Why Regulation Is a Tricky Business in the Sharing Economy

Sharing platforms such as Airbnb and Uber are disrupting a number of different industries, but they’re also raising a lot of policy questions. Across the country and around the globe, governments are grappling with particular problems created by these platforms. Residents complain to their local elected leaders that homes in their neighborhoods should not be rented out as hotels, and taxicab operators have asked regulators to stop what they think is unfair competition from ride-sharing companies.

New research from Sarah Light, Wharton professor of legal studies and business ethics, examines what role the federal government should play in regulating these organizations. Her paper is titled, “The Role of the Federal Government in Regulating the Sharing Economy,” and it will appear in the forthcoming book, Cambridge Handbook on the Law of the Sharing Economy. Light recently joined Knowledge at Wharton to discuss what she’s uncovered.

An edited transcript of the conversation follows.

Knowledge at Wharton: What was the inspiration for this research?

Sarah Light: The inspiration was kind of multifaceted. Being a law professor at Wharton, I’m constantly asking questions about the relationship between law and business because I think it’s really important for business school students to understand the legal and regulatory environment in which they’re operating. There are a lot of different visions of the relationship between the two, including the most cartoonish, caricature view that says that business firms passively receive legal rules dictated from above. That’s a very compliance mindset. What I wanted to look at in this research is what happens when business innovation leads to regulatory change. What are the kinds of business innovations that lead to regulatory change, and how should regulators respond?

Knowledge at Wharton: Why is it important to make a distinction between local and federal regulation on this topic?

Light: There are two important questions when we’re thinking about how to regulate business innovation or technological innovation. The first question is, what should legal rules look like? The second question is, who should get to decide what those legal rules look like? It could be the federal government. It could be the states. It could be local governments. One of the reasons why it’s important in the sharing economy context is because we have the aggregated impact of so many local actions — things that were once considered local in nature are beginning to have an impact nationwide. If you think about the kind of incumbent industries or taxis or hotels, those tend to be regulated by a local or municipal taxi and limousine commission, or by a local zoning board that sets rules on where hotels are allowed and restrictions on short-term rentals. When you have the scaling up that Airbnb or Uber and Lyft have created, these are potentially national issues in scope.

Knowledge at Wharton: You say in the paper that sharing platforms have an uneasy relationship with traditional local government. Could you explain why?

Light: I wouldn’t actually limit the claim just to local government. I would say that sharing platforms are a form of combined business and technological innovation that upend existing rules designed with a different vision of the economy in mind. In one of the papers, which is a piece that I wrote with Eric Biber, J.B. Ruhl and Jim Salzman, we try to unpack the different ways in which sharing economy platforms disrupt policy. We’ve come up with four primary ways.

The first way in which business or technological innovation can lead to policy disruption is through what we call and end-run. An end-run is a situation in which, notwithstanding similarities to the incumbent industry, the innovative business argues that it is not subject to regulations governing the incumbent. This is Uber saying, “We’re not a taxi. You shouldn’t require us to get medallions to operate in New York City.”

“What are the kinds of business innovations that lead to regulatory change, and how should regulators respond?”

The second type of policy disruption is what we call an exemption. This is a case in which there is existing law, and the innovative business fits into an explicit legal exception. But it raises the same kinds of policy considerations that the legal regime was originally designed to address. As an example of an exemption, I would talk about Airbnb. Under federal civil rights laws — both the Civil Rights Act and the Fair Housing Act — there are rules that say that hotels and landlords are not allowed to discriminate on the basis of race, gender, religion, etc. Those are very strong norms about preventing discrimination. But those laws have express exceptions for a homeowner renting out space in their home. The challenge is when you have Airbnb, which now is renting out more rooms per night that many major hotel chains, you’ve scaled up the nature of this exemption in a way that it is having the same negative public interest effects of potentially discriminating against people that the Civil Rights Act and the Fair Housing Act were originally designed to prevent.

The third type of policy disruption is what we would call a gap. This is where there’s literally no existing regulatory regime. The thing is so new that nothing clearly applies. That would be something like automobiles at the turn of the century.

The final category of policy disruption that we articulate is a solution. This is the idea in which an innovator solves a regulatory problem but may be blocked by over-inclusive legal rules. There are two real examples of this. One is distributed solar generation, which is potentially blocked by existing legal rules governing utilities in many states. The second example is the idea that Tesla wants to forward-integrate, to sell its cars directly to customers. Many state laws actually prohibit auto manufacturers from direct sales to consumers out of a concern to protect the franchise operations. Tesla wouldn’t be undercutting its franchises because it doesn’t have any. But these laws would potentially block those direct sales.

Knowledge at Wharton: How do we figure out where the federal government should step in and where it should back off and let the local governments take the stage?

Light: I think there are three primary areas in which the federal government should be playing a role. One relates to anti-discrimination. A second relates to privacy. The third relates to the general diffusion of policy experimentation by state and local government actors.

With the example that I gave about Airbnb, we have anti-discrimination laws at a national level for a reason. When we’re thinking about the hotel industry or rides in vehicles, these are classic examples of interstate commerce. If, as some emerging empirical scholarship is demonstrating, these platforms mimic or amplify discriminatory sentiment in the United States, then there is a concern that failing to regulate them at the national level will interfere with interstate commerce. So, I think those are national anti-discrimination norms that need to address the problem.

The second area relates to privacy and security. You provide a whole lot of information to Uber and Lyft when you sign up on their platforms. They have access to a lot of information that you’re not providing but simply by tracking your movement. A very well publicized, kind of sad example was a few years ago, when it became known that Uber had something called God View. It was tracking where people were taking an Uber on a Friday night and figuring out where they were going on a Saturday morning to track people’s dating behavior. This led to articles in the press, and people were very concerned. It doesn’t depend on whether you live in Miami or New York: Generally speaking, consumers want their privacy protected.

The final area where I think the federal government should play an important role relates to the diffusion of policy experimentation. Many people use the phrase that the states are “laboratories of experimentation.” This comes from a Supreme Court opinion by the late, great Justice Louis Brandeis. Obviously, there’s a lot of experimentation going on right now in cities and municipal governments around the country. Some are partnering with Uber and Lyft in order to provide access to public transportation. They’ll provide a discount if you take an Uber ride that drops you off at a regional transit hub. Where there are policy successes, it’s not necessarily always clear that those policy successes can be known by other municipalities around the country. I think the federal government should be playing a coordinating role to help publicize local policies that are successful, and possibly ones that are not successful, so people don’t have to reinvent the wheel.

“I’ve argued that it’s important to have a kind of adaptive, dynamic regulatory system that allows for learning and change.”

Knowledge at Wharton: How is the federal government doing so far at regulating the sharing economy?

Light: I think the federal government is moving quite slowly in this regard. Honestly, there’s just a lot more action happening at the municipal level, and in some cases at the state levels, because that’s often how the incumbent industries have been regulated.

For example, many cities have had to make the determination about Uber and Lyft. Are these taxis or not? Uber and Lyft lobbied heavily to not face 500 different municipal rules but rather have this issue resolved at the state level. More than a dozen states have passed laws that permit Uber and Lyft to enter the market statewide. They set uniform rules on the insurance coverage that’s required. They govern the background checks and safety checks that are needed. In some cases, many of these rules actually preempt local governments from adopting additional regulations, with the exception of maybe rules about who gets to be picked up at the airport. But that’s where the action is with respect to Uber and Lyft. With Airbnb, a lot of the action is at the local level with zoning boards, more than at the federal government level.

Knowledge at Wharton: You point out that when something innovative is in its infancy, who should regulate and how to regulate should be dynamic as that industry is developing. But are governments generally comfortable with that?

Light:  It’s a really good question. In my work I’ve argued that it’s important to have a kind of adaptive, dynamic regulatory system that allows for learning and change, when conditions warrant it.

The example that I give in the paper is on environmental law policy. There’s very little empirical research right now on the environmental impact of Uber and Lyft. If you think about it, there’s great environmentally positive potential. If I decide not to buy a car because I generally take public transportation, and if take Uber or Lyft to get somewhere that I can’t go with public transit, then that’s good for the environment. On the other hand, if you just take a Lyft every day, then you’re replacing public transportation with more rides. The question of whether Uber and Lyft are increasing or decreasing vehicle miles traveled and what their implications are for local pollution and greenhouse gas emissions is not determined yet. But they’re out there. We can’t leave them unregulated.

From my perspective, it’s important to allow more policy experimentation, often at the local level, while we’re gathering the empirical data about what the impacts are. Once there’s more data, then maybe the regulatory approach will converge on a single solution or two solutions. But I think the experimentation is very important when there’s uncertainty about what the impacts are.

Knowledge at Wharton: Are consumers comfortable with experimentation on the regulatory front? Do they even notice?

Light:  I’m not sure how much consumers are necessarily aware of it. There was a point at which Uber and Lyft were illegal here in Philadelphia, but they were still operating. It’s not necessarily the case that the legal regime governs what consumers do. It might govern what drivers do if vehicles are impounded, as they were in Philadelphia a couple of years ago. The entities that I think notice are Uber and Lyft. They care a great deal about what the regulatory environment is. From their perspective, uniform rules are better. Less regulation is better.

The city of Austin, Texas, provides a really interesting example of how local experimentation doesn’t necessarily lead to a happy outcome — or maybe it leads to a very happy outcome, depending upon your perspective. There was a push for a statewide rule that would preempt local and municipal governments from adopting any kind of standards with respect to Uber and Lyft. It didn’t pass, which meant that municipal governments were allowed to experiment with their own rules. Austin insisted that Uber and Lyft drivers be fingerprinted and undergo the same kinds of criminal background checks that taxi drivers undergo. Uber and Lyft threatened to pull out if this became the law, and they ultimately did pull out of the city. Other ridesharing platforms entered the city shortly thereafter.

From the perspective of the public interest, maybe it’s a very good thing that the city of Austin insisted on fingerprinting. We’ve heard stories about unfortunate incidents involving criminal activity with these platforms, just as you hear about unfortunate incidents involving criminal activity with other transportation services. But from the perspective of consumers, I think there was a lot of support for having these platforms in the city. The loss for them may have been a significant one. Thinking about it from the viewpoint of the regulator, I think the security interest is very important. From a city’s perspective, if they want to insist on criminal background checks, they should be entitled to. The challenge becomes when there’s more of a patchwork across different jurisdictions.

“Let’s let cities and local governments and states experiment. Then at a certain point, maybe some uniform rule may be more appropriate when we have our information.”

Knowledge at Wharton: These sharing platforms also put ordinary people in a position of having to be aware of regulations that maybe they’d never been aware of before. An Airbnb host probably was not schooled in hotel regulations prior to becoming a host.

Light: Absolutely. This is another example of the scaling-up problem and the way in which the regulatory state is designed with a particular vision of what the hospitality industry looks like. If there are local rules in a city that say that there must be ingress and egress, there must be a monthly fire inspection, that makes perfect sense if you have a full-time hotel staff able to meet with your fire inspector. But that makes a lot less sense for Aunt Florence in Peoria, who may not have the ability to meet with a fire inspector or expend all the resources to meet with the same kinds of regulatory burdens that a hotel could.

Knowledge at Wharton: What is the key takeaway of this research for policymakers as they continue to navigate this space?

Light: On the question of who should regulate, I think the key takeaway is that when we’re dealing with something really new, experimentation is good. As we’re learning more information about what the potentially negative implications of a business are that would trigger some kind of regulatory response, let’s let cities and local governments and states experiment. Then at a certain point, maybe some uniform rule may be more appropriate when we have our information. I think experimentation and dynamism and adaptive learning are very good when businesses are new.

On the question of how to regulate, the work that I’ve done argues that right now the law is drafted in a way that is not neutral between the incumbent industry and the innovative business, so we should strive for some kind of neutrality based on the function. If Uber acts like a taxi, even if it’s organized in a different way and uses a smartphone app and rents cars from people rather than owning fleets of vehicles, it’s performing the same kind of service and raises the same kinds of public interest concerns as taxis. We should think about how to regulate them in a way that doesn’t cause a massive drop in the price of medallions and essentially exclude taxis from the market. Nor should it exclude Uber from the market. We want to level the playing field between them. The research that I do focuses on how to operationalize that idea of neutrality across the two organizational forms of the incumbent and the innovator, while balancing the public interest concerns and the desire to promote innovation.

Knowledge at Wharton: What’s next for this research?

Light: I am continuing to think about the relationship between business innovation and technological innovation and the regulatory state. I’ve been working on a project on autonomous vehicles, or self-driving vehicles, looking at the question of who should get to regulate. Should it be at the state level? Should it be at the federal level? The House and Senate have both introduced legislation that would preempt states from regulating in this area.