Uber's Bastards II
On the proliferation of algorithmic overseers in the on-demand service industry.
Hello everyone! Today we are going to talk about my favorite subject: the impact of Uber and its bastard spawn. Last time we did this, we talked about the (im)moral economy of on-demand labor platforms and their transformation of care work via A Roosevelt Institute report by Katie J. Wells and Funda Ustek Spilda. Today, we’ll look at another report from Wells, Spilda, Veena Dubal and Mark Graham—this time at Fairwork US, investigating the industries corrupted by algorithmic management tools (and the firms deploying them).
As always, if you enjoy my writing then consider a free subscription. If you want to support me further so I can do this writing more regularly, consider a paid subscription. For $7 a month (a pint at the pub) or $70 a year (a new video game that you'll lose interest in after 10-15 hours), you can help me keep writing.
Last time, I talked at length about my core issues with Uber and the firms taking after its business model before opening up the conversation to talk about its transmogrification of care work. I’ll summarize some key points here before we move on, but revisit my previous essay to read more:
The so-called “gig economy” is an altar to greed and misery. Mass immiseration, starvation wages, mental health crises, horrendous working conditions, the offloading of every conceivable cost onto workers and the public, the narrowing of political imagination, racial discrimination for workers and consumers, the acceleration of surveillance pricing, these and many more social ills are key consequences of the on-demand business model.
On-demand labor did not proliferate because it was profitable or innovative. This business model metastasized across the economy because of a stringent commitment to worker misclassification, algorithmic discrimination, anti-competitive capital-intensive strategies, impressive public relations, robust political lobbying, and shoddy journalism.
The core ambition of the on-demand ride-hail model is to make previously illegal profits legal again. It's about subverting urban governance, breaking New Deal labor regimes that prohibit certain types of exploitation, subjecting workers and workplaces to more authoritarian forms of control and discipline, and reshaping drivers, markets, consumers, firms, and governments into forms more amenable to privatized profits and socialized losses.
On-demand nursing promises to be even worse for everyone involved. Gig nursing platforms in the health-care industry have introduced new and considerable risks: more work for less pay; unpredictable scheduling and compensation; discontinuous care that puts patients at risk; an algorithmic off-hands approach that puts workers at risk; and a general decline in the standards one can expect in care work as a practitioner or recipient.
The first entry in Uber’s Bastards talked about on-demand nursing, but there are a plethora of other Uber clones across our economy that are worth looking at, especially as they encroach into more of our daily life.
I’ll be going through key parts of Fairwork’s report “When AI Eats the Manager” in hopes of drilling into you—like with the Roosevelt Institute report—how bad things are in this sector for nearly everyone.
A key part of the on-demand business model is the deployment of algorithmic overseers. By taking in as much information as possible through pervasive surveillance and give as little information as possible, the hope is that you can create information asymmetries that give you the power to set fares and compensation at optimal levels (maximal and minimal, respectively). Veena Dubal, one of the authors of this Fairwork report, has called this algorithmic wage discrimination (also cited above). In an LPE Project blog post introducing the concept, Dubal writes:
As a labor management practice, algorithmic wage discrimination allows firms to personalize and differentiate wages for workers in ways unknown to them, paying them to behave in ways that the firm desires, perhaps for as little as the system determines that they may be willing to accept. Given the information asymmetry between workers and the firm, companies can calculate the exact wage rates necessary to incentivize desired behaviors, while workers can only guess as to why they make what they do.
In addition to being rife with mistakes that are difficult or impossible for workers to ascertain and correct, algorithmic wage discrimination creates a labor market in which people who are doing the same work, with the same skill, for the same company, at the same time, may receive different hourly pay. Moreover, this personalized wage is determined through an obscure, complex system that makes it nearly impossible for workers to predict or understand their frequently declining compensation.
You cannot understand the extent to which on-demand labor is exploitative, dehumanizing, disruptive, unpredictable, and immiserating until you appreciate the extent to which it is managed by black-box algorithms primed to extract as much as possible—and how big of a boon this has been for firms which were previously unable to achieve profits except through accounting tricks.
Editorial
Algorithmic overseers are not only deployed to extract as much from both ends of a transaction, but to flense as much off cost structures as possible. Scheduling, payment processing, paperwork, notifications, customer service, HR, Q&A, and more are all being eliminated or consolidated or automated under the pretense of algorithmic efficiency. When these systems fail, as they constantly do, you’re left with a chatbot to handle your complaints and queries.
The new algorithmic management technologies have direct implications for operational cost-cutting, efficiency metrics across the supply chains, and measurable KPIs (key performance indicators). This hyper-quantification of work can expedite decision-making, scaling, and meeting the performance indicators set by investors and funders. However, the implications for work that has been re-structured and re-imagined through these technologies are far reaching for workers, consumers, and societies. Automated technologies are displacing traditional management relationships, transparency, and accountability in the workplace. The data that platform workers produce may shape how platform companies set prices, manage workers, create personalised pay structures, and even offset financial liabilities. Indeed, one company named Argyle has amassed the employment records of 40 million platform workers in the US, and sells this data as its primary source of profits. The reality of working with AI is different from the promises associated with it. AI is, to put it descriptively, eating the managers.
To that end, the report looks at 11 on-demand platforms across ride-hail, food delivery, elder care, and healthcare to see what the impact of these algorithmic overseer technologies is shaping out to be. Are they replacing managers? Are they creating new management practices and institutions? Are they allowing these firms to develop new relations with consumers, workers, other companies, or governments? And, most important, what are they doing to our notions of work and fairness.
Fairwork's report looks at 11 platforms (CareRev, Clipboard Health, Doordash, Grubhub, Instacart, Lyft, Papa, ShiftKey, ShiftMed, Uber, and Ubereats) in the United States. They’re assessed on a 10 point scoring system that evaluates working conditions according to five principles:
Fair Pay: Workers, irrespective of their employment classification, should earn a decent income in their home jurisdiction after taking account of work-related costs. We assess earnings according to the mandated minimum wage in the home jurisdiction, as well as the current living wage.
Fair Conditions: Platforms should have policies in place to protect workers from foundational risks arising from the processes of work and should take proactive measures to protect and promote the health and safety of workers.
Fair Contracts: Terms and conditions should be accessible, readable and comprehensible. The party contracting with the worker must be subject to local law and must be identified in the contract. Regardless of the workers’ employment status, the contract should be free of clauses which unreasonably exclude liability on the part of the service user and/or the platform.
Fair Management: There should be a documented process through which workers can be heard, can appeal decisions affecting them, and be informed of the reasons behind those decisions. There must be a clear channel of communication to workers involving the ability to appeal management decisions or deactivation. The use of algorithms should be transparent and result in equitable outcomes for workers. There should be an identifiable and documented policy that ensures equity in the way workers are managed on a platform (for example, in the hiring, disciplining, or firing of workers).
Fair Representation: Platforms should provide a documented process through which worker voice can be expressed. Irrespective of their employment classification, workers should have the right to organise in collective bodies, and platforms should be prepared to cooperate and negotiate with them.
9 companies scored 0 points, meaning they failed to adhere to any of these minimum standards for fair work. One scored one point, another scored two points. This is abysmal but unsurprising given the aggression with which these companies pursue profits illegally and immorally.
Key Findings
FAIR PAY
Fairwork found that only one company (ShiftMed) out of the 11 studied paid its workers a minimum wage. Why? ShiftMed is the only firm to classify workers as employees, which protects them under federal labor laws and provides some basic guarantees about working conditions and compensation. Every other company, however, classifies workers as independent contractors who:
are responsible for significant work-related costs and spend parts of their workdays engaged in unpaid activities, such as driving long-distances to get to a shift or waiting for a customer to receive an order.
One thing to understand about substandard wages is that they’re a great anti-competitive strategy and were understood as such back when the Fair Labor Standards Act (which establishes standards for minimum wages, overtime pay, and youth employment) was passed in 1938. As Eamon Coburn writes in Yale Law Review:
These concerns animated early minimum-wage advocates who expressed two theories of substandard wages as unfair competition: substandard wages as competition by subversion of public norms, and substandard wages as taking an implicit subsidy from workers and the public. In the 1930s, political, business, and labor leaders continued to frame labor abuses as a competition problem throughout the consideration and implementation of national wage-and-hour legislation, repeating both of these theories to justify federal action.
The first theory sees substandard wages as a violation of public standards around living wages or industry norms, allowing a firm outcompete others on prices by reducing (labor) costs. Firms like Uber might respond to this by saying—as they have in the past—that adhering to public norms would make their service too expensive and rob the public of its benefits. Little over a century ago, this was understood as such an immoral response that you had radicals like President Theodore Roosevelt saying: "We will not submit to that kind of prosperity any more than we will submit to prosperity obtained by swindling investors or getting unfair advantages over business rivals."
The second theory, Coburn writes, sees substandard wages as a way for firms to "pad their bottom line with money owed to workers and gain the full benefit of their employees' productivity while shouldering only part of the cost." This also manifests as a tax burden shouldered by other firms as workers paid substandard wages then rely on social programs that their employers won’t pay into. Uber and Lyft have long been the gig economy’s vanguard here, using worker misclassification to avoid paying hundreds of millions of dollars in unemployment taxes, workers’ compensation, and paid family and medical leave insurance—undermining the social safety net and pushing up tax rates for other firms.
To make their unit economics work, on-demand labor platforms need to embrace anti-competitive strategies in hopes of attracting enough capital, customers, and political power to reforge relevant markets into forms more able to generate and sustain profits: price-gouging customers and underpaying (and misclassifying) workers are key.
FAIR CONDITIONS
Unsurprisingly, each of these firms all trade in horrific work conditions.
Fairwork was unable to fi nd suffi cient evidence to award a point to any of the platforms in this study. Workers report signifi cant task-specifi c risks and lack of a safety net.
Across 11 of the largest on-demand labour platforms in the US, workers reported physical assaults, verbal abuse and stressful working conditions. Fairwork finds that safety is a major issue for on-demand nursing companies, on-demand elderly care companies, on-demand delivery companies, and on-demand ride-hail companies. In healthcare, significant changes are needed to orient, train, and manage on-demand workers so that they can protect both themselves and their patients.
The gig economy is, out of necessity, built on a business model that necessitates worker misclassification. Again, they are seeking profits and returns that are locked behind rules and regulations aimed at protecting workers. Both cursory glances and in-depth examinations quickly make this feature painfully clear.
In 2022, a national survey of gig workers by the Economic Policy Institute painted a bleak picture:
19 percent of gig workers went hungry because they couldn't afford enough to eat (compared to 14 percent of service sector employees)
31 percent didn't have earn enough to pay their utility bills (vs. 17 percent of service sector employees)
18 percent lived in a household where someone didn't seek medical care in the last month because of the cost (vs. 13 percent of service sector employees)
30 percent used SNAP (vs. 15 percent of service sector employees)
In 2021, The Verge and New York Magazine collaborated on an investigative report looking at New York City’s 65,000 delivery workers who were left to fend for themselves by the apps that exploited and misclassified them, as well as a city that seemed uninterested in their safety or livelihoods, forcing them to gerry-rig solutions themselves:
Workers developed the whole system — the bikes, repair networks, shelters, charging stations — because they had to. To the apps, they are independent contractors; to restaurants, they are emissaries of the apps; to customers, they represent the restaurants. In reality, the workers are on their own, often without even the minimum in government support. As contractors and, often, undocumented immigrants, they have few protections and virtually no safety net. The few times city authorities noted the delivery worker’s changing role, it was typically with confused hostility. Until recently, throttle-powered electric bikes like the Arrow were illegal to ride, though not to own. Mayor de Blasio heightened enforcement in 2017, calling the bikes “a real danger” after an Upper West Side investment banker clocked workers with a speed gun and complained to him on The Brian Lehrer Show.
The NYPD set up checkpoints, fining riders $500, seizing their bikes, and posting photos of the busts on Twitter. The police would then return the bikes because, again, they were legal to own. It was a costly and bewildering ritual. For years, bike activists and workers pushed for legalization, though the apps that benefited from them were largely silent. It was only when another group of tech companies — hoping to make scooter-sharing legal — joined the fight that a bill moved forward in Albany. Then the pandemic hit, restaurants were restricted to takeout, and the mayor had to acknowledge that the bikes were an essential part of the city’s delivery infrastructure. He halted enforcement. The bikes were officially legalized three months later.
In 2022, the Markup reported it had independently tracked a total of 361 ride-hail and delivery drivers who'd been victims of attempted and successful carjackings since 2017—28 drivers had been killed as a result. That number was likely much higher though. Uber revealed that year there were at least 24,000 "safety incident reports" from 2017 to 2020 that involved a passenger physically assaulting a driver, but specifics were declared confidental and kept under seal:
Our data isn’t comprehensive, however. Most police departments don’t gather statistics on carjackings specifically against gig workers, and the gig companies have repeatedly declined to provide their own data on carjackings. The Markup compiled its database through police reports and local news articles that cite police reports, and by conducting interviews with drivers, family members, and lawyers representing drivers (or their families) who were victims of carjackings.
As I’ve written over the years, Uber and Lyft have repeatedly seen driver shortages at key moments despite incentives that make the pay livable because the working conditions are intolerable for most drivers!
FAIR CONTRACTS
Contracts do not fare any better among the platforms:
Two of the evaluated platforms – ShiftMed and Papa – have clear and accessible terms and conditions. But the widespread use of liability clauses on the platforms included in this year’s study place nearly all the risk of negligence on workers rather than companies.
Ethical and responsible data protection measures for worker data are needed for the 11 platforms in this study, and more transparency and accountability are needed for workers to understand how their data is collected, processed and stored. Fairwork finds that class action waivers and arbitration clauses are commonly used, and they limit workers’ ability to bring legal claims collectively or have their cases decided by a court of law.
For years, gig companies have pushed for mandatory arbitration because it is incredibly good at stymying class action lawsuits and legal precedents—again, if your business model relies on skirting the law, regulatory arbitrage, and aggressive lobbying, you need to stop angry workers in deplorable conditions from collectively demanding the right to earn a livable wage or be safe in the course of their work or get health insurance or other indications of dignified work.
Uber and Lyft recently said as much when they asked the Supreme Court in 2024 to block state officials from using their enforcement powers to seek money for workers and customers who've signed the companies' mandatory arbitration agreements.
Technically, the companies want the justices to review a 2023 California state court appellate ruling, opens new tab that allowed California’s attorney general and labor commissioner to continue litigating claims that Uber and Lyft owe money to drivers who were misclassified as independent contractors. They contend that the California Court of Appeal – like state appellate courts in five other states – misread a key 2002 Supreme Court decision when it concluded that state officials are not bound by workers’ arbitration agreements.
But make no mistake: The theory espoused by Uber and Lyft would preclude all kinds of litigation by states attorneys general, from consumer protection and unfair competition litigation to antidiscrimination suits. If Uber and Lyft are right, state AGs and other officials simply would not be permitted to bring lawsuits seeking monetary relief for anyone who signed an arbitration agreement.
The companies said as much in their petitions, arguing that in response to Supreme Court decisions allowing companies to impose mandatory arbitration on workers and consumers, states have become increasingly likely to adopt “creative devices,” in the words of Lyft’s lawyers from Munger, Tolles & Olson, to undermine arbitration. And unless the Supreme Court steps in, Uber and Lyft said, state officials will continue to expand such loopholes until the Federal Arbitration Act is effectively nullified.
The cases in question that Uber and Lyft are petitioning trace back to a 2020 San Francisco state court case "seeking (among other relief) unpaid wages and benefits for allegedly misclassified drivers" and argues that because drivers waived their rights when signing these contracts, whether or not they read them closely, the state's officials cannot weigh in. And the success that Uber and Lyft have enjoyed with blocking drivers from costly class action lawsuit over wage theft, misclassification, liability for injuries or death, and so on have inspired others across the gig economy to emulate the forced arbitration strategy. As Jacqueline Vanacore of Washington College of Law's Arbitration Brief lays out in 2024:
The substantial damage awards won by workers who litigated misclassification claims laid the groundwork for employers’ burgeoning preference for mandatory arbitration agreements and class action waivers. In Estrada v. FedEx, a California Appeals Court held that FedEx drivers are employees instead of ICs because FedEx maintained control over drivers by requiring them to pay for business expenses such as uniforms, fuel, insurance, and truck maintenance. The workers were awarded $14 million in damages from 2005-2008 and $12 million in legal fees. Moreover, in 2014, the Ninth Circuit held that FedEx drivers were misclassified as ICs and were owed back wages for missed meals, breaks, and overtime and were entitled to worker’s compensation and unemployment insurance. FedEx settled the class action dispute with over 2,000 drivers for $228 million.
The FedEx cases illustrate the financial risk of misclassification lawsuits to Gig employers, whose workforce mainly consists of ICs. Although arbitration can attempt to address individual employment misclassification, the broader problem of misclassification for an entire workforce will persist. The outcome of private arbitrations is secret, non-precedential, and only applies to a small group or an individual worker. Statistics demonstrate that arbitrators who frequently rule in favor of an employer are more likely to be hired to resolve future employment disputes. Misclassification claims brought to arbitration will likely have a low dollar value, especially when workers are prohibited from filing a class action or must represent themselves in arbitration. An Uber driver who represented herself in arbitration received only $4,000 backpay for reimbursed expenses. The decision resulted in significant financial savings for Uber but inadequate compensation or justice for the worker.
It is a win-win situation for the firm, freeing it of the costs (fiscally and reputationally) of a court battle by disaggregating claimants, preserving misclassification, and insulating itself from laws and regulations it’s running afoul of elsewhere.
FAIR MANAGEMENT
As I’ve talked about above, algorithmic overseers are central to the gig economy business model—in fact, here and elsewhere in the world of algorithmically mediated labor, I’ve argued plantation logic is integral to disciplining and controlling a workforce that is both “independent” but persistently surveilled and highly regimented/structured/guided with insights gleaned from digital tools.
Fairwork’s assessment:
Fairwork was unable to award a score for this principle to any of the assessed platforms. We were unable to find sufficient evidence of a due process for decisions affecting workers.
Improvements are needed for workers to meaningfully appeal low ratings, report issues of non-payment, late-payment, deactivations, other penalties, and disciplinary actions. Although many of the platforms offer public statements in support of equality, diversity and non-discrimination, more evidence is needed to confirm that these policies are put in practice.
The deployment of algorithmic overseers that helps force on-demand workers to accept unfair and unchallengeable firings and dehumanizing work conditions. There are a litany of second-order consequences, but it’s important to keep in mind just how destabilizing algorithmic management is to the worker. Zephyr Teachout echo’s Dubal’s concerns in another article on the ascent of algorithmically targeted wages:
Algorithmic management also transforms the nature of supervision, and the power and sentimental relationships between supervisors and mid-level decisionmakers and locates the decision-making in a combination of the upper-level management and the results of hyperindividualized tools that rely on spying and psychological updating. Not only do the direct supervisors have little power, but the workers are then employed in a state of rational paranoia, where they know that they are being punished and rewarded and experimented upon, but they have no way of knowing whether any given decision they are faced with is a result of a game, an experiment, a punishment, or reward, or changing circumstances on the ground and changing needs at the job.
Such insecurity is necessary for highly personalized wages to work, as is an extensive surveillance apparatus that arbitrarily savages workers. It also encroaches on privacy concerns, in deep and under-appreciated ways that will come for all of us eventually. As Teachout goes on to write:
The right to privacy in one’s thoughts and actions is fundamental, a basic right that implements the democratic commitment to human dignity. Privacy implicates the right of people to have control over the boundaries of what is known about them—both what they want to protect from view, and what they want to project in the public arena. While employers have always supervised and monitored and—since the 1990s—recorded an enormous amount of data about employees, the opportunity to collect data that allows for targeted wages increases the incentive to monitor substantially, and the opportunity to collect data that allows for experimentation and extreme Taylorism does the same.
Therefore, the privacy concerns which have long attended the workplace—and never been adequately addressed—move from second order concerns to first order concerns as the scope of monitoring increases, and the arenas which are monitored move from the superficial to the intimate. As Michael Selmi has argued, privacy constitutes the person, and while “it is one thing to give an employer broad dominion over its own workplace, but it is quite another to extend that dominion wherever the employee goes.”
As Teachout, Dubal, and Sarah Jaffe also remind us—these tools of surveillance, control, domination, privacy violation, and algorithmic discrimination find earliest adoption and persistent experimentation in industries dominated by Black and brown workers—such as care work, ride-hail, and food delivery. Plantation logic through and through: we can return to older levels of extraction in modern non-white workforces by making the (digital) factory resemble the plantation with discipline, control, the treat of starvation, piece-pay, and a persistent commitment to dehumanization.
FAIR REPRESENTATION
Fairwork’s assessment:
Collective organisation and representation is a fundamental right for workers and employees. Fairwork was unable to evidence that the 11 platforms in this study assure freedom of association or expression of worker voice in line with the Fairwork Fair Representation principle thresholds.
As shown in the report, various models of contracting labour are used by digital labour platforms; these can either hinder or enable workers to act on their right to collectively organise. We were unable to evidence that the 11 platforms in this study assure freedom of association or the expression of worker voice in line with the Fairwork Fair Representation principle thresholds.
All of the elements I’ve discussed conspire to disempower workers at every possible turn. It is pretty hard to have solidarity, let alone representation, if your employer has deployed an impressive technology stack aimed at preserving your precarity with personalized starvation wages, pervasive surveillance, a lack of privacy, or arbitrary firings and punishments. Worker solidarity groups do emerge despite all of this but, as I write above, many of those groups (necessarily so) spend their energy ameliorating how unsafe and how extractive this line of work is. That is: helping bring down operating costs that the company dumps on them (bikes, scooters, cars, fuel, weather gear, maintenance of equipment, etc.), retrieving stolen equipment when cops don’t help (which is most of the time), doing patrols to deter attacks on delivery workers, and so on. Successful union drives are far and few between despite constant valiant efforts made to organize workers, take over company facilities, partner with legal advocates to push back against gig economy lobbyists, and work with lawmakers to push through pro-worker legislation.
Silicon Valley’s New Partnerships
A key insight offered by this report is that these digital labor platforms are partnering with major institutions to embed their services. There are three strategies at play here:
Step 1: Ignoring the law
A fundamental element of the on-demand business model. Ignoring the law doesn't only allow firms to roll back the clock on labor laws in hopes of realizing profits that are otherwise illegal, but by insisting the law itself is outdated—it wasn’t made with The Digital Economy in mind. As Fairwork’s researchers write:
In this phase, the rule of law is secondary to the holy aura of innovation. To advance this worldview and to subvert governance structures, labour platforms exploit real problems caused by years of austerity, from decaying public transit infrastructure and neighborhood disinvestments to struggling social services and wage stagnation. Platform firms argue that they, rather than a recalcitrant government or any of its under-resourced programmes, should be at the center of solutions. When Obama’s 2008 campaign manager and White House senior advisor, David Plouffe, joined Uber as its new senior vice president for policy and strategy, Plouffe said Uber would help workers put money “back in their pocket” and receive the “pay raise that they’ve been denied for years.” At the same time, he offered assurances that the company was self-reliant. “We are not asking for special tax breaks like those who want to build a factory or headquarters in a city often do.” Legal precedents and government institutions, Plouffe suggested, were an impediment to progress, not evidence of it. Other platforms have mimicked these Silicon Valley ideas about how change happens – with powerful outsiders – and followed suit with their own arguments against regulation as a common good.
Step 2: Securing permanent exemptions by rewriting local and state laws
As Fairwork writes:
One of Uber’s greatest innovations is its argument that technologically-mediated business models are so unique that they merit brand new business categories. What’s so important about having a new business category? The new categories are the very tool that helps platforms win carve-outs from existing rules. After Uber or DoorDash convince policymakers that they deserve their own category, the platforms then argue that they should operate wholly free of government interference or any standing regulatory body. To wiggle themselves out of this public oversight, labour platforms draw on campaign language that is eerily similar to the Koch brothers’ deregulatory efforts in the 1990s, and the contemporary efforts of groups like the ultraconservative American Legislative Exchange Council. In this phase, platform companies act as “regulatory entrepreneurs,” companies for which rewriting laws, as opposed to simply currying favor through traditional lobbying, is a significant part of their development plans.
That wildly successful strategy has already started to see action with on-demand care workers, with firms arguing that "Digitally-dispatched healthcare workers" and their employers should be exempt from local and state laws much like Uber's legal category of "Transportation Network Company" allowed the firm to flirt established regulations in place for ride-hail dispatch operators that did not operate through an app. For nurses and nursing assistants, this would mean any worker deployed through an app or website would be misclassified as an independent contractor—a move that, again, removes the host of protections and guarantees getting in the way of profitability for these firms.
In this phase, if a platform meets city-level resistance in these category-making efforts, the platform turns to state preemption – the nullification of municipal ordinances by state legislatures. Since 2017, Uber and its peers have pressured 34 state legislatures to prohibit governments at the city and county level from setting labour standards such as a minimum wage, raising tax revenues on ridehailing services, or mandating safety or accessibility measures. Hawaii’s law, for instance, preempts “any ordinance or other regulation adopted by a political subdivision that specifically governs transportation network companies, transportation network company drivers, or transportation network company vehicles.” An economic development expert pointed out the irony of these state-level interventions: “I frankly think it’s hypocritical of Uber and Lyft to say ‘We are partners of cities’ while systematically undermining the ability of their elected officials to actually manage how these services fit into the milieu.”
Step 3: Embrace government and institutional partnerships
In this phase, platforms from Uber and DoorDash to Instacart and Papa secure partnerships with a host of institutions, from insurance providers and non-governmental social service providers to the federal government itself. Starting with Arizona in 2019, a handful of southern Republican states changed their laws to allow patients to use Medicaid funds to pay Uber and Lyft for rides to nonemergency medical appointments. In 2021 Joe Biden’s administration partnered with Uber to provide free rides to Covid-19 vaccination appointments and installed Seth Harris, who wrote an influential study about the benefits of Uber’s worker treatment, in a top labour position. Uber has also worked with traditional unions to legislate sectoral or industry-wide bargaining for rideshare drivers while exempting workers from established labour protections, like the right to strike. Last year, Uber issued $30 million to one of California’s largest single-funded PACs, while a partnership between Uber and the Minnesota Department of Human Services to provide transit for disabled and elderly residents, especially in rural areas, threatened to derail minimum wage campaigns in that state. After a decade of disregarding laws and deceiving policymakers, now the company is, as a spokesperson told Bloomberg, “pitching proposals to state legislators that add benefits while protecting flexibility.”
DoorDash has built partnerships with food banks, churches, Meals on Wheels, veteran non-profits, hundreds of anti-hunger organisations, and major grocery chains. Instacart has not only followed suit by emulating DoorDash's move to accept platforms or Uber's work with public transit agencies, but has gone even further: it enjoys partnerships with universities to back PR talking points with research it funds, as well as partnerships that go so far as to use public funding for stipends for Instacart groceries. Papa partners with Medicare Advantage plans that let public funds go to Papa, works with insurers like Allstate to provide itself as a benefit for employee plans, and partners with Uber to offer ride-hail for its workers and their clients.
What is wrong with gig firms choosing collaboration instead of antagonism? Well they aren’t. They’re still impoverishing their workers and subjugating them to starvation wages, eroding labor standards for people outside their industries that find inspiration in Silicon Valley’s fight for the right to exploit workers, and persisting as a public parasite on multiple fronts:
they evade taxes that pay into our social safety net while pushing more of their workers onto it
they degrade the quality of public goods and services in cities they purport to partner with
they degrade the working conditions of employee and independent contractor working conditions outside of industries infected by gig economy logic
they introduce algorithmic governance tools that use surveillance to enthusiastically violate basic norms around privacy, fairness, dignity, and morality when it comes to worker agency and autonomy, prices, wages, and the boundary between our personal lives and our work lives
they consolidate market power and political power off a project that is explicitly aimed at starving and disempowering workers, price gouging consumers, and eviscerating obstacles to monopolies and the rents that come with them—whether that be public institutions, market competitors, or their workers’ livelihoods
Or, as Fairwork writes:
Collectively, the past 15 years have seen a strategic move among Silicon Valley’s platforms and their relationships with government, institutions, and civic organisations. It is a shift from antagonism to collaboration. Platform partnerships can generate dependencies, help companies gain institutional legitimation, and secure market power.