The License

The bottlenecks of the earlier plates were given. Land's scarcity was fixed by nature; the great talent's by the accident of birth; even the machine's, while it lasted, by the plain fact that building another took time. None of them was anyone's decision. The bottleneck of this plate is different in its origin, and therefore in its character, for it is made on purpose, by law. A licence, a permit, a medallion, a charter, a quota: each is a legal instrument whose function is to limit the number of people allowed to do a thing — to drive a cab in this city, to sell liquor on this street, to practise this trade, to broadcast on this frequency, to bring in this many tons of that good. Before the law, the number was whatever the market made it; after the law, it is whatever the law decrees. The scarcity is manufactured, and like every scarcity in this volume it throws off a rent. The only questions are how large the rent is, who collects it, and whether the public the licence is always said to protect is in truth its beneficiary or its prey.

The mechanism is the plainest in the volume, because it is scarcity in its most deliberate form. Suppose a trade that anyone may enter, in which competition has driven the price down to the cost of doing the work — the ordinary result of free entry, with no rent left over for anyone. Now let the law cap the number of providers below what the market would supply. The cap makes the service scarce; the scarce service commands a higher price; and the difference between the new, higher price and the old cost is a rent, collected by whoever holds one of the limited permits and paid by every customer who must now buy at the inflated price.

The rent is a private tax — a charge levied on the public for the benefit of the permit-holders — with the peculiarity that it is collected not by the state that imposed it but by the lucky parties the state has licensed. And like every restriction of quantity below the efficient level, it destroys as well as transfers: the customers who would gladly have bought at the old price but will not pay the new one go without, and the mutually beneficial trades they would have made simply do not happen. That loss enriches no one.¹ There is nothing subtle in any of this. What makes the licence interesting is not the mechanism but the disguise — that this simple engine of extraction is, almost without exception, justified, and often sincerely believed, to exist for the protection of the very people it taxes.

Once a permit confers a rent, the permit itself becomes a valuable thing — as valuable, precisely, as the stream of rents it carries. A licence that lets its holder earn, year upon year, some sum above what the unrestricted trade would yield is worth, to anyone who would buy it, the whole of that sum gathered up and brought to the present: the rent capitalised into a price.² This is the same capitalisation by which the rent of land becomes the price of land, met three plates ago; here it has produced one of the strangest assets a modern economy has known. The clearest case is the New York taxi medallion. The city created the medallion in 1937 and fixed the number of them, and for the better part of a century that number scarcely moved while the city around it grew without cease.

By 2014 a medallion sold for more than a million dollars — more than the price of a house, for a tin badge whose only power was the city's permission to do what anyone with a car could otherwise have done.³ The medallion became, for decades, the best investment in America, and was treated as the safest: drivers, many of them immigrants, borrowed heavily to buy one, regarding it as a retirement saved in metal, an asset that only ever rose. They were buying a rent — paying its full capitalised value up front, in the expectation that the scarcity which produced it would hold forever. It did not.

Set against this is the honest case for the licence, which is real and must be stated at full strength. Some trades genuinely require it. We license doctors and pilots and electricians because incompetence in such work is not merely disappointing but lethal, and because the customer cannot judge the competence for herself until it is too late; the licence certifies a minimum the buyer could not verify, and bars from practice those who fall below it. Some scarcities, too, are genuine and not manufactured: the broadcast spectrum is a real and finite thing, only so many frequencies fitting in the air without ruinous interference, and a licence that parcels it out is managing a true commons, not inventing a false one.⁴ In these cases the restriction earns its keep, and the only quarrel is over design — how high to set the bar, and who should capture the rent that a real scarcity yields.

But the honest case has been stretched, over the last seventy years, to cover a vast territory it does not fit. In the early 1950s about one American worker in twenty needed a licence to do their job; today the figure is between a fifth and a quarter, and the number of occupations licensed in at least one state has risen from around thirty to some eleven hundred.⁵ Much of this growth is in trades where no one's life is at stake — the florist, the hair-braider, the interior decorator, the auctioneer — and where the licence's effect, study after study finds, is to raise the earnings of those inside the trade by something on the order of a tenth or more, to raise prices for consumers, to thin the number of jobs, and to shut out the mobile, the poor, and the immigrant, all with little measurable gain in quality or safety. The pattern is so regular that it invites a generalisation George Stigler made into a theory in 1971: that regulation is frequently acquired by the industry it regulates and operated for that industry's benefit — the regulated capturing the regulator, and the licence that claims to guard the public serving in fact to cartelise the trade.⁶ Milton Friedman had drawn the decisive distinction a decade earlier, between certification, which tells the buyer who has met a standard and leaves her free to choose otherwise, and licensure, which forbids the unlicensed to practise at all. Certification supplies the information the honest case demands; licensure adds the restriction the cartel desires. A great deal of what passes for consumer protection, Friedman argued, is licensure where certification would have served — protection of the producer in the mask of protection of the public.

Why, then, do such licences survive, decade upon decade, once their character is understood? The answer is the cruelest property of the capitalised rent, and Gordon Tullock named it in 1975: the transitional gains trap.⁷ The moment a rent is created and capitalised into the price of a permit, the people who first received the permits — who got their rent as a windfall, for nothing — sell out and depart with their gains. Those who come after buy the permit at its full inflated price, and so earn no windfall at all, only the ordinary return on a costly asset; the rent is now embedded in what they paid for it. And this means that to abolish the restriction — to throw the trade open and dissolve the rent — would not be to strip a privileged class of an unearned gain. It would be to bankrupt the current holders, who paid good money in good faith for an asset the state had created and seemed to promise to protect. The original beneficiaries are beyond reach; the present holders are the sympathetic victims of any reform; and so the inefficient restriction, enriching no one any longer, defends itself from behind the people trapped inside it. The act of capitalisation converts a transfer into a fortress.

The New York medallion shows the trap, and then shows the grimmer thing that happens when the fortress finally falls. For decades the scarcity held and the price rose. Then, after 2011, the smartphone did what no reformer had managed: ride-hailing services put tens of thousands of cars on the streets with no medallion at all, and the protected scarcity simply evaporated. The rent vanished, and with it the capitalised value of the medallion, which fell from over a million dollars to a few hundred thousand and, in many sales, far less. The drivers who had bought at the top — who had borrowed against a lifetime to purchase a rent the world then abolished — were left owing far more than the medallion was worth, ruined through no fault of their own. A later investigation found the bubble had been swollen by years of reckless lending and by a city that quietly inflated the prices at which it sold; the human cost was severe, including a series of suicides among drivers crushed by the debt.⁸ It is the whole argument of the plate written in a single tragedy: a rent manufactured by law, capitalised into an asset, sold to latecomers at its height, and annihilated when the scarcity that sustained it broke — leaving the public a little richer in cabs, and a community of drivers destroyed.

If the disease is well understood, so are the cures, and they follow from the diagnosis. Where the honest aim is to protect the buyer from incompetence, certification will do it without the rent: let a credible body attest who has met the standard, publish the fact, and leave the buyer free to hire the uncertified if she wishes and bear the risk. The information that genuine protection requires is supplied; the restriction that the cartel requires is withheld.⁹ Where the scarcity is real — spectrum, the landing slot, the fishery, any true and finite commons — the remedy is not to give the permits away, manufacturing private fortunes out of a public resource, but to auction them, so that the rent the scarcity yields accrues to the public that owns the resource rather than to whatever private parties the state happens to favour. This is exactly what the United States began to do with the airwaves in the 1990s, selling spectrum licences at auction and raising tens of billions for the treasury — the very principle, applied to a real scarcity, that the volume's ninth plate will urge upon the rent of land itself.¹⁰

And where the licence is simply a barrier that protects a trade at the public's expense, the remedy is to remove it — to delicense, or at the least to subject every licence to periodic review and a sunset, so that each must justify its restriction afresh or lapse. The remarkable thing about occupational licensing today is that hostility to it spans the political spectrum, uniting free-market economists, anti-poverty advocates, and antitrust enforcers in a rare agreement that much of it serves no one but the licensed. That so little has been dismantled even so is the transitional gains trap at work — a measure of how stoutly a capitalised rent defends itself, long after everyone has seen it for what it is.

The licence is the volume's purest case of a manufactured scarcity, and it sharpens the distinction the whole book is drawing to its finest point. A rent can arise from a scarcity that nature or genuine necessity imposes — the land, the gift, the finite spectrum — and such a rent, whatever one concludes about who should pocket it, is at least unavoidable, the price of a real constraint. Or a rent can arise from a scarcity that a statute invents for no purpose but to enrich those it favours, and such a rent is pure extraction, a tax the public pays so that a protected few may charge what free competition would deny them. The two wear the same clothes, for every licence is justified as the first kind even when it operates as the second; and so the test can never be the justification, which is always public-spirited, but only the effect. Does the restriction demonstrably buy the public something — safety, competence, the husbandry of a real commons — worth the rent it costs? Or does it buy the public nothing, and the permit-holders everything?¹¹

There is a final reason the manufactured rent deserves its own plate, and it is that a rent created by law is a prize, and a prize worth winning is worth fighting for. If a stroke of legislation can hand a trade a permanent stream of rents, then the trade has every reason to spend money — on lobbying, on campaigns, on the quiet capture of the boards that write the rules — to obtain that stroke and to defend it. The expense of that fighting is the subject of the next plate, where the scramble to win and keep a rent becomes a category of waste in its own right; and the question of whether any of this restriction can be defended as efficient — as a signal or a guarantee the market could not supply for itself — is the subject of the plate after that.¹² For now it is enough to have seen the simplest of the made bottlenecks plainly: a line drawn in a statute book, a scarcity conjured out of nothing, a rent collected ever after by those on the right side of the line — and a public assured, every step of the way, that the line was drawn for its own good. Whether it was is a question to be answered licence by licence, and answered by what each one does, not by what it says.¹³