Received: 30 June 2017 Revised: 1 August 2019 Accepted: 5 August 2019

DOI: 10.1111/1748-8583.12258

REVIEW ARTICLE

Algorithmic management and app-work in the gig

economy: A research agenda for employment
relations and HRM

James Duggan | Ultan Sherman | Ronan Carbery |

Anthony McDonnell

Cork University Business School, University

College Cork, Cork, Ireland Abstract
Current understanding of what constitutes work in the
Correspondence
James Duggan, Cork University Business growing gig economy is heavily conflated, ranging from con-
School, University College Cork, Cork, Ireland. ceptualisations of independent contracting to other forms
Email: [email protected]
of contingent labour. This article calls for a move away from
Funding information problematic aggregations by proposing a classification of
Irish Research Council
gig work into three variants, all based strongly upon key
technological features: app-work, crowdwork, and capital
platform work. Focusing specifically on the app-work vari-
ant, this article's more delineated focus on the textured
dimensions of this work proposes new lines of enquiry into
employment relationships and human resource manage-
ment. Examining the crucial role of algorithmic manage-
ment, we critically discuss the impact of this novel
mediation tool used by gig organisations for the nature of
employment relations within app-work, work assignment
processes, and performance management. In so doing, we
propose a series of research questions that can serve as a
guide for future research in this increasingly important field.

KEYWORDS
algorithmic management, app-work, employment relations, gig
economy, HRM, precarious employment

Hum Resour Manag J. 2019;1–19. wileyonlinelibrary.com/journal/hrmj © 2019 John Wiley & Sons Ltd 1
2 DUGGAN ET AL.

1 | I N T RO D UC T I O N

Working arrangements are increasingly precarious, with independent contracting and temporary work becoming
more commonplace (Bonet, Cappelli, & Hamori, 2013; Harvey, Rhodes, Vachhani, & Williams, 2017). The emergence
of the “gig economy”—an economic system that uses online platforms to digitally connect workers, or “individual ser-
vice-providers,” with consumers—represents a new form of contingent labour (Harris, 2017). The “gig” business
model bypasses many of the regular responsibilities and costs of employment, leading to widespread legal ambiguity,
which has resulted in challenges as to whether workers should in fact be classified as employees (Collier, Dubal, &
Carter, 2017; Fabo, Karanovic, & Dukova, 2017). Discourse around work in the gig economy, or “gig work,” traverses
from the positive, with emphasis on the apparent autonomy and flexibility afforded to workers, to the negative, with
critics viewing it as a means by which businesses lower costs and erode employment standards and labour regulation
(Friedman, 2014; Stewart & Stanford, 2017).
The gig economy is disruptive to our traditional understanding of work, as its digital on-demand, or work-as-
required, principle sees personnel, as subordinate workers, becoming increasingly disposable (Todolí-Signes,
2017). It produces less long-term jobs, as people are hired to complete hyper-flexible “gigs,” working to com-
plete tasks for a defined, short period of time, often with low commitment existing between workers and orga-
nisations (Friedman, 2014; Harvey et al., 2017). Gig workers tend to be classified as independent contractors,
with the numbers legally employed by organisations operating in the gig economy significantly smaller (Todolí-
Signes, 2017). For example, Uber, a prominent “ride-hailing” or transportation gig organisation, currently has
almost four million drivers across 700+ cities worldwide (Madrigal, 2019) but only legally employs 22,000 in
total (Uber, 2019). Similarly, Lyft, another ride-hailing service, operates in 600+ locations across the United
States and Canada with almost two million drivers on the platform but employs less than 5,000 (McNeill, 2019).
Likewise, Deliveroo, a food-delivery company, has over 35,000 “riders” in 200 cities but only directly employs
an estimated 2,000 (Hurley, 2018).
Across Europe, Eurofound (2017) estimate that the number of people engaging in gig work as their main labour
market status makes up less than 0.5 per cent of all employment. The United Kingdom has the highest incidence of
gig work within Europe, with estimates of 4.4 per cent, or approximately 2.8 million people, engaged in this form of
work in 2017 (Lepanjuuri, Wishart, & Cornick, 2018). Of these, 25 per cent reported that some form of gig work was
their main job (Eurofound, 2017). Based on these figures, it is suggested that over 5 million people could be working
in the UK's gig economy by 2022 (Dupont, Hughes, Wolf, & Wride, 2018). The United States appears similar to
Europe, with figures of 0.5 per cent estimated to be participating in this form of work as of 2015 (Katz & Krueger,
2019). However, measuring the overall size of this new economy proves difficult because organisations are not
obliged to publish figures, and most gig working arrangements fall outside existing capabilities of labour market mea-
surement tools.1 Also, despite boasting large worker numbers, little is known regarding how many individuals regu-
larly engage in this work, rather than being one-off or periodic workers.
Uncertainty exists in respect to what gig work does and does not involve, with different forms of contingent
labour commonly subsumed into gig classifications (Bernhardt & Thomason, 2017; Howcroft & Bergvall-
Kåreborn, 2019; Kuhn, 2016). This, we argue, is erroneous. Despite outward similarities with nontraditional
forms of work, key differences exist that warrant specific consideration (e.g., number of parties involved and the
influence of technology). Likewise, there is no one universal work classification or set of “rules” that can be
implemented in the gig economy. Individuals who occasionally boost income by renting out apartments on Air-
bnb, an online platform for property rental, are very different from those who make a living by working for
ride-hailing or food-delivery services like Uber or Deliveroo (Rozzi, 2018). Again, each of these is strikingly dif-
ferent from crowdworking platforms, such as Amazon Mechanical Turk, that connect businesses with skilled
freelance workers (Rozzi, 2018). Because work and conditions are hugely individualised across platforms,
employment relations and human resource management (HRM) implications along with policy and union
DUGGAN ET AL. 3

responses will vary and need to be tailored accordingly. This article's first objective is to provide conceptual
clarity on gig work by moving away from monolithic perspectives and differentiating it into three key variants:
capital platform work, crowdwork, and app-work. This, we argue, enables a more textured understanding of the
idiosyncratic nature of gig working arrangements (Liao, Wayne, & Rousseau, 2016). Accordingly, this
conceptualisation of gig work provides the article's first contribution.
The article's second objective is to examine the criticality of management-by-algorithm, a relatively novel and
alternative means of utilising the influential role of technology to monitor, manage, and control workers (Rosenblat &
Stark, 2016). By focusing on the app-work variant of gig work, our second contribution involves nuancing the key
implications of algorithmic management for employment relations and HRM. Algorithmic management automates
HR-related duties and functions traditionally undertaken by human managers. Therefore, our app-work focus stems
from it being a gig work variant that is particularly problematic from an employment relations and HRM perspective,
as evidenced by the burgeoning academic, practitioner, and press narrative (e.g., widescale protests, debates, and
legal issues surrounding app-work organisations; Schmidt, 2017; Stewart & Stanford, 2017; Tran & Sokas, 2017;
Wright, Wailes, Bamber, & Lansbury, 2017).
The article elucidates several research questions around the nature of employment relations and HRM responsi-
bilities in app-work organisations. Given the relative infancy of academic enquiry on gig work, we argue that an
appropriate point of departure is to examine the impact that algorithmic management has on the formation and
development of the working relationship in the first instance. Gaining insight into the process through which workers
establish a working relationship, as well as investigating how they adhere to and are managed by the algorithm's
instructions, holds the prospect of better understanding the intricacies of labour within app-work. Building on this,
we identify several areas strongly impacted by algorithmic management, specifically the nature of employment rela-
tions, work assignment processes, and performance management procedures (Prassl, 2018; Rosenblat, 2018), as cen-
tral strands of the proposed research agenda.

2 | C L A S S I F Y I N G A N D D I S T I N G U I S H I N G G I G WO R K

Building on Wolfe (1941), many conceptualisations of the terms “work” and “employment” have been offered by
legal and organisational scholars. These reflect the changing nature of working relationships and the emergence of
nonstandard forms of employment over time. Most point to the idea of an employee being “controlled” by the orga-
nisation, “in the service” of an “employer,” “directing” the work process in exchange for remuneration, or the recipro-
cal obligations involved (Cappelli & Keller, 2013; Connelly & Gallagher, 2004; Schein, 1980). Although researchers
conceive of an employee in different ways, the means by which gig organisations view their workers has been con-
troversial. In distancing themselves from legal responsibility towards “employees,” many gig organisations have
adopted other monikers for worker classification (e.g., “taskers” with TaskRabbit and “riders” with Deliveroo). There
are, however, several cases being taken by gig workers, with the support of unions, challenging the legal status of
their employment (McGaughey, 2018).
Working arrangements are traditionally classified as either “employment” or “contract work.” Situating gig work
in the literature on employment classifications, it is a hybrid of contingent work types (Kuhn, 2016). Figure 1 illus-
trates how gig working compares to other types of employment. It has obvious similarities to independent contra-
cting (Carr, Hall, Mason, & Varney, 2017) and forms of subcontracting with the involvement of at least three parties.
Gig work also bears similarities to temporary employment, which is neither full-time nor open-ended (Friedman,
2014); temporary agency work, which distributes work via third-party labour intermediaries (Ward, Grimshaw,
Rubery, & Beynon, 2001); and 0-hr contracts, where no guaranteed hours are offered (O'Sullivan et al., 2015). The
precarity of these arrangements makes them comparable to aspects of gig work in terms of a lack of commitment to
long-term relationships, flexible working hours, project-based work, and piece-rate payments.
4 DUGGAN ET AL.

FIGURE 1 Summary of working arrangements, adapted from Cappelli and Keller (2013)

The common denominator across all gig work, and a critical distinguishing feature from other contingent labour,
is the presence of an intermediary in the form of a digital platform organisation. Digital platforms present a new way
of organising work and offering services, functioning as online businesses that facilitate commercial interactions
between at least two parties—workers and customers (Gramano, 2019). This enables the meeting between the
worker and customer, and, in doing so, mediates this relationship (Gandini, 2018). Building on work by De Stefano
(2016), this distinctive feature allows us to identify three different variants of gig work: capital platform work,
crowdwork, and app-work.
Capital platform work exists where individuals use a digital platform to sell goods peer-to-peer or to lease assets
(e.g., Airbnb and Etsy). In this context, the role of the digital platform is to connect customers with a form of capital
owned by an individual. The term “sharing economy,” often used interchangeably with the gig economy, is most
closely associated with capital platform work. However, the critical distinction is that capital platform work relies on
the “sharing” of underutilised assets, such as accommodation, for financial gain, rather than the completion of work
(Schmidt, 2017). Consequently, the arrangement between capital platform workers and the gig organisation is more
akin to an e-commerce or business-to-business relationship. This is sometimes referred to as a form of micro-
entrepreneurship, in that these “workers” share greater similarity with small businesses than employees (Vandaele,
2018; Zervas, Proserpio, & Byers, 2017).
Contrastingly, crowdwork and app-work exist on platforms that allow workers to connect with customers to sell
their services by performing tasks. Crowdwork refers to work-mediating digital platforms, such as Amazon Mechani-
cal Turk and Fiverr, through which workers remotely complete tasks (De Stefano, 2016). In a typical crowdworking
scenario, an organisation or individual posts a task or project to be completed via a platform. Tasks can range from
software coding, to survey completion, to audio transcription. Any potential number of contributors can attempt to
undertake the task from any geographic location, with the most suitable crowdworkers selected (Berg, 2016). On
some crowdworking platforms, individuals may work simultaneously on the same task, with the customer selecting
and paying for only the best product (De Stefano, 2016). Different crowdwork platforms set minimum compensation
for certain tasks, and others let the compensation be set by their requester. In other cases, no clear relationship
DUGGAN ET AL. 5

exists between the customer and the worker: they complete the task autonomously and are paid by the platform,
which then provides the result to the customer. Crowdwork has proved popular for both workers and customers.
For example, figures show that 10,000 new tasks are published and 7,500 are completed per hour on Amazon
Mechanical Turk (Berg, 2016).
Although crowdwork is generally conceptualised as one form of work, it can be subdivided into four distinct
types. “Cloud-based crowdwork” is where tasks can be completed remotely via the Internet. If the task is not given
to a specific individual but to an undefined group of people online, it is “crowdwork.” If the task is further subdivided
into smaller units for piecemeal work, with each individual remunerated with an equally small amount of money, it is
“micro-tasking crowdwork.” If tasks cannot be subdivided but work is carried out simultaneously, by a large group of
individuals, and in the end only one result is used and paid for, it is “contest-based crowdwork” (Schmidt, 2017).
However, as all types of crowdwork allow for the completion of tasks remotely, thereby lacking a discernible
employer, we argue that it is less likely that these workers will develop a transparent working relationship with the
gig organisation (Berg, 2016).
The third variant of gig work, app-work, refers to service-providing intermediary digital platform organisations (or
“apps”) that utilise workers to perform tasks locally (e.g., transport and food-delivery) for customers who pay for
these services, with the organisation retaining a percentage of the exchange (De Stefano, 2016). An app (short for
“application”) is a software program designed to perform a specific function directly for the user on mobile devices,
most typically smartphones (Dickinson et al., 2014). App-work is a form of labour wherein the offering of traditional
working activities in local markets is conducted through apps, managed by intermediary digital platform organisa-
tions, that intervene in setting minimum quality standards of service and in the selection and management of individ-
uals who perform the work. Howcroft and Bergvall-Kåreborn (2019) suggest that app-work is a distinct form of
crowdwork (which they combine with capital platform work and label as “asset-based services”). We, however, argue
that when work is to be completed locally at a specific location and time, by an assigned worker who is managed and
subjected to minimum performance standards by a single intermediary digital platform, it is wholly distinctive to
crowdwork. In app-work, an algorithm quickly identifies and offers labour to one person; whereas in capital-platform
work and crowdwork, it is the customer or requester who decides and selects whose services to pay for.
The speed at which the working relationship is created in app-work is strikingly unique. That is, many workers
are hired almost instantly once they have agreed to the terms and conditions set out in the app and uploaded the rel-
evant documentation (e.g., driving licence and proof of identity). Within some digital platform organisations, individ-
uals must electronically accept the platform's terms each time they pursue work opportunities or gigs—in the case of
an Uber driver, for example, they must accept the associated terms and conditions each time they log in to the app
(Tran & Sokas, 2017). Consequently, app-work participants may, over time, develop a working relationship with digi-
tal platform organisations, due to the agentic relationship being more transparent.
App-work relationships are generally not rooted in traditional employee–employer dyads, but rather involve
multiple parties contributing to the dynamic exchange agreement. Trilateral, and sometimes quadrilateral, rela-
tions exist in app-working arrangements. For example, Deliveroo describes itself as offering a service that links
workers with partner restaurants to provide a food-delivery service to customers (Deliveroo, 2018). A customer
uses the Deliveroo app to order food from a participating restaurant or supplier. The app notifies the restaurant,
which prepares the order. At the same time, the app notifies a registered courier or worker that a delivery from
the restaurant to the customer is required. Thus, the four italicised parties indicate that this is a multiparty
working relationship. Indeed, all app-work involves a minimum of three parties, with platforms facilitating and
mediating transactions between workers, customers, and occasionally suppliers (e.g., partner restaurants, in the
case of Deliveroo). These multiple parties are illustrated in Figure 2. It is the app that connects and directs each
party in the work arrangement, thereby functioning as a centralised mechanism governing the dynamics of this
working relationship. The next section explores the algorithmic architecture of apps and the extent to which it
manages app-workers who utilise platforms.
6 DUGGAN ET AL.

F I G U R E 2 Parties involved in app-

work. *A broken line connects the supplier
to other parties to indicate that the
supplier only exists in certain app-working
arrangements

3 | ALGORITHMIC MANAGEMENT AND APP-WORK

Technology is changing how organisations manage work. The management practices found in app-work arrange-
ments borrow extensively from how digital media platforms, such as Facebook and Netflix, manage and influence
their users. An algorithm is a computational formula that autonomously makes decisions based on statistical models
or decision rules without explicit human intervention (Eurofound, 2018). It is a sequence of instructions telling a
computer what to do within a set of precisely defined steps and rules designed to accomplish a task. However,
instead of repeatedly processing a stable set of consistent instructions, algorithms rewrite themselves as they work.
Technological advances have allowed organisations to utilise artificial intelligence that simultaneously learn and solve
problems in increasingly complicated domains, from creating products to autonomously managing business processes
(Mann & O'Neil, 2016). Algorithms are presented as objectively and mathematically correct and, as a result, people
trust and abide by them (Lee, Kusbit, Metsky, & Dabbish, 2015). However, Jago (2019) finds that people believe
technological agents lack the same level of moral authenticity as human agents. Algorithms increasingly make deci-
sions that have tended to be the remit of managers and HR professionals. On hiring platforms such as LinkedIn, algo-
rithms sort through thousands of profiles to recommend promising job candidates to company recruiters (Carey &
Smith, 2016). Likewise, in the gig economy, algorithms match customers or requestors with workers (Lee et al.,
2015), and also function to evaluate worker performance on many platforms (Rosenblat, 2018).
Management-by-algorithm, or algorithmic management, is commonplace within app-work. We define algorithmic
management as a system of control where self-learning algorithms are given the responsibility for making and exe-
cuting decisions affecting labour, thereby limiting human involvement and oversight of the labour process. It replaces
some of the tasks and processes that workers typically engage with by using algorithms that are developed by the
very same individuals' data on the platform. Consequently, workers and consumers contribute, unremunerated, to
the stock of intangible capital of the platform. This has seen stratospheric valuations placed on app-work platforms
such as Lyft, valued at $24 billion at IPO in 2019 despite losing over $900 million in 2018 (Ovide, 2019).
DUGGAN ET AL. 7

For app-workers specifically, algorithms embedded within digital platforms govern the rules used to select and
manage labour when facilitating a transaction between a customer and worker (see Figure 2). A key feature of app-
work platforms is the automatic coordination and matching of the transaction through a set of advanced algorithms,
creating a space where supply and demand integrate through automatic management and enforced mechanisms
(Lehdonvirta, 2018). In other words, platforms use algorithms to match supply-and-demand in the market, while also
mediating and closely monitoring the work performed (Gandini, 2018). The use of algorithms allows platforms to
track workers' movements and assign work using workers' smartphone-based global positioning system (GPS). Algo-
rithms are used to undertake typical HR processes like work assignment and performance management without the
need for face-to-face interaction. Within this context of HR processes, a particularly apposite question concerns the
creation of the app itself. Certainly, as digital platform organisations continue to expand, marketing specialists, sys-
tem designers, programmers, and data scientists are recruited to work exclusively on these platforms (Meijerink &
Keegan, 2019). These core staff are effectively responsible for the design, maintenance, and development of the
app. In the absence of HR professionals to coordinate app-workers, policies for recruiting, monitoring, and managing
workers are mainly developed by these core staff (Meijerink & Keegan, 2019). In many ways, then, their role in the
organisation is akin to a rudimentary HR function, as the platform that they are responsible for overseeing manages
the working arrangement for app-workers.
Algorithmic management keeps marginal and labour costs relatively low (Schmidt, 2017), with considerable sav-
ings stemming from platforms functioning as virtual automated managers, thus negating the need for human supervi-
sors and managers (Lee et al., 2015). App-workers are typically managed via tracking mechanisms and customer
ratings, thus forming one of the fundamental principles of the gig economy in that most core HR processes (i.e., the
assignment of tasks and performance evaluation) are fulfilled by one of the two groups of users, the worker or the
customer, through the medium of the app-work platform (Schmidt, 2017). For example, Uber and Lyft drivers are not
directly supervised, nor are they required to wear a uniform or display organisational signage in their vehicles. How-
ever, the platforms have been known to provide drivers with precise instructions on cleanliness of their workspace,
how to behave with customers, and guidelines for maintaining proper hygiene (Steinberger, 2018). Likewise, both
platforms monitor the quality of work based on anonymous customer ratings (Steinberger, 2018).
Algorithmic management tracks, disciplines, and sets expectations for workers without human supervision or
recourse, often potentially to the detriment of workers' social protection (Vandaele, 2018). Through its very purpose,
algorithmic management eliminates the more interpersonal and empathetic aspects of people management. Without
an organisational partner advocating their needs and maintaining a balanced working relationship (Gilbert, De Winne,
& Sels, 2011), app-workers may lose trust and confidence, resulting in a reduced sense of well-being. As a means of
exerting control, algorithms essentially automate management practice. This, however, does not extend to protecting
workers or facilitating their supposed autonomy: platforms determine the eligibility of workers (e.g., the newness of
vehicles is a factor in working with Uber and Lyft); some platforms allow workers to rate customers, but these ratings
rarely influence customers' ability to use the platform, and levels of demand and time-pressures across many plat-
forms mean workers have less choice to work entirely on their own schedules (Kuhn & Maleki, 2017). Algorithmic
management also extends to pay, with many app-work platforms allowing workers to operate wherever they want.
The platform then relies on “economic nudges” (Rosenblat & Stark, 2016) in the form of algorithmically determined
surge-pricing (Gandini, 2018)—called “boosts”–to entice workers to areas of high demand, a process Woodcock and
Johnson (2018) term “gamification-from-above.”
The disparity that exists between those who design the app and those who physically provide the service it sells,
and between those who offer the infrastructure for labour but no stability or benefits to accompany it, are a defining
feature of many app-work platforms. Algorithmic management therefore facilitates asymmetric information in the
working relationship by controlling the supply of labour, targeting different workers with variable incentives, remov-
ing workers from platforms without remedy, and mediating disputes at its discretion. Slee (2017) compares facets of
the algorithmic management structure to the “boss from hell,” describing it as erratic and bad-tempered with the
potential to fire workers on a whim with no recourse to appeal. The nascent literature is gradually addressing
8 DUGGAN ET AL.

algorithmic management and control, but considerable knowledge gaps remain in terms of the employment relations
and HRM implications of this approach, to which we now turn.

4 | A P P - W O R K ' S I M P L I C A T I O N S F O R EM P L O Y M E N T R E L A T I O N S A N D
HRM: A RESEARCH AGENDA

Algorithmic management, as one of the defining features of app-work, significantly impacts on the functioning of the
working relationship. Although the relationship initially appears to be an entirely transactional, economic exchange,
recent research on the nuances of app-working arrangements signals the potential existence of a more complex, tex-
tured psychological contract, wherein app-workers view their association with the digital platform organisation in a
broader, more relational sense (Ashford, Caza, & Reid, 2018). For instance, some app-workers are reported as seek-
ing professional development opportunities from organisations (Graham, Hjorth, & Lehdonvirta, 2017), social interac-
tion and support from colleagues and managers, and mentoring from more senior colleagues (Ashford et al., 2018).
This aligns with emerging research on the psychological contract, which calls for a “multifoci” perspective to better
understand the individualised nature of contemporary working arrangements (Alcover, Rico, Turnley, & Bolino,
2017). This indicates that modern working relationships, like those found in app-work, are not simply based on an
agreement between two parties but are instead derived from the multiple parties in the employment network.
Of course, one of the parties in this working arrangement is the app itself, which, as explained, is controlled by
the organisation and utilises algorithmic management. Organisations do not have psychological contracts, but their
human agents do (Rousseau, 1995). When one examines employment relationships in app-work, the agency question
becomes especially apposite. Emerging research highlights that Uber drivers successfully make demands of the app
in terms of encouraging and facilitating tipping from customers (Kuhn & Maleki, 2017; Riesman, 2014) or improved
performance ratings for engaging in citizenship behaviour (e.g., returning a passenger's lost item; Rosenblat & Stark,
2016). Likewise, Ravenelle (2019) argues that the capricious nature of apps may impact workers' experiences of psy-
chological contract violation or fulfilment. Findings like this support the argument that the app-worker, in part at
least, may develop expectations, even if the digital platform organisation rejects any notion of individual workers
being employees.
Despite this, the fragmented nature of app-work, through its reliance on technology via digital platforms and
governing algorithms, may erode the reciprocity found in traditional employment relationships. Therefore, app-work
challenges our understanding of HRM concepts and practices. Certainly, it is evident that established practices such
as employment relations, work assignment processes, and performance management exist in the app-work context.
However, the overall approach towards HRM appears to differ significantly from established models, in both the
context of its strategic purpose and the way in which activities are delivered and implemented.
App-work seems to pass the risks of employment almost entirely onto individual workers, predicated by the
organisational view that these workers are independent, self-employed contractors. Thus, on the surface, the denial
of employment status and associated rights for app-workers indicates little relevance for HRM as a profession and
function. However, with multiple court rulings making determinations in favour of individual workers gaining employ-
ment status, HRM naturally becomes increasingly relevant. Notwithstanding this, we highlight that HRM practices
are implemented for app-workers, despite the lack of an official HR function. We now critically discuss several of
these key areas of HR practice, commencing with considering the nature of employment relations in app-work, and,
in so doing, outline worthwhile research questions for future scholarly enquiry.

4.1 | Employment relations

As the term implies, the centrality of the “app” in app-working relationships is absolute. That the algorithm underpin-
ning the app is designed, developed, and implemented heavily or entirely by marketing specialists and system
designers within organisations, with little, if any, worker contribution, challenges established thought on the HR
DUGGAN ET AL. 9

function's role in effectively managing working relationships (Meijerink & Keegan, 2019). Although gig organisations
have a HR function, this largely serves the core staff (i.e., employees) of these firms, rather than app-workers. This
raises many challenges around the existence of an employment relationship and its dynamics, while also raising ques-
tions for the future role of the HR function within the gig economy's business model.
A heavily transactional relationship appears to be at the heart of app-work, with workers paid for the quantity of
work undertaken rather than the time spent working. The seemingly nonexistent focus on the development of
mutual trust and commitment in the working relationship further solidifies the transactional nature of this exchange.
At the recruitment stage, most roles are advertised on the basis that workers have the autonomy to work when they
wish with considerable independence, indicating that there are little or no expectations of a long-term relationship
unless desired (Jabagi, Croteau, Audebrand, & Marsan, 2019). Likewise, workers are typically onboarded quickly, via
a prompt screening process, ensuring a readily accessible source of labour for the organisation (Kuhn & Maleki,
2017). Although this approach to onboarding workers certainly reduces costs by eliminating many of the labour and
time costs involved (Healy, Nicholson, & Pekarek, 2017), it may prove problematic when looking beyond short-term,
transactional cost benefits towards recruiting motivated workers who are likely to succeed in roles. Traditionally,
organisations seek to strategically identify workers who “fit” well with the organisation, as a means of improving and
sustaining performance in the long term (Ahmad & Schroeder, 2002). Fit theory argues that workers who recognise
that their values align with those of the organisation are more likely to be intrinsically motivated, which, in turn,
impacts on positive work outcomes and intentions to remain with the organisation (Ahmad & Schroeder, 2002).
However, app-work predominantly ignores this concept, in terms of both “person-job” and “person-organisation” fit
(Carless, 2005). For example, there is an evident lack of consideration of cultural fit and a limited consideration of
technical expertise, with seemingly little interest in worker retention on a long-term basis (Friedman, 2014). Of
course, the socialisation literature identifies the importance of determining the alignment between employee and
organisation early in the working relationship (Woodrow & Guest, 2017). Perhaps if gig organisations made more of
a concerted effort to ensure fit between worker and platform, they would move towards reducing labour turnover
and retaining its elite workers (Campbell, 2018), although their business model currently appears unconcerned with
such matters.
Furthermore, given the focus on the completion of specific, narrow tasks within most app-work, investment in
training and development opportunities for workers is severely limited or nonexistent (Meijerink & Keegan, 2019). In
some cases, workers attend an information session or watch online induction videos prior to commencing work. This,
in effect, acts as the entirety of the training and socialisation processes. As such, aspects of app-work can be viewed
as bearing resemblance to Taylorism and Fordism, wherein work is fragmented into on-demand tasks and workers'
performance is measured closely (McGaughey, 2018). Furthermore, the crucial role of algorithmic management at
the core of app-work arguably allows greater complexity and management of tasks and people than established,
long-standing concepts (Wood, Graham, Lehdonvirta, & Hjorth, 2019). This raises several research questions per-
taining to the transactional foundations upon which app-work relationships are formed. For example, are digital plat-
form organisations entirely unconcerned with recruiting workers that fit well within the organisation? If so, how
sustainable is the gig economy's business model in terms of retention and high worker turnover, particularly as most
organisations withhold any development opportunities?
Although the transactional aspect of app-work is clearly evident, we argue that this conceptualisation may under-
appreciate the inherently nuanced structure of app-work, which forms a working relationship with a minimum of
three parties involved (Wood & Lehdonvirta, 2019). In particular, arguments that app-work relationships are centred
solely around economic gain should be challenged, particularly as this ideal may be perceived differently by each
party involved. For example, emerging evidence signals that app-workers are seeking a more codetermined and rela-
tional work arrangement, populated by opportunities to develop new skills useful in furthering their careers, in addi-
tion to craving social interaction and networking opportunities (Graham et al., 2017; Petriglieri, Ashford, &
Wrzesniewski, 2018). Consequently, psychological contract theory may be useful in this context, as the perception
of each party is of critical importance and the terms may differ significantly through each individual's “eyes”
10 DUGGAN ET AL.

(Rousseau, 1995). This theory appreciates that one party can perceive a particular relationship and expectations that
the other does not recognise. Baruch and Rousseau (2019) note that the norm of reciprocity, found in traditional
employment relationships, may not always be clear when multiple parties are involved, each holding different posi-
tions and resources. This can conceivably be problematic if one party feels that a violation or breach has taken place.
Thus, although a legal employment contract may be denied by the gig organisation, it is feasible that psychological
contracts exist and develop, at least from the worker's perspective. The idea that app-workers only seek to work for
financial gain appears at odds with recent research and wider scholarship on the sociology of work, around the rele-
vance of relational arrangements (Wood & Lehdonvirta, 2019; Woodcock & Johnson, 2018).
Shared understandings and reciprocal contributions for mutual benefit are at the core of traditional, functional
exchange relationships between workers and organisations (Dabos & Rousseau, 2004). From the app-worker's per-
spective, having no say in how work is assigned and how performance is assessed means that the working relation-
ship is less representative of one that has been mutually codetermined and could instead be perceived as a working
arrangement of subjugation (Harvey et al., 2017). According to organisational support theory, employees interpret
organisational policies, practices, and treatment—such as HRM—as indicators of the organisation's support and com-
mitment to them (Vanhala & Ritala, 2016). But, for app-workers, traditional understandings around reciprocity and
organisational support no longer apply or, at a minimum, are considerably different. In its role as intermediary, the
digital platform organisation is the only party with full access to and control over the data, processes, and rules of
the platform (Jabagi et al., 2019).2 If this trend is indicative of future developments in the world of work, it represents
a significant shift from established conceptualisations of the employment relationship. Again, this form of labour
management raises several worthwhile research questions: Besides the obvious financial implications, why do gig
organisations typically adopt a “hard” approach to managing their workers? What effect would a more codetermined
work arrangement, whereby workers have a greater voice, have on the working relationship?
The relationship between app-workers and their organisations strays significantly from a conventional employ-
ment relationship. With multiple, unique sources of dependence across at least three parties, app-work constitutes a
working relationship in which platforms simultaneously generate dependence and determine the rules that shape,
afford, and limit worker agency (Wood & Lehdonvirta, 2019). Likewise, for most app-workers who encounter issues
of any type in their roles, their primary point of contact with the platform is by email. In the absence of a human
manager, enquiries are handled by support representatives, the email equivalent of a call centre, often managed by
third parties. Responses to queries, issues, or concerns are typically met with automated responses based on an algo-
rithmic assessment of key words within emails (Rosenblat, 2018). Automated responses may be appropriate for fac-
tual enquiries but lack the ability of a human to understand and respond to nuanced issues. As debate increases
around this complex reimagining of work, controversy surrounding working conditions and workers' rights has led to
legal issues and attempts at collective industrial action. Despite the provision of some autonomy and flexibility for
app-workers, ongoing power asymmetries and struggles undoubtedly exist between workers, those customers
requesting labour, and the platforms that act as both intermediaries and algorithmic managers by enabling and man-
aging work processes (Lampinen, Lutz, Newlands, Light, & Immorlica, 2018).
The industrial relations literature has long focused on the power dynamics of the employment relationship
(Atchison, 1991; Fudge, 2017; Smith, 2006). In comparison to more traditional employment relationships, the power
distance between digital platform organisations and app-workers appears to be weighted in favour of the former. In
most cases, app-workers represent the party with the least significant power in many aspects of the working
arrangement. App-workers are subjected to sophisticated forms of control, continuous surveillance, and anonymous
customer ratings, with organisations and customers holding the most dominant positions (Pichault & McKeown,
2019). Most platforms and apps are designed to give the appearance of offering choice to the worker, but this is usu-
ally executed in a manner wherein platforms directly shape the engagement in such a way that workers are disadvan-
taged by not agreeing, that is, penalising workers for rejecting tasks or for not working during peak periods
(Flanagan, 2018). Likewise, many app-work platforms fully exploit the possibility to minutely monitor the activities of
app-workers in real time, rendering them heavily at the mercy of the platform's instructions, expectations, and
DUGGAN ET AL. 11

demands (Kuhn & Maleki, 2017). This creates an “always-on” form of control that some workers perceive as unfair,
given their employment classification (Gandini, 2018). This has significant implications, in that research already indi-
cates that when a power imbalance exists between two or more parties, those in the more vulnerable position find it
difficult to have their voice heard (Mowbray, Wilkinson, & Tse, 2015), resulting in perceptions of unfair treatment
(Van Buren & Greenwood, 2008). Algorithmic management fragments work into individual tasks, closely monitors
workers, and links pay to performance as a means of segregating workers (Healy et al., 2017), leading scholars to lik-
ening the gig economy to a modern-day, digitalised Taylorism (McGaughey, 2018).
The well-being of app-workers has emerged as a concern, with app-work frequently characterised by health and
safety issues, inadequate social protection, troubling management structures, and a lack of investment in workers
(Vandaele, 2018). Recent research reports that some app-workers feel exploited at the hands of organisations as a
consequence of having little input into how work is assigned or completed (Wood et al., 2019). Accordingly, worker
morale and well-being undeniably decline (Fleming, 2017). Some digital platform organisations refute these claims,
arguing that they provide workers with a voice through in-house surveys and support groups (Dupont et al., 2018).
Uber, for instance, contends that it provides opportunities for involvement to its workers through arranging quality
improvement courses where workers can have their voice heard (Rosenblat, 2018). Likewise, if drivers are at risk of
deactivation from the platform due to low customer ratings, Uber sometimes offers a training course to facilitate
improvement. However, there is a fee for attending. Effectively, the cost of low ratings for drivers is to pay for your
own training. From the organisation's perspective, initiatives like this are in lieu of recognised HR processes, such as
training or appraisals, and serve as the most appropriate means of managing the unique nature of app-work (Kuhn &
Maleki, 2017). Such features intimate that HRM in app-work is not focused on the maintenance of a relationship
between employer and employee, but rather on control and coordination (Meijerink & Keegan, 2019). It can there-
fore be argued that through algorithmic management, the traditional notion of the “tyranny of the clock” has been
replaced by the “tyranny of the algorithm” (Lehdonvirta, 2018).
The power imbalance in employment relations was central to the origins and rise of the union movement and col-
lective bargaining. Collective bargaining is a process whereby workers come together to address pressures for under-
cutting the price of each other's labour. Although collective action and reform is an option for gig workers, the very
nature of app-work makes this especially difficult: the manager is an algorithm, coworkers are independent contrac-
tors (potentially geographically dispersed and in competition with each other), and the work is often carried out in
isolation or in contact only with the customer (Prassl, 2018). This undoubtedly complicates the organisation of collec-
tive action, as workers can be difficult to engage with, find and reach. Despite varied success, attempts have
highlighted that collective industrial action is far from structurally impossible for app-workers (Schiek & Gideon,
2018). Many trade unions have sought to engage with app-workers, most famously including GMB in the UK and
the New York Taxi Worker Alliance, both of whom have brought multiple cases against Uber (Johnston & Land-
Kazlauskas, 2018). Additionally, there are a host of ongoing legal cases proceeding through the courts surrounding
app-work platforms, with the majority seeking to determine who is an “employee” and who is a “worker”
(McGaughey, 2018).

4.2 | Work assignment

Algorithmic management dictates how work is assigned and commences as soon as app-workers log-in to the plat-
form. The supposed nonfinancial reward of app-work is that workers are afforded a high degree of flexibility and
autonomy by choosing when and where to work (Healy et al., 2017). The realities of these claims are, we argue,
highly debatable. It is an algorithm that identifies the worker who is better or more quickly able to cater to a certain
customer, thereby allocating gigs amongst workers to ensure the highest level of speed and efficiency (Gramano,
2019). Most app-work platforms suggest that they are merely in the business of “matching” workers with customers
that request services (Prassl, 2018). However, this portrays a further power imbalance. For example, although app-
workers operating on courier and ride-hailing platforms are not paid while waiting to be assigned tasks, platforms still
12 DUGGAN ET AL.

gather and generate useful data and analytics from workers on “dead miles” (i.e., driving while waiting for tasks to
appear, but not being paid to do so), thereby enhancing its algorithms (Thomas, 2018). Accordingly, with potential
issues of a power imbalance in the context of work assignment, what are the ethical implications of the intense algo-
rithmic control exercised by platforms? More specifically, how transparent is the process by which work assignment
algorithms are formed and updated using app-workers' data?
Although regularly advertised as a key benefit of app-work roles, the realities of the autonomy in choosing when
to work and which tasks to accept are more limited than often claimed (Wood et al., 2019). Most platforms employ a
range of strategies to ensure that jobs are accepted as quickly as possible. For example, ride-hailing platforms often
do not reveal passengers' destinations to drivers until after the trip has been accepted (Cunningham-Parmeter, 2016;
Prassl, 2018). Likewise, many platforms are known to penalise workers who do not maintain high acceptance rates
for tasks offered. For example, Uber has been known to designate drivers as unavailable for work if three consecu-
tive rides are declined (Steinberger, 2018). Uber also stimulates oversupply in times of potentially high demand by
algorithmically urging drivers to temporarily relocate and benefit from “surge-pricing.” During periods of excessive
demand when there are not enough drivers on the road and customer wait times are long, Uber increases its normal
fares. Drivers, however, may be offered “non-surge” priced trips once they arrive at the location and may subse-
quently be penalised if they decline the trip, despite having absorbed the costs of travelling to the location (Ros-
enblat, 2018). Algorithms also identify future periods of potentially high demand, with drivers notified via the app
that, for example, they are likely to earn above average fares if they work the upcoming Friday and Saturday. If the
monetary benefit is not quite as substantial as promised, this may constitute a breach of their psychological contract.
The gig economy has pioneered algorithmic techniques of worker control, such as behavioural nudges and sched-
uling prompts, based on volatile real-time and predictive analytics that produce economic incentives intended to allo-
cate workers to high-demand areas (Van Doorn, 2017). In theory, workers can choose when to work and they can
also operate on more than one platform simultaneously, which may pose health and safety concerns given the lack
of oversight on hours worked. However, many app-workers are encouraged to keep high acceptance rates through
occasional promotions that offer higher pay (Lee et al., 2015). Likewise, when a worker is ready to finish work, algo-
rithmic control again becomes apparent, as workers on some platforms are often enticed to earn more money by
staying active on the app and customer demand is high (Prassl, 2018).
As supposed independent contractors, app-workers should be free to avail of promises to choose when to work
and which tasks to accept. Yet, for most app-workers, platform control via algorithmic management heavily limits
autonomy. In reality, workers must work long hours and at peak times to garner high earnings and maintain good rat-
ings and are often left with little choice but to accept whichever tasks are offered (Prassl, 2018). Thus, for most plat-
forms, rather than simply “matching” workers and customers, they instead act as digital work intermediaries that use
algorithms to tightly manage a large, invisible workforce (Prassl, 2018). Therefore, any ideal of individuals possessing
full flexibility of their own work schedule may be rare, though empirical evidence is required to establish how exten-
sive this is. Accordingly, we pose the following research questions: As worker autonomy often appears more limited
than advertised, what are app-workers' experiences of work assignment processes? Is there variability in these expe-
riences across different organisations? Likewise, if app-workers are penalised or disadvantaged for exercising the
autonomy promised to them, what are the implications of this potential psychological contract breach for the work-
ing relationship?

4.3 | Performance management

The means through which platforms exercise algorithmic control over app-workers' performance are many and var-
ied, and, as a result, are central in approaches to performance management. Once “active” on an app, platform con-
trol and associated monitoring takes hold in an intensive way (Prassl, 2018). For example, GPS systems can be used
to monitor the speed and position of vehicles, along with tracking every movement of app-workers, and, conse-
quently, can verify what an individual is doing. The Uber app tracks workers' GPS locations and has acceleration
DUGGAN ET AL. 13

sensors built into the driver's version of the app to detect heavy braking and speeding (Prassl, 2018). These data,
along with customer ratings and reviews, are then used to identify the best performing workers and to alter the algo-
rithm that assigns tasks to workers. However, much like the work assignment processes, there appears to be limited
transparency and understanding about how exactly these algorithms are altered, by whom, and under whose
instruction.
Several platforms issue performance metrics to app-workers, which often include comparisons to other workers
and to overall performance rankings, as a means of ensuring high levels of productivity (Van Doorn, 2017). Addition-
ally, most platforms utilise customer ratings of workers via anonymous systems as a means of performance evalua-
tion. Before being shared with workers, assessments are forwarded to platforms, who have the opportunity to
verify—albeit indirectly—the quality and punctuality of the service rendered (Gramano, 2019). The importance of
these ratings ultimately leads to further informal or “soft” platform control (Wood et al., 2019), by way of driver
behaviours in terms of offering customers bottled water, phone chargers, and similar “extras” in order to receive high
ratings. All such extras are at the expense of individual workers.
“Harder” platform control approaches are utilised to create high-performance expectations and the need to sat-
isfy organisational and customer needs. For example, Uber previously operated immediate 10-minute deactivation
periods for drivers who repeatedly refused less profitable trips (Prassl, 2018). This sanction is derived through data
generated via the platform, and as such is automatic and immediate. These approaches appear to be highly effective
in eliciting strong performance levels, with workers striving to maintain above-average ratings for fear of being
deactivated from platforms (Kuhn & Maleki, 2017). The pursuit of high-performance levels may come at the expense
of worker safety, with recent research highlighting the implications of pressurised working environments on some
app-workers' health and well-being (Wood et al., 2019). Consequently, regulation may be required to mitigate against
the potentially detrimental effects of monitoring on app-work platforms.
As the platform has the opportunity to exercise penetrating control over all aspects of the service delivery, the
worker can be described as being in a situation comparable to a permanent probationary period (Gramano, 2019).
Indeed, such is the potential intensity of monitoring and control in app-work that it could be described as possessing
characteristics of an electronic panopticon (Prassl, 2018) that goes far beyond what one thought possible in the past
(Dagnino, King, & Tienari, 2017). The panopticon, associated with the philosopher Bentham, refers to a central tower
that can observe occupants without them knowing fully whether they are being watched. This may lead to an
assumption amongst workers that they are under constant observation, thereby impacting their behaviour.
A potential danger arising from the use of algorithms in managing workers is that such extensive data become
viewed as a reliable and objective truth, eroding any scope for human interpretation of the nuances of workplace
behaviour. This is problematic, given the inherent weaknesses in simplistic and entirely quantitative measures of per-
formance. Likewise, we are also learning that algorithms can contain significant biases (Guszcza, Rahwan, Bible,
Cebrian, & Katyal, 2018). Given the fundamental role of customer ratings, and that workers in receipt of low ratings
are potentially subjected to being deactivated for future work (Tran & Sokas, 2017), such concerns are magnified.
For example, in some cities, Uber drivers with an average rating, calculated by an algorithm, of lower than 4.6 out of
a possible 5 are at risk of disbarment from the platform (Kuhn & Maleki, 2017). These levels of control perhaps pro-
vide context as to why the majority of drivers do not seek to remain with the organisation in the long-term, as exem-
plified by reports suggesting that 11 per cent of drivers become inactive within the first month, and only 4 per cent
remain on the platform for longer than one year (McGee, 2017).
When app-workers are disbarred, the reasoning is often unclear, as is any potential scope for reinstatement with
the lack of an appeals process. Indeed, the way that sanctions are applied based on ratings is often unpredictable
and arbitrarily employed (Prassl, 2018). The lack of human interaction in seeking to address such issues is notewor-
thily absent and troubling (Kirven, 2018). Given that app-workers are unlikely ever to actually meet their “employers”
or a line manager, performance management in the gig economy is a substantially different concept compared with
more traditional work roles (Barley, Bechky, & Milliken, 2017). This leads us to ask what processes, if any, exist to
allow workers to challenge customer evaluations? The accuracy of these processes raises concerns for worker
14 DUGGAN ET AL.

protection and for allowing recourse to a perceived unfair rating (Vandaele, 2018). If ratings are treated as sacrosanct
and app-workers have no ability to provide context or an alternative account, their earning capacity is ultimately at
the peril of unknown customers.
There are many occasions where an app-worker's performance evaluation may be strongly impacted by circum-
stances outside of their control, given that multiple parties exist in the working relationship and success is dependent
on all parties fulfilling their obligations (Alcover et al., 2017). Consider the working arrangement within food-delivery
platforms, such as Deliveroo or UberEats, for example. For illustration, if the restaurant supplied the courier with the
wrong order, he or she would then be confronted with an unhappy customer upon delivery. If the courier had not
been sufficiently trained to deal with difficult customers, he or she may perceive that this error and subsequent
aggravation from the customer constitutes a violation of the exchange agreement. This encounter may then disrupt
delivery metrics or result in a complaint, and as a result the worker's performance report is impacted (Kuhn & Maleki,
2017), potentially resulting in the worker voluntarily ending his/her relationship with the platform or being less likely
to receive future gigs. From the organisation's perspective, an error from another party in the working arrangement
could result in the untimely exit of a worker or in less use being made of a strong performer.

5 | C O N C L U S I O N: L O O K I NG F O R W A R D

The gig economy is a small, but growing component of the global labour market. This “new” economy has been
embraced by customers. Although welcomed by some workers, it does, however, represent a further trend towards
precarious working. Existing conceptualisations of gig working have, we argue, resulted in some misleading classifica-
tions and conflation of what gig work entails and the size of this economy. Having identified the crucial role of tech-
nology in facilitating work in the gig economy, we propose that distinguishing app-work from other variants of gig
work represents an important first step to a more coherent typology of gig workers, which, in turn, is central to bet-
ter understanding the issues such workers face and the implications for HRM and the employment relationship.
Moreover, this is also important for public policy-making bodies, who are being asked to consider appropriate pro-
tections in several jurisdictions (Berg, 2016; Friedman, 2014).
Although most gig organisations deny the existence of an employment relationship, the strict levels of control
enforced by many app-work organisations via algorithmic management raises noteworthy contradictions that war-
rant developed empirical consideration (Kuhn & Maleki, 2017). Conflict between app-workers and organisations on
issues surrounding control, dependency, and working conditions, for example, would suggest that there is more to
the working relationship than simply remuneration and flexibility (Meijerink & Keegan, 2019). In other words, app-
workers may view their working relationship beyond a merely economic or transactional exchange. Although scholar-
ship is beginning to consider and address issues around the employment relationship and HRM in the gig economy,
there is an undertheorisation, which is not unexpected in a new domain such as this. According to Jabagi et al.
(2019), the detached and distributed nature of the gig economy signals a radical reinvention of work, embodied by a
significant shift towards novel management tools enabled by technology. Although digital platform organisations typ-
ically position themselves as providers of an app, functioning simply as a neutral technology platform to match
workers with customers, the reality appears more likely to be that most platforms act as digital work intermediaries
in the business of tightly managing a large, invisible workforce (Prassl, 2018). From an employment relations and
HRM perspective, the gig economy creates a new dynamic where these technologies do not just mediate economic
and social relation, but also help to coconstitute them (Wood & Lehdonvirta, 2019).
The practice of HRM, serving to effectively manage people within organisations, has typically been based on the
notion of traditional employment. In the absence of a conventional employer–employee relationship, app-work
arrangements disrupt this conceptualisation. Despite the lack of a discernible employment relationship, gig work still
incorporates HRM-like activities, including, but not limited to, managing working relationships, work assignment pro-
cesses, and performance management and evaluation (Meijerink & Keegan, 2019). The preceding discussion has, for
DUGGAN ET AL. 15

example, considered how the working relationship in app-work is established and managed in ways that raise con-
cerns around power imbalances, fairness, and worker well-being, and in a way that places little value on the develop-
ment of a deep relationship with associated levels of commitment (Healy et al., 2017). Similarly, work assignment
processes form a complex reimagining, with algorithmic mechanisms contradicting claims of supposed flexibility and
autonomy (Collier et al., 2017; Harris, 2017). Likewise, with its primary focus on quantitative customer ratings and
tracking mechanisms, performance management is strikingly different from, but no less important to our traditional
understanding. App-workers are subject to tight levels of control, where the customer evaluation is often critical
(Healy et al., 2017), and without recourse to question or refute performance scores. This arguably results in a gener-
ally one-sided process (Flanagan, 2018), which leads to a further shift in power away from workers.
The implications of app-work and the pervasiveness of algorithmic management for the HR profession need
greater consideration. Some have noted that the gig economy redefines the role of HRM away from upholding
employment relationships towards the governance of exchanges on platforms that serve to allow the co-creation of
value for all parties involved (Meijerink & Keegan, 2019). Others question the long-term viability of the role of HRM
in the gig economy, particularly within the strategic context of motivating workers, ensuring high-quality perfor-
mance, and providing social support (Jabagi et al., 2019). With algorithmic management replacing human supervision
across the majority of platforms, organisations must instead rely on workers to self-organise and self-motivate to
advance task performance and organisational goals (Jabagi et al., 2019). HRM activities tend to be outsourced to the
system designers that develop and manage algorithms, which removes many of the costs of HRM and passes
employment risks to the individual (Snyder, 2016). This data-dominated approach moves work into an inhuman form,
with algorithms undertaking roles that were traditionally the preserve of HR professionals. The concomitant rise of
interest in big data and HR analytics may further indicate the removal of the more interpersonal and empathetic
aspects of people management (Angrave, Charlwood, Kirkpatrick, Lawrence, & Stuart, 2016). This raises questions
around what the HR function serves to do, or how it should act, in these organisations (i.e., a HR function that serves
permanent staff within the organisation, but has no responsibility for “independent” gig workers)? Although gig work
may bring increased organisational productivity, how ethical and appropriate may this be? With the tendency to
refuse employee rights, HRM may be sleepwalking into the eradication of ethical responsibility (Greenwood, 2013),
for example, in relation to the level of protection and use of individual data, transparency and accountability of algo-
rithmic processes, and worker well-being. All things considered, the rise in app-work and the all-pervasive role of the
algorithm brings a need for a more critical consideration of the consequences of this for employment relations and
of how HRM, as a function and discipline, should react and evolve.

ENDNOTES

1
There is a lack of clarity as to what precisely gig work entails and this in turn means that estimates of the size of the gig
economy are difficult to corroborate and compare. Different estimates exist because of significant variation in definitions
of what is included or excluded in the gig economy. Consequently, caution is necessary in interpreting the size, scope, and
spread of the gig economy.
2
In March 2019, a number of Uber drivers in the United Kingdom initiated legal action under the General Data Protection
Regulation, requiring that the company provide data it holds about them. Specifically, the drivers want to know how Uber
uses its algorithms to assign work and to track working time. The drivers have requested this data to enable them to accu-
rately calculate holiday pay and other pay owed.

ORCID

James Duggan https://orcid.org/0000-0002-1928-4723

Ultan Sherman https://orcid.org/0000-0002-2639-746X
Ronan Carbery https://orcid.org/0000-0001-8836-2038
Anthony McDonnell https://orcid.org/0000-0002-9127-2452
16 DUGGAN ET AL.

RE FE R ENC E S

Ahmad, S., & Schroeder, R. G. (2002). The importance of recruitment and selection process for sustainability of total quality
management. International Journal of Quality & Reliability Management, 19, 540–550. https://doi.org/10.1108/
02656710210427511
Alcover, C. M., Rico, R., Turnley, W. H., & Bolino, M. C. (2017). Multi-dependence in the formation and development of the
distributed psychological contract. European Journal of Work and Organisational Psychology, 26, 16–29. https://doi.org/
10.1080/1359432X.2016.1197205
Angrave, D., Charlwood, A., Kirkpatrick, I., Lawrence, M., & Stuart, M. (2016). HR and analytics: Why HR is set to fail the big
data challenge. Human Resource Management Journal, 26, 1–11. https://doi.org/10.1111/1748-8583.12090
Ashford, S. J., Caza, B. B., & Reid, E. M. (2018). From surviving to thriving in the gig economy: A research agenda for individ-
uals in the new world of work. Research in Organizational Behaviour, 38, 23–41. https://doi.org/10.1016/j.riob.2018.
11.001
Atchison, T. J. (1991). The employment relationship: un-tied or re-tied? Academy of Management Perspectives, 5(4), 52–62.
https://doi.org/10.5465/ame.1991.4274750
Barley, S. R., Bechky, B. A., & Milliken, F. J. (2017). The changing nature of work: Careers, identities, and work lives in the
21st century. Academy of Management Discoveries, 3, 111–115. https://doi.org/10.5465/amd.2017.0034
Baruch, Y., & Rousseau, D. M. (2019). Integrating psychological contracts and ecosystems in career studies and manage-
ment. Academy of Management Annals, 13, 84–111. https://doi.org/10.5465/annals.2016.0103
Berg, J. (2016). Income security in the on-demand economy: Findings and policy lessons from a survey of crowdworkers.
ILO: Conditions of Work and Employment Series No. 74. Available from http://www.ilo.org/wcmsp5/groups/public/—
ed_protect/—protrav/—travail/documents/publication/wcms:479693.pdf
Bernhardt, A., & Thomason, S. (2017). What do we know about gig work in California? An analysis of independent
contracting. UC Berkeley Labour Centre. Available from http://laborcenter.berkeley.edu/what-do-weknow-about-gig-
work-in-california/
Bonet, R., Cappelli, P., & Hamori, M. (2013). Labour market intermediaries and the new paradigm for human resources. The
Academy of Management Annals, 7, 341–392. https://doi.org/10.1080/19416520.2013.774213
Campbell, H. (2018). The Rideshare Guide. New York: Simon & Schuster.
Cappelli, P., & Keller, J. R. (2013). Classifying work in the new economy. Academy of Management Review, 38, 575–596.
https://doi.org/10.5465/amr.2011.0302
Carey, D., & Smith, M. (2016). How companies are using simulations, competitions, and analytics to hire. Harvard Business
Review. Available from https://hbr.org/2016/04/how-companies-are-using-simulations-competitions-and-analytics-to-hire
Carless, S. A. (2005). Person-job fit versus person-organisation fit as predictors of organisational attraction and job accep-
tance intentions: A longitudinal study. Journal of Occupational and Organisational Psychology, 78, 411–429. https://doi.
org/10.1348/096317905X25995
Carr, C. T., Hall, R. D., Mason, A. J., & Varney, E. J. (2017). Cueing employability in the gig economy: Effects of task-relevant
and task-irrelevant information on Fiverr. Management Communication Quarterly, 31, 409–428. https://doi.org/10.
1177/0893318916687397
Collier, R. B., Dubal, V. B., & Carter, C. (2017). Labour platforms and gig work: The failure to regulate. Institute for Research
on Labour and Employment: Working Paper 106–17.
Connelly, C. E., & Gallagher, D. G. (2004). Emerging trends in contingent work research. Journal of Management, 30, 959–
983. https://doi.org/10.1016/j.jm.2004.06.008
Cunningham-Parmeter, K. (2016). From Amazon to Uber: Defining employment in the modern economy. Boston University
Law Review, 96, 1673–1728. Available from https://search.proquest.com/docview/1851662998?accountid=14504
Dabos, G. E., & Rousseau, D. M. (2004). Mutuality and reciprocity in the psychological contracts of employees and
employers. Journal of Applied Psychology, 89, 52–72. https://doi.org/10.1037/0021-9010.89.1.52
Dagnino, G. B., King, D. R., & Tienari, J. (2017). Strategic management of dynamic growth. Long Range Planning, 50,
427–430. https://doi.org/10.1016/j.lrp.2017.06.002
de Stefano, V. (2016). The rise of the ‘just-in-time’ workforce: On-demand work, crowdwork and labour protection in the
gig economy. International Labour Office: Conditions of Work and Employment Series, 71. Available from http://
heinonline.org/HOL/P?h=hein.journals/cllpj37&i=507
Deliveroo (2018). Legal information. Available from https://deliveroo.ie/legal
Dickinson, J. E., Ghali, K., Cherrett, T., Speed, C., Davies, N., & Norgate, S. (2014). Tourism and the smartphone app: Capabil-
ities, emerging practice and scope in the travel domain. Current Issues in Tourism, 17, 84–101. https://doi.org/10.1080/
13683500.2012.718323
Dupont, J., Hughes, S., Wolf, R., & Wride, S. (2018). Freedom and flexibility: the relationship Deliveroo riders have with the
labour market. London: Public First. Available from https://www.scribd.com/document/392753154/Public-First-
Freedom-and-Flexibility#download
DUGGAN ET AL. 17

Eurofound (2017). Non-standard forms of employment: recent trends and future prospects. Dublin: Eurofound. Available from
http://eurofound.link/ef1746
Eurofound (2018). Automation, digitalisation and platforms: Implications for work and employment. Luxembourg: Publications
Office of the European Union. Available from https://www.eurofound.europa.eu/sites/default/files/ef_publication/
field_ef_document/ef18002en.pdf
Fabo, B., Karanovic, J., & Dukova, K. (2017). In search of an adequate European policy to the platform economy. Transfer:
European Review of Labour and Research, 23, 163–175. https://doi.org/10.1177/1024258916688861
Flanagan, F. (2018). Theorising the gig economy and home-based service work. Journal of Industrial Relations, 61, 57–78.
https://doi.org/10.1177/0022185618800518
Fleming, P. (2017). The human capital hoax: Work, debt and insecurity in the era of uberization. Organisation Studies, 38,
691–709. https://doi.org/10.1177/0170840616686129
Friedman, G. (2014). Workers without employers: Shadow corporations and the rise of the gig economy. Review of Keynesian
Economics, 2, 171–188. https://doi.org/10.4337/roke.2014.02.03
Fudge, J. (2017). The future of the standard employment relationship: Labour law, new institutional economics and old
power resource theory. Journal of Industrial Relations, 59, 374–392. https://doi.org/10.1177/0022185617693877
Gandini, A. (2018). Labour process theory and the gig economy. Human Relations, 72, 1–18. https://doi.org/10.1177/
0018726718790002
Gilbert, C., De Winne, S., & Sels, L. (2011). The influence of line managers and HR department on employees' affective com-
mitment. International Journal of Human Resource Management, 22, 1618–1637. https://doi.org/10.1080/09585192.
2011.565646
Graham, M., Hjorth, I., & Lehdonvirta, V. (2017). Digital labour and development: Impacts of global digital labour platforms
and the gig economy on worker livelihoods. Transfer: European Review of Labour and Research, 23, 135–162. https://doi.
org/10.1177/1024258916687250
Gramano, E. (2019). Digitalisation and work: Challenges from the platform economy. Contemporary Social Science, https://
doi.org/10.1080/21582041.2019.1572919
Greenwood, M. (2013). Ethical analyses of HRM: A review and research agenda. Journal of Business Ethics, 114, 355–366.
https://doi.org/10.1007/s10551-012-1354
Guszcza, J., Rahwan, I., Bible, W., Cebrian, M., & Katyal, V. (2018). Why we need to audit algorithms, Harvard Business
Review. Available from https://hbr.org/2018/11/why-we-need-to-audit-algorithms
Harris, B. (2017). Uber, Lyft, and regulating the sharing economy. Seattle University Law Review, 41, 269–285. Available from
http://heinonline.org/HOL/P?h=hein.journals/sealr41&i=277
Harvey, G., Rhodes, C., Vachhani, S. J., & Williams, K. (2017). Neo-Villeiny and the service sector: The case of hyper flexible
and precarious work in fitness centres. Work, Employment and Society, 31, 19–35. https://doi.org/10.1177/
0950017016638023
Healy, J., Nicholson, D., & Pekarek, A. (2017). Should we take the gig economy seriously? Labour and Industry: A Journal of
the Social and Economic Relations of Work, 27, 232–248. https://doi.org/10.1080/10301763.2017.1377048
Howcroft, D., & Bergvall-Kåreborn, B. (2019). A typology of crowdwork platforms. Work, Employment and Society, 33,
21–38. https://doi.org/10.1177/0950017018760136
Hurley, J. (2018, February 26). Boss determined to deliver the right ingredients for success. The Times. Available from
https://www.thetimes.co.uk/article/boss-determined-to-deliver-the-right-ingredients-for-success-6gtczs8xq
Jabagi, N., Croteau, A. M., Audebrand, L. K., & Marsan, J. (2019). Gig workers' motivation: Thinking beyond carrots and
sticks. Journal of Managerial Psychology, 34(4), 192–213. https://doi.org/10.1108/JMP-06-2018-0255
Jago, A. S. (2019). Algorithms and authenticity. Academy of Management Discoveries, 5, 38–56. https://doi.org/10.5465/
amd.2017.0002
Johnston, H., & Land-Kazlauskas, C. (2018). Organising on-demand: Representation, voice, and collective bargaining in the
gig economy. Conditions of Work and Employment Series No 94. Geneva: International Labour Office. Available from
https://www.ilo.org/wcmsp5/groups/public/—ed_protect/—protrav/—travail/documents/publication/wcms:
624286.pdf
Katz, L., & Krueger, A. (2019). The rise and nature of alternative work arrangements in the United States, 1995–2015. ILR
Review, 72, 382–416. https://doi.org/10.1177/0019793918820008
Kirven, A. (2018). Whose gig is it anyway? Technological change, workplace control and supervision, and workers' rights in
the gig economy. University of Colorado Law Review, 89, 249–292. Available from http://heinonline.org/HOL/P?h=hein.
journals/ucollr89&i=267
Kuhn, K. M. (2016). The rise of the ‘gig economy’ and implications for understanding work and workers. Industrial and
Organisational Psychology, 9, 157–162. https://doi.org/10.1017/iop.2015.129
Kuhn, K. M., & Maleki, A. (2017). Micro-entrepreneurs, dependent contractors, and Instaserfs: Understanding online labour
platform workforces. Academy of Management Perspectives, 31, 183–200. https://doi.org/10.5465/amp.2015.0111
18 DUGGAN ET AL.

Lampinen, A., Lutz, C., Newlands, G., Light, A., & Immorlica, N. (2018). Power struggles in the digital economy: Platforms,
workers & markets. In CSCW '18. Companion of the 2018 ACM Conference on Computer Supported Cooperative Work and
Social Computing (pp. 417–423). New York, NY: ACM. https://doi.org/10.1145/3272973.3273004
Lee, M.K., Kusbit, D., Metsky, E. & Dabbish, L. (2015). Working with machines: The impact of algorithmic and data-driven
management on human workers. Association for Computing Machinery: Conference on Human Factors in Computing Sys-
tems, 1603–1612. https://doi.org/10.1145/2702123.2702548
Lehdonvirta, V. (2018). Flexibility in the gig economy: Managing time on three online piecework platforms. New Technology,
Work and Employment, 33, 13–29. https://doi.org/10.1111/ntwe.12102
Lepanjuuri, K., Wishart, R., & Cornick, P. (2018). The characteristics of those in the gig economy. London: Department for
Business, Energy & Industrial Strategy. Available from https://assets.publishing.service.gov.uk/government/uploads/
system/uploads/attachment_data/file/687553/The_characteristics_of_those_in_the_gig_economy.pdf
Liao, C., Wayne, S. J., & Rousseau, D. M. (2016). Idiosyncratic deals and contemporary organisations: A qualitative and
meta-analytical review. Journal of Organisational Behaviour, 37, 9–29. https://doi.org/10.1002/job.1959
Madrigal, A.C. (2019). The Uber IPO is a landmark. The Atlantic. Available from https://www.theatlantic.com/technology/
archive/2019/04/ubers-ipo-historic-despite-its-10-billion-loss/586999/
Mann, G., &O'Neil, C. (2016). Hiring algorithms are not neutral. Harvard Business Review. Available from https://hbr.org/
2016/12/hiring-algorithms-are-not-neutral
McGaughey, E. (2018). Taylorooism: When network technology meets corporate power. Industrial Relations Journal, 49,
459–472. https://doi.org/10.1111/irj.12228
McGee, C. (2017, April 20). Only 4% of Uber drivers remain on the platform a year later, says report. CNBC Technology.
Available from https://www.cnbc.com/2017/04/20/only-4-percent-of-uber-drivers-remain-after-a-year-says-report.
html
McNeill, J. (2019, March 26). Introducing Lyft driver services. Medium. Available from https://medium.com/@jmaclyft/
introducing-lyft-driver-services-ac1ab9488ac6
Meijerink, J., & Keegan, A. (2019). Conceptualizing human resource management in the gig economy: Toward a platform
ecosystem perspective. Journal of Managerial Psychology, 34(4), 214–232. https://doi.org/10.1108/JMP-07-2018-0277
Mowbray, P. K., Wilkinson, A., & Tse, H. M. (2015). An integrative review of employee voice: Identifying a common
conceptualisation and research agenda. International Journal of Management Reviews, 17, 382–400. https://doi.org/10.
1111/ijmr.12045
O'Sullivan, M., Turner, T., McMahon, J., Ryan, L., Lavelle, J., Murphy, C., O'Brien, M., & Gunnigle, P. (2015). A study on the
prevalence of zero hours contracts among Irish employers and their impact on employees. Kemmy Business School, Uni-
versity of Limerick. Available from https://dbei.gov.ie/en/Publications/Publication-files/Study-on-the-Prevalence-of-
Zero-Hours-Contracts.pdf.
Ovide, S. (2019). It's the first day of the rest of Lyft's life. Bloomberg. Available from https://www.bloomberg.com/opinion/
articles/2019-03-29/lyft-ipo-it-s-the-first-day-of-the-rest-of-its-life
Petriglieri, G., Ashford, S. J., & Wrzesniewski, A. (2018). Agony and ecstasy in the gig economy: Cultivating holding environ-
ments for precarious and personalised work identities. Administrative Science Quarterly, 64, 124–170. https://doi.org/
10.1177/0001839218759646
Pichault, F., & McKeown, T. (2019). Autonomy at work in the gig economy: Analysing work status, work content and work-
ing conditions of independent professionals. New Technology, Work and Employment, 34, 59–72. https://doi.org/10.
1111/ntwe.12132
Prassl, J. (2018). Humans as a service: The promise and perils of work in the gig economy. Oxford: Oxford University Press.
Ravenelle, A. J. (2019). “We're not Uber:” Control, autonomy, and entrepreneurship in the gig economy. Journal of Manage-
rial Psychology, 34, 269–285. https://doi.org/10.1108/JMP-06-2018-0256
Riesman, A. (2014). We asked 10 black-car drivers if they prefer working for Lyft or Uber–Here's why Lyft won by a land-
slide. Intelligencer. Available from http://nymag.com/intelligencer/2014/12/lyft-uber-drivers.html
Rosenblat, A. (2018). Uberland: How algorithms are rewriting the rules of work. Oakland, CA: University of California Press.
Rosenblat, A., & Stark, L. (2016). Algorithmic labour and information asymmetries: A case study of Uber's drivers. Interna-
tional Journal of Communication, 10, 3758–3784. https://doi.org/10.2139/ssrn.2686227
Rousseau, D. M. (1995). Psychological contracts in organisations: Understanding written and unwritten agreements. London:
Sage Publications.
Rozzi, F. (2018). The impact of the gig economy on US labour markets: Understanding the role of non-employer firms using
econometric models and the example of Uber. Junior Management Science, 3(2), 33–56. https://doi.org/10.5282/jums/
v3i2pp33-56
Schein, E. H. (1980). Organisational psychology. Englewood Cliffs, NJ: Prentice-Hall.
Schiek, D., & Gideon, A. (2018). Outsmarting the gig-economy through collective bargaining – EU competition law as a bar-
rier to smart cities? International Review of Law, Computers & Technology, 32, 275–294. https://doi.org/10.1080/
13600869.2018.1457001
DUGGAN ET AL. 19

Schmidt, F.A. (2017). Digital labour markets in the platform economy: Mapping the political challenges of crowd work and
gig work. Friedrich-Ebert-Stiftung. Available from http://library.fes.de/pdf-files/wiso/13164.pdf
Slee, T. (2017). What's yours is mine: Against the sharing economy. London: Scribe. https://doi.org/10.2307/j.ctv62hf03
Smith, C. (2006). The double indeterminacy of labour power: Labour effort and labour mobility. Work, Employment and Soci-
ety, 20, 389–402. https://doi.org/10.1177/0950017006065109
Snyder, B. (2016). The disrupted workplace: Time and the moral order of flexible capitalism. Oxford: Oxford University Press.
https://doi.org/10.1093/acprof:oso/9780190203498.001.0001
Steinberger, B. Z. (2018). Redefining employee in the gig economy: Shielding workers from the Uber model. Fordham Journal
of Corporate & Financial Law, 23, 577–596. Available from https://ir.lawnet.fordham.edu/jcfl/vol23/iss2/5
Stewart, A., & Stanford, J. (2017). Regulating work in the gig economy: What are the options? The Economic and Labour Rela-
tions Review, 28, 420–437. https://doi.org/10.1177/1035304617722461
Thomas, C. (2018). Ride oversharing: Privacy regulation within the gig economy. Cardozo Arts and Entertainment Law Journal,
36, 247–276. Available from https://heinonline.org/HOL/P?h=hein.journals/caelj36&i=257
Todolí-Signes, A. (2017). The gig economy: Employee, self-employed or the need for a special employment regulation?
Transfer: European Review of Labour and Research, 23, 193–205. https://doi.org/10.1177/1024258917701381
Tran, M., & Sokas, R. K. (2017). The gig economy and contingent work: An occupational health assessment. Journal of Occu-
pational and Environmental Medicine, 59(4), 63–66. https://doi.org/10.1097/JOM.0000000000000977
Uber (2019). Company information. https://www.uber.com/newsroom/company-info/
Van Buren, H. J., & Greenwood, M. (2008). Enhancing employee voice: Are voluntary employer-employee partnerships
enough? Journal of Business Ethics, 81, 209–221. https://doi.org/10.1007/s10551-007-9489-y
Van Doorn, N. (2017). Platform labour: On the gendered and racialised exploitation of low-income service work in the ‘on-
demand’ economy. Information, Communication & Society, 20, 898–914. https://doi.org/10.1080/1369118X.2017.
1294194
Vandaele, K. (2018). Will trade unions survive in the platform economy? Emerging patterns of platform workers' collective
voice and representation in Europe. Working Paper, European Trade Union Institute.
Vanhala, M., & Ritala, P. (2016). HRM practices, impersonal trust and organisational innovativeness. Journal of Managerial
Psychology, 31, 95–109. https://doi.org/10.1108/JMP-03-2013-0084
Ward, K., Grimshaw, D., Rubery, J., & Beynon, H. (2001). Dilemmas in the management of temporary work agency staff.
Human Resource Management Journal, 11(4), 3–21. https://doi.org/10.1111/j.1748-8583.2001.tb00048.x
Wolfe, J. H. (1941). Determination of employer-employee relationships in social legislation. Columbia Law Review, 41, 1015–
1037. https://doi.org/10.2307/1117747
Wood, A. J., Graham, M., Lehdonvirta, V., & Hjorth, I. (2019). Good gig, bad gig: Autonomy and algorithmic control in the
global gig economy. Work, Employment and Society, 33, 56–75. https://doi.org/10.1177/0950017018785616
Wood, A. J., & Lehdonvirta, V. (2019). Platform labour and structured antagonism: Understanding the origins of protest in
the gig economy. Paper presented at the Oxford Internet Institute Platform Economy SeminarSeries, 5 March.
Woodcock, J., & Johnson, M. R. (2018). Gamification: What it is, and how to fight it. The Sociological Review, 66, 542–558.
https://doi.org/10.1177/2F0038026117728620
Woodrow, C., & Guest, D. E. (2017). Knowledge acquisition and effective socialisation: The role of the psychological con-
tract. Journal of Occupational and Organizational Psychology, 90, 587–595. https://doi.org/10.1111/joop.12178
Wright, C. F., Wailes, N., Bamber, G. J., & Lansbury, R. D. (2017). Beyond national systems, towards a ‘gig economy’? A
research agenda for international and comparative employment relations. Employee Responsibilities and Rights Journal,
29, 247–257. https://doi.org/10.1007/s10672-017-9308-2
Zervas, G., Proserpio, D., & Byers, J. W. (2017). The rise of the sharing economy: Estimating the impact of Airbnb on the
hotel industry. Journal of Marketing Research, 54, 687–705. https://doi.org/10.1509/jmr.15.0204

How to cite this article: Duggan J, Sherman U, Carbery R, McDonnell A. Algorithmic management and app-
work in the gig economy: A research agenda for employment relations and HRM. Hum Resour Manag J. 2019;
1–19. https://doi.org/10.1111/1748-8583.12258