also explained by the on-demand

nature of platform work, which requires platform workers to compete in a large pool of

others locally or globally, while the availability of work is uncertain and often limited.

When tasks do appear, they dictate short deadlines, which can be stressful (EU-OSHA,

2017; Huws et al., 2016; Maselli et al., 2016). Speed pressure can also be problematic

for platform workers in transportation services. This depends on the organisation of the

platform, however, and in particular on its payment and evaluation system. For example,

platform workers earning a high hourly wage will be less likely to accept a

disproportional number of tasks in a limited period, compared with workers who are paid

per task (Lehdonvirta, 2018). This may also apply to household services, even though

Pesole et al. (2018) find that these workers are the least likely to face tight deadlines.

Professional online workers, who do mostly creative, profession-based tasks that are

rarely repetitive, are also often required to meet regular tight deadlines and may have to

win tasks through online competition (Pesole et al., 2018). The need to constantly

compete to have sufficient work causes stress. Nevertheless, these platform workers

tend to have a different perception of this situation. Those engaging in online contests

are aware of the insecurity that this type of work entails, and tend to approach it as a

way to build up a portfolio and only participate in contests when they have sufficient

time to do so (Eurofound, 2018).

Both work intensity and speed pressure can be high for platform workers, and is likely to

be aggravated by further increases in the competition between workers for tasks and

between platforms for clients. Tasks requesting undifferentiated skills, high work

intensity and speed pressure may provide reasons for workers to stop doing platform

work, and some workers state that they only accept these conditions because the work

is temporary (Eurofound, 2018; Lenaerts, 2018). There are substantial negative effects

on job quality (Lamberts et al., 2016), especially when the work itself is repetitive, not

set in a well-structured organisation, and without the necessary support, such as

obligatory rest periods or specialised materials. This holds for platform work and other

types of irregular employment, for example casual work, homeworking and temporary

agency work (Vereycken and Lamberts, 2018).

Emotional demands 4.2.6

Emotional demands are an indicator of the emotional pressure workers feel during the

execution of tasks, including dealing with direct demands from clients, or having to hide

your feelings (Eurofound, 2017b). Personal contacts have been proven to influence the

psychological well-being of workers (Boekhorst et al., 2017; Huang et al., 2011).

Emotionally demanding jobs have a lower overall job quality, bringing, for example,

higher risk of sleep problems, low general and physical heath, and low mental well-being

(Szekér et al., 2017). Jobs that involve dealing with and supporting people have been

found to be especially emotionally demanding, for example those in in commerce and

hospitality, healthcare, and the education sector (Eurofound, 2017b).

In platform work, emotional pressure can be two-sided. Platform workers are hired

for individual tasks that are not usually emotionally demanding. Not all platform workers

have direct contact with clients, for example those in contest-based online work.

However, because of the fierce competition between workers, constant monitoring of the

work and online evaluations that are available to all users of the platform, platform

Study to gather evidence on the working conditions of platform workers

62

workers are required to be flexible, friendly, and customer-oriented regardless of the

type of work they do. The application of rating systems requires platform workers to

have a service mentality, making these jobs far more emotionally demanding than their

counterparts outside the platform economy (Lee et al., 2015; Raval and Dourish, 2016).

Lee et al. (2015) also emphasise that, in the context of ride-sharing, clients tend to

underestimate the importance and impact of their ratings on the worker. In this way, a

‘platform job’ is broader than the tasks paid for, as it also involves constant customer

care and public relations. The same accounts for household service and professional

workers facing an imbalanced power relationship, in which they have to meet and please

the client directly and sometimes work under their direct supervision. This is not

necessarily specific to platform work, but rather depends on other circumstances, such

as the type of contract or the task at hand. A parallel can be seen with casual workers,

for example, those who have no guarantees to have a job the next day or week, and aim

to keep their employer satisfied in the hope of receiving a stable contract (Vereycken

and Lamberts, 2018). Similar tendencies are found among temporary agency workers

(ibid.).

Platform workers’ dependence on the client for payment and good ratings causes stress.

While this applies to any type of platform work, it is especially the case for online and

on-location workers with differentiated skills who are trying to sell their expertise and

portfolio to potential clients. Although some platforms allow workers to negotiate with

clients on the specificities of the task, and to have a final say in whether they accept a

task or not, in many cases the client plays a large role in the allocation of the work.

Schmidt (2017) finds that creative online workers especially are in a vulnerable position,

as their work is more loosely defined and often vaguely described. This requires social

skills to maintain the relationship with the client, notably when payment depends on the

approval and acceptance of work by the client, and where evaluation by the client is a

crucial factor for future work (ibid.). Similar issues arise for portfolio work, where

freelancers simultaneously work for a variety of clients and depend on good customer

relations for future work (Vereycken and Lamberts, 2018). In the context of platform

work, the lack of support from direct colleagues increases the risks related to high

emotional demands compared to regular employees (EU-OSHA, 2017; Pesole et al.,

2018). That being said, not all platform workers report this as an issue (Eurofound,

2018).

As a final point, being in emotionally disturbing situations is often discussed along with

emotional demands (Eurofound, 2017b). In this regard, the moderation of harmful or

violent online content – which occurs in some forms of online platform work – warrants

attention. Exposure to emotionally damaging images is known to have a considerable

impact on workers’ well-being and mental health (Lamberts et al., 2016). When this type

of content is encountered in official instances, the workers involved usually receive

extensive psychological counselling. Platform workers, however, may perform this type

of work without appropriate psychological and social support and in a solitary situation,

which increases its negative impact (Solon, 2017; Taylor et al., 2017).

Physical environment 4.2.7

The physical environment in which work is set refers to the exposure to ergonomic,

ambient and biochemical risks, the material required and acquired, and other

physical occupational health and safety risks (Eurofound, 2015). These indicators have a

direct impact on job quality outcomes such as the physical or psychological health of

workers. OSH is a priority for EU policymakers, and the improvement of working

conditions to protect workers’ health and safety is enshrined in Article 153 of the TFEU.

The physical environment and health and safety of platform work have been discussed

extensively, both in the literature and in public and policy debates. National experts

consulted for this study also identified health and safety as an important challenge for

platform work in the survey, and most of the focus group discussions also emphasised

it. Note, however, that the focus has largely been on mental and physical health and

Study to gather evidence on the working conditions of platform workers

63

accidents at work, and there is no evidence on occupational diseases in relation to

platform work. Most attention regarding health and safety issues is paid to on-location

platform work carried out at the client’s premises or in public spaces (e.g.

transportation, household or professional services). Far less attention has been devoted

to the health and safety risks of online work.

For most tasks carried out by platform workers, the related risks are comparable to

those for regular employees doing similar tasks. On-location workers such as drivers will

face traffic and ergonomic risks, bikers have the increased risk of deadly traffic

accidents, cleaners are exposed to chemical substances and experience ergonomically

straining situations. Online workers face risks related to long working hours on

computers, such as visual fatigue and musculoskeletal problems. In the case of platform

work, however, protective occupational health and safety regulations are not necessarily

guaranteed and are hard to regulate (Tran and Sokas, 2017). Platform workers are

responsible for their own health and safety, which includes the provision of protective

equipment and the materials and tools needed to perform the task (Huws, 2016;

Eurofound, 2018).

However, several studies find that platform workers are not always able to assess the

health and safety risks related to a task or are unaware of and uninformed about

potential risks, and receive limited or no training on health and safety (Huws,

2016; Pesole et al., 2018; Eurofound, 2018). This is problematic, considering that

platform workers are, on average, younger and identified as being subject to a higher

risk of occupational accidents and injuries (Huws, 2016; Tran and Sokas, 2017). Another

issue is that platform workers do not necessarily do this type of work regularly, and thus

may be less experienced with the task, and frequently switch between activities. In this

sense, there are similarities with temporary agency workers, who also face higher risks

than employees in similar jobs (Benavides et al., 2006; Howard, 2017; Tran and Sokas,

2017; Wilde, 2016). In addition, competition between platform workers may encourage

them to accept tasks for which they do not have the right equipment or have no

experience or, or continue to take on when sick (EU-OSHA, 2017; Pesole et al., 2018). A

related issue that was raised in the focus groups is that platform workers are generally

paid by task and not by hour. This may lead workers to work long hours and to work as

fast as they can, which forces them to take risks such as crossing the street on a red

light. Working long hours and combining multiple jobs can also cause stress and fatigue

(Cottini and Lucifora, 2010).

Previous research and the fieldwork conducted for the study confirm that many platform

workers are unaware of or unconcerned by health and safety risks. Exceptions include

on-location platform workers such as food delivery riders and worker-initiated and

client-determined moderately skilled platform workers (EU-OSHA, 2017; Pesole et

al., 2018; De Groen and Kilhoffer, 2019). Food delivery riders indicate concerns about

traffic accidents and having insurance (EU-OSHA, 2017; Pesole et al., 2018). One food

delivery rider participating in a focus group for France, for example, highlighted the

importance of accident insurance, noting that this was necessary because delivery riders

feel pressured to take risks, and indeed knew colleagues that had had grave accidents as

a result of doing so. This platform worker argued that the eligibility criteria of the

insurance should be fit for purpose; preconditions such as having paid social security

contributions for at least one year exclude workers with no or limited labour market

experience.

Some platforms offer protective gear or equipment to their workers, either for

rent or free of charge. Some food delivery platforms, for example, offer bikes, helmets,

backpacks and jackets.66 Other platforms leave this up to the worker. Deliveroo

66

Interestingly, several food delivery platforms initially obliged their workers to wear specific gear, but later

came back on this decision as it called into question whether the workers involved were in fact self-employed or rather fell under the supervision and direction of the platform – taking on the role of employer.

Study to gather evidence on the working conditions of platform workers

64

riders, for example, use their own bikes and are responsible for their maintenance,

whereas riders working for Takeaway can use a bike provided by the platform. Some of

the experts and stakeholders consulted argued that the materials or equipment provided

by platforms might not suffice to fully protect workers. On the other hand, when

platform workers have to provide their own equipment, there is the risk that they might

try to save costs and, therefore, do not foresee the cost of high-quality tools or the exact

materials required for a specific task. Platform workers who can set their own price can

account for the costs of the equipment and materials needed to perform the task

(Eurofound, 2018). Platform workers who cannot do so indicate that this makes it

difficult to make a living out of platform work and are forced to (temporarily) quit when

their equipment or material breaks down, finding, for example, that they are unable to

pay for bike repairs (ibid.).

Experts at focus groups in Estonia and France mentioned that the responsibility for

health and safety measures usually falls on the employer. With platforms refusing to

take on this role, or registering the workers as self-employed, this responsibility falls on

the platform workers. Experts emphasised that platforms do not always provide the

necessary incentive to their workers to prevent accidents, and that in general preventive

measures are being overlooked. In the Estonian focus group, the issue of insurance was

discussed at length. Estonia is a small country without large international insurance

companies, and providing specific insurance for platform workers is too expensive for

local insurance companies. A further complication is that insurance firms only allow

employers to insure workers, but platforms are not necessarily employers, and neither

do insurance products typically cover temporary, short-term assignments.

In short: challenges related to the work dimension 4.2.8

Of the three dimensions distinguished in the WES model, the work dimension is most

strongly affected by technology or algorithms. These may radically reshape how work is

allocated, organised, monitored and performed. In this regard, however, it is important

to distinguish between completely new tasks that have been made possible through

platforms, such as online micro-tasks, and those that existed before. In the latter case,

the working conditions may not differ much from those in more traditional work

environments, depending on the platform. The related physical and psychological risks

are not necessarily very different, but rather are aggravated in the context of platform

work. This can be attributed to the non-conventional work environment, the absence of

an organisational structure, and the use of algorithmic management. Responsibility for

health and safety often falls upon the platform workers themselves, who in addition use

their own materials and equipment.

In this area, the allocation and organisation of work as well as surveillance, direction and

performance appraisal, and the physical environment, emerge as the main challenges.

Platform workers performing low-skilled on-location tasks face higher physical risks,

while high-skilled work is associated with higher levels of autonomy and complexity and,

for online work, with risks linked to excessive screen time. Whereas typically the

employer is responsible for health and safety provision and providing the necessary

equipment and materials, this is left to the platform worker, who executes tasks outside

a conventional workplace. For these reasons, this challenge is labelled high and specific

to platform work. Depending on the platform, platforms, clients or workers may allocate

tasks and can determine the method, order and speed in which work should be done.

Many platform workers are not ‘their own boss’, but instead fully depend on the tasks

allocated to them, which they execute as instructed by the platform or client. This is

problematic, especially with regard to the allocation of tasks, and is moreover an issue

specific to platform work. For these reasons, this challenge is categorised as high.

Autonomy in work organisation is a medium challenge, as it is linked to the complexity

of the task at hand. Those performing more complex tasks will (necessarily) have more

autonomy. The use of technology, apps and algorithms in surveillance, direction and

performance appraisals is unique to platform work. This may, however, present major

Study to gather evidence on the working conditions of platform workers

65

challenges as platform workers are monitored continuously and receive online ratings or

reviews that they often cannot counter. Such ratings nevertheless have a substantial

impact on platform workers’ future work opportunities. This challenge is also specific to

platform work and categorised as high, notably as very little is known about the

underlying algorithms platforms use (the ‘black box’), which complicates any

intervention by policymakers and other actors, such as inspectorates. Task complexity

and emotional demands are not considered challenges. Although some platform workers

do need to manage relationships with clients, challenges related to emotional demands

appear to arise in individual cases rather than on a wider scale. Work intensity and

speed pressure are minor issues that pertain in particular to some types of platform

work.

The expected evolution of these challenges is generally unclear because this depends on

the types of platform work and which platforms will grow and the extent to which

competition between platform workers will become more intense. This, however, also

depends on the national and EU-level responses introduced. These are handled in

subsequent sections.

Table 8: Summary table of challenges related to the work dimension

Indicator Importance of

challenge

(high, medium,

low, none)

Specificity of challenge

(specific to platform work, common for

non-standard work,

general labour market)

(Most) Affected types of platform work

(all types, online vs. on-location work, low- vs. high-skilled, client-, platform- or worker-

determined)

Autonomy in the allocation of tasks

High Specific to platform work

All types, though most problematic for:

- platform- or client-determined

work

- low-skilled work

Autonomy in work organisation

Medium

General labour market All types, though most problematic for:

- low-skilled platform work

Surveillance, direction and

performance appraisal

High Specific to platform work

All types, however:

- algorithmic surveillance and

direction most prevalent with low- skilled platform work

- rating systems are universal

Task complexity None General labour market Low-skilled platform work

Work intensity and speed pressure

Low

General labour market All types, but especially problematic for:

- low-skilled platform work

Emotional

demands

None

General labour market On-location platform work where

there is direct (face-to-face) contact with clients

Physical

environment

High Specific to platform

work

All types, but especially problematic

for:

- on-location platform work

Source: authors’ own elaboration, based on the literature consulted and the fieldwork performed in this study.

Study to gather evidence on the working conditions of platform workers

66

Figure 7: Stylised representation of work challenges for different types of platform work

Note: a higher score (= more on the outside of the spider chart) indicates a better situation regarding job quality. However, keep in mind that a final job quality score would be defined by the combination of and balance between multiple job characteristics. Source: authors’ own elaboration.

Figure 8: Stylised representation of work challenges for different types of platform work

Note: a higher score (= more on the outside of the spider chart) indicates a better situation regarding job quality. However, keep in mind that a final job quality score would be defined by the combination of and balance between multiple job characteristics. Source: authors’ own elaboration.

allocation of tasks

autonomy in work organisation

surveillance, direction & performance appraisal

task complexity work intensity & speed

pressure

emotional demands

physical environment

online, low-skilled online, high-skilled

on-location, low-skilled on-location, high-skilled

Study to gather evidence on the working conditions of platform workers

67

4.3 The employment dimension

The employment dimension encompasses aspects relating to the formal context in

which a platform worker performs tasks. This includes the legal employment status,

contract, social protection, and the composition of a worker’s earnings, among other

topics. Next to these legal aspects, the employment dimension also discusses work-

related issues that directly affect a platform worker’s personal life, such as working time,

training or career opportunities. In relation to platform work, the determination of the

employer and the termination of the relationship with the platform also have to be

considered. The use of technology, apps or algorithms has implications for the

employment dimension, though to a lesser extent than in the work dimension. Ratings

and rankings, for example, may severely affect the earnings potential of platform

workers. This is discussed in more detail below.

Employment status 4.3.1

The employment status or labour market status indicates the status of a person as either

working in the framework of an employment relationship (employee) or working on their

own behalf and for their own account (self-employed).67 It refers non-exhaustively to the

contractual aspect of employment in terms of the autonomy and the authority that

workers have in their jobs, the duration and number of working hours, incorporating

economic risk (European Commission, 2019). The distinction between the categories of

self-employed and employee is generally based on subordination, whereas the concept

of economic dependency is often used as a criterion for intermediate categories, such as

dependent self-employed.

The employment status of platform workers is one of the most discussed topics in the

public and policy debates on the platform economy in Europe, both at the Member State

and EU levels. This interest is mirrored in both the academic and grey literature, with a

large number of articles discussing the issue. There is a consensus in the literature and

among the surveyed experts and stakeholders that the uncertainty on the employment

status of platform workers is a pressing issue that is expected to remain stable or

increase, as platform work gains in importance. In fact, based on available

literature and expert inputs, the employment status of platform workers appears to be

the most important challenge that needs to be addressed (Garben, 2019). This focus

on the employment status follows from the notion that it determines, to a large extent,

what rights and obligations workers have, for example concerning labour protection,

social protection or taxation.68 This applies to both individual and collective rights.

In the EU, platform workers typically do not have a separate status and, therefore, have

to be classified under one of the employment statuses recognised in the country. This

has proven difficult for a number of reasons. First, platform work is characterised by

complex employment relationships that involve multiple parties (i.e. platform,

client and potential additional actors, such as restaurants for food delivery riders)

(Eurofound, 2018; Lenaerts et al., 2018). Platforms typically argue that they play the

role of intermediary, matching supply and demand for specific services. Although such

clauses are generally legally not enforceable, platforms often indicate explicitly in their

terms and conditions that they are not the employers of the workers using the

platform, often without specifying the employment status of the platform worker or by

classifying them as self-employed. There are exceptions, for example platforms offering

employee contracts or other types of contracts. But these are limited.

Another important complication as regards the employment status of platform workers is

that it is difficult to apply the principles laid down in labour law or case law to

67

Council of the European Union, Proposal for a Council Recommendation on access to social protection for

workers and the self-employed, Article 9 (b), 15394/18, LIFE.1.C (10.12.2018).

68 Taxation of platform work income was in fact indicated as the main challenge in the Danish focus group.

Study to gather evidence on the working conditions of platform workers

68

distinguish employees from the self-employed in the context of platform work. Platform

work is blurring the boundaries between ‘employee’ and ‘self-employed’, which

complicates the classification of workers into statuses and can lead to

misclassifications.69 In general, the self-employed do not have the same labour rights

as employees. As explained in detail below, in some countries labour rights that are

typically attributed to employees are extended to individuals who find themselves in the

grey area between being self-employed or an employee, for example the French El

Khomri Law, which grants the right to vocational training to self-employed platform

workers. Some experts consulted for this study have indicated that it should be

examined as to whether classification issues in platform work are part of a larger trend

that is also found in other cases, and whether the uncertainty about the legal status is

deliberately used to decrease labour costs or push responsibilities from one party to the

other.

A related issue is that under certain circumstances, platform work may not be

considered an economic activity, for example because of the low number of hours

worked or the low income generated.70 In such cases, platform workers do not

necessarily have access to social protection through their platform work activities.

However, this is mitigated by the fact that most platform workers have a main activity

besides platform work through which workers have access to social protection (EU-

OSHA, 2017; ILO, 2016; Eurofound, 2018a; Pesole et al., 2018). Nevertheless, this does

mean that those who depend the most on platform work – i.e. those who do not have a

main activity – are covered the least (see infra).

Against this background, platform workers are generally considered self-employed or

independent contractors. Whereas this may certainly reflect the real employment

relationship for some platform workers, for example on-location or online workers with

differentiated skills providing expert services, others may find themselves in a grey zone

(Eurofound, 2018). Platform workers doing low-skilled on-location tasks (e.g.

transportation or household services) or who are active on platforms that exercise

considerable control over the work allocation and work organisation, set the transaction

price and have extensive surveillance and control mechanisms, are most likely to be

misclassified. Furthermore, platforms such as Foodora who are active in several

countries seem to adapt to the national legal and regulatory framework by, for example,

only working with self-employed workers in one country while offering multiple options

elsewhere. For these reasons, bogus self-employment is a much-discussed challenge for

platform work. It was one of the main topics covered in the Slovenian focus group, for

example.

Although, as indicated above, challenges related to workers’ employment status are

among the most salient issues, they are not, however, specific to platform work. In

contrast, the increased flexibility and workers’ autonomy, but also the facilitated forms

of (extensive) control, made possible through digitalisation and technological change,

have blurred the boundaries between different employment statuses more generally (De

Groen et al., 2017a). Also, different types of employment relationships are found in

other non-standard forms of work, and determining workers’ employment status is not

always straightforward (Vereycken and Lamberts, 2018). Given the current challenges,

some platforms are believed to veer towards deliberate misclassification of platform

workers. Indeed, this was suggested by one of the consulted experts in an interview. In

many ways, platform work is seen as a test case, highlighting how changes in the world

69

In France, for example, subordination is key to determining the employment status. Workers should be

classified as employees when platforms have the power to direct, control and sanction. To be classified as self- employed, platform workers should be free to accept or refuse tasks, the platform should not be able to give directions or instructions, and the worker should receive no sanction even in case of misconduct. In the UK, the concept of economic dependency came into play in recent court cases (see following sections).

70 This is the case in Belgium, when platform workers earn less than a specific cut-off level through an officially

licensed platform.

Study to gather evidence on the working conditions of platform workers

69

of work put pressure on the existing regulatory and legal frameworks. For that reason,

many have called for an update of national labour law (ibid). As will become clear, some

countries have either attempted to regulate the conditions of the platform work indirectly

by defining the status of platform work or by expanding the personal scope of national

legislation. Portugal is one example.

Determination of the employer in platform work 4.3.2

Alongside the discussion on the employment status of platform workers, the role of the

platform has been subject to debate. Platforms generally claim to be intermediaries,

connecting workers to clients, or even tech start-ups, and often explicitly refute the

role of employer in their terms and conditions (Kilhoffer et al., 2017). In several

countries, for example Belgium, Bulgaria, Denmark, Finland, the Netherlands, Slovakia,

Sweden and the UK, it has been debated whether Uber should be categorised as a taxi

service or as another type of organisation. If classed as service providers themselves (as

opposed to mere information society intermediary), the risk for platforms of being

classified as ‘employer’ increases (while not being certain, and depending on the exact

nature of their relationship with the platform workers). The unclear status of platforms

has been identified as a major challenge by the experts and stakeholders consulted for

this study as well as the literature. This issue is likely to become more severe as the

platform economy continues its development. In most of the countries, platform

businesses are not recognised as a separate business sector and/or represented in a

separate employers’ representative body. This issue is hardly discussed for other types

of work; it is specific to platform work and hence categorised as a challenge of ‘high’

importance in this study.

The determination of the employer in platform work relationships is difficult for a

number of reasons. The lack of available data on platform work and the high level of

heterogeneity in the types of platform work and in the relationships between platforms,

platform workers and clients all complicate the analysis of whether platform workers are

employees with platforms as their employers. As for the employment status of platform

workers, it is difficult to generalise conclusions about the role of the platform because of

this heterogeneity, and a case-by-case assessment is needed. Related to this issue, the

status of platforms is not legally defined, nor is the concept of employer in most EU

Member States.

Where there is a legal definition, it tends to be linked to the concept of an employee or

to the employment relationship (as explained by several of the experts consulted). In

Slovakia, for example, an employer is a legal or natural person who employs at least one

natural person in a labour-law relationship, and, if so stipulated by a special regulation,

also in similar labour relations. The determination of whether a platform is an

employer is thus intertwined with the question of whether a platform worker is

an employee. In Germany, the concept of employer is primarily derived from that of

employee. According to German case law, an employer is someone who employs at least

one employee. Despite the link between the determination of the employer in platform

work and the employment status of platform workers, the latter has received

significantly more attention than the former, probably because of the strong connection

to working conditions and social protection. Or, in other words, for platform workers it

may be more important to understand whether they are employees or not than to

identify their employer.

The determination of the employer in platform work relationships can present a

challenge in any type of platform work, similar to the difficulties in determining the

status of workers. However, the issue is mostly discussed for on-location platform work,

notably transportation services, where there is doubt over the status of platform

workers and on which the debate on bogus self-employment has centred. This

topic has been somewhat less discussed for other types of on-location platform work, for

example household or professional services, but could also be pertinent there. It has

received even less attention for online workers. Nevertheless, the lack of clarity on the

Study to gather evidence on the working conditions of platform workers

70

employer in platform work is problematic because it creates confusion as to who workers

can receive orders from, or contact when faced with an issue, and in general leaves

platform workers in the dark about their own status and contracts.

To overcome these issues, Prassl and Risak (2016) suggest a functional concept of

employer could be adopted, and identify five main functions that employers have: (i)

inception and termination of the employment relationship, (ii) receiving labour and its

fruits, (iii) providing work and pay, (iv) managing the enterprise-internal market, and

(v) managing the enterprise-external market. Prassl and Risak (2016) explain that the

contractual identification of an employer could be replaced by a functional

conceptualisation that assesses whether the platform exercises the five functions. Taking

Uber as an example, the authors find that although the platform’s terms and conditions

explicitly refute the role of employer, Uber does carry out the five functions mentioned

above.

Contracts: type of employment contract, contractual information provision and 4.3.3

termination of the employment contract

Along with the challenges to determine the employment status of platform workers and

the employer, contracts have been much discussed in platform work. The employment

contract lays out the employment relationship between an employer and an employee,

making explicit the subordination of the employee to the employer’s command or control

(Eurofound, 2018). Contracts stipulate the duration of the relationship (e.g. permanent

or temporary), working time (e.g. full-time or part-time) and other aspects.

Traditionally, employment contracts were open-ended and full-time. Non-standard

contracts do not conform to this traditional form, but rather include part-time, fixed-

term, temporary, casual and seasonal work (Broughton et al., 2010). Non-standard

contracts are common for non-standard forms of employment. Examples are zero-hour

contracts, non-written contracts and voucher-based work.

Except for those platform workers who are genuinely self-employed, platform workers

are often confronted with non-standard contracts. Most of them do not receive a formal

contract but are instead referred to the terms and conditions stipulated by the platform,

which workers have to accept. These terms and conditions, however, are often

ambiguous. They may, for example, identify platform workers as self-employed and

simultaneously have elements that are at odds with self-employment. AMT’s terms and

conditions, for example, state that there is no legal relationship between the platform

worker and the platform, but that platform workers are not allowed to sub-contract

tasks. Prassl (Prassl, 2013) refers to this as ‘the contractual denial of employee or

worker status’ and explains that platform workers are often expected to sign

documentation that sets their status as self-employed. In addition, it has been noted

that platforms change their terms and conditions without notice to the platform workers,

or even with retroactive impact.71 The terms and conditions may also have clauses that

are not legally enforceable. This all runs against the principle of fair contracts, which

have to be transparent, concise and available to workers, and reflect the genuine nature

of the employment relationship (Graham et al., 2019). This issue applies in particular for

platforms who exercise a lot of control over their workers.

Furthermore, platform workers typically have no protection against the termination

of their relationship with the platform (Berg, 2016). Platforms can end relationships

with platform workers by deactivating their account, often without giving workers prior

notice or information as to why this decision was made. On some platforms, including

Uber, dismissal is linked to performance; workers whose ratings are too low are no

longer allowed to work through the platform. Platform workers are not protected when a

platform ceases its activities, for example in case of insolvency (i.e. collective dismissal

71

Examples can be found on faircrowdwork.com

Study to gather evidence on the working conditions of platform workers

71

of platform workers). This issue has been flagged in the literature and by the experts

consulted in the study. On this note, it has to be recalled that platform workers usually

bring their own equipment, which they have bought, and that some of them need

specific licences in order to work. Uber drivers, for example, need a taxi licence in

Ireland. Some platform workers thus run up significant costs to be able to perform

platform work activities, but are not protected against termination of their relationship

with the platform, that is, the equivalent of ‘dismissal’ in an employment relationship.

Another issue that requires attention is that according to the Transparent and

predictable working conditions (TPWC) Directive, workers have the right to be

informed about their working conditions, which should occur in a timely manner

and in written format to which workers have easy access. In practice, platform workers

may not receive much information when starting work. They are usually referred to a

webpage, which may not be detailed or clear (note that the TPWC Directive does allow

the provision of information by electronic means). Conditions such as the start or end

date of the relationship or working times may not be explicit or very detailed. Platform

workers typically acknowledge receipt of this information by ticking a box on the

platform’s website that states they accept the platform’s terms and conditions.

Social protection 4.3.4

Social protection is understood as the coverage of risks and needs associated with

unemployment, sickness and healthcare, old age, invalidity, parental responsibilities,

loss of a spouse or parent, housing, and social exclusion. The social protection coverage

of platform workers has received quite a lot of attention in the literature and debate,

along with the challenges related to the workers’ status. This social protection challenge

has received significant attention from policymakers, social partners and other actors as

well, not only in relation to platform work but also in the context of non-standard work

more generally.

The literature and the experts consulted in this project have suggested that platform

workers tend to have less access to social protection (De Stefano, 2016; ILO,

2016; Berg et al., 2018; OECD, 2018). This is attributed to the classification of platform

workers as self-employed, their short work contracts, or even the fact that some

platform work is not considered an economic activity (‘not work’) (ibid.).

Workers’ access to social protection is partially linked to their employment status in the

labour market (Forde et al., 2017). Social protection systems, however, were designed

at a time when the traditional employment relationship with one employer and one

employee was the norm and are hence tailored towards the model of a single, stable,

full-time employment relationship (OECD, 2018). Other forms of work, in particular non-

standard forms of work and self-employment, may not fit into these schemes so easily.

As a result, these individuals are generally less protected and often responsible for their

own insurance. Several experts consulted for this study have, therefore, pointed out that

the challenge is not necessarily to improve the social protection of platform workers but

rather to improve social protection of the self-employed and non-standard

workers in general. This challenge is, therefore, clearly not specific to platform work

and is associated with the challenge of the unclear employment status and work

relationships in platform work.

Platform workers, being classified as self-employed, are typically less protected than

employees in the EU (both statutory and effective access), though they may have access

to non-contributory and some contributory social benefits. In terms of effective access,

the self-employed face issues with eligibility conditions, for example under-insurance due

to low minimum required incomes, short duration of benefits, waiting periods or

possibilities for opt-out exemptions. In some countries, the self-employed are covered

by the same social security scheme as employees, though not necessarily for all

branches (e.g. Austria, Hungary, Italy, Slovakia, Slovenia and Sweden) (ESIP, 2019).

Other countries have separate schemes for the self-employed (mandatory or voluntary),

Study to gather evidence on the working conditions of platform workers

72

granting access to some or all branches. One expert, however, rightfully pointed out that

platform workers often do not register as self-employed, which implies that in practice

they are not covered at all through their platform work.

Most platform workers have a main activity besides platform work, through which they

have access to social protection (EU-OSHA, 2017; ILO, 2016; Eurofound, 2018a; Pesole

et al., 2018). ILO (2016) argues that this indicates an inverse relationship between a

platform worker’s dependence on platform work and their social protection coverage:

those who depend the most on platform work are covered the least. Those

platform workers without labour market activities besides platform work face higher

occupational risks, as this lack of social protection may encourage workers to continue

working when sick (EU-OSHA, 2017b). In addition, there is the constant psychological

burden for workers of ‘not falling sick’ because they do not have access to societal

protection (EU-OSHA, 2017b). This is especially problematic as this group of platform

workers is the most vulnerable. This group is likely to have lower labour market

participation overall or combine platform work with other forms of non-standard work.

These platform workers may not fully realise the consequences of not having access to

social security, nor be aware of their rights and obligations. With platform work

becoming more popular, experts believe that the number of platform workers whose

only source of income is platform work will grow, thus aggravating this challenge.

In this context, access to social protection and who pays social security contributions are

concerns that have been raised by platform workers, social partners, policymakers and

academic experts, though these challenges are part of a broader issue affecting the self-

employed and those in non-standard forms of work. Platforms seem less concerned

about this issue, arguing that platform workers as self-employed should take on their

responsibilities. On top of the issues described above, policymakers and social partners

have noted that platform work may contribute to the underfunding of social security

schemes or to shifting the burden of funding social protections from the employer and

worker to only the worker (consulted experts and stakeholders). In the French focus

group, there was a debate about whether or not platforms could or should contribute to

social protection. While a representative of a smaller platform argued that platforms’

business model is not suited to this and that they do not make enough revenue to

contribute, especially in the case of smaller platforms, other participants referred to

platforms’ responsibilities in this area. Some platform workers indicate that they are

worried about social security (accident insurance in particular), while others are less

worried because they can fall back on an outside activity (Eurofound, 2018). The focus

group discussion in the Netherlands, for example, showed that platform workers are

most concerned about their earnings and would rather opt for higher remuneration and

less social protection than vice versa.

Earnings: price-setting, wages and additional fees 4.3.5

The (additional) income that can be earned is one of the main reasons why platform

workers engage in this type of activity. The potential earnings and payment

conditions are, therefore, key elements that platform workers consider when deciding

which platforms to offer their services upon (Eurofound, 2018). Earnings, moreover, are

part of almost every job quality index (Eurofound, 2017b). Paid work provides people

with the means for material needs and decent living conditions, as well as the possibility

for personal development and a place in society. The level of remuneration influences

the impact of work on the living standards of the worker and is formed by a core wage

that can be increased by additional fees.72

In addition to the level of pay, other features concerning platform work warrant further

review: the speed and periodicity of payments, wage setting and negotiations, fees

72

Examples are extra pay for overtime, Sunday work, dangerous working conditions, performance-related pay

(e.g. based on individual performance, company shares, profit-sharing schemes, etc.).

Study to gather evidence on the working conditions of platform workers

73

withheld by the platform, tips received from clients, costs related to platform tasks,

unpaid time, surge pricing and similar issues have been discussed in the literature and

were raised by experts, stakeholders and platform workers consulted in the study.

Issues related to earnings in platform work have been identified as a challenge that

affects all types of platform work, but particularly those workers who cannot set their

own prices (mostly those engaged in low-skilled online or on-location work). These

issues are expected to remain stable or grow in the future, as flagged by the surveyed

national experts. Concerns about earnings and payment conditions in platform work

have been voiced by trade unions and platform workers in particular (Daugareilh et al.,

2019; Huws et al., 2019). Other actors seem less concerned about earnings, typically

arguing that platform work is merely a secondary source of income.

To gain further insight into the challenges related to earnings in platform work, the level

of remuneration is examined first. A recent survey by Pesole et al. (2018) finds that

for 40% of platform workers the share of income obtained through platform work is less

than 25% of their total income; and for 30% the share of income gained from platform

work falls between 25% and 50% of the total income.73 This means that for 30% of

platform workers the income obtained through platform work makes up more than 50%

of their total income. A significant group of platform workers thus strongly depends on

the income gained through platform work.74 Platform workers typically do not have

access to additional fees (such as additional pay for night work).

These results corroborate previous surveys and studies, though reliance on platform

income appeared to differ significantly across countries (Berg, 2016; Huws et al., 2016;

Ilsøe and Madsen, 2018). A study by Smith (2016) concludes that lower income earners

are more likely to do platform work and that 56% of platform workers state that the

income from platform work is essential to them. The experts and stakeholders consulted

for this project have emphasised that special attention needs to be paid to the group of

platform workers who depend on this income. Several experts have indicated that

this group is likely to grow as the platform economy proliferates. In general, however,

there is only scarce evidence about what platform workers earn. This was pointed out by

several of the consulted experts (Bulgaria, Greece, and Estonia). In France, however,

platforms have to report what they pay to workers.

Most platform workers do not earn sufficient income through their platform work

activities to make a living (Eurofound, 2018). Low pay is an initial factor that may cause

this issue, and for several reasons: platform workers are paid by task rather than by

hour; work is broken down into small tasks (piecework); there is high competition

among platform workers with not always enough tasks offered; platform work may

not fall under minimum wage settings;75 and platform workers may lack bargaining

power (Eurofound, 2018; ETUI, 2018). That platform workers are paid by task rather

than by hour was flagged as a major challenge in the French focus group. It was noted

that this may lead platform workers to take unnecessary risks in an attempt to maximise

the number of tasks performed per hour. Remuneration for online micro-work in

particular is reported to be extremely low, sometimes just a few cents per task done

73

Since this is measured as a percentage of the total personal income, a risk for overestimation of the

importance of platform work as a source of overall income exists. The reason for a potential overestimation is that many student workers without any other income and depending on their parents, use platform work to earn extra pocket money (Pesole et al., 2018). These workers are included in the calculations.

74 Similarly, Hall and Krueger (2016) report that for 24% of the Uber drivers in the US, earnings gained

through Uber are the only source of income.

75 National minimum wages may only apply to employees. In countries without a national minimum wage (e.g.

Austria), minimum wages are usually set in collective agreements. However, the personal scope of collective agreements is typically limited to employees (though there are exceptions, e.g. collective agreements extend to all workers in the Norwegian cleaning sector, including platform workers). The enforcement of the minimum wage is difficult due to difficulties in the classification of platform workers into employee or self-employed and as supervisory authorities may have too little staff to follow up.

Study to gather evidence on the working conditions of platform workers

74

(Berg et al., 2018). Larger-scale tasks, performed online or on-location, generally have

high remuneration because they often require specific skills at market prices (Eurofound,

2018). As a result, platform workers could be earning well below the national or sectoral

minimum wage (mentioned by experts from Estonia, Germany, Latvia, Poland, Slovenia

and Spain). Some platform workers depend on the tips they receive.

Sometimes platform workers have to pay a fee to the platform.76 Some platforms

charge a fee upon registration, but it seems more common for a fee to be charged on

each transaction (a certain share or a fixed amount) (Leimeister et al., 2016). This is

remarkable, given the ILO standards on charging fees to workers for intermediation. ILO

Conventions 96 and 181 pertain to employment agencies. Both explicitly contain articles

stating that such agencies or intermediaries cannot charge any fees or costs to workers

(Lenaerts et al., 2017). When asked about these fees, which can be 10% per

transaction, some workers indicate that they find them too high in comparison with the

services that the platform offers (De Groen et al., 2018a; Lenaerts, 2018), especially

when the income gained per transaction is relatively low.

Platform workers are typically expected to cover costs associated with platform

work, such as gas, insurance and other costs, and to bring their own materials and

equipment. These costs have to be covered by the income earned through platform

activities. Workers who are able to set their own prices take these costs into account

when determining the price (Lenaerts, 2018). This issue was pointed out in the

literature. It was also brought up in interviews with trade unions, workers and other

stakeholders, discussed during focus groups and suggested by a few national experts in

their questionnaires, for example in Belgium and Italy.

Box 2: Price setting in platform work

With regard to earnings in the platform economy, an issue that warrants further attention is price setting: are platform workers able to set their own price, or does this fall on the platform or client? On those platforms where the platform or client allocates the work, have significant control over the organisation of work, and exercise surveillance and direction upon the worker, platform workers in general cannot set their own prices. On the platforms where the platform

worker can decide on the work allocation and work organisation, and there is less surveillance and direction, workers typically can determine their own prices.

Importantly, the first type of platform is common among low-skilled on-location and online platform work (notably performing transportation services), and particularly the group of workers engaged in low-skilled on-location work receives significant attention in (Western) Europe. It is this group for which there is most doubt about the nature of the employment relationship with the platform (are these workers self-employed or in fact working in

subordination to the platform?). The notion that these platform workers cannot set their own prices could be interpreted as another sign that these platform workers may not be self- employed, as the self-employed typically can set their own prices. It also raises the question as to what extent these workers can bargain for the price.

The second group of platforms allow workers to set their own price more frequently, for example platforms intermediating on-location work based on (moderately) differentiated skills

such as household or professional services or intermediating online work requesting differentiated skills. Remuneration per task is significantly higher for these workers, as they can account for their skills and time, but also for fees paid to the platform and costs incurred with regard to materials and equipment needed to do the work.

When platforms set prices, this can take many forms, such as a fixed amount per task (with potential top-ups for working during specific times or performance incentives), standard or

76

There has been some discussion on this especially in relation to temporary agency work, as temporary work

agencies are not allowed to charge such fees to their workers but rather ask clients to pay (Lenaerts et al., 2018). According to Article 7 of the ILO Convention 181 which applies to private employment agencies, private employment agencies shall not charge directly or indirectly, in whole or in part, any fees or costs to workers. ILO Convention 96, on fee-charging employment agencies – abolishes “fee-charging employment agencies conducted with a view to profit”.

Study to gather evidence on the working conditions of platform workers

75

minimum prices for a specific task (e.g. graphic design or in contest-based work) (Eurofound, 2018). Workers on such platforms additionally face changes in earnings due to changes in the prices set by the platforms, the ratings received, next to other factors such as fluctuations in the

hours worked, and over which platform workers often have no control. Many platforms use dynamic pricing schemes (De Groen and Maselli, 2016; Eurofound, 2018). Uber and Lyft, for example, have a system of ‘surge pricing’ that encourages workers to be available during peak hours; Deliveroo and Foodora have similar schemes to stimulate bikers to work during busy lunch and dinner shifts. There has been some discussion on dynamic pricing and workers have been trying to game the system.

Regarding workers’ sense of being fairly paid, professional online workers agree that they receive fair pay for their work more so than non-professional online workers and on-location workers in transportation and household services (Pesole et al. 2018). This can potentially be attributed to the question of whether there is competition between platform workers on prices or on the quality of the tasks delivered. An interesting example here is the Danish cleaning

platform Hilfr, whose workers can be covered by a collective agreement that sets a minimum wage per task. Workers can decide for themselves whether they take part in this collective

agreement, but the platform encourages this to guarantee fair wages (interview with platform representative).

In addition to low remuneration, the lack of income security is a significant stressor

for platform workers. The financial security of workers is strongly related to their

dependence on platform work. Workers with a stable job outside the platform economy

have higher income security than those who rely on their platform earnings. Some

workers face insecurity as regards the number of hours they can work through the

platform, which may result in low earnings and limited income security (pointed out by a

Romanian expert and discussed in the section on working time). The frequent changes in

the prices set by the platforms and in the platforms’ business models contribute to this

income uncertainty.

Along with the level of remuneration, platform workers also face issues related to non-

payment and unpaid time. Platform workers are often eager to receive their payment

soon after completing the task. Typically, the payment runs from the client to the

platform worker through the platform. Platforms can withhold (part of) the payment in

case of a complaint or poor rating by the client. Platform workers may have few

opportunities for recourse in such cases. Platform workers additionally spend time on the

platform for which they are not paid, for example, waiting time and time spent looking

for new tasks (Berg, 2016; Eurofound, 2018). Similar issues have been noted for other

types of non-standard work as well. Casual work, for example, is characterised by

limited and occasional working hours, resulting in income insecurity and low wages

(Eurofound, 2018a).

Working time 4.3.6

One of the features that makes platform work attractive for workers is the flexibility

platform work offers. Workers have the flexibility to decide whether they want to

work or not and to work at the times that best suit them, allowing workers to

ensure that the work fits into their schedule. This flexibility provides workers with

autonomy that they may not have in other jobs and is therefore one of the main reasons

to engage in platform work (Berg, 2016; Pesole et al., 2018). Most platform workers

combine platform work with another job or care tasks and may find it difficult to

work during regular hours or at fixed times. Many workers need this flexibility to ensure

that platform work fits into their schedule. Other workers choose platform work because

they prefer to work from home or because they consider it paid leisure time (Berg,

2016). Some workers even do platform tasks during their main paid job (Berg, 2016).

What the different motivations to engage in platform work have in common is that

workers want to work when it suits them (Pesole et al., 2018).

Research on platform workers’ working times suggests that most do not work regularly

on platforms and do it as a part-time activity (Balaram et al., 2017; Brawley and Pury,

Study to gather evidence on the working conditions of platform workers

76

2016; Daugareilh et al., 2019). According to Pesole et al. (2018), platform workers have

low labour market participation degrees. Over 40% of the workers surveyed work less

than 10 hours per week through platforms. 75% of them works less than 30 hours per

week through platforms (ibid.). Other studies have attained similar findings. Combining

the hours of platform and regular work, around 30% of platform workers work less than

10 hours per week in total, and over 50% of the workers work less than 30 hours per

week (Pesole et al., 2018). This suggests that platform work may very well be a second

job, but one that is potentially combined with other forms of precarious work,

leaving workers in a difficult position.77 Next to this group, there is a small but significant

group of platform workers (>5%) that report working over 60 hours per week on

platforms (Pesole et al., 2018). These workers are fully dependent on platform work.

The challenges with regard to working times in platform work are manifold. In

practice, the flexibility in being able to determine whether, when and how long to work

may be much more limited than expected, and platform workers are confronted with

unstable working hours as these are largely dictated by demand. Working time

challenges could pertain to all forms of platform work but seem particularly relevant for

on-location workers and for those active on platforms determined by the client or the

platforms themselves. Workers involved in worker-initiated platform work report

experiencing less problems with working time flexibility demanded by the platform and

higher levels of planning autonomy (i.e. being able to control working times, work

schedule and when to take breaks or leave).

Depending on the type of work, various elements push platform workers to longer

working hours. For online workers, for example, the large competition pushes down

wages, creating the need for longer working hours to earn a certain amount (ILO, 2016).

The same accounts for platform workers in transportation, where earnings fluctuate

depending on the supply and demand of drivers (Chen et al., 2015; De Groen and

Maselli, 2016; Horton and Zeckhauser, 2010; Wilde, 2016). Jiang et al. (2015) further

describe how platform workers are sometimes unable to stop working, even though they

have achieved their preferred number of working hours, because of an unpredictable

overload of work and the pressure to remain available. On the other hand, some

platforms, for example Uber, allow workers to announce the last task of their ‘shift’, to

match them up with a client whose location is close to or on the worker’s way home. The

long and unpredictable working hours in platform work raise platform workers’

probability of suffering from mental health problems (Cottini and Lucifora, 2010).

While a small group of platform workers works a (very) high number of hours, many

workers indicate they would like to work more (Berg et al., 2018; Eurofound,

2018). Platform workers on food delivery platforms who have to register for ‘shifts’ have

especially claimed that the algorithm distributes shifts rather unequally among the

workers, with some workers getting a lot of hours and others very few, without any

explanation as to why this is the case (Eurofound, 2018; Lenaerts, 2018). To receive

sufficient paid tasks, many platform workers indicate having to be on standby and

reachable at all times (Berg et al., 2018; Eurofound, 2018). On the same note, what

platform workers in food delivery spend to travel to the meet-up points where they wait

for tasks to come in is typically not remunerated (ibid). This implies that many workers

report long working hours and that not all time spent on the platform is paid time.

This issue was mentioned explicitly in the focus group held in Slovenia. The

interconnectedness of low pay and long working hours was seen as a challenge by the

77

This discussion raises the question as to whether platform work results in more labour market inclusion.

Pesole et al. (2018) find that platform workers are more likely to have a lower degree of inclusion in the traditional labour market. Platforms are thus not facilitating labour market inclusion in the traditional economy, but rather create a growing division in the labour market between a core group of workers with decent working and employment conditions and a newly arising, growing proportion of workers with questionable contractual agreements.

Study to gather evidence on the working conditions of platform workers

77

Slovenian focus group participants: low pay forces platform workers to work very long

hours to obtain a sustainable income.

From a legal point of view, it remains an open question as to whether the time online

waiting for a task is also considered working time, as is the case for other forms of work,

for example hospital staff available on call. Related to this, being classified as self-

employed, platform workers run the risk of not being covered by national legislation and

regulation on working time if this is only applicable to employees (e.g. paid public

holidays, minimum resting time). The need to be constantly available affects workers’

work-life balance, hampering workers’ ability to schedule or enjoy their leisure time

(European Commission, 2015; Huws, 2016; Martin et al., 2016; Smith and Leberstein,

2015). Schörpf et al. (2017) finds that platform workers worry that temporary

unavailability will cause job loss and damage their online reputation, affecting future

opportunities.

Platform workers’ autonomy over choosing when to work could be limited by several

characteristics of the platform. Most importantly, platforms need to guarantee the

availability of workers for clients, especially in the case of on-location work or for

those types of work where the platform or client selects a worker (platform- or client-

determined work) (Eurofound, 2018). Some platforms, therefore, impose a minimum

number of hours for workers to be active on the platform, which sometimes have to

be fulfilled during specific time slots. Workers who are less active are requested by

the platform to explain why or even removed from the platform if this inactivity persists.

Other platforms steer workers towards the most convenient working hours through

bonus systems such as surge pricing on passenger transportation platforms or

increased prices per ride on food delivery platforms, or because a lack of demand

means that there is little or no money to earn at other times (Smith and Leberstein,

2015). Lehdonvirta (2018) reports that a lack of available work is not necessarily only

due to a lack of demand, but also depends on how platforms allocate work (e.g.

maximum earnings and competition).

On-location workers in particular indicate less autonomy over whether, when and how

much to work, compared to online workers (Pesole et al., 2018). There are plenty of

examples of food delivery riders who prefer other working hours but are motivated to be

available for work and to work evenings and weekends because there are fewer tasks at

other times (Eurofound, 2018). For these platform workers, the flexibility of working

time loses its meaning, and this issue is aggravated for workers who depend on the

income gained through platform work (Berg, 2016; Lehdonvirta, 2018; Standing,

2015). Non-professional online workers, such as clickworkers, are perceived to have the

highest degree of autonomy (Pesole et al., 2018). This is confirmed by Lehdonvirta

(2018), based on interviews with platform representatives and workers of three

platforms intermediating online tasks. Lehdonvirta (2018) finds that workers are

formally free to set their own working times and generally could determine when and

how much they work and how to divide their time across tasks. Structural constraints

such as the availability of work and workers’ dependency on the income gained,

however, do matter. Workers who depend on their platform work income reported being

on call continuously.

Since the business model of platforms is to fulfil ad hoc requests of clients, platform

work is typically associated with atypical working hours (e.g. working at night, during

weekends and on holidays), unstable and unpredictable work schedules (e.g.

working during different hours every day or week), and a high demand of availability

and flexibility towards the platform (Forde et al., 2017). All these features can be

problematic from a job quality perspective (Széker et al., 2017). As regards the atypical

working times, however, platform workers usually do not find this problematic and most

workers are satisfied with their work-life balance (Eurofound, 2018). These observations

from the literature were confirmed in the fieldwork conducted for this study, for example

in expert interviews and the focus group in the Netherlands. It could be explored to see

Study to gather evidence on the working conditions of platform workers

78

what the added value of the TPWC Directive would bring as regards the predictability of

work schedules (see infra).

Regarding working times, there are parallels between platform work and other

non-standard forms of work. Workers in casual jobs, for example, report combining