According to the New York Times, Airbnb yesterday “released” data about their business in New York City. As I first reported on Airbnb in New York two years ago, when that business was a lot smaller, I was interested. Airbnb’s Chris Lehane says “Our hope is that people will understand 99 percent of people on Airbnb in New York City are using it as an economic lifeline,” and who could object to that? Would the real numbers show that we critics are wrong?
My work has been based on scrapes of the Airbnb web site (now done better by Murray Cox at Inside Airbnb), so it’s necessarily less accurate than Airbnb’s own data. On the other hand, I don’t have a $25 billion market valuation at stake in the answer, so it may be easier for me to be honest in my reporting.
I hoped that “released” meant that I could get the data, but I was quickly disillusioned. It turns out to mean “made available only by making an appointment to visit Airbnb’s New York City office”, which is a bit of a joke. Instead, all we get is the summary statements from Airbnb PR. Still, it is better than nothing. So I read on…
My first response to the New York Times article was dismay at this statement: “From November 2014 until November 2015, some 93 percent of revenue earned by active hosts in New York City who share their entire home came from people who have only one or two rental listings on the platform”. That is a number far higher than I had seen, and suggests that a much bigger portion of the Airbnb business is their archetypal “regular New Yorkers occasionally renting out the home in which they live” than I had thought. I had reported about 40% of Airbnb’s business coming from people with more than one rental listing and the numbers suggested 20% of business coming from people with more than two listings. Have I and other critics been getting it wrong? In the absence of complete data we have to make some estimates about income after all. This would be unfortunate as I have just PUBLISHED A BOOK ABOUT THE SHARING ECONOMY THAT MAKES A GREAT CHRISTMAS PRESENT and that is critical of Airbnb.
But today the New York Times ran a correction: “From November 2014 until November 2015, some 75 percent of revenue earned by active hosts in New York City who share their entire home came from people who have only one or two rental listings on the platform.” (my emphasis). The change from 93% to 75% is significant: that’s almost four-fold increase in the proportion that comes from three-or-more listers. All of a sudden the Airbnb numbers look much more like those collected by myself and other external investigators, which Airbnb routinely say are inaccurate.
So what’s the real picture? Yes, 25% of Airbnb revenue in NYC comes from people renting out “more than two” listings. My own estimate actually comes out below that at 20% so my estimates are more friendly to Airbnb’s “regular people” pitch than reality. My numbers also show that about 40% of total revenue comes from people with more than one listing, which is just what I reported two years ago. It’s likely that the real number is closer to half, given the way my estimates seem to underestimate revenue from “more than three” listers. In short, far from showing that the critics were wrong, Airbnb’s numbers show that our data, which they have been rubbishing, is pretty good and even generous to them: their numbers suggest that even more of the business comes from multiple listers than we have been claiming.
So here’s the right way to say it. “From November 2014 to November 2015, about half of Airbnb’s revenue in New York City comes from multiple-listing hosts. Hosts with three or more listings contribute 25% of the total.” That’s a much more commercially-focused operation than the original claim.
The 93% number that Airbnb gave is, by the way, their projection on next year’s figures, to which I can only say – if you’re going to release data, maybe talk about the data and not about your dreams and aspirations? So far their supposed efforts to clamp down on hosts with many listings have been half hearted, and given that it may cut into their revenues we should not give it a lot of credibility. Airbnb has been talking the talk a long time about this challenge on their site, and yet so far they have done basically nothing about it (I’m travelling and don’t have access to the full data set at the moment, or I’d show you).
Maybe more on this later. But for now, the new Airbnb numbers do nothing to undermine the critics’ case.
For some time I have been working on a simulation of reputation systems: a computational model I can use to think through some of the issues they raise. A first pass at this model is now available, together with a fairly long document describing how it works and some results, on GitHub as a Jupyter notebook here.
I was particularly interested in a seeming paradox in what we have learned about real-world reputation systems. As I say in the introduction:
In the few years since they have become widespread, reputation systems have shown two seemingly contradictory characteristics:
(Lake Wobegon effect) Most ratings are very high. While ratings of Netflix movies peak around 3.5 out of 5, ratings on sharing economy websites are almost all positive (mostly five stars out of five). The oldest and most widely-studied reputation system is eBay, in which well over 90% of ratings are positive; other systems such as BlaBlaCar show over 95% of ratings as “five out of five”.
(Panopticon effect). Service providers live in fear of a bad rating. They are very apprehensive that ratings given for the most frivolous of reasons by a customer they will never see again (and may not be able to identify) may wreck their earnings opportunities, either by outright removal from a platform or by pushing them down the rankings in search recommendations. Yelp restaurant owners rail at “drive-by reviewers” who damage their reputation; Uber drivers fear being “deactivated” (fired), which can happen if their rating slips below 4.6 out of 5 (a rating that would be stellar for a movie).
So are reputation systems effective or not? Here’s the seeming contradiction:
The Lake Wobegon effect suggests that reputation systems are useless: they fail to discriminate between good and bad service providers (my take on this from a couple of years ago is here). This suggestion is supported by quite a bit of recent empirical research which I have summarized in MY NEW BOOK!. Customers are treating reviews as a courtesy, rather than as an opportunity for objective assessment. Rather like a guest book, customers leave nice comments or say nothing at all.
The Panopticon effect suggests that rating systems are extremely effective in controlling the behaviour of service-providers, leading them to be customer-pleasing (sometimes extravagantly so) in order to avoid a damaging bad review.
If you are not a fan of computer models, or just have better things to do, here are my main conclusions, paraphrased:
The model demonstrates the important role of social exchange compared to a pure market or transactional exchange in most customer–service provider exchanges. It is this social exchange that is at the root of the Lake Wobegon effect, where all providers are above average. Reputation systems do indeed fail to discriminate on the basis of competence (quality).
A small number of entitled customers can induce a Panopticon effect. Service providers who engage in Give & Take exchanges with their customers (even very competent ones) risk being given a negative review, which will damage their business. The incentives of the reputation system encourage providers to indulge their customers, in order to avoid this unlikely but damaging judgement.
If reputation systems spread and customers become used to rating people in an “honest” fashion, we are building a terrible world for service providers. They must engage in emotional labour, catering to customer whims, or risk their livelihood. The Panopticon is here. The reputation systems continue, it should be noted, to fail to discriminate based on the competence of the service provider — instead of changing quality, they change attitude.
The Lake Wobegon effect and the Panopticon effect can coexist, and are coexisting. Reputation systems as they currently stand are failing to discriminate based on quality. But there is only one thing worse than a reputation system that doesn’t work, and that’s a reputation system that does work: Reputation systems promise a dystopic future for service providers, in which their careers are being shaped by reputation systems that are not working as advertised, but are working to compel compliance.
Susan Crawford wrote about “Getting Over Uber”. Swimming against the tide as a technophile and Internet enthusiast, she has come to believe that transport and communications networks in cities are about more than the market exchange of getting a ride. Also that Uber — a company that already squeezes its drivers as tightly as it possibly can — will squeeze even more tightly if it becomes unconstrained. Uber, Crawford says, is not a good idea for American cities.
Tim O’Reilly responded with Getting Over Taxis. He found Crawford’s arguments puzzling and unconvincing. He did some back-of-the-napkin math to show that Uber can be better for drivers than taxis. He concludes that while “common carriage” (uniform and universally accessible transport) is a noble goal, “when the private sector is doing a better job of providing that service than the previous government-chartered monopolies, government needs to get out of the way.”
Here, I want to do two things:
I think Tim O’Reilly’s back-of-the-napkin math about driver income gets some things wrong and I want to put another point of view.
That said questions about driver income are probably not going to make the difference in this debate, which takes us back to Susan Crawford’s post.
(Aside: I’ll sometimes call them “O’Reilly” and “Crawford” below. Californians may feel this looks hostile, but that is not the intent. I’ve just never met either of them, so take it as old-style British formality.)
(Do the Math: Taxi vs Uber)^2
Start with the questions about driver income. O’Reilly notes that taxi drivers typically rent his or her taxi from the owner, usually for a fee of just over $500 per week, after which, the driver keeps 100% of all fares and tips (but has to pay for gas). He compares this “gate fee” to the following Uber driver expenses:
Uber’s 25–30% that it keeps of every fare.
A $109 per week lease from Toyota, provided by Uber.
While it’s not easy to translate Uber’s cut into a weekly amount, O’Reilly notes that for this to equal the $500 per week gate fee for taxis, the driver would be making $2000 per week, which “seems unlikely”.
The equation, that Uber fee + lease is the equivalent of taxi lease and operating expenses (save gas) is off the mark. But I want to be constructive about this, so before I spell out an alternative, a few disclaimers:
There is no one taxi driver. It’s a complicated industry; even within one city, there is complexity. In Toronto, for example, there have been moves to permit more owner-operators (ambassadors) to take some power away from fleet owners, and then some modifications to let ambassadors have one other driver (to get the most use out of their license) and so on. Different cities have different rules. Small towns are different from the big metropolises.
There is no one Uber driver. The company sets very different rates in different cities ($2.15 per mile + $0.40 per minute in New York; $0.75 per mile + $0.15 per minute in Detroit), takes a different percentage of the fare, and even sets different “safety fees” ($0 in New York, $2.50 in Gary, Indiana, according to Biz Carson). And that’s before the whole surge pricing thing.
I’m not an expert. If you want to take a look at the complexities of driver expenses by someone who is, see Lawrence Meyers’ dense 27-page epic “Towards A Cost Estimate of A NYC UberX Driver”. Of course, NYC is only one city and the picture will be different elsewhere; details clarify the picture, but details also muddy the picture.
So with all those caveats, here is what Tim O’Reilly missed: the $500 per week “gate fee” that he talks about includes maintenance, repairs, depreciation and insurance in addition to the fee that the vehicle owner takes. The Uber “fee plus lease” misses the cost of maintenance (except, I believe, for oil changes and tire rotations), repairs, and insurance. Once we include those costs, things look different: the short version is that most Uber drivers probably get about the same as most taxi drivers.
Here’s the longer version. From what I could see last year, each dollar of a taxi fare gets split very roughly four ways: a quarter goes to the leaseholder, a quarter to the costs of car operation (including insurance), a quarter to gas, and a quarter to the driver. The “gate fee” is the leaseholder and the operation parts, so about half of the overall income.
When it comes to Uber, a quarter (or over) goes to Uber, about half goes to gas and costs of operation (minus commercial insurance) leaving about a quarter for the driver.
Where does that “half” come from? Two places: one is a table in Meyers’ paper that lists the revenue per mile that a driver is earning, and the percentage of revenue that is lost. A percentage of 40 to 50% is in the middle of this chart:
A second is that Meyers’ cost estimate is a bit higher than the numbers calculated by Justin Singer and lower than the 57c per mile that the IRS allows, so it’s in the right ballpark (but remember, there are many different ballparks).
So from what I can see, the overall split is fairly similar between Uber and taxis.
But there are some other differences to remember, one in favour of Uber and one against:
In its favour: Uber claims greater utilization (more rides per hour) which would lead to better incomes. The data it has provided in support of this is partial.
On the other hand, there’s nothing here about commercial insurance. Adding commercial insurance is expensive (Meyers suggests an additional 8c per mile, which amounts to somewhere around 5% or so of the fare). Of course, most Uber drivers don’t take out this insurance: part of Uber’s cost advantage is that passengers and drivers are often taking uninsured rides.
(Aside: I recently attended an Uber driver information session. They did not mention insurance at all until an audience member asked about it, at which point they said it’s between the driver and the insurance company. They would have had to wink broadly to make it any more clear that they aren’t checking insurance and won’t ask questions.)
What this leads to is that the Uber driver’s position is not so different from that of the taxi driver: both keep somewhere around a quarter of the fare, and increase utilization on Uber rides gets eaten up by the per-mile costs Uber drivers have to pay. While Tim O’Reilly says the amount you can make as an Uber driver is “almost surely higher than the median income for taxi and limousine drivers in 2012” I would suggest that it’s probably about the same, with quite a bit of variance both ways.
I admit that the estimates above remain full of holes and the conclusions are wrapped in caveats, but there’s one other reason I have confidence in my overall conclusion that most of Uber’s drivers are not making significantly more than taxi drivers. If Uber had comprehensive data that proved drivers were making a good income after expenses they would shout it from the rooftops. The fact that they haven’t (all their posts and papers talk about “before expenses” income) tells us a lot.
Crawford argues that “ Uber consistently squeezes its drivers as tightly as it possibly can; new drivers are paying an even higher cut to Uber than the first generation did.” And I agree: the future is likely to be more difficult for Uber drivers.
Uber is currently losing money in its efforts to attract drivers and passengers. It’s possible that its Xchange car loan program is also a driver subsidy. But while losing money may help build the company pre-IPO, when accounts are private and growth is everything, it is not a sustainable strategy.
Within cities, the company keeps its own slice of the pie small when it gets started in a new location, to get riders and drivers onto the platform and to push the aggressive growth strategy it has adopted. It has then increased its cut in many places. Once the taxi companies have been pushed to the side, why should it continue to pay its drivers as much as it does now?
Where is the real dividing line?
I might be wrong, but I suspect that all the above is beside the point.
Susan Crawford talks about “My tribe — the technophiles, the Internet enthusiasts” being thrilled about Uber, and it’s the technophile part of this that is key: people who identify with the technology will generally identify with Uber. How we feel about Uber is shaped by how we feel about free markets and civic governance; who we identify with.
For all the argument, I suspect most people would have the same view of Uber whether their drivers make more than taxi drivers or make less. If taxi drivers make more than Uber drivers, then to some that would simply prove that taxi drivers are fat-cat monopolists who need to stop overcharging their customers and adjust to the new world. If Uber drivers make more than taxi drivers, then that just shows that the efficiency of new technology is taking us into a win-win world. And yes, I’m aware that identity-driven thinking goes the other way too.
Crawford owns up to her bias: “I’m a fan of taxis wherever I find them.” So I should make my own bias clear. I work in the private sector technology industry, but I’m a big fan of democracy. Public transit and city-provided public services matter, warts and all. Personally, I’ve met fascinating people (travelling from San Francisco airport on the BART I met one of the authors of the Kyrgyzstan constitution) and it takes you to interesting places (the M60 bus from La Guardia into New York city takes you to 125th Street and Lexington Avenue— it took me a while to place the intersection, but who wouldn’t want to go there?) I live in a house near public transit and did so while my kids grew up so they could be mobile and independent.
Does technology drive business or does business drive technology? I see business as the lead here. And given the incentives at work, I cannot trust Uber. Its success is rooted not only in its technology, but in avoiding costs like sales tax (in many cities), like insurance, and (despite all the claims) like providing universal service. It succeeds (as Crawford hints) by avoiding the costs of being a constructive partner in the cities where it operates.
It’s easy to forget that Uber is not yet a publicly traded company, so any information about its business comes from leaks or press releases. Venture capital needs its successful exit, and as anyone who has looked for a date on Ashley Madison or a blood test from Theranos now knows, there are big incentives to paint a selective picture when billions are at stake. Once Uber is public, the incentives change: balanced books and cost control become more important. What happens when Uber has displaced taxis and then needs to squeeze its drivers a little harder?
There’s one thing that Crawford and O’Reilly (and I) can agree on, which is that the sudden interest in urban transit that Uber has sparked may be valuable. I side with Crawford, because Uber is not just about getting a ride from A to B, it’s about our cities and the scope of our democratic institutions. Cities are more than the site for consumer-driven market exchanges. Where the line gets drawn between community, government and marketplace will differ from country to country, but what is constant is that people need a say in the decision: as citizens, not just as consumers. Local government, flawed as it is, is important — and it can be innovative. Cities need tending and democracy, not venture capital, is the best tool for that job.
Self-promotion alert: I have more to say about Uber and other sharing economy businesses in my book, “What’s Yours is Mine: Against the Sharing Economy”, coming soon from OR Books.
(Attention conservation notice: the most interesting paragraph is the one about project NiFi, starting “Finley also writes…”)
According to Klint Finley in Wired, the lesson from the Volkswagen testing scandal is to use more open source software in more places. In particular, that Internet of Things (IoT) devices should be driven by open source software (even though the VW was not an IoT case). Here is Finley:
To protect consumers and realize its true promise, the Internet of Things must go the direction of the software and hardware that supports the Internet itself: it must open up…
Today, the vast majority of smart home gadgets, connected cars, wearable devices, and other Internet of Things inhabitants are profoundly closed… Ostensibly, this is for your own protection. If you can’t load your own software, you’re less likely to infect your car, burglar alarm, or heart monitor with a virus. But this opacity is also what helped Volkswagen get away with hiding the software it used to subvert emissions tests.
This seems wrong on two counts.
Finley writes about initiatives like the OpenWrt operating system for embedded devices as an alternative, but a lot of IoT devices already run on Linux. What stops individuals from being able to exert control over their gadgets is the use of Linux permissions structures, not the openness of the OS code. IoT security frameworks will be much like security frameworks on Android and other mobile operating systems: sandboxed applications running in their own user space, using the security features of the operating system. The open source/closed source distinction is essentially irrelevant to the problem.
Finley also writes that the closed nature of some IoT devices “makes it harder to trust that your thermostat isn’t selling your personal info to door-to-door salesmen or handing it out to the National Security Agency.” Which is ironic, because the software that might be handing out your personal info to the NSA is already open source. The NSA NiagaraFiles project provides routes data among different computer networks and protocols. The NSA recently released this software as open source, and it is now hosted as an Apache project called NiFi. So that is the open source community (in the form of Apache) actively assisting the NSA in its data collection activities. The core developers on the project are all from the NSA and defense contractors (link). And NiFi is being touted as a big thing for IoT applications, so that your personal info can be more effectively routed to more destinations. The NiFi project is one more step in the active collaboration of Apache with the NSA, which I discussed back here and here, and tangentially in my FORTHCOMING BOOK.
The VW case is important and raises some big questions, but open-source vs closed source software is not one of them. For a better take, see Zeynep Tufekci here.
(Full disclosure and openness: in my day job I have some involvement in IoT projects. My employer — FOR WHOM I DO NOT SPEAK –uses a mixture of open source and proprietary code in its work.)
Boston lawyer Shannon Liss-Riordan has brought two new lawsuits against on-demand delivery companies, claiming that those doing the deliveries should be classified as employees rather than as contractors. Cyrus Farivar writes about these newest cases, against GrubHub and DoorDash), while Davey Alba lists the companies sued so far:
A federal judge in San Francisco recently granted class action status to a group of four Uber drivers, making this the first case to be certified as a class action.
(If you want some background on these cases, try Susie Cagle‘s illustrated op-ed from June.)
The main argument against employee status has been the flexibility that supposedly comes with service provider status: service providers choose when to work and when not to work. Benjamin Sachs describes a new Interpretation of the Fair Labor Standards Act by the US Department of Labor, which makes it clear that there is more to the definition than just flexibility:
employee status under the FLSA is to be determined according to an “economic realities” test. With respect to the economic realities test, moreover, the Interpretation emphasizes that the test turns on a determination of whether the worker is “economically dependent on the employer (and thus its employee) or is really in business for him or herself (and thus its independent contractor).
A driver putting in 50 hours a week in a car leased by Uber, to take one extreme, is clearly dependent on Uber. A driver putting in a few hours here or there, and also driving for Lyft, is much less dependent.
More from Sachs:
What is the Nature and Degree of the Employer’s Control” Control is the last factor in the six-prong test, and it’s the last one the Interpretation discusses, but it may be the most relevant for Uber and Lyft. Why? Because the Interpretation takes up, and then dispenses with, two of the most common views about why on-demand workers ought to be considered independent contractors. First, the Interpretation states that the lack of direct supervision over how work is carried out is “largely insignificant” when workers work offsite. And, second, the Interpretation states that workers’ ability to determine when they work is also “not indicative of independent contractor status.” Citing the Third Circuit’s DialAmerica Marketing decision, the Interpretation thus concludes that “the fact that the workers could control the hours during which they worked and that they were subject to little direct supervision was unsurprising given that such facts are typical of homeworkers and thus largely insignificant in determining their status.” In other words, you can be an employee even if you set your own hours and are never directly supervised. This is a conclusion with unmistakable relevance to the on-demand debate.
Those defending companies’ rights to classify their workers as independent contractors often warn that if the court cases succeed, then service providers will lose the flexibility that is part of the appeal of “gig economy” work:
“The way we look at it, the laws governing employers require [them] to exert much more control over their employees, monitor, make sure they’re taking break times,” Ted Boutrous, Uber’s lawyer, said in a press conference last week. “It’s inevitable the flexibility and autonomy that drivers crave would have to be limited.” A spokesperson added that managing overtime would be another reason Uber would have to assign shifts.
That’s from Carmel DeAmicis of re/code, who writes very smartly about this. She goes on:
They’re stretching the truth. Labor laws don’t prohibit flexible working conditions. If drivers were legally employees, they could still drive one hour one week and 40 the next. In a business like Uber’s, where apps track when workers are logged in, it would be easy for a company to send a push notification to people after four hours of work, requiring them to take a 15 minute break, or for the app to turn off after a 40-hour workweek to prevent overtime. Monitoring drivers would be easy for a company whose algorithms have optimized pricing at all hours.
Benjamin Sachs also writes about the supposed loss of flexibility that would come with employee status:
If a court determines that these facts are consistent with a finding of employment, the drivers would be “employees.” But Uber would not somehow then be required to exercise additional control over when and how long the drivers worked, or over other aspects of the job that are currently flexible. Uber would be required to comply with minimum wage laws, safety and health laws, and anti-discrimination laws, and it would be required to contribute to unemployment insurance and withhold payroll taxes and the like. But it could do all of this without taking away the flexibility that the drivers currently enjoy.
The big problem for Uber and others is that right now they have a cheap labour force who take all the risks associated with providing the service. Uber’s business model is based on avoiding regulations: it knows full well that most of its drivers do not have proper insurance for the work they do (I am sure this is one of the reasons that they never break down driver expenses in their claims about income), and may or may not pay full taxes on their income.
Nick Carr writes about sharing power drills, following up on (my mention yesterday of) Sarah Kessler’s article about this prototypical sharing transaction, following the drill meme back to its origin and adding some interesting reflections on the complexity of sharing transactions.
A drill is a fairly inexpensive commodity. It’s easy to buy one, and it doesn’t take up much room in your house. And once you own it, you can use it any damn time you please. (The upside of low utilization is high availability.)… Buying and owning a drill, in other words, doesn’t involve much in the way of transaction costs, either up-front or ongoing.
Now if, instead of buying the drill for yourself, you decide to share it with some other people, whether through a neighborhood co-op or some rental arrangement, suddenly you face all manner of transaction costs. You have to hash out the financial arrangements, you have to figure out where the drill happens to be at the moment you need it, and you have to go out and pick it up and bring it home (burning gas, perhaps, as well as time). And if somebody else wants to use the drill at the same time you need it, then you’re in for some negotiations and probably some irritation. And if the drill breaks or gets lost (or a “little screw head” gets misplaced), a whole new set of transaction costs kick in. And, don’t forget, your knuckleheaded neighbor never recharges the battery after he uses the drill, so you’re going to wedge yourself into the closet only to find that the drill is dead. More irritation.
Transaction costs, in this context, might also be called pain-in-the-butt costs, and pain-in-the-butt costs don’t have to get very high before you say, “Screw it, I’m buying a drill.” You accept, even welcome, low levels of utilization in order to avoid onerous transaction costs. And, yes, you are being totally rational. Utilization is not everything.
Read, as they say, the whole thing.
Jathan Sadowski and Karen Gregory have a really good piece in the Guardian asking “Is Uber’s ultimate goal the privatisation of city governance?” It is sparked by Uber’s experiments in shared rides, and schedule rides through fixed stops, in San Francisco. Some have argued that Uber is looking to privatize public transport, but Sadowski and Gregory think there is something else:
the company wants to be involved in city governance – fashioning the new administrative capacities of urban environments. Rather than follow government rules, like any other utility, Uber wants a visible hand in creating urban policy, determining how cities develop and grow, eventually making the city itself a platform for the proliferation of “smart”, data-based systems.
FWIW I think they are on the right track: as cities become data-driven, city governments will increasingly be looking to new software systems to run them; but why should cities continue to be run by governments? Why not just outsource the whole thing to Silicon Valley, and specifically Uber, where the expertise in number crunching and algorithmic delivery lies? So long as we think in terms of a consumer model of citizenship where we consume services, rather than a democratic model in which we participate in the shaping of our cities, the prospect is dangerously tempting.
focused on regulating the so-called “on-demand economy” of tech-driven gig employment, to put forward concrete policy models that can help restructure the “1099” contractor relationship to offer workers greater protection. One potential model is the statutory employee framework, under which contractors are for certain regulatory purposes considered workers, generally for tax laws. NELP notes that local and state policymakers can expand this structure to provide “portable” benefits by “directly requir[ing] that companies that use IRS-Form-1099 workers abide by labor standards such as the minimum wage and others, and pay into Social Security and state workers’ compensation and unemployment insurance funds.”
She looks at the prospects for workers’ rights in the sharing economy, and with a lot of links she points out some places where action is being taken to push back against the efforts of some sharing economy companies to push all the risk and uncertainty of their business onto the shoulders of their service providers. A valuable resource.