Until now, that is. In their short book, First Thing We Do, Let’s Deregulate All the Lawyers, Brookings Institution economists Clifford Winston and Robert W. Crandall and University of Houston economist Vikram Maheshri make a systematic attempt—largely successful in my view—to assess two conflicting claims. The first is that the current government restrictions on legal services serve to assure quality. The second is that the restrictions are mainly an attempt by existing lawyers to prevent competition and maintain lawyers’ incomes. The authors give evidence against the first claim and in favor of the second.

Limiting competition | As I noted, the authors build a strong empirical case and don’t depend on anecdotes. But even anecdotes can be persuasive. As the late economist George Stigler supposedly said, “The plural of ‘anecdote’ is ‘data’.” And the authors lead with a particularly strong anecdote originally told in more detail by deregulation advocate and regular Regulation contributor George Leef. It is about Rosemary Furman, a legal secretary in Florida who made a living filling out divorce papers for her lawyer boss until she realized that she could do the same on her own. And so she went into business for herself, cutting the price of filing for divorce substantially; yet she still made more money per client than she had previously. The Florida Supreme Court ordered her jailed for practicing law without a license. Although Florida’s governor intervened to prevent her from going to jail, she never provided the service again. The government’s point was made. Interestingly, note the authors, none of her clients ever complained about her work.

The authors carefully build their case, first telling of the various restrictions on who can be a lawyer. All but a few state governments, they note, require prospective lawyers to have graduated from a law school that the American Bar Association has accredited. One notable exception is California, where one can become a lawyer simply by passing the bar exam and a competency exam. Every state government but Wisconsin’s requires all would-be lawyers to pass a bar exam. The Wisconsin government makes exception only for graduates of the University of Wisconsin Law School! This would make sense, from the viewpoint of quality assurance, only if the University of Wisconsin Law School graduates are, on average, better than those of any other law school, including Stanford, Harvard, Chicago, and Yale.

Quality control? | The authors’ argument proceeds in three steps. First, they argue that regulation of entry is not about assuring quality. Second, they show that, as would be expected when competition is limited, lawyers earn a substantial premium, currently on the order of 50 percent. Third, they show that lawyers have an incentive to lobby for legislation and regulations that increase the demand for their services.

Why do the authors think regulation is not about assuring quality? One reason is that the American Bar Association, which accredits law schools—essentially a fox in charge of the hen house—has not even considered accrediting foreign law schools or online law schools. If the ABA’s true motive were to assure quality, it would seriously consider accrediting such schools. Another reason, they write, is that “the ABA has refused to provide further information about a law school’s quality beyond its accreditation status and has continually issued disclaimers of any law school rating system.”

The authors devote a big part of the book to measuring the income premium that lawyers make because of the restriction on the number of lawyers. Their bottom line: “[W]e find that by 2004 lawyers’ earnings premiums amounted to $64 billion—or an eye-popping $71,000 per practicing lawyer—and that those premiums were widely shared among the legal profession.” They find that the premiums were substantial for the whole period they studied, 1975–2004, but rose a lot in the 1980s and 1990s.

Could this increased premium over that time period reflect an increase in the quality of lawyers rather than a restriction of competition per se? Their answer is no. They point out that the grade point average of people admitted to at least one law school rose only slightly, from 3.25 in the late 1970s to 3.34 in 2004. Moreover, these data don’t account for grade inflation, which would make the underlying increase in quality even less than the 0.09-point increase would suggest.

The third part of their argument is that the restriction on the number of new lawyers gives existing lawyers an incentive to lobby for regulations that increase the demand for their services. Winston, Crandall, and Maheshri point to expansions in liability for unsafe products as one such lobbying activity. They also note that lawyers from more than 20 law firms “met extensively with commissioners from the federal Commodity Futures Trading Corporation to shape the implementation of new financial regulations under the Dodd-Frank Wall Street Reform and Consumer Protection Act.”

Liberalizing legal services | The authors’ solution is to deregulate the legal profession by allowing people to provide various services without a law license and by allowing different types of organizations, not just law firms, to provide services. They point out two big advantages of deregulation. The obvious one is that with more supply and more competition, prices of legal services would fall. What about the fear that consumers would not have quality assurance without the ABA as a gatekeeper? The authors maintain that consumers can get information about lawyers’ quality in many ways, especially in the Internet age. They should have also noted that anyone who wants to hire only a lawyer who has passed a bar exam and graduated from an ABA-accredited school would still be free to do so.

The authors also point to a subtle benefit of deregulation: it would break down solidarity in the legal lobby, thus undercutting the push for more government regulation. They don’t make this argument totally clear, but it seems to be an application of the late Mancur Olson’s theory of collective action. The big problem with collective action, noted Olson, is the free-rider problem: those who don’t pay for the benefits of lobbying still get the benefits. The free-rider problem, therefore, leads to less lobbying than otherwise. The American Bar Association and the American Association of Justice (formerly the American Trial Lawyers Association), which both lobby for regulations and legislation that benefit lawyers, would have a bigger free-rider problem if there were more lawyers and fewer restrictions on who could become a lawyer. I find this argument persuasive, but it would have been helpful if the authors had elaborated on it somewhat.

One caution: although most of the book is well-written, there are a few key parts in which the authors write as if their main audience is economists who are sophisticated in econometrics. It’s clear that they dig into the econometrics to drive home their high degree of confidence that their empirical findings are robust. But you might want to skip over those parts.

Of course, the book’s title is a play on the famous quote from Shakespeare’s Henry VI, Part 2: “The first thing we do, let’s kill all the lawyers.” Winston, Crandall, and Maheshri offer a much less violent, and much better, alternative.