THE PIPELINE PROBLEM.(The Talk of the Town)(pharmaceutical industry)

Merck is one of history's most innovative corporations. It devotes three billion dollars a year and ten thousand people to the research and development of new drugs. So here's a question: How many drugs for diabetes do you think all these men and women, this army of scientists, managed to come up with in the past four years? None. How many anti-cancer drugs? Zero. How many drugs that fight infectious diseases? Zero. Since 2000, in fact, Merck has introduced just three new drugs. Drug development is hard, but, by any measure, eking out less than one product a year is no way to make a living in the major leagues.

Profitable as "big pharma" remains--Pfizer made twelve billion dollars last year--a deep sense of anxiety prevails in the industry. That's because Merck is no exception: most drug companies have what's known as a pipeline problem. That is, the patents on the drugs that are now making money for them are about to expire, and they don't have enough new drugs in development. The number of "new molecular entities"--drugs not yet introduced in the United States in any form--approved annually by the F.D.A. has fallen by sixty per cent since 1996, and new drug applications have dropped nearly forty per cent. What's more, many of the drugs that are being invented are of the "me, too" variety--variants of existing drugs.

Big pharma's solution has been a mania for mergers. As the industry joke has it, you know you're in the pharmaceuticals business when you've worked for five companies in the past two years and you're still sitting at the same old desk. If your own pipeline is low, the thinking goes, buy another one. And so Pfizer bought Warner-Lambert and Pharmacia, Glaxo merged with SmithKline Beecham, and Astra merged with Zeneca. Last month, the French drugmaker Sanofi launched a hostile bid for Aventis. When the going gets tough, the tough go shopping.

The traditional pharmaceutical research model harks back to processes developed by German and Swiss chemical firms in the late nineteenth century, when chemists synthesized and screened thousands of compounds in search of a few potential new drug candidates. Although the methodology is more sophisticated now, success is still in many ways thought to be a matter of brute force: throw hundreds of scientists at a problem and hope for the best. It's crapshoot economics; a few great successes can pay for myriad failures. So bigger has always been seen as better.

Today, though, the advantages of size are trumped by what are called "diseconomies" of scale: inertia, bureaucracy, risk aversion, clock-watching, office politics. Joseph Kim saw a lot of this firsthand, as a scientist at Merck for nine years, and now he likes to compare Merck to the Titanic. "Companies like Merck have fantastic scientists working for them, but they also have these middle and upper layers of managers who are just ...