Visible hands: How gig companies shape workers' exposure to market risk

• Published date: 01 January 2024
• Author: Michael David Maffie
• Date: 01 January 2024
• DOI: http://doi.org/10.1111/irel.12337

Industrial Relations. 2024;63:59–79.

|

59wileyonlinelibrary.com/jour nal/irel

The “multi- homing problem,” where people s witch between compet ing companies, is a centr al

challenge for all p latform organiz ations (Evans, 2003; Hagiu & Wrig ht, 2015; Maffie &

Gough, 2023). As these organi zations faci litate the exchang e of goods and serv ices, plat form

companies' c an only disti nguish the ir serv ices from a compe titor if they can of fer unique (or

“marquee”) conne ctions betwe en users1 (Evans & Sch malense e,20 16). For example, should a

customer be able to h ire the sam e set of drivers on eit her Uber or Lyft, the se two compan ies

become fu nctionally interchangeabl e (Bryan & Gans,2019). Accordi ngly, theory suggests that

platform compan ies must build ba rriers th at “lock in” people to the ir serv ices in order to

develop a defensible market po sition and eventually rea ch profitability (Hag iu & Wright,2015).

Despite the i mportance of multi- homing to the op eration of digital platforms, labor s cholars

have largely neglec ted this area of res earch, with some even sugge sting that these compan ies are

generally i ndifferent about worker retention (Duggan e t al.,2019 ; Josserand & Kai ne,2019; van

Doorn,2 017 ). Yet the tools platform companie s use to retai n users are quickly emerg ing in the

1Digital pl atforms medi ate many kinds o f interactio ns, giving r ise to several t erms to des cribe the rel ationship bet ween people a nd

platforms . In this arti cle, I use the te rm “users” or “p eople” when dis cussing ge neral platform d ynamics th at apply to virt ually all

digital p latforms. Whe n describi ng two- sided m arkets involvi ng the sale or excha nge of consume r goods, I use th e terms “buyers ”

and “selle rs.” Finally, when de scribing in stances wh en the seller i s primari ly selling th eir labor, I refer to “c ustomers” and

“wor ker s.”

Received : 19 October 2022

|

Accepte d: 24 May 2023

DOI: 10.1111/irel.12337

ORIGINA L ARTICL E

Visible hands: How gig companies shape workers'

exposure to market risk

Michael DavidMaffie

Cornell Un iversity, Ithaca , New York, USA

Correspondence

Michael D avid Maffie, C ornell Univer sity,

545D Statler Ha ll, Ithaca, N Y 14853, USA.

Email: maf fie@cor nell.edu

Abstract

How do gig platforms prevent workers from defec ting to a

competitor? Drawi ng on 40 original inter views and survey

data from 210 ride- hail drivers, t he author fi nds that plat-

form companie s calibrate workers' exp osure to market

risk using ga mified rewa rd systems. Thes e rewards protect

compliant workers from chang es in market cond itions,

raising the c osts of accepting work from a compet itor. Yet

those who do not comply are “pushe d” to the periphe ry,

increasi ng their market r isk. This ar ticle illu strates how

platform compan ies can use t heir “visible h ands” to har-

ness and control ma rket forces, shaping worker behav ior

within and a cross platforms.

This is an op en acces s article und er the terms of t he Creative Commons Attribution-NonCommerc ial-NoDerivs License, w hich

permit s use and dist ribution in any me dium, provid ed the origi nal work is properly c ited, the us e is non-comm ercial and no

modifi cations or adap tations are ma de.

© 2023 The Author. Indu strial Relat ions publishe d by Wiley Period icals LLC on beh alf of Regents of th e University of Ca lifornia

(RUC ).

60

|

MAFFIE

online g ig economy (Griesbach et al.,2019; Vallas & Schor,2020). Specif ically, many of the larg-

est gig platform comp anies, includ ing Instacar t, Uber, and GrubHub (Davalos & Ben nett,2022),

have adopted the most com mon method by which platforms reta in users— gamifi ed rewards.

Gig platforms' reward s systems are intrigu ing because they ar e far more complex than those

currently re presented i n the literatu re. While m any platforms reta in users th rough a combi-

nation of prici ng plans (e.g., annual membersh ips or pay- as- you- go) and subsidi es (e.g., cash

back) (Rochet & Triole,20 03), gig platforms' rewards syst ems alter nea rly every aspe ct of the

labor process on t hese service s (Griesbach et al.,2019; Mason,2 018; van Doorn & Ch en,2020):

they can change t he allocation of work, information offere d to workers, pre sentation of work-

ers' informat ion to consumer s, and give workers acc ess to organi zational res ources, suc h as

phone- based support or he lp repairing equipment (Caplan & Gi llespie,2020). While it is clear

that gig platforms , such as Uber a nd Lyft, have an economic i ncentive to lock workers i nto

their ser vices (Bryan & Gans,2 019), scholars cu rrently lack the ory to explain how these typ es

of rewards may allow them t o do so.

In this ar ticle, I exami ne two questions. Fi rst, how do gamif ied work systems sh ape work-

ers' movement across c ompetitors? Second , how effective are the se systems at lo cking people

into a platform? Drawing on 40 i nterviews f rom Uber drive rs, I find th at gamif ied rewards

systems man ipulate workers' exposu re to market risk (i.e., drive rs' probability of loss) th ree

ways: (1) suppress ing information from workers, (2) distributing jobs u nevenly across workers,

and (3) insulating some workers from low- value work. These three mech anisms form U ber's

“visible hand s” of market regulation. For those who follow U ber's desire d behaviors, U ber

uses its “vi sible hands” to protect them from un favorable market condition s. As workers must

continually u se Uber to retain this p rotection, however, Uber's rewa rd system rais es the costs

of working on a competi ng ride- hail servic e, locking the m into working for Uber. Yet those

who refuse to follow Ube r's desired b ehaviors are “pushe d” to the peripher y, increasi ng their

market risk. Next, dr awing on or iginal survey data from 210 Uber and Lyft dr ivers, I examine

how effective Ube r's “visible hands” are at locki ng workers into its s ervice. In this st udy, I find

that as drivers i ncreas e their Uber P ro level, they shi ft between 3.5% and 12% points of th eir

working time away from Lyft onto Ube r.

Examin ing how gamified rewar d systems shape workers' ability to mult i- home provi des two

insights into t he nature of control and de pendence i n the gig economy. First, wh ile extent re-

search rec ognizes that plat form companies exter nalize market risk onto workers ( Dubal,2020),

scholars assu me that thes e companies do s o in a uniform fa shion (De Stefano, 2 016; Wel ls

et al.,2 019). In this stu dy, I find that compan ies can uneven ly distribut e market risk acros s

their workforce in orde r to lock compliant workers i nto their ser vice. Yet those who do not

comply with thes e behaviors are “pus hed” to the per iphery and use d as shock absorbers for

compliant workers. Using ma rket risk in thi s fashion repres ents and evolution in the way plat-

form companie s shape worker behavior. Previously, platform companies re lied on direct nega-

tive employment action s, such as dism issals or suspensions to s anction noncompli ant workers

(Ros enbl at,2 018). This st udy finds, however, that platform compa nies can leverage the ir unique

capacity as m arket constructors to incre ase or decrease workers' market r isk, enabling them to

shape workers' actions w ithout relying on more recogn izable forms of discip line or punishme nt.

Second, th is article r eveals a previously u nknown relat ionship betwe en market risk and

worker dependenc e. While platfor m companie s attempt to build worker dep endence on th eir

servic es (Wood & Lehdonvirta,2 021), extent resea rch has yet to consider how these comp anies

can harne ss market forces to do so. By exploring how gami fied reward systems ca librate work-

ers' exposure to m arket risk, this arti cle documents the thre e market mechanisms th at platform

companies c an use to raise workers' dep endence on a serv ice. In doing so, this art icle shows that

dependenc e is not only a function of workers' econom ic alternatives (e.g., Schor et al.,2017 ) and

individua l platform features (e.g., Wood & Lehdonvirt a,2 021), but that platform c ompanies

can wield the ir power as market organizers to c ultivate worker dependence.