I'm a little late to organizing my thoughts here but figured I'd publish anyway. For those of you who still have time to vote in California, I ask that you vote for Prop 22. If you are not in California or are reading this after the vote, I hope this provides a useful perspective on the situation.
Prop 22 is a ballot proposition in California which would overturn the impact of AB5 on the gig economy (ridesharing, food/grocery delivery) which had required companies to reclassify workers as employees. Prop 22 would provide a new class of worker in between independent contractors and employees for gig economy workers, providing a guaranteed minimum wage for on-trip time, healthcare subsidies, and extra insurance protections. If you’d like to read more about Prop 22, see here.
I sympathize with the goals of the legislators behind AB5 and understand they are trying to help Uber drivers, many of whom do not make a lot of money. Having worked at Uber for two years, I care deeply about helping Uber drivers and improving their lives.
But that is precisely why I am so against AB5 as I believe, like all too common with government legislation, that it will greatly harm the people it is trying to help. I will walk through my argument below. For simplicity, I will focus this piece on Uber drivers, but the arguments hold for all gig economy workers impacted.
Uber drivers are far closer to the typical independent contractor than employee
Uber drivers fit squarely within our traditional understanding of independent contractors.
Let us review all the ways in which Uber drivers match the traditional IC, not employees. For simplicity, consider whether each one of these matches more closely with a home repairman on Taskrabbit ( traditional IC) or a retail worker at Best Buy (traditional employee).
Below are things Uber drivers can do which are unheard of for a traditional retail employee:
1) Set your own overall hours each day. Uber cannot demand or restrict your hours.
2) Work for multiple gig economy companies with no recourse from any of them.
3) There is no minimum work threshold. You can work 0 hours a week, 60 hours a week, or anywhere in between.
4) You can work whenever you want. Morning, afternoon, night, weekdays, weekends.
5) You can work wherever you want. City center, suburbs, etc.
6) You can choose to accept or reject whatever rides you want. As of 2020, Uber drivers have full freedom to choose which rides to accept with no impact on their ability to stay on the platform.
7) You can quit at any point and come back to the platform with no consequence.
8) You do not have a direct manager or a manager of any kind. There is no interview, no job requirements of any kind except background check and proper vehicle
The flexibility with driving for Uber is immense, which makes sense if you view Uber as a platform (like Airbnb) and not an employer.
Now Uber drivers don’t match 100% to traditional ICs. But the main argument along these lines from reclassification supporters is that Uber drivers can’t charge their own price. But as of July 2020, this is not true: all Uber drivers can set their own price in California.
Therefore, to the extent you believe in a necessary distinction between ICs and employees, it’s hard to argue that Uber drivers should fall under the employee bucket.
Uber drivers value their flexibility immensely and most do not want to be employees
AB5 supporters severely underestimate the value that flexibility provides to drivers.
Unlike nearly all employees, Uber drivers have total control over their schedule. You tend to find that Uber drivers work hours which are complementary to full-time work, not overlapping (the chart below compares it to the American Time Use Survey claims for work). This makes sense as most Uber drivers are part-time, not full-time.
I’ve heard of many cases of Uber drivers using this flexibility to care for loved ones, attend school, manage their own health issues, or travel to their home countries for long periods of time. Remember, Uber is one of the few options people have if it's the 25th of the month and they are behind on rent. The ability to press a button and make money is a true innovation which is still underrated. This flexibility is especially unheard of among low-wage employees, where unpredictable hours outside of workers control can wreak havoc on their lives.
“A more recent survey of California drivers showed that 88% began driving because they needed a job where they could choose where and when they work — and 68% of drivers said that they would stop working if the flexibility they have today was no longer provided.” (Source)
One paper calculates the value of this flexibility is immense: if you forced drivers to work in a less flexible working arrangement that is short of employees (ex: taxi shifts), you’d need to pay them anywhere between 50-100% more in order to ensure they continue driving!
This matches the survey results commissioned by Uber which indicate 82% of drivers favor Prop 22 and a similar amount prefer staying an IC. Even if we account for the skew of part-time vs full-time drivers, and the fact that Uber paid for the poll, it is still an overwhelming response by drivers which roughly matches other surveys I’ve seen.
When Uber drivers tell us they value this flexibility greatly, often at the expense of taking a job as an employee, we should listen. The answer is not forcing them into a work arrangement they’d rather avoid and eliminating options.
AB5 would destroy valuable driver flexibility
The flexibility that Uber drivers value so much would be destroyed with AB5. Critics say nothing in the legislation demands that Uber restrict driver flexibility, but basic logic and economics tells us it must. No large company with employees operates like the platform that Uber does. This Uber economics blog post puts it well:
“Think about it this way: Starbucks offers one of the most flexible part-time jobs around, but baristas can’t just walk in unannounced, decide they will only make lattes while refusing all orders for cappuccinos, leave during the morning rush to go pick up their kid from school (without permission from their boss), and return to work at a Peet’s Coffee. That would absolutely be allowed under the law, so why doesn’t it happen?
The answer is simple: businesses simply won’t survive if they have zero control over what their hourly employees, whether full- or part-time, actually do. Imagine a restaurant that offered their employees the same freedoms that drivers currently have with Uber. On any given night they might have a single waiter trying to serve 100 customers or 100 waiters all serving one customer. In the middle of preparing a dish, the chef could leave to go cook for the food truck across the street. And anytime their employees decided to be on-shift, they would have to be paid, whether they had any work to do or the restaurant was closed. Why should anyone believe this same logic wouldn’t also apply to companies like Uber?”
In practice, Uber drivers becoming employees would mean working specific shifts, having restricted hours (few will be able to work overtime), only working for one company, taking every trip request they are sent, and not being able to take vacation when they want.
Remember, Uber drivers value flexibility so much that many would need to be paid 1.5-2x more if you forced them to be an employee (otherwise they would just do something else). This means that even if AB5 raises wages for many drivers, it will likely make many worse off because they now lack the flexibility they greatly value and will make it more difficult for anyone to join the platform and access quick income, as I will explain more below.
AB5 would result in most Uber drivers losing access to income
So if all Uber drivers become employees, Uber will be incentivized to have full-time employees driving shifts. This, in combination with the rise in overall rider prices, will result in most California Uber drivers losing access to the platform.
The Uber economics team (whose work is always top-notch) put out an analysis on the impact of AB5. They estimate 76% of California drivers will lose access to work, with the figure as high as 95-100% in some smaller/rural towns.
The drop in work opportunities is so large because most drivers in California work part-time. 74% work less than 25 hours (providing nearly half of trips), while only 9% work at least 40 hours.
Furthermore, "full-time" one week does not mean "full-time" the next:
“Among drivers who typically work the equivalent of full-time hours, 1 in 3 drivers takes off at least one week per quarter, and 1 in 4 takes off two or more weeks... Even those drivers who work every week typically increase or decrease their hours by around 25% from one week to the next.” (Source)
Of course, AB5 would mean many full-time drivers getting a raise (though losing the flexibility they also value). But it’s incredible that many supporters of AB5 are unwilling to acknowledge this tradeoff or admit that their policy preference may put most drivers out of work. I don’t believe the way to help drivers is by ensuring 75% can no longer earn on Uber.
In addition, AB5 would result in significant increases in rider prices 20-60% in major cities. Despite the popular narrative that Uber is mostly about serving rich people, Uber is relied upon by people from all walks of life, including low-income workers getting to work or anyone without access to a car. Uber is also often a complement to public transit (makes it easier to get to bus or train stops), not a substitute. This a very real cost which can’t be ignored, even more so during a pandemic where many rely on Uber to get around safely.
Uber driver pay is comparable to other low-skill jobs
Driving for Uber is a low-skill job. Nearly anyone with a driver’s license can do it, there are no language barriers to entry, and there is no discrimination based on your age, race, ethnicity, or anything else. In many ways, Uber is the fall-back option for everyone. If you need to make money, Uber is there in a way no other company is.
And so while driving for Uber is not likely to be a lucrative job, it pays comparatively to similar jobs in retail, fast-food, and restaurant work. Most studies on driver wages are biased downward, either by relying on bad data, not accounting for dual-apping, or making unreasonably high expense assumptions.
Uber data, however, shows drivers grossing $20-25 an hour on average between 2015-2017 (consistent with recent estimates too). Not that this is among a period of large changes to prices across many major markets, showing that Uber has far less control over driver pay than you think.
Expenses vary by driver and city, but one careful Seattle study puts expenses at around $2-3 per hour for most drivers and at most $8 an hour for the few drivers buying a new car to drive.
While these net earnings may not sound like much, they compare favorably to other types of low-skill jobs (which don’t provide the same flexibility) and they tend to be even higher in major cities (ex: Seattle study estimates net earnings at $23 an hour, much higher than the minimum wage).
While some drivers are bound to make less than minimum wage given the nature of the platform (anyone can work at any time), most drivers do not. Even when making the minimum wage, the benefits of working in an IC job with far more flexibility and often in a part-time capacity means many workers prefer Uber over other comparable low-skill jobs. Uber can often be a boon for many people who need to supplement their income or are having trouble finding work.
Being an Uber driver is not for everyone. But I don’t believe making it illegal for someone to make this decision or trade-off for themselves is in their best interest.
Economically, Prop 22 is a much preferred alternative to AB5
I believe the current situation for Uber drivers could see improvements. But reclassifying Uber drivers as employees does far more harm than good. Prop 22 as a “third way” makes a lot more sense, even if imperfect.
Prop 22 makes some key improvements: payments for benefits, occupational accident insurance, disability and accidental death insurance, all of which make sense to me. The on-trip minimum wage provides a floor while on trip, though I personally prefer letting drivers set their own rates given the lack of ability for Uber to control wages without cutting driver flexibility.
And remember, AB5 requires more than just a minimum wage. It lays out a number of cumbersome requirements like 1.5x overtime, meal and rest breaks, the costs of managing and monitoring employees directly, along with a litany of other regulations which don’t really make sense in the context of Uber’s relationship with its drivers.
If your goal is to raise driver wages, there are much better options. NYC instituted a driver pay law which, although it has reduced flexibility somewhat and raised prices for consumers, nonetheless kept drivers’ IC status and would therefore be a preferred model to AB5. Similarly, Uber has rolled out health insurance, sick pay, parental leave and bereavement payments in Europe while maintaining IC status.
Wanting to make changes to improve Uber drivers’ lives does not mean we should make them employees.
Legally, AB5 is an arbitrary and poorly applied standard for settling the employee/IC question
AB5 is a bad idea economically, but it also doesn’t hold up in a legal context.
The courts have ruled that Uber drivers are employees under AB5’s ABC criteria. I am not here to dispute that legal question, though Uber has.
What I would argue is AB5 is a poorly written law which groups workers which should be classified as independent contractors (ICs) as employees. You don’t have to take it from me though. The California legislature itself has granted exemptions to over 65 industries, including doctors, lawyers, psychologists, broker dealers, real estate, and insurance brokers.
With a new round of exemptions passed in September for many types of freelancers (in response to an overwhelming negative reaction to AB5), it’s easier to count the number of industries not exempt at this point. If the three-prong test makes so much sense, why should it only apply to Uber and Lyft? The law comes off as arbitrarily applying different rules for gig economy companies than everyone else, while also wreaking havoc on other industries which get caught in the enforcement.
And while not perfect, the IC/employee distinction is an extremely important part of how our modern economy functions. 10-15m Americans are independent contractors and many industries rely heavily on independent contractors and simply could not function if all workers needed to be employees (see a wikipedia list below). Moreover, it’s not just about employer interest. Not everyone wants to be an employee given the tradeoffs associated (as we discussed earlier).
Whatever your concerns about employee classification in general, AB5 is not the answer.
Conclusion: Prop 22 is the Best Solution for Drivers and Consumers Alike
I’ve laid out the many reasons why AB5 is a poorly reasoned and economically disastrous law for gig economy workers. I strongly believe that Prop 22 is a good, even if imperfect, middle ground which will add some protections but preserve the flexibility drivers so value.
The argument for AB5 in a large way rests on the notion that drivers don’t know what’s best for them. AB5 destroys the flexibility they all claim to want, forces part-time drivers to become full-time workers, and boxes them into a work arrangement many have been trying to avoid. Putting 75% of drivers out of work, many of whom lack other options during a pandemic, is not the way to help them. Raising prices significantly on riders which rely on Uber to safely get around during these times is also not the solution.
Uber has provided real innovation in the ability for people to work in ways impossible 20 years ago. Let us embrace that innovation, and not eliminate it with paternalistic government action.
I hope voters can make the right choice.