Michael Kremer’s unconventional ideas attract a following and win him a full professorship at Harvard -- at age 35
Eventually, Michael Kremer admits that he’s in economics because it’s so much fun.
Pardon me?
Yes, says Mr. Kremer, sitting in a sparse but slightly jumbled office at the Brookings Institution, where he often spends time as a senior fellow. As he sees it, economic theory and practice must be taken seriously because the insights they lead to are crucial to the health of societies.
But they are also a blast.
That enthusiasm has produced some surprising ideas. Among them:
* Social Security taxes should be lower for young workers than for prime-age workers, since young people are more likely to be pushed into unemployment or crime by such taxes. In addition, many people need the extra income as much when young as when retired -- for example, to finance education.
* Population increases have been a good thing, at least historically, and that might still be the case, at least in the developed countries. In his paper titled “Population Growth and Technological Change: 1,000,000 B.C. to 1990,” he argues that more people generate more ideas. Over most of history, the resulting technological progress has fueled economic growth and allowed the population to grow even faster.
* Governments should offer to buy patents as a way to lower the cost of beneficial inventions such as pharmaceuticals. Mr. Kremer’s system would involve using auctions to estimate the value of patents. Patents, once purchased, would be placed in the public domain so that any company could produce the good. The system would include safeguards against collusion by firms, and would aim to prevent monopoly holders of patents from charging excessive prices.
Of course, many academics have unorthodox ideas. Mr. Kremer is distinct, his fans say, in being able to analyze and convey them in a convincing fashion -- so convincing that the economics department at Harvard University decided to lure him away from Massachusetts Institute of Technology, where he already had tenure, to a full Harvard professorship. He began there this academic year, and he is winning acclaim, at age 35, as one of the country’s top emerging economists. In 1996, President Clinton named him one of the first 60 recipients of the lucrative Presidential Early Career Award for Scientists and Engineers. The next year, he received one of the John D. and Catherine T. MacArthur Foundation’s famous fellowships.
Colleagues use words like “amazing,” “creative,” and “special” to describe him. His hallmark, they say, is that he fashions elegant, conceptually innovative models that make his often-counterintuitive ideas about complex issues at least plausible, and usually convincing.
So much so, that other scholars hesitate to dismiss even the most unlikely Kremer hypotheses. And some of them do seem unlikely. In response to his analysis of population growth, for example, a spokesperson for Zero Population Growth points out that other economists emphasize that, whether or not economies benefit purely in monetary terms, population growth also causes such problems as environmental damage, urban crowding and sprawl, and increased pockets of poverty.
Even more provocative is a paper of Mr. Kremer’s that asks: “Can Abstinence Increase the Prevalence of AIDS?”
Mr. Kremer’s credentials may need to be set out before explaining that idea, which brought much attention -- including that of the anonymous MacArthur panel -- to a young economist who was already getting noticed in academic circles.
Much of his thinking and career have been shaped by just one year, 1985-86, which he spent in Kenya, before heading to graduate school. He had planned to remain only a few months, in a village in Kakamega district. But after a local official asked him to teach mathematics, physics, and chemistry at the district’s first secondary school, Mr. Kremer kept extending his stay. The school had trouble finding a replacement because few qualified Kenyans were interested in teaching in the rural area where it was located.
His eventual solution: “I wrote back to my Harvard dormitory and found there were a lot of people interested in this sort of thing.” He set up WorldTeach, a nonprofit international teacher-placement organization (http://www.worldteach.org/contact.html), which has now placed hundreds of teachers, most of them American, in such countries as China, Costa Rica, Ecuador, Kenya, and Namibia. Mr. Kremer directed the organization from 1986 to 1989, and now serves as its board president.
His year in Kenya changed everything for him. Battling shortages of equipment but buoyed by his students’ enthusiasm for learning, Mr. Kremer became even more interested in his undergraduate specialty, Third World development. And much of his work relates specifically to Kenya.
“Elephants,” for example, is a 1996 paper inspired by his visit to a Kenyan game park. In the paper, he and a colleague, Charles Morcom, developed an economic model for the illicit market for elephant ivory. Among their findings was that, under some circumstances, governments or conservation organizations would do well to stockpile ivory -- culled from sick elephants, for example -- and then threaten to flood the market if numbers of the animal became dangerously low. That would drive down the price for elephant ivory and discourage poaching, they argued.
Mr. Kremer’s own endangerment from malaria, which he contracted while in Kenya, prompted one of his most significant, complex research efforts: to design incentives for companies to develop vaccines to prevent diseases like malaria, tuberculosis, and AIDS, and to find ways to make them affordable to the people in developing countries. There, such diseases kill millions each year. Some 90 percent of malaria cases occur in sub-Saharan Africa alone. Yet, vaccine-research spending on malaria, as compared with spending on diseases more common in the West, is very low, because biotechnology and pharmaceutical firms fear they will not make enough money to cover the cost of research.
In an extension of an earlier paper on patents, Mr. Kremer proposes that the governments of industrial nations, along with international organizations and private foundations, should build up funds with which to entice pharmaceutical and biotechnology firms to develop vaccines. They could do that by promising, in advance, to buy any new vaccines, and to pay a reasonable price.
That approach to encouraging research avoids the need for governments to try to choose which projects to finance. Under Mr. Kremer’s scheme, the funds would spend nothing until a vaccine was developed. The funds might, he adds, also promise bonuses based on the vaccine’s effectiveness.
In a paper now in progress (a draft is available, along with other papers, at http://www.economics.harvard.edu/faculty/kremer/papers.html), Mr. Kremer sets out ways to use this form of “pull funding” to complement traditional financing for basic research. He also argues for global coordination to discourage some countries from taking a free ride on the research incentives provided by other countries.
He and other advocates of the idea, including Jeffrey D. Sachs, the director of Harvard’s Center for International Development, are making headway. The World Bank recently announced that it would build a fund of $1-billion for such purposes. Meanwhile, President Clinton’s budget proposal includes $1-billion over 10 years in tax credits to firms whose vaccines were purchased by organizations like Unicef.
The Economist, in a review of Mr. Kremer’s earlier paper on the subject, called his work “intriguing,” and found fault with it only on the grounds that firms might collude to profit from the purchase system. The system could conceivably “end up subsidising drugs by more than their social value,” wrote The Economist.
Those reservations aside, the scope of such work impresses Mr. Kremer’s colleagues, including Edward L. Glaeser, a professor of economics at Harvard. “He has a certain amount of intellectual courage. A great deal, in fact,” he says.
“At a time when people often are tackling smaller and smaller things, it’s important that people like Michael are there to handle topics like that,” he says.
Similarly broad in scope is another idea that Mr. Kremer came up with in Kenya, prompted by a snafu -- one of his own.
“I don’t know if this is too embarrassing a story to tell.” He hesitates, chuckles. “When we were running our first orientation for incoming WorldTeach volunteers, we found a site that seemed like a good one, and we hired people to teach the volunteers Swahili, and we arranged time for extra coaching sessions which would also be a chance for the volunteers to practice-teach Kenyan children. And we arranged for food and cooks. ...
“But we forgot toilet paper.”
That, he says, made him think “about the difficulty of running a business, or running an operation of any kind, in a developing country, where it’s easy for one thing to go wrong, and that can cause the whole thing to fail.”
Hence the metaphor in the 1993 paper that resulted, “The O-Ring Theory of Development,” referring to the part whose failure caused the explosion in 1986 of the space shuttle Challenger.
Mr. Kremer at first thought about that phenomenon as it applied to development efforts, but then realized it also related to labor markets.
If one low-skill worker could jeopardize the work of many high-skill workers, he reasoned, it followed that high-skill workers would tend to group with high-skill workers, low-skill with low-skill -- and that would result in large wage disparities between firms.
That phenomenon could help explain the large income differences between countries, and the predominance of small firms in poor countries, he reasoned. In another paper, with Eric S. Maskin, a professor of economics at Harvard, he argued that evidence in Third World economies suggests that the “sorting” phenomenon he describes is becoming more pronounced.
Economists continue to debate such issues, but many have praised Mr. Kremer’s approach for emphasizing how many factors combine to create inequality. He has shown, wrote Steven N. Durlauf, a professor of economics at the University of Wisconsin at Madison, in the journal Politics & Society, why “the role of an individual’s education in determining his labor market prospects cannot be known without understanding who his co-workers are and the sort of firm in which he is employed.”
Jeffrey G. Williamson, the chairman of Harvard’s economics department, responds almost giddily to the O-ring paper’s title alone: “It sounds very weird,” he says, but the work “turns out to be an insight about technology that has a very profound influence on the way we think about technological change and growth.”
Moreover, its approach is helping redefine development economics, he says. A few practitioners of “the new microdevelopment,” pursuing empirical work as thoroughly as theory-building, are also influencing such fields as industrial organization and international trade, he explains, and “Michael is leading that pack.”
For example, the O-ring thesis also bears on the economics of education: If one person’s small error can ruin everyone’s work, Mr. Kremer argues, it follows that, from the level of the firm up to even a whole nation, “if everybody else decides to invest in skill and become high-skill, then it’s worthwhile for me to do that. If other people don’t do that, it doesn’t do a lot of good for me to be high-skill.”
Other economists working in the same areas as Mr. Kremer appear unusually receptive to his approaches. In part, that may be because his peers are still taking stock of his unorthodox strategies, which, as is increasingly common in the discipline, make use of methods and approaches from fields outside economics -- epidemiology and engineering, for example.
Certainly, Mr. Kremer himself says, there will be disagreements about the policy implications of his ideas on, say, the AIDS epidemic. To date, however, he appears to have inspired work along lines similar to his, rather than the opposition more typical of academic discourse. For example, his paper on the mutuality of population growth and technological progress prompted Charles I. Jones, an assistant professor of economics at Stanford University, to look specifically at the role of fertility in that process. “My research,” says Mr. Jones, “is very much in line with the approach he takes. I build on it and look at a slightly different question.”
Mr. Kremer’s papers commonly try to build consensus among different schools of thought. He says: “Rather than just say, should taxes be higher or lower -- the straight long-standing debate -- I try and think, ‘Well, is there a way we could restructure the institution in some way to ideally achieve multiple objectives?’”
More generally, he often links issues of economic structures with forms of human decision-making that are more subtle than those considered in classical economic theory -- not just whether, say, a cost is worth incurring, but what are the factors that go into such decisions. “That’s something that economics in general is looking at a lot these days,” he notes.
Most strikingly, those factors are at play in Mr. Kremer’s startling consideration of the dynamics of AIDS-prevention efforts, and his hypothesis that abstinence, under some circumstances, could actually increase the incidence of AIDS.
“One thing that is a standard implication of a lot of epidemiological models is that it’s very important to focus on the highly active people,” Mr. Kremer explains. “Reductions in rates of partner change there, or safe sex there, are going to have a much bigger impact than among people who aren’t very active.”
He reasons: “If high-activity people reduce their activity by a smaller proportion than low-activity people, the composition of the pool of available partners will worsen.” And so he concludes: “There might be some circumstances in which a reduction in the number of partners by one group in the population could actually increase the prevalence of H.I.V. in the long run for society as a whole.” He did, he says, try to determine whether various factors would mitigate against that actually happening. “To my surprise, you couldn’t, really, rule out this possibility,” he says.
So, he believes, AIDS-prevention campaigns might most effectively concentrate on people with many sexual partners, and emphasize safe sex rather than abstinence.
He cautions that his paper is an attempt to encourage more informed responses to the AIDS epidemic, and that it presents a theoretical possibility. He says he published the work in Mathematical Biosciences, to alert mathematical epidemiologists; “but by no means is that at a stage that it should be going to policy people.”
Again, says Harvard’s Mr. Williamson, just coming up with the hypothesis exemplifies what makes Mr. Kremer so compelling an economist: “He is not a prisoner of paradigms. His head is just fully unplugged from everybody else’s strong priors and commitments, so he can see things that the rest of us don’t.”
http://chronicle.com Section: Research & Publishing Page: A20