I just finished reading Daniel Carpenter’s book, Reputation and Power: Organizational Image and Pharmaceutical Regulation at the FDA. I won’t say that it was fascinating all the way through, but for a 700 page book about the history of the Food and Drug Administration in the United States, it was pretty good. I picked it up because, over the space of about 3 months last fall, two people recommended it to me. I thought to myself, what are the chances that two people would independently come up to me and say “you must to read this 700 page book about the FDA” unless it was a really amazing book?
The reason they were recommending it to me was that I’ve been interested in administrative discretion and the way that it is dealt with by public servants (see here). This is part of a more general interest that I’ve developed in the executive branch of government, along with the view that the executive is seriously undertheorized in normative political philosophy.
Carpenter’s book, however, is more interesting as part of the project that it belongs to more properly, which is the recent attempt by a group of academics in the United States (many at The Tobin Project) to think more seriously about regulation – and to break the stale, and highly stylized, confrontation between so-called “public interest” and “rent seeking” or “regulatory capture” theories, which has dominated debates over regulation for the past 50 years.
Here’s roughly how the stale debate works. According to the “public interest” theory, regulation arises in response to market failure. Marketplace competition takes place in accordance with a certain set of rules – the most minimal set being those defined by property rights and contract law. In many cases, however, these rules will be insufficient to promote healthy forms of competition – those which generate benefits for all. Incompleteness in the system of property rights may allow firms to produce negative externalities, such as pollution. Or information asymmetries may lead consumers to purchase something that, were they better informed, they would not purchase. And so a wise and benevolent government will intervene, by adjusting the rules or imposing new ones, in order to prevent market participants from employing these strategies. The intent is not all that different from what sport bodies like the NHL do, when they adjust the rules in order to improve the game (my favorite instance of which is the introduction of the “Sean Avery rule” to hockey in 2008).
This is, however, more of a normative model of how regulation should work than an empirical account – it’s a story about why it would be in the public interest for the state to introduce certain sorts of regulations. As such, it raises a number of important empirical and institutional questions, about how the less-than-fully-wise, poorly informed, or not-entirely-benevolent politicians or bureaucrats who staff the government ever succeed in recognizing the public interest, and what incentive they have to promote it through legislation. There are obviously many reasons to think that real-world regulation is going to be significantly less optimal than what the public interest theory would recommend.
So there is lots of room for doubt about the “public interest” view, and about to degree to which considerations of the public interest actually inform regulatory decision-making (indeed, it is difficult to find any straightforward proponents of the “public interest” view as an institutional theory – this is probably the closest). But rather than pursuing these doubts, developing an empirically informed skepticism about real-world regulation, what critics of the regulatory state have instead gravitated toward is a ridiculously negative caricature, the so-called “regulatory capture” theory. According to this view, regulation is not about the public interest at all, on the contrary, regulation is everywhere a conspiracy against the public. Regulation exists because organized interest groups try to use the coercive power of the state to constrain competition or to achieve non-productive transfers. They succeed because smaller, more concentrated interests have a greater capacity for collective action than more diffuse, less concentrated interests.
According to this “rent-seeking” perspective, something like the dairy supply management system in Canada is the model of all regulatory intervention. The supply management system restricts entry to the dairy business as well as limiting production, thereby raising prices and creating rents for farmers who own a quota. These rents represent a non-productive transfer from consumers, who pay higher prices for dairy products than they would under conditions of open marketplace competition. Farmers are able to get away with this because they are organized, whereas consumers are not, and so any politician who threatens to dismantle the system is threatened with highly organized and persistent protest. Consumers, on the other hand, are simply not agitated enough by paying a dollar “too much” for a litre of milk to do anything about it. (Indeed, the only organized group that I have heard of that really lobbies against the dairy cartel in Canada are the large pizza chains, because they purchase such gigantic quantities of cheese – confirming the general point that it takes a concentrated interest to act effectively against another concentrated interest).
As one can see in the politics of dairy supply management, there is obviously something to the “rent seeking” theory. But to generalize this and say that all regulation, including environmental and consumer protection law, follow the same general pattern, is a fairly wild claim. Furthermore, even though the “regulatory capture” story has been used by right-wing politicians for over 40 years as grounds for deregulation, it’s not clear how many of them actually believe it. After all, if people on the right were so concerned about regulatory capture, you’d think they might be extra-cautious about who they appoint to regulatory agencies, to make sure that their appointees have as much distance as possible from the regulated interests. But right-wing parties tend to do the exact opposite, stuffing agencies full of people with extremely close ties to industry – and thereby, it would seem, positively encouraging capture.
Part of the reason for this, I suppose, is that the “capture” theory doesn’t really have a theory about how to improve regulation. Since regulation is, at its heart, just an exercise in rent-seeking, the only solution is deregulation. “Better regulation” is, from this perspective, something of a contradiction in terms. This encourages a certain unhealthy cynicism towards the regulatory agencies.
Now suppose one were to take a step back from this old debate and adopt the sensible view, viz. that there is a pressing need for regulation, in order to correct various forms of market failure, but that it may be difficult to align the machinery of government in the right way to produce such regulation, and there is always the danger that this machinery will be captured by various special interest groups, thereby subverting the intent of regulatory intervention. One might then be tempted to ask a series of fairly straightforward practical questions, such as: How does one go about trying to ensure that regulation does serve the public interest? How does capture occur, and what conditions make it more or less likely? How can one tell when an agency has been captured, and how can the situation be repaired? What forces make regulatory agencies more or less effective in advancing a public interest mandate? And so on.
These are the questions that the Tobin Project group have been taking on (e.g. here). Astonishingly, there has been practically no serious empirical research done on these questions. (Proponents of the capture theory relied almost entirely upon synthetic a priori reasoning rather than empirical investigation – they just “deduced” the reality of capture from a set of quasi-economic models.)
Carpenter’s book on the FDA fits into this broad current of thought, and is in this respect very important. I didn’t find his explicit theoretical concepts, viz. “power” and “reputation” to be particularly illuminating. His implicit framework, however, is of considerable interest. His core conviction seems to be that organizational culture is extremely powerful in determining the way that an agency like the FDA confronts and processes new situations, i.e. whether it will be aggressive, obstructionist, confrontational, conciliatory, supine, etc., in dealing with industry.
For example, I learned a great deal from Carpenter’s discussion of the thalidomide catastrophe. I was aware of the broad outlines of the story, but I hadn’t realized that, thanks to the FDA, there was not a single case of thalidomide-induced birth defects in the United States. On the other hand, the scientist who blocked the approval of thalidomide in the U.S. market (Frances Kelsey) got a little bit lucky, in that she had no specific reason to suspect that the drug caused birth defects. She dragged her feet on the approval mainly because she distrusted the company that was putting it forward, and had concerns about a set of much more minor issues. She delayed it for long enough, however, that the birth defects began to show up in Europe, where the drug was in widespread use. This made her into a national hero, and gave the FDA practically unassailable prestige and authority for the next two decades at least.
There is also an interesting discussion of the public pressure that the FDA withstood in the ‘70s, in refusing to approve the sale of Laetrile/amygdalin (a quack cancer therapy derived from apricot pits). The story, basically, is that the FDA stood firm against a huge public campaign (including the development of a large-scale smuggling network, as well as celebrity pressure from none other a dying Steve McQueen), but was ultimately vindicated when a large-scale study proved the drug useless. This in turn disposed the organization toward skepticism when the first platinum-based chemotherapy agents were submitted for approval, and bolstered its tendency toward resistance in the face of public pressure for rapid approval (which turned out to be the wrong thing to do). So again, the key to understanding how the organization behaved is to look at the organization’s own history.
Overall, I think no one who reads this book could possibly take the capture theory seriously as a general account of how regulation works. What comes through quite clearly is that the FDA has always had a strong, independent conception of where the public interest lies, as well as substantial discretion in terms of how it interprets its mandate (for my purposes, the discussion of how the agency justified its commitment to studying the efficacy of drugs, prior having received the explicit mandate to do so in 1962, is particularly interesting). Carpenter doesn’t really take sides on whether the FDA does its job well or poorly. My suspicion is that he underplays a bit how frustrating the FDA’s approval process can be. Nevertheless, the book provides an extremely clear picture of the forces that determine how the FDA goes about its work – whether it moves quickly or slowly, whether it is legalistic or flexible, whether it is proactive or reactive, etc. In this respect, Carpenter’s book is an instance of serious scholarship on the subject of regulation, something that has been sorely lacking.