The Toll Booth Society: Liberty Dies of Rents, Not Revolutions

We picture the death of liberty as a scene: tanks in the square, a decree read on the radio, a crowd that goes home and stays there. It makes for good cinema and bad history. In the free countries I care about, France and the United States, nobody is going to abolish the freedom to work, to build or to sell. They will simply charge for it. A little here, a little there, at a gate with a uniformed attendant and an excellent reason.

I call this the toll booth society. Not a tyranny, nothing so honest. A road that still runs from wherever you start to wherever you want to go, except that every few miles a striped bar comes down and someone asks for a license, a permit, a quota, a stamp, a fee, a diploma that has nothing to do with the job, or the blessing of the people already doing it. Each booth, taken alone, is defensible. Taken together, they decide who gets to travel.

Start with the most innocent-looking booth of all, the occupational license. In the early 1950s, fewer than 5 percent of American workers needed a license from a state government to do their jobs legally. By 2008, according to the labor economist Morris Kleiner, writing for the Hamilton Project at Brookings in 2015, the share was estimated at almost 29 percent. A Harris survey of roughly ten thousand workers in 2013 put it at 28.4 percent licensed or certified. In two generations, the country that likes to call itself the land of opportunity quietly turned more than a quarter of its labor force into a guild.

The case for each license writes itself: the public cannot judge a dentist's skill, a mortgage broker's honesty or an electrician's wiring before the house burns down. I take that argument seriously. Kleiner took it seriously too, and then he looked at what licenses actually deliver. Tougher licensing requirements for dentists did not improve dental outcomes; they raised prices, most likely because there were fewer dentists. Stricter licensing of mortgage brokers had no effect on foreclosures and made mortgages more expensive. My favorite study in his paper is an old one about television repair. Researchers brought sets with known defects to repairers in three places: Washington, D.C., which did not license TV repair at all; New Orleans, where Louisiana licensed the individual repairman; and San Francisco, which licensed the shop but not the worker. Parts fraud turned up in about half of the repair attempts in New Orleans and in Washington alike. The license on the man changed nothing. The difference showed up in San Francisco, at about 20 percent, where it was the facility that answered for the work.

That is the whole problem in miniature. A license sounds like quality and often works like a fence. Kleiner's summary is blunter than mine: licensing transfers income from consumers, in the form of higher prices, to licensed workers, in the form of higher wages, which it can raise by as much as 15 percent. Using standard models, he and his co-authors estimated that occupational licensing could mean up to 2.85 million fewer jobs and $203 billion a year in costs to consumers. I want to be honest about that number, because it is the kind people like to wave around. It is a ceiling, not a measurement: it assumes the entire wage premium comes from market power rather than from better skills. The truth is lower. But even a fraction of $203 billion is a large toll for the privilege of having your hair cut by someone who passed an exam in hair.

The French booths have better architecture

Now cross the Atlantic, where the booths are older and built of stone. In March 2013, France's own Inspectorate General of Finance, hardly a nest of anarchists, examined 37 regulated professions and activities, from notaries and court bailiffs to pharmacists, veterinarians, architects and plumbers. In 2010 they had combined revenue of €235.8 billion, employed 1.1 million salaried workers and generated value added equal to 6.4 percent of GDP. Their net profit before tax averaged 19.2 percent of revenue, which the inspectors computed as 2.4 times the profitability of the rest of the economy. For five of these professions, the median practitioner earned more than €10,000 net a month; for three, more than €20,000. Between 2000 and 2010, their net profit grew by 46 percent while the country's tax take grew by 29 percent.

The report did not say these people were crooks. It said something more uncomfortable: that their incomes "do not always" correspond to the length of training, the size of the investment or any real risk of failure, and that in some cases the regulation in force was the only explanation. It also noted that a 20 percent cut in some regulated tariffs would still leave practitioners a reasonable margin. When a ministry of finance writes that sentence about its own country, it is not making a political point. It is reading a ledger.

Here is where honesty forces me to slow down, because the same report contains the best objection to my argument. The inspectors found that for doctors, lawyers, chartered accountants and auditors, the most numerous of the professions they studied, there was "real competitive intensity" despite the rules: free establishment, no lasting quotas, mostly free fees. Regulation, they wrote, is economically justified when it answers precisely and narrowly to real problems, like the public's lack of expertise or conflicts of interest. So the lesson is not that rules are bad. A rule that protects the customer is a guardrail. A rule that protects the incumbent from the customer is a toll booth. The difference is not in the text of the law. It is in who would scream if the rule disappeared, and why.

The national myths do not survive the data

Every country tells itself a story about its booths. The French story, told by Americans and by a good many French liberals, is that France is a museum of protected trades and America a frontier. The American story, told by Americans to themselves, is that red tape is a European disease. The OECD has spent years building indicators of product market regulation, scoring the laws on the books from zero to six. Its latest wave, based on rules in force on January 1, 2023, for France and January 1, 2024, for the United States, says something neither myth predicts. Economy-wide, France performs better than the average OECD country. The United States performs worse.

That does not make France a free-market paradise. The same country note flags high state ownership, high barriers in retail and in the retail sale of medicines, and professions regulated more strictly than the OECD average. The American note, for its part, finds a licensing and permit regime more burdensome than the OECD average, high tariff barriers, and architects, civil engineers and real estate agents more strictly regulated than elsewhere. A detail I find instructive: where rules are set by the states rather than Washington, the OECD scored New York. The booths are not where the slogans put them. They are wherever an organized group has had time to build one.

Explaining the rain

The economist who understood this best was Mancur Olson. In The Rise and Decline of Nations, published in 1982, he argued that stable societies accumulate what he called distributional coalitions: small, well-organized groups that lobby for privileges whose gains to them are smaller than the losses they impose on everyone else. None of them stops growth. Each one shaves a little off it. Over decades, the economy does not collapse; it silts up. Olson had a lovely image for what conventional growth accounting misses. Economists, he said, trace the water in the river back to the streams and lakes, but they do not explain the rain.

Olson's rain is the slow accumulation of booths. He used it to explain why Germany and Japan, whose interest groups had been smashed by war and occupation, grew so fast afterward, and why free trade matters even more than textbooks say: foreign competition is the one thing a domestic cartel cannot lobby out of existence. He even had a word for France, whose labor movement was split among communist, socialist and Catholic unions, so that no single one held a monopoly over a workplace. Division, in that case, protected the public. I do not recommend civil war or occupation as a growth strategy. But it is worth noticing that the only reliable cure Olson found for accumulated rents was a shock large enough to knock the booths down.

The most expensive booth of all, I suspect, is the one that stops people from building homes where the jobs are. Chang-Tai Hsieh and Enrico Moretti estimated, in a paper published in 2019, that housing restrictions in high-productivity cities like New York and the San Francisco Bay Area lowered aggregate American growth by 36 percent between 1964 and 2009, because they kept workers away from the places where they would have produced the most. It is a model, with the assumptions every model carries, and I hold it at arm's length. But look at its shape. No one passed a law against prosperity. Thousands of local boards, each protecting a view, a parking ratio or a property value, collectively told millions of people that the most productive cities in the world were full.

That is how liberty actually dies in a democracy. Not by abolition but by accretion, and always with the consent of the people who are already inside. The neighbor who opposes the apartment building is not a villain. The notary who defends the tariff is not a thief. The licensing board that adds a year of training is sincerely worried about quality. That is what makes the toll booth society so durable: every booth is staffed by someone who believes in it, and the people who pay are scattered, unorganized and, worst of all, mostly invisible, because the cost of a toll is paid by those who turned back before reaching the gate.

Who builds and who collects

I am sometimes asked where I stand, left or right. The question assumes the line runs where it ran in 1950. The line I care about runs elsewhere. On one side are the people who build: the founder who wants to open a pharmacy-style shop for over-the-counter medicines, the immigrant with a trade and no local diploma, the young contractor who would put up housing if the permit took weeks instead of years, the startup that wants to serve customers a bank would rather keep. On the other side are those who collect, and they are found in every party, every union, every chamber of commerce and every incumbent firm with a good lobbyist. Rents are not a left-wing or a right-wing vice. They are a human one, and they flourish under any government that confuses protecting a profession with protecting the public.

Some of the fights I have written about on this site are, at bottom, fights over a booth. When JPMorgan decided which AI agents could touch its customers' accounts, the real question was who holds the key to the gate. When the CFTC drew a line between casinos and prediction markets, it was deciding which regulator gets to run the booth. When 148 French deputies voted to keep a €1.4 billion tax break for pensioners, or when the Assembly rewrote a tax on business owners at midnight by 25 votes to 23, you could watch distributional coalitions do exactly what Olson said they would do, in real time and with the best of intentions.

What would a free country do instead? Not burn the booths down. Some protect us, and a liberal who forgets that the customer needs protection from fraud is not a liberal but a lobbyist for the other side. It would put every booth on trial, regularly, with the burden of proof on the gatekeeper. Kleiner proposed cost-benefit analysis before any new licensing requirement, licenses that travel across state lines, and lighter forms of certification where a full license protects no one. The French inspectors proposed letting the competition authority push regulated tariffs toward real costs. Poland, in 2011, set out to liberalize access to about 250 of its 380 regulated professions. None of this is revolutionary. That is the point. A liberty that dies by accretion can only be defended the same way: gate by gate, fee by fee, with patience and a ledger.

The same logic runs through the two other questions this site keeps asking. A nation's independence is also a matter of who holds the switches it depends on, which I take up in Sovereignty Is a Supply Chain. And the energy and the honest measurement that make it possible to build anything at all are the subject of Progress Is Measured in Kilowatt-Hours and Error Bars.

A society that stops building rarely notices the moment it happens. There is no square, no tank, no decree. There is only a road where the bars come down a little more often each year, and a traveler with a cart full of tools who looks at the line of booths ahead, does the arithmetic, and turns around. Liberty is the right to keep walking. The least we owe that traveler is to count the booths.