William Hamovitch and Richard Morgenstern (H-M) are certainly correct in saying that experience is at least as important as research output in the determination of salaries. However, professor will probably never gain any considerable experience at prestigious university unless he publishes; instead, he will be fired. This is demonstrated by the fact that only 6.4 percent of the associate professors at the time of promotion to that rank had published nothing as compared to 22.4 percent for assistant professors who had not yet been promoted. Only 1.6 percent of the full professors at the time of promotion had published nothing. The existence of diminishing returns to publishing lends support to H-M's conclusion that experience is possibly the most heavily weighted factor in the reward process. To ignore, though, the very large contribution which publication makes in the salary and promotion process would significantly reduce the predictability of the model. Given the diminishing returns to publishing and the high relative rewards to experience, why do professors continue to publish after gaining tenure? Nonpecuniary rewardsprestige and self-satisfaction-would then explain their H-M suggest that a larger part of the salary differential is unrelated to rational market behavior. Actually the opposite may be true. If professors are willing, to do research at low pay because of the nonmonetary rewards, why pay them any more than is necessary? * University of Missouri, Rolla.

The question of whether controls on the importation of foreign oil into the United States should take the form of tariffs or quotas has been a topic of recent public debate and investigation bv econonmists. Under static competitive conditions it is well known that equivalent tariffs and quotas can be constrtucted. Hence in this context, there is no choice to be mlade on economic efficiency grounds.' Hlowever, in a recent isstue of this Review, George Hay poinlts out that the actual market for oil in the United States differs fromii the required textbook conditions for equivalence. Under the U.S. oil import program which prevailed until recentl-, each refiner's quota for inmport of foreign oil is a positive function of his refinery input. Since import tickets are allocated free of charge, rather than auctioned, the form of the quota lowers the marginal cost of domestic refiners. Hay goes on to show that when combined with other static competitive assumptions, this quota mechanismi could generate greater consumer benefits in terms of lower prices than would an equivalent tariff (equivalent in the sense that the same percentage of imports is admitted).2 Hay expresses a preference for tariffs in a real world context and warns that his analysis of the price effects of the U.S. oil quota system holds only under very restrictive conditions. However, he does not address what is perhaps an even more important deviation of the domestic oil industry from the standard textbook model: crude oil production in the United States was limited in the major producing states by regulatory commissions that practiced market demand prorationing under the old oil quota program. Under this system, an-y price set by the industry is ratified by the commissions by limiting production to a level that will not result in the accumulation of undesired inventories.' We are not addressing the issue of the level of price in the oil industry in this paper. Rather, we wish to review the effects of tariffs and quotas on resource allocation, an issue which Hay omits from his analysis; and, for this purpose we make use of the simple model of a profit-maximizing monopolv as a characterization of the domestic oil industry. The assumption of profit maximiza-

Macro-finance theory implies that trend inflation and the equilibrium real interest rate are fundamental determinants of the yield curve. However, empirical models of the term structure of interest rates generally assume that these fundamentals are constant. We show that accounting for time variation in these underlying long-run trends is crucial for understanding the dynamics of Treasury yields and predicting excess bond returns. We introduce a new arbitrage-free model that captures the key role that long-run trends play in determining interest rates. The model also provides new, more plausible estimates of the term premium and accurate out-of-sample yield forecasts.

The probit is generally considered to be one of the easiest nonlinear maximum likelihood problems. Nonetheless, in the course of at-tempting to replicate G. S. Maddala’s (1992, pp. 335–38) probit example, Houston Stokes (2003) encountered great difficulty. Of the six coeffi-cients, five coefficients/standard errors he could duplicate, but the sixth was off by more than rounding error. So he tried another package. And another. And another.... Finally, five dif-ferent packages had declared convergence to five solutions that differed only in the sixth coefficient. Estimates of the sixth coefficient ranged from 4.4 to 8.1, and estimates on its standard error ranged from 46 to 114,550.

Linear regressions with period and group fixed effects are widely used to estimate treatment effects. We show that they estimate weighted sums of the average treatment effects (ATE ) in each group and period, with weights that may be negative. Due to the negative weights, the linear regression coefficient may for instance be negative while all the ATEs are positive. We propose another estimator that solves this issue. In the two applications we revisit, it is significantly different from the linear regression estimator.