Rideshare Licensing Fails the Consumer Protection Test

By Ryan Bourne and Nathan Miller of Cato.

"Occupational licensing is sold as consumer protection that screens out incompetent or dangerous workers. Today, about 22 percent of the American workforce needs a government license to do their jobs, and that safety and quality argument is what state and local lawmakers use to justify the entry barriers, licensing fees, and training hours that typically accompany it.

The trouble is that the claimed benefits have historically been difficult to test or measure, so the debate over whether licenses really protect anyone was fueled mostly by anecdotes. That’s why a new NBER working paper by Jonathan Hall, Jason Hicks, Morris M. Kleiner, and Yun taek Oh is so interesting. It leverages Uber data and finds no consistent evidence that a government occupational license improves riders’ ratings or drivers’ behavior on the road. The results suggest that, once a platform is already screening and monitoring its drivers, licensing adds no detectable, additional consumer protection.

Why Uber Is a Good Testing Ground

Uber’s vast datasets from trip tracking and consumer ratings have proven a goldmine for researchers on an array of economic questions. 

One of us has written before about how data on female and male drivers’ earnings gave good insights into non-discriminatory causes of gender pay gaps. Now similar data can be used to examine the effects of occupational licensing laws for rideshare drivers.

Uber dispatches drivers mainly by proximity. So when a licensed and an unlicensed driver are both near a rider, which one is dispatched is essentially a coin flip. The researchers use that quasi-random assignment along with data Uber collects about the ride quality, such as the rider’s star rating and telematics data on how carefully the car was driven (by measuring things like hard braking and hard acceleration, which are tied to greater crash risk).

The authors exploit two settings that allow a test of licensing’s effects. First, New York City licenses rideshare drivers through the Taxi and Limousine Commission, which, at the time covered by the study, required a fingerprint background check, a defensive-driving course, and a 24-hour vehicle for hire course and exam. Neighboring New Jersey does not have these requirements, yet drivers from both areas serve the same New Jersey riders. 

Second, in 2017, Texas preempted local occupational licensing of rideshare drivers, abolishing Houston’s restrictive licensing requirements. That let the researchers compare previously licensed drivers with those who entered the market after the entry restrictions were lifted.

The researchers sampled 213,000 trips in the New York area from April to August 2017 and nearly 497,000 in Houston from September 2017 to January 2018 to look for differences between drivers with and without an occupational license. They checked for differences across seven metrics:

1. Star rating (1–5)
2. Proportion receiving ratings less than 5 stars
3. Proportion receiving 1‑star ratings
4. Prevalence of hard braking
5. Whether more than 20 percent of brakes in a trip were hard brakes
6. Prevalence of hard acceleration
7. Whether more than 20 percent of accelerations in a trip were hard accelerations.

No Sign Licensing Helps Anyone

Across both metro areas, the paper finds no consistent evidence that licensing improved consumer outcomes. 

In the New York vs. New Jersey comparison, licensed drivers actually earned slightly lower ratings than their unlicensed counterparts—0.0228 fewer stars on Uber’s five-star scale, or less than half a percent off the unlicensed mean score of 4.78. 

Just one of the seven metrics compared came out in licensing’s favor. Licensed drivers logged fewer trips with a high share (20 percent or more) of hard-braking events, yet even that result didn’t survive robustness checks the authors undertook. It faded once the authors widened their sample window and swung wildly depending on whether they controlled for vehicle model and year. 

The results were similar in Houston. The researchers could not detect a statistical difference between the drivers who entered after deregulation and previously licensed drivers on all seven outcomes, despite the two groups differing sharply in experience (288 prior trips versus roughly 2,584 for the licensed) and age. This is strong suggestive evidence against the common fear that ending entry requirements would flood a market with worse providers.

When previously licensed drivers were instead compared with unlicensed drivers who had already been on the platform, the only differences ran against licensing: on three of the four driving-behavior measures, the previously licensed drivers braked and accelerated harder, with roughly 15 percent more of the high-hard-braking trips that flag risky driving.

Licensing Is a Costly Regime

A large body of research has already established the costs associated with occupational licensing. A strict licensing regime restricts entry into a market, thins the supply of workers, and thus tends to push prices up. 

As was summarized earlier this year in Cato’s Handbook on Affordability, stringent nurse-practitioner scope-of-practice rules raise child checkup costs between 3 and 16 percent, home improvement jobs are between 15 and 50 percent more expensive (depending on the job type) in states with the strictest home technician licensing regimes, and in one Virginia case study of hair braiding, the number of beauty shops grew 7 percent faster than in bordering states after Virginia deregulated in 2012.

So, although this paper doesn’t examine how occupational licensure affects fares, it’s reasonable to think based on existing literature that licensing rideshare drivers simply means fewer drivers and (absent substitute platforms outside the license regime) higher fares. This paper supplies evidence that despite that added cost, licensure often buys nothing in the way of improved quality or safety. 

Rideshare markets already have methods that go a long way to vetting drivers’ quality—reputational feedback, ratings, reviews, or the threat of account deactivation. That makes further government licensure redundant. The new paper is a strong complement to what Cato argued in our handbook earlier this year: the cheapest way to lower the price of a service is often just to let people provide more of it."

The Food Revolution Nobody Planned

By Byron Carson.

"Long before policymakers conceived New York City Groceries, entrepreneurs fed people. They did it cheaply, and they made people better off in unimaginable ways. 

The approximately 63,000 grocery stores in the US (according to 2023 Census estimates) indicate a great deal of competition, which encourages owners to lower prices and improve food quality. The figure below indicates that these stores must also compete with approximately 650,000 domestic food and beverage retailers.

That’s a lot of competition, and it helps lower prices while producing a dizzying array of goods and innovation that makes the mythical land of Cockaigne look paltry. While higher prices understandably attract attention — recent CPI data for food indicates a 3 percent rise over the previous year — innovation takes place in myriad ways that are easy to overlook.

Open-air marketplaces and food halls are just some of the ways people innovate. To clarify, think about the mundane problems associated with acquiring food. There are myriad transaction costs associated with finding a seller interested in selling what you want, assessing the quality of an item, trusting that the seller isn’t cheating you, and so on. Entrepreneurs develop public or open-air markets and food halls to earn profits, and they do so by lowering transaction costs for consumers and providing innovative services. 

Historical and modern examples alike indicate the extent of innovation. Since 1742, Boston’s Faneuil Hall, or “The Cradle of Liberty” (and later the Quincy Market), has provided organized spaces for people to buy and sell food in mutually beneficial ways. The Redding Terminal Market is now a popular indoor marketplace and food hall, but it developed from the open-air markets in Philadelphia during the late nineteenth century. Similar kinds of commercial activity even developed in New York, perhaps ironically, in places now being touted as sites for government-owned grocery stores. La Marqueta, or the Park Avenue Market, emerged from the haphazard commerce underneath a viaduct, where people sold goods to customers willing to pay for them. Such commercial activity expanded so much that by 1936, the city enacted a more orderly marketplace. 

Food halls, food courts, food truck parks, and similar ventures are now common, reflecting entrepreneurs’ efforts to earn profits by providing goods and experiences customers value.

On recent trips to Tampa and London, I learned of two interesting examples of people developing food halls through private enterprise. In Tampa, the Oxford Exchange (opened in its current form in 2012) was once home to an arcade of shops in the 1920s. It was converted from a bookstore in the 2010s and now hosts several dining rooms, a coffee and wine bar, lounges, meeting spaces, and shops. London’s Mercato Mayfair (one of several food halls owned by Mercato Metropolitano) is located in a deconsecrated church and houses various food and beverage vendors.

From my brief visits, these places seem like vibrant focal points where eating is perhaps the least interesting thing to do. They offer comfort and service, novel and quality foods, places to meet, and more. The profit motive and the search for innovation, not governmental food policy, drive these efforts.

WorldFoodTrucks offers another case of private food innovation. Located in Kissimmee, Florida, WorldFoodTrucks is the first and largest food truck park in the US, where customers can find more than 100 trucks, each offering different culinary options, including more than 80 world cuisines. The park is open every day of the year (until 4 am on weekends), and its operators report serving more than 13 million customers since the park opened around 2023. Not only does the park offer food, but it also provides a convenient and inexpensive way to feed larger groups of people with interesting cuisine at corporate and celebratory events, such as graduations and weddings. Such market-driven opportunities are innovations because they satisfy several goals people value beyond simply providing cheaper food. 

The economist Steven Horwitz wrote that grocery stores are indicators of American progress. We should broaden these indicators to include food halls, food trucks, and other ways entrepreneurs try to feed people. Such progress benefits picky eaters, the fitness-conscious, people who want to grill out on a nice day, those trying to meal prep for the week, people making dinner for date night, those preparing a family meal on a tight budget, and many others. Progress is progress for rich and poor alike, and it serves myriad individual goals.

Rather than devise policies that consistently fail to make people better off, perhaps we should understand how entrepreneurial activities actually put food on the table and make people better off in ways they value."

  

Health-care costs for typical Canadian family will reach over $21,000 this year

By Nadeem Esmail, Nathaniel Li and Milagros Palacios of The Fraser Institute.

The Price of Public Health Care Insurance, 2026

• Canadians often misunderstand the true cost of our public health care system. This occurs partly because Canadians do not incur direct expenses for their use of health care, and partly because Canadians cannot readily determine the value of their contribution to public health care insurance.
• In 2026, preliminary estimates suggest the average payment for public health care insurance ranges from $6,464 to $21,115 for six common Canadian family types, depending on the type of family.
• Between 1997 and 2026, the cost of public health care insurance for the average Canadian family increased 2.3 times as fast as the cost of food, 1.7 times as fast as the average income, and 1.5 times as fast as the cost of shelter. It also increased much more rapidly than the average cost of clothing, which has fallen in recent years.
• The 10 percent of Canadian families with the lowest incomes will pay an average of about $637 for public health care insurance in 2026. The 10 percent of Canadian families who earn an average income of $88,572 will pay an average of $8,644 for public health care insurance, and the families among the top 10 percent of income earners in Canada will pay $66,350.

 

The cost of lighting has decreased by over 1000x

Tweet from Paul Graham. 

"One data point in technology making everyone richer: the cost of lighting has decreased by over 1000x."

  

 

The Tax Gains from Moving Across State Lines

Comparative tax burden · Tax year 2026

By Daniel Di Martino of The Manhattan Institute. 

"Americans have been moving from high-tax states to low-tax states for a long time, but the trend has become more acute since the Covid-19 pandemic hit the world in 2020. The increasing availability of remote work allowed many workers to move elsewhere and keep their jobs, and many companies also chose to relocate or reduce office occupancy. In addition, Americans are increasingly self-sorting according to political preferences. Most coverage of this trend has focused on the rich and how much they have to gain by moving from high-tax to no-income-tax states. Obviously, multimillionaires can keep more of their income if they move from a high-tax jurisdiction like New York City to Palm Beach, where there is no state or local income tax. My new income tax tool shows that not only the rich, but also low- and middle-income Americans, have a lot to gain from moving across state lines.

Take a couple earning $120,000 in New York City. The husband has a decent job paying $100,000, and his wife makes $20,000 working part-time. They have two children still in school. That couple does not benefit from itemizing deductions in their federal tax return, so they take the standard deduction and owe $10,040 in federal income taxes. Since they have two minor children, they will receive a $4,400 child tax credit to offset their tax liability.

Since they both have traditional jobs, their employers will owe $9,180 in Social Security and Medicare payroll taxes, while they will pay the same amount from their salary in payroll taxes, out of an effective compensation cost to their employers of $129,180. Since they live in New York State, they owe $5,186 in state income tax and $581 in payroll taxes for paid family and disability leave (both spouses contribute), but they also benefit from an $800 state child tax credit. Finally, since they live in New York City, they will pay an additional $3,727 in city local income tax.

All in all, out of a compensation cost of $129,180, the couple pays $32,693 in income and payroll taxes, or 25.3% of their income, leaving them with a take-home pay of $96,487. This couple faces an effective 36.3% marginal tax rate.

Would this couple be better off if they moved from New York City’s metropolitan area to the Nashville metropolitan area? Imagine the cost of this move is that the wife loses her $20,000 job and thus their income falls. This is a big hit, but they would pay no state and local taxes, and their federal income tax would also be much lower due to the progressive tax structure. Their federal income tax would be, net of the child tax credit, just $3,240, while their payroll tax liability would fall proportionally. Out of a new employer compensation of $107,650, this couple would pay a total of $18,540 in payroll and income taxes, for a take-home pay of $89,110. But every dollar goes much farther in Nashville than in New York City, as housing and everyday goods and services are cheaper—specifically, 14.48% cheaper. While New York City is 12.6% more expensive than the average U.S. territory, Nashville is 3.7% cheaper. Thus a take-home pay of $96,487 in New York City is equivalent to $85,690, while one of $89,110 in Nashville is equivalent to $92,534. In other words, even with a $20,000 lower nominal income, a married couple with two kids can still increase their real take-home pay by nearly 8% by moving across state lines. If they managed to keep their full income, or the wife later found another job, their real income would actually increase to $109,221—over $23,000 in additional real income, a 27% increase.

Now take the case of a middle-income single worker in Los Angeles, making $60,000 per year. His take-home pay would be $47,961. Say he lives in the Los Angeles metropolitan area, so his price-adjusted take-home pay is lower, at $42,219. If he moves to Orlando in Florida, his take-home pay will rise to $50,390, and adjusted for cost of living it would be $49,694. In other words, his after-tax pay rises by $2,429 per year, or over $200 per month, and his price-adjusted after-tax pay rises by $7,475, or over $600 per month—a nearly 18% increase.

While tax and cost-of-living gains are increasing with income, even the lowest-income earners in the United States can see increases of over 10% by moving across state lines, while the richest can see gains of over 30%." 

Does Bank Consolidation Harm Customers?

By Jeffrey Miron of Cato. 

"Antitrust policy presents a challenge for both libertarians and policymakers. On the one hand, competitive markets are good, which might suggest policy should limit firm mergers. On the other hand, mergers can have beneficial effects (such as economies of scale and scope, or disciplining unproductive firms), so broad opposition to mergers is likely counterproductive.

New research on bank consolidation offers evidence on this tradeoff. Contrary to the belief

that bank mergers reduce competition, increase borrowing costs, and limit households’ access to credit, … [the study finds that m]ergers have no meaningful effect on interest rates, approval rates, or late payments. Merged banks do not appear to use their increased size to charge borrowers more or restrict access to mortgages.

This may be due to

the intense competition in local mortgage markets. The typical county has more than 130 active mortgage lenders per quarter, and the median lender controls just 0.4 percent of its local market. Therefore, even when two banks merge, borrowers generally continue to have many other lending options. In some cases, local competition actually increases after mergers.

Whether these conclusions apply in markets with only a few firms, where mergers might substantially increase market concentration, is harder to know. But this evidence should still remind antitrust and banking regulators to consider the full range of effects from mergers, not just the impact on concentration per se."

By the Time Governments Are Regulating AI, They’re Regulating the Past

By Mark Jamison of AEI.

"AI is changing fast. And spreading fast. Both are problems for people seeking to regulate it.

Regulation works best when regulators understand what they are regulating. That is not the case for AI. Costs are collapsing, having dropped 1000-fold for large language models since 2023. At the same time, capabilities are expanding, business models are changing, and market leaders are turning over rapidly.

These dynamics create both opportunities and problems. Now almost anyone can use AI to manage their household or launch a business. But, as Bill Gates recently noted, they can also create deep fakes, launch phishing attacks, or break into internet sites, as happened to Hugging Face.

This also makes regulation hard: Rules written for the AI that regulators see today will no longer exist by the time the rules take effect.

Nevertheless, many people want regulations that would control AI. The EU has embraced what it calls comprehensive AI regulation, in which regulators judge the relative riskiness of AI applications and systems and then apply controls ranging from outright prohibitions to light-touch oversight. Some people are calling for mandated surveillance of AI users, restrictions on model capabilities, product standards, and computer code review. Gates recommends an international organization layered on top of national all-of-government regulators to oversee all AI risks people imagine. All of these approaches assume overseers who would control innovation.

While it is true that whenever a technology’s costs fall and its abilities grow, people use it more. Sometimes for evil. When Henry Ford put cars within reach of every family, some families created new businesses, but others became bank robbers. When internet service providers spread access across the country, e-commerce exploded, but so did criminal activity on the dark web. In these instances, successful regulatory responses were not to limit cars or the internet, but to use the technologies for regulatory purposes.

The examples of automobiles and the internet illustrate a path forward for AI policy: Let the technology evolve for the good it can do. At the same time, officials and entrepreneurs can protect citizens by developing their own innovations based on a deep understanding of the technologies and their markets.

Recent research published in the Journal of Economic Perspectives provides insights into AI and its markets. The researchers examined the AI most people use, LLMs. LLMs are growing in complexity, now operating in three layers: The Model Layer, where creators such as OpenAI and Meta design and train LLMs; the Inference Layer, where AI providers like OpenAI and Together AI host and run models to respond to user requests; and the Application Layer, where a large ecosystem of startups and established firms embed LLM capabilities into user-facing applications for accounting, legal, retail, and other services.

Activity in each layer has exploded in multiple directions. The Model Layer grew from 1 model in early 2023 to 668 by the end of 2025. There are curious dynamics in this layer. Open-weight models—which allow users to customize systems for specific tasks—charge users 90% less than do closed weight models. Open-weight providers effectively give away their models after spending billions in development and training. Despite what looks like bad economics, there are over twice as many open-weight models as closed-weight models, 449 versus 219.

Customers in the Model Layer also make choices that appear counter intuitive. Even though open-weight providers charge 90% less than do their closed weight counterparts, customers use closed-weight models more than twice as often.

The complexity doesn’t stop there. At the Inference layer, the number of providers grew from 30 to 90 in 2025. These providers largely host open-weight models and their non-price capabilities vary. Closed-weight model creators are more likely to have vertical relationships at this layer.

The growth and interplay of these two layers illustrate why controls can be counterproductive. They limit innovators’ abilities to experiment, meaning that there would be fewer models in both levels. Fewer models at this stage of development means fewer opportunities for customers to express their preferences. And it is unclear whether the two layers will remain separate.

Model diversity is growing in several ways. Measured by the Artificial Analysis Intelligence Index, 80% of the models fell between 0.1 and 0.29 on the scale at the beginning of 2025. By the end of the year, they fell between 0.22 and 0.61, an increase in spread of over 100%.

Market leadership changes often. In the Application Layer, the market leader for science-oriented models changed eight times in 2025, while the market leader for legal services changed five times.

What does this mean for regulators? The innovators, investors, and customers driving AI are creating tremendous value, estimated to be approaching $1 trillion. Regulatory controls handicap legitimate AI providers and create market opportunities for those less inclined to follow the rules.

The lesson isn’t that government has no role in AI’s evolution. It is that seeking to control AI is counterproductive. AI policy should let the government be a leading AI user without limiting legitimate users of AI. This is more like what good governance has always done: punish harmful conduct and protect citizens by adapting its own capabilities as the world changes."