Finn E. Kydland
Although the issue of time consistency might sound arcane, it is crucial for economic policy. In their classic 1977 article, Kydland and Prescott pointed out that even governments that care deeply about their citizens often have “time-consistency” problems. A government may decide, for example, in the interest of all its citizens not to subsidize people’s losses if they are flooded. The reason is that if people expect to be bailed out when flooded, they will locate in areas where it is inefficient for them to locate, because they are not bearing the whole cost of that location decision. So the government announces that it will not bail out people if their homes are damaged by floods. Then a flood comes, and the government decides that the optimal strategy is to bail people out: there is no danger of the bailout causing them to locate there because they already have. Here is the problem: people figure out that the government has this time-consistency problem—making a decision later that contradicts the earlier announced policy—and so do locate in the flood-prone area in the first place. Of course, people do not talk about the government in these terms, but they will try to estimate how likely it is that the government will cave in to political pressure and change its policy. If there is even some substantial likelihood, people’s decisions to live in the flood-prone area will be distorted by it. In short, unless they can make a binding commitment regarding future policies, even governments that care about their people will have a credibility problem.
Kydland and Prescott modeled this problem mathematically and showed that it applies to many policy issues. For example, “time inconsistency” could explain why governments had trouble ending inflation. Although low or zero inflation is an optimal long-run policy, a government that announces a monetary policy to achieve it will be tempted to increase the growth rate of the money supply so as to reduce unemployment. But each year, government will find itself in that position and may never take the necessary painful short-run measures to end inflation.
Kydland’s and Prescott’s answer to the problem of persistent high inflation was to make central banks follow rules that would prevent them from inflating. But many economists and most government officials want the central bank to have some flexibility to deal with unanticipated events. The solution: structure the incentives so that central banks will care about keeping inflation low, but will still have the power to deal with unanticipated events. Economist Kenneth Rogoff,1 building on Kydland’s and Prescott’s framework, showed that this optimal balance between credibility and flexibility could be achieved if monetary policy were delegated to an independent central bank and if the central bank were managed by someone more averse to inflation than the citizens in general. Interestingly, Alan Greenspan, an inflation “hawk,” became chairman of the Federal Reserve Bank just two years after Rogoff’s research. Also, reforms of central banks in New Zealand, Sweden, and the United Kingdom were based on academic economists’ research that drew on the Kydland-Prescott framework—and the result was a substantial decline in inflation in those three countries.
Another example of the time-consistency problem is patents. Patents involve a trade-off between the short-term welfare loss from monopoly exploitation of a patent and the long-term welfare gain from the patent-induced incentive to innovate (see intellectual property). A government that does not credibly commit to enforcing patents will often violate the intellectual property rights of patentees. Many governments around the world are doing this with patents on drugs. Potential innovators, anticipating this, will innovate less. So a government that wants optimal innovation needs to figure out some way to lock itself in to protecting patents in the future.
One final example is tax policy. The incentive to accumulate capital will depend on the anticipated tax rate on capital in the future. A government may commit to a low tax rate in order to encourage capital formation. But once people have invested in capital, governments will be tempted to raise the tax rate on capital because the capital is already “formed.” Again, though, many people will anticipate this and, unless the government is tied in to its commitments, will invest less in capital than otherwise. One can see the U.S. Constitution, along with an independent judiciary that enforces it, as a way of handling this credibility problem.
Kydland is a citizen of Norway. He earned his B.S. from the Norwegian School of Economics and Business in 1968 and his Ph.D. from Carnegie Mellon University in 1973. He is currently a professor at the University of California at Santa Barbara and at Carnegie Mellon University.
Kenneth Rogoff, “The Optimal Degree of Precommitment to an Intermediate Monetary Target,” Journal of International Economics 18 (1985): 1169–1190.