Dan Carpenter's massive new study of the Food and Drug Administration (FDA) is a definitive study of regulatory politics and administrative behavior destined to stand alongside other classic studies of administrative agencies, such as Herbert Kaufman's The Forest Ranger (1960) or Martha Derthick's Policy Making for Social Security (1979). Like Carpenter's earlier work, Reputation and Power is marked by deep erudition, thorough scholarship, painstaking attention to detail, and a wide-ranging attention to alternative disciplinary paradigms. And it is argued with great craft, subtlety, and creativity both in developing its historical narrative and in its cogent theoretical analysis.

You might be tempted to think that all these gifts are lavished too generously on a narrow topic—a study of one administrative agency among many in the alphabet soup that flourish in Washington. But you would be wrong. For Carpenter has actually put his powerful spotlight on the agency that is arguably the pivotal regulatory body in the modern liberal state: an agency whose mission it is, essentially, to regulate access to your body—to patrol access to health care or quackery—and this mission requires it to reach far into the basic structures of American social and economic life. The author shows how, in its pursuit of this mission, the FDA has shaped vast parts of the modern economy, mapped the path of medical innovation, and made citizens the world over increasingly dependent on the quality and substance of its work. Moreover, he demonstrates that the FDA has generally accomplished these things on the basis of a surprisingly modest statutory basis, while operating with little of the fanfare, or budget, of the immense New Deal and Great Society pillars of the American welfare state. But it has not been less successful in changing the political and economic landscape of the nation for all that.

People are living longer nowadays, and those of us of a certain age have long since accustomed ourselves to dependence on new pharmaceutical products, many of which are not used intermittently as acute issues arise but are embedded in our daily routines. We grew up with the “Pill” (and various other contraceptive devices), which structured almost everyone's sex lives and transformed marriage and family formation. And we have aged with estrogen treatments, drugs to lower cholesterol, and various other medications and medical devices aimed at ameliorating the effects of living longer (especially high blood pressure, heart disease, and incipient diabetes). And if (when) we happen to have some kind of acute health crisis (a stroke, a cancer, or a heart attack, for example), we learn that a steady stream of new inventions are produced to treat these issues, with varying degrees of effectiveness. These things are built into our lives and our budgets, and we not only take them for granted but also expect the stream of innovation to continue. But many of these new medications and medical devices, because they are used for long periods of time, pose new and unusual health risks—risks that are subtle and take a long time to emerge and are difficult to evaluate. And because the stakes are so high for each of us, it is not surprising that our society would find ways to collectivize these tasks. Of course, nothing so rational, from a collective viewpoint, ever happens so simply or straightforwardly.

The first half of Carpenter's book shows the tortuous path by which the modern FDA emerged. Early drug regulation was aimed mostly at combating “snake oil”—remedies that did little or no good, were marketed by ephemeral fly-by-night entrepreneurs, and could kill you. This was really a job for a police force that aimed to keep ambitious predators away from law-abiding and credulous people. But over time, especially after the passage of the 1938 Food, Drug and Cosmetics Act, the agency turned away from its early focus on policing to find ways to regulate new drugs increasingly produced by respected firms, staffed by well-trained professionals. This turned out to be enormously challenging both technically and politically, and required people in the agency to devise methods of evaluating drugs at various stages of research though production and use. Doing this required recruiting (and partly inventing) a new profession, clinical pharmacology, into the key decision-making apparatus of the agency and marginalizing other professional competences at the same time. It also involved colonizing universities and other organizations to manage the research of human subjects, and essentially transforming the internal competences of the organizations in the regulated market.

Carpenter shows that meeting these challenges regularly required the agency to move far beyond its statutory authorization at any moment in time (the 1906 act establishing the FDA, the 1938 Food, Drug and Cosmetics Act, or the 1962 Harris-Kefauver Act, for some examples) and to develop research protocols and practices that drug firms would need to implement if they wanted access to the increasingly lucrative marketplace. Indeed, he shows that statutes tended to follow, rather than lead, agency action. William Eskridge and I have labeled creative agency activity of this kind “administrative constitutionalism,” and have argued that it is the central activity of the modern administrative (constitutional) government (A Republic of Statutes [2010]). We have argued (implicitly against Friedrich von Hayek, Carl Schmitt, and other critics of the modern administrative state) that agency creativity of this kind is not inconsistent with the rule of law but is, indeed, required by the constitutional entitlements that lie at the basis of legality in a constitutional state. In the case of the FDA, rather than authorizing new regulatory actions, statutes tended to follow and ratify earlier administrative determinations that had already been created and announced in the form of speeches, regulatory rules, and agency precedents.

Carpenter also traces some of the effects of the FDA's regulatory posture, and here one can quibble a bit with his interpretation. Early in the book, he dismisses capture theory—or at least the cruder forms of it advanced by Gabriel Kolko on the left and George Stigler on the right—according to which firms actively build or recruit an agency to implement their own private purposes (the creation and maintenance of monopolies). And on this terrain I think he is right. But capture may be a much subtler process than this. If a regulatory agency's policies support an industry structure in which incumbent firms are protected from the entry of competitors and regularly earn monopoly rents on their products, that is itself evidence of a kind of “capture,” whether or not the firms brought this situation about by forming a cartel or it somehow fell into their laps.

We should not ignore the fact that the industry (as well as its major firms) finds it necessary to maintain a very active political presence in Washington, and to lobby extensively on legislative and administrative proposals. And it contributes actively to congressional and presidential campaigns, and operates a regular revolving-door relationship between the FDA and client firms, industry organizations, and political supervisors. In this sense it is active in maintaining what I would call the circumstances of capture (in the broad sense), even if it (or its component firms) did not bring these circumstances about. Capture used in this way is free of any sense of a firm creating and colonizing an agency in any active sense. It is descriptive of a static or stable evolved situation and is, in this regard, history independent. I am convinced that Carpenter is right that pharmaceutical industry structure, especially firm structure, is endogenous to the (historical) regulatory choices of the FDA, and so cannot be said to have been captured by a preexisting set of firms in the Kolko-Stigler sense. But I think that endogeneity goes both ways in complex strategic situations and, in that sense, policies, agency structures, and industrial organization are in a kind of equilibrium relationship in which incumbent firms enjoy some of the benefits of cartelization and where this permits the agency to successfully pursue its own goals.

That Carpenter has documented a kind of capture helps make sense of what is otherwise a rather puzzling aspect of modern FDA regulation. After subjecting firms to extremely exacting and costly procedures prior to investigating or marketing new molecules, the FDA has been very reluctant institutionally to pull drugs off the market on the basis of postapproval clinical experience. It does not, for example, have any really systematic way of collecting information on clinical uses, and indeed, is not able to “see,” in any regular way, how it is that doctors (who remain free to prescribe drugs for off-label uses) actually prescribe approved drugs. Moreover, such information as does dribble in is effectively marginalized and paid little systematic attention (this is partly because the agency has delegitimized clinical, as opposed to experimentally produced, information). Thus, once a firm has gotten a drug through the regulatory gauntlet to the promised land of patent-protected marketing, it has in effect real money in the bank that is not likely to disappear (short of an outright catastrophe; and Carpenter shows that the bureaucratic routines inside the agency make it hard for a catastrophe to be recognized).

If one wanted to devise a regulatory system that would protect incumbent firms (however that incumbency was initially achieved), it would be pretty hard to improve on the FDA as it actually operates. So, even if the FDA created the regulatory winners (the Pfizers and Mercks, etc.), or (more likely) created a competitive ecosystem in which certain kinds of firms (like Merck especially) would flourish, those winners operate profitably through the exploitation of monopolies created by the FDA's regulatory process.

The driving force of Carpenter's account is reputation. On his account, the FDA succeeded at various points in developing and maintaining two kinds of reputation: a reputation for expertise and a reputation for exercising discretion. Every significant actor in the FDA ecology—political leaders, journalists, the scientific community—has bought into these reputations. Thus, firms that wish to market new molecules must devote enormous effort, money, and time to develop evidence of the effectiveness and safety of a new drug, using the FDA's demanding requirements. And even after this effort, they could not be assured of a positive result, at least not immediately. Firms that could survive this gauntlet had to have very deep pockets and, hopefully, a diversified portfolio of patent-protected drugs already on the market. The agency's twin reputations had an enormous effect on the incentives and internal structures of the regulated firms. But, of course, these same requirements meant that drug development and marketing would be systematically biased toward diseases that were widespread and chronic.

Ironically, the development of these crucial reputations (for assessing the features of drugs aimed at treating chronic conditions) turned on scandals associated with acute drugs. The initial authority under the 1906 act was largely aimed at destroying the patent medicine industry, which had produced a stream of dangerous and often deadly potions. The approval of the landmark 1938 act followed a series of deaths attributable to elixir sulfanilamide, another patent medicine used against infections. And the 1962 act was triggered, at least in part, by the crisis over thalidomide (which was aimed at a chronic problem but had acute effects on fetal development). The author warns against attributing strong direct effects to these legislative acts—though each did expand the statutory authority of the agency—and I completely agree that agency actions were more critical. But still, it is hard not to notice that the expanded capacity of the agency to regulate drugs treating chronic life-cycle conditions and diseases occurred politically because of a crisis associated with an acute effect of a marketed drug.

Carpenter shows that the effect of the regulatory procedures for investigational and new drugs has enormous consequences for the structure of the pharmaceutical industry—leading to a situation in which a few large firms that are really specialized in regulatory approval dominate the industry. These large firms operate by buying smaller firms or their patents, and pushing the potentially profitable drugs through the regulatory process, thereby maintaining a portfolio of patent-protected (i.e., profitable) drugs. Of course, the drugs that merit purchase and marketing by the big companies capable of managing the regulatory process must have big and lucrative markets (birth control, cholesterol, and blood pressure management drugs are prototypes). Thus, rare diseases or conditions and those that affect small minorities do not attract industry effort. And it seems to me that this also leads to underinvestment in acute conditions that can be resolved by a short treatment. The result is that the drug industry focuses on protecting chronic conditions of those of us who are in some sense typical. Another consequence is that much of the fundamental research leading to these drugs is not done by the big firms; it is done in government-sponsored labs or by small start-ups that lack the resources or expertise to get them to market.

Indeed, the recent crises faced by the FDA appear to have been consequences of its success in creating a reputation-enforced regulatory regime and associated industrial structure. The HIV/AIDS crisis was one of the first manifestations of the vulnerability of the FDA regime. HIV/AIDS drugs probably did not initially look like winners for Big Pharma, whose members may have been slow to investigate and develop new drugs to treat the epidemic's effects. The slow response, however, incubated a vigorous mobilizational effort in the AIDS community to put pressure on the agency and its client firms. And, partly as a result of the pressure put on the FDA and its congressional patrons, the agency managed to respond flexibly to this pressure by allowing rapid market entry to new drugs in the absence of effective treatments on the market. But as new diagnostic abilities locate new and potentially treatable diseases that affect minorities within the population, this kind of situation seems bound to recur. However, there is no guarantee that every affected group will be as successful in political mobilization. Maybe the FDA system is the best regulatory regime possible in a market economy. In many ways, it has had impressive successes in developing treatment regimes and fostering the development of many important medications. But it seems nevertheless chronically prone to political crises when compelled to address minority populations with unusual and unprofitable health issues.

Eskridge and I have argued that the real American constitution—which we call the Republic of Statutes—is generally vulnerable to normative criticisms of this kind. And under some conditions (as in the case of the HIV/AIDS crisis), these critiques can produce political backlash. New treatments for cancers for which existing medicine offers little hope can also provide fertile ground for normative or political reactions. The Republic of Statutes is pretty good at delivering to most of us the kinds of statutory rights that make modern life tolerable. But it is not so good at protecting minorities. The formal (Big-C) Constitution, as imperfect as it is, is needed for this. I do not believe that the FDA itself can be relied on to systematically address such problems of inequality. But if the FDA were to be recast through the enactment of a statutory or constitutional right to health care, and if that right were to be regulated by some notion of equal protection, these issues could be addressed more systematically—and not merely through the regulatory instruments controlled by the FDA but also by the use of other governmental instruments.

These last speculations are not intended to take anything away from Carpenter's achievement in Reputation and Power. It is a masterful study in the best tradition of political science and will stand as a definitive treatment of regulation, and not merely of the FDA's policies and practices. Along with his earlier work, this book will be an essential part of the emerging study of the American administrative state, whether that study takes place in political science, history, sociology, law, or, indeed, in schools of medicine and pharmacology.