Barking Cats

From Barking Cats by Milton Friedman in his column in Newsweek, 19 February 1973.

In a recent column (Newsweek, Jan. 8), I pointed out that approval of drugs by the Food and Drug Administration delays and prevents the introduction of useful as well as harmful drugs. After giving reasons why the adverse effects could be expected to be far more serious than the beneficial effects, I summarized a fascinating study by Prof. Sam Peltzman of UCLA of experience before and after 1962, when standards were stiffened. His study decisively confirmed the expectation that the bad effects would much outweigh the good.

The column evoked letters from a number of persons in pharmaceutical work offering tales of woe to confirm my allegation that the FDA was indeed “Frustrating Drug Advancement,” as I titled the column. But most also said something like, “In contrast to your opinion, I do not believe that the FDA should be abolished, but I do believe that its power should be” changed in such and such a way—to quote from a typical letter.

I replied as follows: “What would you think of someone who said, ‘I would like to have a cat, provided it barked'? Yet your statement that you favor an FDA provided it behaves as you believe desirable is precisely equivalent. The biological laws that specify the characteristics of cats are no more rigid than the political laws that specify the behavior of governmental agencies once they are established. The way the FDA now behaves, and the adverse consequences, are not an accident, not a result of some easily corrected human mistake, but a consequence of its constitution in precisely the same way that a meow is related to the constitution of a cat. As a natural scientist, you recognize that you cannot assign characteristics at will to chemical and biological entities, cannot demand that cats bark or water burn. Why do you suppose that the situation is different in the social sciences?”

The error of supposing that the behavior of social organisms can be shaped at will is widespread. It is the fundamental error of most so-called reformers. It explains why they so often believe that the fault lies in the man, not the “system,” that the way to solve problems is to “throw the rascals out” and put well-meaning people in charge. It explains why their reforms, when ostensibly achieved, so often go astray.

The harm done by the FDA does not result from defects in the men in charge—unless it be a defect to be human. Most are and have been able, devoted and public-spirited civil servants. What reformers so often fail to recognize is that social, political and economic pressures determine the behavior of the men supposedly in charge of a governmental agency to a far greater extent than they determine its behavior. No doubt there are exceptions, but they are exceedingly rare—about as rare as barking cats

I like the point that complex systems have their own inherent rewards and punishments. Trying to bolt on some additional attribute (such as the capacity to bark) is simply not possible. To change what the system produces, you have to change its integral rewards and punishment mechanisms.

You can breed a cat to bark but you cannot train it to do so. You have to tackle the system, not the individual.

Deterministic social engineers always miss this. They want to mandate reality onto complex systems without first understanding the system, its rewards/punishment mechanisms, and the emergent order of the system. It is why so many centrally imposed solutions fail.