It’s not a crime if we do it (to nurses) with an app
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
If I could abolish one piece of received wisdom about tech policy, it would be this: "Tech moves at the speed of innovation and regulation moves at the speed of government, so regulation will always lag behind tech."
(If I could abolish two pieces of received wisdom about tech policy, the other one would be "If you're not paying for the product, you're the product." Decent treatment is not a customer reward program, and "voting with your wallet" only works if you're a billionaire whose wallet is thicker than all the other wallets put together.)
To be clear, there are times when tech enables new forms of conduct that don't fit neatly into the existing policy framework. For example, we apply copyright to anyone who makes or handles a copy of a creative work, and that used to be a pretty good proxy for "someone in the supply chain of the media industry."
The problem is that computers work by making dozens and dozens of copies every time you click your mouse, and we all use computers for everything, and clicking a mouse doesn't make you part of the entertainment business. The fact that we've had hyperinflation in "making and handling copies" but continued to apply an esoteric industrial framework to pretty much everything everyone does all the time is a huge problem that desperately needs fixing:
Copyright notwithstanding, tech generally does not outrun our capacity to regulate it. Rather, tech bosses come up with incredibly flimsy reasons why their business doesn't fit into the existing regulatory framework, and policymakers accept these ridiculous excuses so readily that one can only assume they're in on the racket.
Take "fintech," all those neobanks and the cryptocurrency junk and shitcoins and stablecoins and NFTs and so on that a group of pump-and-dumpers, money launderers and stock swindlers have pushed for more than a decade now. As Trashfuture's Riley Quinn says, "Whenever you hear 'fintech,' you should think 'unregulated bank.'" It's not hard to apply existing regulations to these companies: they fall under banking law, usury law, securities law and gambling law.
There's no (good) reason not to apply these legal frameworks to the crypto industry – but there are plenty of bad reasons not to. The most obvious reason not to apply those regulations is that you are on the same side as the pump-and-dumpers, money launderers and stock swindlers. The reason we struggle to regulate fintech is that we just don't want to.
Then there's Uber, which claimed that it wasn't a taxi company, it was a "transportation network company," which meant that none of the regulations we apply to taxis should apply to Uber. To call this a transparent ruse is to do great violence to the good, hardworking transparent ruses putting in the hard yards to run honest scams. "Uber isn't a taxi company, it's a transportation network company" is about as plausible as those t-shirts that read "It's not a bald spot, it's a solar-panel for a sex-machine."
Emboldened by the success of the "transportation network company" wheeze, Uber launched Uber Eats, claiming that it wasn't a "food delivery company" but rather a "delivery network company." This set up the template for a remorseless tide of new sex-machine solar-panels that have pushed Uber's system of wage-theft and worker misclassification into an expanding constellation of labor categories.
From fintech to price-fixing to gig-work, the entire industry runs on the very stupid proposition that "it's not a crime if we do it with an app":
One of the worst of these sex-machine solar-panels is to be found in nursing, where a cluster of heavily capitalized apps that nurses must rely on to get shifts insist that they aren't "healthcare staffing agencies," rather, they are "healthcare worker platforms" that should be exempted from the regulations that we started applying to the former after a string of calamities and disasters.
This phenomenon is detailed in eye-watering detail in "Uber For Nursing," a must-read new report by Katie J Wells, Maya Pinto, and Funda Ustek Spilda for the AI Now Institute:
If "Uber for nursing" rings a bell, you might be thinking of "Uber for Nursing: How an AI-Powered Gig Model Is Threatening Health Care," an earlier report that Wells and Spilda wrote for the Roosevelt Institute in late 2024:
The Roosevelt Institute report contained many eye-popping findings, most notably that at least some of the leading national nursing gig-work platforms were using data-brokers to find out how much debt nurses were carrying, and offered lower wages to the nurses with the most debt, on the grounds that the most economically desperate nurses will accept the lowest pay:
The new report describes how, in the absence of a muscular policy response, these nursing gig-work companies have raised fantastic sums of money, some of which they have diverted to regulatory capture projects in a bid to states to recognize their solar-panel sex-machines, with great success. These companies haven't merely refined their lobbying game, either – as a sphincter-puckering appendix detailing the experience of nurses with these apps shows, they have also made great strides in immiserating nurses and transferring their earning power to gig platforms and the hospitals that rely on them.
This degradation of the work experience is characteristic of the new world of AI-powered jobs. AI isn't taking workers' jobs, but it is enshittifying them, with degrading, neurosis-inducing surveillance and high-handed discipline:
But gig-work companies remain laser-focused on healthcare workers, likely because that is one of the only growing professions left in America. They're trying to screw over healthcare workers for the same reason Willie Sutton robbed banks: "That's where the money is." The implication here is that the 15% of the American workforce that is employed in the healthcare industry is on the front lines of the battle against gig-work and algorithmic management.
Like parasites that attack the sick and weak, gig-work and algorithmic management come first for industries that are already bad for workers and the people they serve, making things much worse while insisting that they're just trying to apply a cool digital fix to a broken analog system. That, too, was Uber's playbook: attacking the medallion taxi system as corrupt and sclerotic – while replacing it with a system that's corrupt, extractive and dynamic, able to evade all attempts to improve things for drivers and riders (such as drivers' unions).
That's what's happened with healthcare staffing agencies. These have long been a fixture in healthcare, partly because there was always a large cohort of skilled healthcare professionals who valued the flexibility of short term contracts (for example, "travel nurses") and partly because hospitals love hiring contractors who aren't part of their workers' unions.
Staffing agencies weren't good. A string of scandals led to waves of regulations in states like Colorado, Minnesota and New York that required agencies to "register annually, disclose shareholders and executive officers, certify worker credentials, report to state authorities on the number of workers employed, document service rates charged to facilities, and list average wages paid to workers by job category." These regulations also banned staffing agencies from locking up workers with noncompete agreements and ripping them off with finder's fees.
Rather than strengthening these protections, gig nursing platforms avoid them. Where staffing agencies secure multi-week contracts for travel nurses, gig platforms typically assign workers to single-day shifts. Where staffing agencies let nurses bargain for their scheduling needs, gig platforms present take-it-or-leave-it offers and no opportunities to speak to a human when things go wrong. And where staffing agencies evaluated the workers on their roster based on employer feedback, the gig platforms install apps that continuously surveil and evaluate workers, downranking them and cutting their hours and pay based on algorithmic judgments that are never explained and cannot be appealed.
Platforms match nurses with shifts, claiming to regulators that they're little more than a "job-notice board." But when they pitch hospitals, they tell a different story, about their ability to use algorithms to erode wages and blacklist workers who make trouble. Healthcare gig-work apps push workers to accept shifts that require more travel and pay less, at facilities they don't want to work at. Refusal to accept a shift can permanently compromise your ability to get future shifts, and/or lower the wage you're offered in future.
In addition to these poor working conditions and low wages, gig platforms have resurrected the prohibited practice of charging workers "finder's fees," by layering on junk fees that take money out of every paycheck. Staffing agencies aren't allowed to do this, but the gig-work platforms' "solar panel for a sex-machine" gambit transforms the finder's fee into a "platform fee" that somehow escapes regulators' grasp.
How is it that a regulator can't see that a "platform fee" is exactly equivalent to a "finder's fee?" This is not a case of technology outpacing regulation – it's a case of lawmakers colluding with profitable firms to evade regulation in order to steal from workers.
The platforms are aslosh in investor cash – Clipboard Health, Intelycare, and Shiftkey are all valued at more than $1b, and Shiftkey just completed a $300m private equity raise. This leaves them with lots of ready cash to spend on regulatory entrepreneurship. In Georgia, Clipboard lobbied "to exempt gig nursing platforms from state unemployment insurance and workers’ compensation laws." In Ohio, Shiftkey and Clipboard are pushing a bill "to classify gig nurses as independent contractors, exempting gig platforms from minimum wage and other worker protection laws." In Utah, Nursa is praising a bill that a state senator called "lightest-touch regulation." All in all, 17 states have nurse gig platform deregulation bills underway.
In 2022, the healthcare gig-work platforms tried to get a California ballot measure to carve nursing platforms out of all state labor laws. They withdrew it, but pursued an "under the radar" approach to get the same thing by seeking changes in administrative rules, rather than state laws. Lobbying for administrative law changes to exempt healthcare gig-work platforms from regulation is also underway in Missouri, Louisiana and Utah.
One bright light in all this comes from New York state, where a 2025 law "affirmatively recognizes gig nursing platforms as entities that must comply with the state’s healthcare staffing agency rules." The existence of this law proves that the crisis of gig-work healthcare platforms is not an example of tech racing ahead of regulation. If New York's state leg can figure out that a gig-work platform is just a staffing agency in app form, then other states can do so as well. If they don't figure that out, that's because they don't want to.
Sometime in this century, our political class and our financial class arrived at a consensus that Douglas Rushkoff describes as "go meta," in his 2022 book Survival of the Richest:
The "go meta" ethos insists that the most important, smartest and most valuable move is always away from productive labor. Don't drive a cab: go meta and own a medallion that you rent to a cab driver. Don't own a medallion, go meta and start a gig-work ride-hailing company. Don't start a gig-work ride-hailing company, go meta and invest in a gig-work ride-hailing company. Don't invest in a gig-work ride-hailing company, go meta and buy options in a gig-work ride-hailing company – and so on and so on, into ever more abstracted forms of gambling and rent-collection.
The reorganization of the economy around parasitic middlemen and financial gamblers (but I repeat myself) is the real reason that we can't regulate tech. Once you've decided that the most important party to a transaction is the person who has the option on the share on the platform on the license that the worker who actually does the job requires, of course you're going to see a solar-panel for a sex-machine in every bald spot.
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
Operating a business is risky: you can't ever be sure how many customers you'll have, or what they'll show up looking for. If you guess wrong, you'll either have too few workers to serve the crowd, or you'll pay workers to stand around and wait for customers. This is true even when your "business" is a "hospital."
Capitalists hate capitalism. Capitalism is defined by risk – like the risk of competitors poaching your customers and workers. Capitalists all secretly dream of a "command economy" in which other people have to arrange their affairs to suit the capitalists' preferences, taking the risk off their shoulders. Capitalists love anti-competitive exclusivity deals with suppliers, and they really love noncompete "agreements" that ban their workers from taking better jobs:
One of the sleaziest, most common ways for capitalists to shed risk is by shifting it onto their workers' shoulders, for example, by sending workers home on slow days and refusing to pay them for the rest of their shifts. This is easy for capitalists to do because workers have a collective action problem: for workers to force their bosses not to do this, they all have to agree to go on strike, and other workers have to honor their picket-lines. That's a lot of chivvying and bargaining and group-forming, and it's very hard. Meanwhile, the only person the boss needs to convince to screw you this way is themself.
Libertarians will insist that this is impossible, of course, because workers will just quit and go work for someone else when this happens, and so bosses will be disciplined by the competition to find workers willing to put up with their bullshit. Of course, these same libertarians will tell you that it should be legal for your boss to require you to sign a noncompete "agreement" so you can't quit and get a job elsewhere in your field. They'll also tell you that we don't need antitrust enforcement to prevent your boss from buying up all the businesses you might work for if you do manage to quit.
In practice, the only way workers have successfully resisted being burdened with their bosses' risks is by a) forming a union, and then b) using the union to lobby for strong labor laws. Labor laws aren't a substitute for a union, but they are an important backstop, and of course, if you're not unionized, labor law is all you've got.
Enter the tech-bro, app in hand. The tech-bro's most absurd (and successful) ruse is "it's not a crime, I did it with an app." As in "it's not money-laundering, I did it with an app." Or "it's not a privacy violation, I did it with an app." Or "it's not securities fraud, I did it with an app." Or "it's not price-gouging, I did it with an app," or, importantly, "it's not a labor-law violation, I did it with an app."
The point of the "gig economy" is to use the "did it with an app" trick to avoid labor laws, so that bosses can shift risks onto workers, because capitalists hate capitalism. These apps were first used to immiserate taxi-drivers, and this was so successful that it spawned a whole universe of "Uber for __________" apps that took away labor rights from other kinds of workers, from dog-groomers to carpenters.
One group of workers whose rights are being devoured by gig-work apps is nurses, which is bad news, because without nurses, I would be dead by now.
A new report from the Roosevelt Institute goes deep on the way that nurses' lives are being destroyed by gig work apps that let bosses in America's wildly dysfunctional for-profit health care industry shift risk from bosses to the hardest-working group of health care professionals:
The report's authors interviewed nurses who were employed through three apps: Shiftkey, Shiftmed and Carerev, and reveal a host of risk-shifting, worker-abusing practices that has nurses working for so little that they can't afford medical insurance themselves.
Take Shiftkey: nurses are required to log into Shiftkey and indicate which shifts they are available for, and if they are assigned any of those shifts later but can't take them, their app-based score declines and they risk not being offered shifts in the future. But Shiftkey doesn't guarantee that you'll get work on any of those shifts – in other words, nurses have to pledge not to take any work during the times when Shiftkey might need them, but they only get paid for those hours where Shiftkey calls them out. Nurses assume all the risk that there won't be enough demand for their services.
Each Shiftkey nurse is offered a different pay-scale for each shift. Apps use commercially available financial data – purchased on the cheap from the chaotic, unregulated data broker sector – to predict how desperate each nurse is. The less money you have in your bank accounts and the more you owe on your credit cards, the lower the wage the app will offer you. This is a classic example of what the legal scholar Veena Dubal calls "algorithmic wage discrimination" – a form of wage theft that's supposedly legal because it's done with an app:
Shiftkey workers also have to bid against one another for shifts, with the job going to the worker who accepts the lowest wage. Shiftkey pays nominal wages that sound reasonable – one nurse's topline rate is $23/hour. But by payday, Shiftkey has used junk fees to scrape that rate down to the bone. Workers have to pay a daily $3.67 "safety fee" to pay for background checks, drug screening, etc. Nevermind that these tasks are only performed once per nurse, not every day – and nevermind that this is another way to force workers to assume the boss's risks. Nurses also pay daily fees for accident insurance ($2.14) and malpractice insurance ($0.21) – more employer risk being shifted onto workers. Workers also pay $2 per shift if they want to get paid on the same day – a payday lending-style usury levied against workers whose wages are priced based on their desperation. Then there's a $6/shift fee nurses pay as a finders' fee to the app, a fee that's up to $7/shift next year. All told, that $23/hour rate cashes out to $13/hour.
On top of that, gig nurses have to pay for their own uniforms, licenses, equipment and equipment, including different colored scrubs and even shoes for each hospital. And because these nurses are "their own bosses" they have to deduct their own payroll taxes from that final figure. As "self-employed" workers, they aren't entitled to overtime or worker's comp, they get no retirement plan, health insurance, sick days or vacation.
The apps sell themselves to bosses as a way to get vetted, qualified nurses, but the entire vetting process is automated. Nurses upload a laundry list of documents related to their qualifications and undergo a background check, but are never interviewed by a human. They are assessed through automated means – for example, they have to run a location-tracking app en route to callouts and their reliability scores decline if they lose mobile data service while stuck in traffic.
Shiftmed docks nurses who cancel shifts after agreeing to take them, but bosses who cancel on nurses, even at the last minute, get away at most a small penalty (having to pay for the first two hours of a canceled shift), or, more often, nothing at all. For example, bosses who book nurses through the Carerev app can cancel without penalty on a mere two hours' notice. One nurse quoted in the study describes getting up at 5AM for a 7AM shift, only to discover that the shift was canceled while she slept, leaving her without any work or pay for the day, after having made arrangements for her kid to get childcare. The nurse assumes all the risk again: blocking out a day's work, paying for childcare, altering her sleep schedule. If she cancels on Carerev, her score goes down and she will get fewer shifts in the future. But if the boss cancels, he faces no consequences.
Carerev also lets bosses send nurses home early without paying them for the whole day – and they don't pay overtime if a nurse stays after her shift ends in order to ensure that their patients are cared for. The librarian scholar Fobazi Ettarh coined the term "vocational awe" to describe how workers in caring professions will endure abusive conditions and put in unpaid overtime because of their commitment to the patrons, patients, and pupils who depend on them:
Many of the nurses in the study report having shifts canceled on them as they pull into the hospital parking lot. Needless to say, when your shift is canceled just as it was supposed to start, it's unlikely you'll be able to book a shift at another facility.
The American healthcare industry is dominated by monopolies. First came the pharma monopolies, when pharma companies merged and merged and merged, allowing them to screw hospitals with sky-high prices. Then the hospitals gobbled each other up, merging until most regions were dominated by one or two hospital chains, who could use buyer power to get a better deal on pharma prices – but also use seller power to screw the insurers with outrageous prices for care. So the insurers merged, too, until they could fight hospital price-gouging.
Everywhere you turn in the healthcare industry, you find another monopolist: pharmacists and pharmacy benefit managers, group purchasing organizations, medical beds, saline and supplies. Monopoly begets monopoly.
(Unitedhealthcare is extraordinary in that its divisions are among the most powerful players in all of these sectors, making it a monopolist among monopolists – for example, UHC is the nation's largest employer of physicians:)
But there two key stakeholders in American health-care who can't monopolize: patients and health-care workers. We are the disorganized, loose, flapping ends at the beginning and end of the healthcare supply-chain. We are easy pickings for the monopolists in the middle, which is why patients pay more for worse care every year, and why healthcare workers get paid less for worse working conditions every year.
This is the one area where the Biden administration indisputably took action, bringing cases, making rules, and freaking out investment bankers and billionaires by repeatedly announcing that crimes were still crimes, even if you used an app to commit them.
The kind of treatment these apps mete out to nurses is illegal, app or no. In an important speech just last month, FTC commissioner Alvaro Bedoya explained how the FTC Act empowered the agency to shut down this kind of bossware because it is an "unfair and deceptive" form of competition:
This is the kind of thing the FTC could be doing. Will Trump's FTC actually do it? The Trump campaign called the FTC "politicized" – but Trump's pick for the next FTC chair has vowed to politicize it even more:
Like Biden's FTC, Trump's FTC will have a target-rich environment if it wants to bring enforcement actions on behalf of workers. But Biden's trustbusters chose their targets by giving priority to the crooked companies that were doing the most harm to Americans, while Trump's trustbusters are more likely to give priority to the crooked companies that Trump personally dislikes:
The movie "Bakuage Sentai BoonBoomger VS King-Ohger" starts screening today!! It's been more than half a year since filming... It's the best Bakuage ever to have made it to this day safely!! 🔥 Please come to the theater🛞👑
