Profits Uber everything? The gig economy and the morality of category work

Abstract

In this essay, we address the question of how the strategic and organizational activities of on-demand sharing economy companies such as Uber are labeled and classified. We approach this question through a categorization lens and explore in particular whether sharing economy companies can legitimately frame the individuals who work for them as “independent workers” and what this implies for the nature of the employment relationship in such on-demand business models. Our overall aim in doing this is twofold. First, we highlight and address an important categorization issue in our current society, which has potentially far-reaching consequences for the nature of employment and the securities and protections that workers used to enjoy in many parts of the world. Second, we advance prior research in the strategy and organizational domain by elaborating how acts of categorization are inherently moral and political in nature. In this way, we aim to provoke researchers toward studying the moral basis of categorization work and we provide pointers in this essay for how they might do so.

In 2016, an employment tribunal in the United Kingdom vehemently criticized Uber and lambasted its framing of itself as an intermediary digital platform, labeling it as utter “fiction” (Hickey, 2016). As the judges wrote in their ruling, “The notion that Uber in London is a mosaic of 30,000 small businesses linked by a common ‘platform’ is to our minds faintly ridiculous.” The judges ruled that any common sense reading would consider Uber as a company that employs its drivers, or “partner drivers,” as Uber names them and cannot simply be classed as an intermediary technology platform. In their words,

We are satisfied that the supposed driver/passenger contract is a pure fiction which bears no relation to the real dealings and relationships between the parties. It is not real to regard Uber as working “for” the drivers . . . the only sensible interpretation is that the relationship is the other way around. (Hickey, 2016)

The suggestion of the judges is that Uber cannot have its cake and eat it too, effectively profiting from its taxi, or taxi-like, operations while absolving itself from its fiduciary duty of care toward its employees and the customers who make use of its service.

The central issue that is at stake in this new context is how on-demand sharing economy companies, such as Uber, are labeled and classified as businesses. This is far from a trivial matter, as such categorization has significant economic and societal consequences. It determines how such companies are regulated and whether they can indeed legitimately frame the individuals who work for them as “independent workers,” repudiating the need to contribute their fair share toward employee protection, benefits, and pay. In this essay, we focus on the political struggles between companies, such as Uber, Deliveroo, and others, on the one hand, and their employees, legislators, and policy makers, on the other. We focus in particular on the struggle around the categorization of Uber workers as “employed” or “independent”, and what this implies for the nature of the employment relationship in this kind of an on-demand business model.

In doing so, our aim in this essay is twofold; first, we aim to highlight and address an important categorization issue in our current society, with far-reaching consequences for the nature of employment and the securities and protections that workers used to enjoy in many parts of the world. Second, we show how these kinds of categorization struggles are inherently political and moral in nature. The veneer of corporate rhetoric or legal jargon oftentimes seems to cast instances of categorization as a natural and automatic affair that appears to be value-free. By elaborating the intrinsically moral nature of such instances of categorization work, we call upon category researchers to cast and study them in such terms instead. While there is some limited categories-related research on the moral aspects of product and market categories, the categorization battles that are raging over employment rights and other high-stakes issues in the economy ask for a much more vigorous normative and morally charged approach of category researchers in the management and organizational domain.

The moral nature of category work

How new categories such as a “ride-hailing company” or a “platform business” emerge and become conventionalized as a common fixture in the economy and society is a subject of interest in ongoing strategic management and organizational research. Over the last decade or so, researchers have explored the emergence of new categories through the blending of the labels and contents of antecedent categories in a market context (e.g. Kennedy et al., 2010). Others have observed how categories may arise on the back of creative analogical connections that individuals draw between market categories and cultural ideas elsewhere (e.g. Navis and Glynn, 2010), and still other studies suggest that new categories are created through individuals inductively generalizing a one-off insight into a common frame of reference (e.g. Kennedy and Fiss, 2013). As Grodal et al. (2015) have outlined, new categories either emerge in the form of a compound, as a conceptual pairing of existing words and categorical contents, or through an inductive process in which a new label is derived from a specific experience. In either of these instances, a new category comes into being on the basis of structured and oftentimes predictable combinations between cognitive categories that are then realized linguistically and expressed in words and grammar (cf. Lakoff, 1987).

Most of the existing categories-related research has been focused on the discursive and cognitive processes through which new categories emerge and affect the producers and consumers of these categories (Grodal et al., 2015). One key tenet of discursive categorization processes is that these are oftentimes politically motivated, based on the interests of the actors involved (Edwards, 1991). As Granqvist et al. (2013) have shown, companies in emerging areas of market activity can often try to manage and manipulate their association with a category, by entirely dissociating from it, claiming it, or hedging it (thus creating some ambiguity around their association with the category itself), depending on what they consider most beneficial to their interests. The kind of rhetorical battles that then in turn ensue between various actors has distributive consequences for those involved. They may get a bigger part of the pie, based on the success of their own company-related labeling efforts, or they may in fact succeed collectively in ensuring the establishment of a newly labeled profitable market category for everyone involved.

Inherent to such rhetorical battles is, however, also a moral dimension to categorization. Not only the distribution of outcomes for market actors and for society is a moral matter, but also the very labeling of a category of activity is value-based and evaluative in nature. Whatever we decide to be in a category, as well as what overall purpose or good (as an ideal) the category serves, are inherently moral judgments. In a recent paper, Arjaliès and Durand (2019) draw attention to the moral constitution of product categories. Instead of simply accepting categories as they are, they argue that we should interrogate their moral basis and consider what they “should” do or “ought to do” from the perspective of society. Following this line of reasoning, category researchers should in particular, we believe, interrogate the, at first glance, seemingly neutral rhetoric of the proponents of new market, product, or employment categories and critically examine what ideals in society these new categories reveal. For example, is Uber’s claim for a new employment category an example of an “innovation” that fosters economic activity or simply a ploy to limit or externalize the responsibilities they have for their workers?

This moral dimension is one that, with a few exceptions (Arjaliès and Durand, 2019), has not been a main concern in category research to date. Perhaps this general neglect reflects an overarching unease with morality and moral matters from a social science perspective (Selznick, 1992). However, it is a perspective that we believe is much needed now to address the “big” categorization battles that are taking place around gig economy companies and the models of employment that they are proposing. In this article, we take a first step in this direction and explore the substantive and moral basis of a new category of employment—the so-called “independent worker”—that is promoted by gig economy companies. We offer our analysis of this specific issue below, after first introducing the broader categorization of gig economy companies.

Categorizing the gig economy

The broader issue of how gig economy companies are categorized can perhaps best be explored by looking at one specific example: Uber. Uber originated as UberCab in 2009, but after an initial legal challenge in California changed its name to Uber in order to “separate itself from the taxicab industry” (Uber homepage). The company currently bills itself as an “online transportation network company,” with a mission to “bring transportation for everyone, everywhere.” Even if the specific label of an “online transportation network company” does not conventionalize as a shorthand for its activity, Uber has in all of its communications and legal submissions vied for a label and an associated category that differentiates and delimits the scope of its business—and with that Uber’s responsibilities and liabilities—to that of being an intermediary platform that brings customers and drivers together. Rather than considering drivers as employees, it considers them as “partners” on its platform and as essentially self-employed independent contractors (Jones, 2018). Alternative labels and categories would inevitably bring the company closer to the category of taxi services with employed drivers, which would obviously challenge its current business model (Brown, 2015) and lead to more stringent regulation as well.

For its part, Uber builds in all of its official communication on neutral, non-descript terms to describe its category, by drawing on the technology (the app) and by then expanding the app to an entire business model and even an imagined form of organization. By using such compound terms as being an “app-based platform,” “transportation network,” or “sharing-based marketplace” that simply connects buyers and suppliers, Uber apparently aims to stake out a legitimate claim to a new category, as a technology company (as opposed to a taxi services company), and to normalize its existence as just one exemplar of the “new” sharing economy.

Reflecting a broader trend with businesses in the new gig economy, our desk research of publicly available material on Uber also suggests that Uber appears to be striving to claim a new category label that marks the experience of ordering and taking an Uber ride (“hailing a ride”) as distinct, and as in effect, in a class of its own. Besides implicitly contrasting itself with existing taxi service categories (cf. Kuilman and Wezel, 2013), the terminology of being a platform-based “ride-hailing” company appears to be designed to trigger an inductive process of categorization (Grodal et al., 2015; Kennedy and Fiss, 2013). Such inductive generalization involves the ad hoc construction of class-inclusion statements (Glucksberg, 2008), in which a particular target—such as the experience of an Uber ride—is positioned as a central or prototypical instance of novel, ad hoc constructed categories—“hailing a ride via an app”—that are created on the fly. In principle, this process best explains situations where a new category has to be created with nothing remotely similar existing (Kennedy and Fiss, 2013), with one of the easiest routes being to figuratively construct a new category around a set of abstracted features or characteristics.

While each of us may question whether this “experience” is indeed that different from booking a taxi, the bigger point is that if and when this new category of “hailing a ride via an app” becomes established, the obvious advantage for a company such as Uber is that it will be seen as the prototype example. The ideal scenario for Uber, strategy-wise, is, we believe, to become the category-defining prototype so that every time the category comes to mind, people will think of Uber, just like every time we need to search online, we think of Google. By the same token, and similar to Google and other Silicon Valley firms, we assume that Uber is also trying to outcompete Lyft and others so as to monopolize this new category and in such a way that it basically becomes synonymous with it. There is indeed some evidence to suggest, given that the derived phrases of “uberization” and “to uberize” have been adopted in the English dictionary, that this tactic is bearing fruit.

Reflecting on these category-defining claims by Uber, the question that we should ask ourselves is whether there is sufficient ground to actually believe that hailing a ride is a new realm of experience and distinct in all of its facets from taking a cab. Similarly, should we as a society accept a company’s claim to be incorporated as a platform, to simply act as an intermediary between customer requests and work demands through an app, and thus be allowed to comply with far fewer corporate or employment laws and have no liability whatsoever when things go wrong? It seems to us that the neglect of these kinds of political and moral dimensions to category work is to a large extent the result of the fact that the gig economy has emerged under the umbrella of the “sharing economy” and has thus benefited from a sort of positive halo effect without more rigorous questioning. As Rinne (2017) has indicated, this is unfortunate as the sharing economy focuses on “underutilized assets, monetized or not, in ways that improve efficiency, sustainability and community,” whereas the gig economy has rather focused specifically on “income generation via ‘gigs’, single projects or tasks for which a worker is hired.” This distinction, however, is important. and we will illustrate it further in the next section of this essay, where we challenge the category-defining claims of many of the gig economy players on the grounds that their activities share too much overlap with existing market categories. We demonstrate this specifically in relation to the discussion of whether Uber drivers can legitimately be framed as “independent workers,” as opposed to seeing them more conventionally as employees. This discussion is particularly important in the context of the employment model and the kind of precarious labor that the categorization efforts of gig economy companies may create, and we thus delimit our focus in the remainder of this essay to this particular issue.

“Independent workers” in the gig economy

The process of category work becomes particularly salient and politically relevant when we focus on how Uber and other “intermediaries” in the so-called gig economy are—as evidenced by court proceedings (cf. Calo and Rosenblat, 2017)—trying to convince politicians, legislators, and administrative agencies that they should categorize the individuals who work for them as “independent workers,” whom they effectively contract for short-term engagements. In particular, many of these companies strive toward establishing this new category of “independent workers” as legally distinct from the “two legal statuses currently available under US labor, employment, and tax law: employees or independent contractors” (Harris and Krueger, 2015: 6). At stake, it seems to us, is the ability for companies like Uber to continue to operate their employee-base as a no-frills market, where individuals get cheaply contracted to carry out the work, but unlike “employees” are not protected under the Fair Labor Standards Act (FLSA) in the United States, do not fall under compensation laws, and enjoy no unemployment insurance cover. Given that alternative work arrangements, such as those within the gig economy, have been argued to form “all of the net employment growth in the U.S. economy from 2005 to 2015” (Katz and Krueger, 2016:7, as cited in Petriglieri et al., 2018), it is especially critical to ensure that legislative efforts develop hand-in-hand.

While no doubt some of the casual drivers for the likes of Uber, or indeed Deliveroo or Foodora, value the additional income and the flexibility that these companies offer, there is still a sizable proportion of people who drive for Uber or participate in the gig economy because of financial hardships and a lack of full-time employment options (Manyika et al., 2016). Ironically, while Uber’s business model, as a sharing platform, is supposed to help empower these disadvantaged workers by providing access to additional income (Calo and Rosenblat, 2017; Manyika et al., 2016), considering them as “independent workers” does in fact only aggravate their plight. These independent workers become, we argue, so-called “precarious” workers—a new category of workers who cling precariously to temporary but largely low-paid jobs with no benefits and with the constant uncertainty of insolvency (Burton, 2015). Uber already contracts its drivers in this manner (ahead of legal verdicts in jurisdictions around the world), where in fact further political and legal support for this categorization will give it even more bargaining power with its drivers and allow it to continue to evade payments for their protection, cover, and training as employees.

Given the significance of the issue, it is important to have a look at how the public debate around this new category has been unfolding and whether the new category is proposed on the basis of substantive grounds. A good place to start with is Harris and Krueger’s (2015) work, two economists who have been working closely with Uber and who argue that “[independent] workers have some similarities to independent contractors and some similarities to traditional employees” (p. 6) as they form in effect a “middle ground between traditional employees and independent contractors” (Harris and Krueger, 2015: 5). Uber drivers, they argue, are “independent workers” who do not fall squarely in the traditional category of employees as they are not economically dependent on a single employer, do not have an indefinite relationship with such an employer, and do not relinquish control as a regular employee does over their work hours.

At the same time, such workers are in their eyes also not independent contractors “because some aspects of the methods and means of work—including the price of their services—are controlled by the intermediary and because they are integral to the business of the intermediary” (Harris and Krueger, 2015: 10). And thus, Harris and Krueger (2015: 10) argue, “independent workers are, in some respects, like individuals working for others, and in other respects are like independent businesses (e.g., they use their own equipment and control their own hours),” a seemingly hybrid constellation, which they argue legitimizes “a new legal and economic category of independent workers.” The conclusion they reach is that this in-between category is a “gray area” that creates ambiguity in the market, where a newly created category instead would offer a more efficient and transparent labor market, facilitate innovative companies such as Uber, and give clear instruction to the courts who otherwise might be forced to choose between two ill-fitting categories and in biased ways given that “the weighing of factors is often left to individual decision makers” (Harris and Krueger, 2015: 6).

Harris and Krueger (2015) base their overall recommendation for this newly constricted category on the idea of the “immeasurability of work hours” in the gig economy and whether these can be clearly assigned to an intermediary or employer. Specifically, they argue that it is ambiguous whether waiting time can be qualified as labor as individuals might be doing personal tasks while working and whether such hours can be legitimately apportioned to a specific “employer.” This ambiguity, they suggest, violates the very notion of a traditional employment relationship. The idea is that when one or more apps are turned on, the driver is “waiting to be engaged,” to, in effect, become economically active, and might be doing personal chores in the meantime. An Uber driver is not in comparison “engaged to wait” as an employee (say as a firefighter would) during work hours.

Eisenbrey and Mishel (2016) challenge this very assumption and argue that the immeasurability claim is empirically flawed, as Uber tracks its drivers’ work hours and effectively assumes that their drivers are continuously “active” when the app is on, as drivers are expected to accept rides within 15 seconds and face serious consequences if they ignore requests or are too slow to respond. Ultimately, not accepting rides may lead to being deactivated (i.e. fired) for having too low an “acceptance rate.” This acceptance rate threat in effect means that a driver is in fact “engaged to wait” and demonstrates that the driver’s time is controlled by Uber, pressurizing drivers to accept rides when they are on the clock. It is also not feasible to assume, given this policy, that drivers can do anything else during that time, including personal tasks. It is thus questionable whether Uber drivers can even be considered as “independent” workers, who are free to “decide which assignments to accept based on criteria such as fee, the desirability of the client, or the timing” and “can change those choices over time” (Manyika et al., 2016: 2). Since the company’s entire model directly relies on having a seamless flow of drivers ready to respond to passengers’ requests, Uber has in fact resorted to a wide range of psychological tactics to keep drivers engaged (Scheiber, 2017). For example, to “lock” them in for as long a period as possible into an active state, the app sends drivers a push notification before an existing trip is completed, showing them that there is a new ride waiting for them. In principle, this can reduce idle time, which is seen as beneficial for the drivers. However, as the company wants to ensure that users get a response as quickly as possible, it withholds information from the drivers as to whether this trip will be economically advantageous to them (e.g., its area, duration, and rate are not disclosed before accepting), which effectively diminishes any “independence” or freedom of choice that the drivers may have. Even if they opt to close the app, it is the “Keep driving” rather than the “Go offline” button that is already pre-selected for them, alongside the nudge of telling them how much in dollar amounts they are from achieving a certain total of net earnings—all in an effort to increase the time they stay connected and thus remain “on call” (Scheiber, 2017).

Another category-related claim that Harris and Krueger (2015: 13) offer in support of the “independent worker” category is that a driver can in principle have multiple apps (such as Lyft alongside Uber) simultaneously turned on, making it legally and “conceptually impossible to assign [work hours including waiting time] to an employer.” However, given that the intermediary who ultimately gets a ride accepted by a driver benefits, it can be argued that “the one that should pay is the one that will benefit—the company whose app provides the first accepted rider” (Eisenbrey and Mishel, 2016). In other words, the various reasons that are offered in support of the new category are not sufficiently strong to warrant a separate, exclusive category in the labor market. Uber drivers can in fact rightfully be classified as employees and have most of the prototypical characteristics of that category:

They don’t set their own fares or freely choose their own customers, their performance is measured and controlled by Uber, their driving is essential to Uber’s business, and the economic reality is that they are not independent businesses but small cogs in Uber’s powerful multinational business. (Eisenbrey and Mishel, 2016)

A different way of looking at this is that if there is some conceptual or legal ambiguity regarding the nature of employment, there is no need, we argue, to immediately run for the hills or resort to a “legislative fix” along the lines offered by Harris and Krueger (Eisenbrey and Mishel, 2016). The interests of the business owners are currently disproportionally favored over those of employees in the proposed category of “independent workers.” What is telling in this respect is that Harris and Krueger (2015) argue for a “middle ground” position for this new category based on positive similarities to, or correspondences with, the categories of employees and independent contractors. As such, it would be natural, we argue, to create a novel hybrid compound that would incorporate such similarities. Compounding, or the creation of novel combinations between words or labels that take the form of noun–noun or adjective–noun modifications (whereby a modifier (noun or adjective) is applied to a head concept (usually a noun)), is a standard way of creating new categories in contexts such as this one where two established categories already exist (Wisniewski, 1997).

Thus, rather than “independent worker,” the compound that would more naturally arise would be something like “contracted employee,” who moves in and out of employment, based on the rides, as jobs, that he or she takes and the amount of time he or she is waiting to be engaged. In fact, this overall category could then offer provisions for some kind of adaptive regulation around specific sub-categories of “contract employees,” whereby, for example, those driving for more than 30 hours per week could receive a status closer to that of a full-time, regular “employee” (see Brown, 2015), whereas those working fewer hours could receive a comparable “social safety net” that can offer at least some protection and security (Sundararajan, 2016). A recent ruling in California, called Assembly Bill 5 (AB5), is a step forward in helping to grant some gig economy workers such “employee” categorization, thus giving them a certain set of protections and benefits (Campbell, 2019). Yet, this does not come without some pressure from the companies again, with Uber and Lyft allegedly telling their employees that they should oppose the bill as otherwise the flexibility of their driving schedules will suffer. In effect, Lyft is no longer allowing some of its drivers to start work from locations the company considers less busy as they would then need to pay a minimum wage for an “idle” worker (Campbell, 2019).

Obviously, granting “employee” status would bring added costs for platform businesses in the gig economy, and thus, we argue, Uber and other gig economy businesses are intentionally lobbying for their own category creation that strategically accentuates dissimilarities with existing categories and thus negates any direct correspondences, or analogies, with these categories. But in supporting the legislation for a completely new category, Harris and Krueger (2015), as their chief proponents, also contradict themselves; they claim a “middle ground” between existing categories as the premise for the new category but then suggest that the resulting category can have none of the positive attributes of those constituent categories. Instead, they strategically invent a “negative” qualification, which while fictitious (Hickey, 2016) is legally and financially in favor of the business owners and then reason backward for its viability.

Here, we take a moral stance on this particular issue. We consider the reasoning in support of a new category of “independent worker” logically flawed and dishonest. We also believe that the new proposed category undermines the post-war established ideal of an employment relationship, with support and protections being provided to workers. In fact, we believe that if this new category becomes institutionalized in jurisdictions around the world, it will not only directly harm individual workers, but also instigate across the labor market a new precariat class of workers who have no job security, are slaving away for poor pay, and whose material and psychological welfare is seriously compromised (Standing, 2011). There may well be differences in the impact of this new category on workers in different countries and jurisdictions around the world (and studying this in a comparative manner would be a worthwhile effort), but we nonetheless expect that the net effect for society will be negative.

The moral matter of categories

The key question, therefore, is, are we able to consider this newly claimed employment category as sufficiently legitimate from a moral perspective? The stakes could not be higher. If Uber and other gig economy companies will get their way in establishing a new category of “independent worker” across jurisdictions around the world, we believe that there is a real danger that the long-established post-war link between employment and worker benefits and protection will be broken, heralding a new era of more flexible, but precarious labor. For us as a society at large, increasingly more attention and scrutiny should therefore be paid to gig companies’ active engagement in politically motivated category work through their lobbying, advertising, and legal submissions.

Our call in this respect is made, we realize, at the eleventh hour. Gig economy companies already operate as if de facto their employees are “independent workers.” These companies also ensure that their language is consistent with this category of employment and appear to suggest that it is very much already an institutionalized fixture. Companies such as Deliveroo and Foodora, for example, meticulously scrutinize any elements of their description that may “mistakenly” have them categorized as traditional food delivery employers. As a part of their efforts, they provide limited to non-existent training to their “riders,” referring to them as “cycling micro-businesses” (Jones, 2018) or even more generically as “agents” (DeGroen and Maselli, 2016) instead.

And while these classic gig economy companies embrace the lingo, other companies that traditionally classified their personnel as employees are now also joining the movement and switching to the “independent worker” label to reap the same kind of crass economic benefits (cf. Fleming, 2017; Marks, 2018). These broader developments in the economy should be more than enough encouragement for management researchers, we believe, to become intellectually and politically involved in defining the appropriate organizational and employment categories for Uber and other sharing economy companies. When researchers do, they inevitably start to consider categorization processes as not only intrinsically political but also moral in nature (Arjaliès and Durand, 2019; Hsu and Grodal, 2015)—a feat that we believe should receive more focused attention.

When management researchers do become involved in this manner, we hope that besides studying categorization processes from the perspective of the distributive consequences they have and the values that they foster, they may themselves also become more normative in how they approach their research. We see far too little instances where in relation to these kinds of big societal topics management researchers take a moral stance (cf. Selznick, 1992), either in their own scholarly work and publications or in the outreach they do through public speeches or consulting in society. Yet, such a normative stance may be exactly what is needed to ensure that the “sharing economy” does not simply turn into a “taking economy” for the party that commands more power (Calo and Rosenblat, 2017: 1660).