Box, George E. P., and David R. Cox, An Analysis of Transformations, Journal of the Royal Statistical Society, series B, 26 (Apr. 1964), 211-243. Boyes, W. J., An Application of Specification Error Tests-The Liquidity Trap, Manchester School of Economics and Social Studies (1978), 139-151. Chang, Hui-shyong, Forms and the Demand for Meat in the United States, this REVIEW 59 (Aug. 1977), 355-359. Greenfield, J. N., Effects of Price Changes on the Demand for Meat, Monthly Bulletin of Agricultural Economics and Statistics 23(12) (1974), 1-8. Ramsey, James B., Limiting Functional Forms for Demand Functions: Tests of Some Specific Hypotheses, this REVIEW 56 (1974), 468-477.

The strength of the relationship between schooling and the logarithm of earnings at different levels of experience can be measured by coefficients of determination or by residual variances. For tracking unobserved post-school investments in cross-sections, the latter are clearly preferable; the greater reliance on the former in the literature is misplaced. If the fraction of earning capacity devoted to postschool investment is uncorrelated with earning capacity at school-leaving, the relationship between schooling and the logarithm of earnings should be strongest at overtaking, which would plausibly be placed in the first decade of work experience. Mincer reported coefficients of determination which followed this pattern, at least for full-year workers. An analysis of more recent data failed to detect this pattern for coefficients of determination, but found more favorable evidence for residual variances using weekly (but not hourly)

T HE Employee Retirement Income Security Act of 1974 (ERISA) has provoked considerable debate about the desirability of various retirement plan provisions and the appropriate role of government in regulating the private pension plan contract. Among the issues debated is the question of who presently pays for private retirement benefits, and who should. An answer to at least the first of these questions is provided by the theory of differences. In competitive markets, a firm that provides pension benefits should pay lower wages than one that does not, thereby offering the same equilibrium value of total compensation to all workers of equal productivity. By the same token, firms that offer pension benefits on relatively desirable terms are expected to offer lower wage rates than firms offering pension benefits on very restricted terms. These simple notions imply that (1) workers pay for their own pensions by accepting lower wages, and (2) government regulation of the content of private pension plans may not significantly alter labor costs or the expected lifetime income of workers. From this perspective, contributions to private pension plans serve the exclusive purpose of enabling individuals to reallocate their resources over time according to their diverse tastes, and do not affect total labor compensation. The primary purpose of this paper is to examine the empirical validity of the equalizing differences hypothesis. By examining the relationship between wages and pension plans, we also hope to improve our ability to account for wage differentials. Contributions to private retirement plans have grown rapidly in recent years-from 1.7% to nearly 4.0% of private sector wages between 1950 and 1979, and coverage has increased from 22% to more than 45% of all private wage and salary workers.' Most previous studies of wage determination have ignored fringe benefits, let alone pension plans.2 But if the growth of pension plans continues and the equalizing differences hypothesis is correct, continuing neglect of pension provisions implies an increasing inability to account for wage differentials across firms, industries, occupations, race, sex, and age, or by the same token, an increasing tendency to attribute such differences to the wrong factors. The paper begins by reviewing the properties of competitive equilibriuni in labor markets in which compensation consists of current and deferred wages.3 Building on this foundation, we demonstrate how the annual cost of a pension plan depends on its various provisions, including vesting, early retirement, normal retirement, and benefit formula. With data on the earnings and pension provisions of individual workers in 133 large firms, we find some support for the equalizing differences hypothesis. We also observe that the extent of equalization diminishes with age, suggesting a redistribution of compensation from younger to older workers.

7 The F-statistic for the first test is F(4,107) = 0.50 while for the second test it is F(4,11 1) = 0.81. Both of these values are well below the critical-F at conventional levels of significance. 8 Further experimentation showed that one additional parameter could be eliminated by, for example, disposing of the last join point (setting d2 = 0.). However, since the purpose here was simply to show that fewer than 12 parameters are required, the equation has been left in the present form.

Regional migration is analyzed utilizing a model that develops a system of urban areas. The areas differ in their endowment of a site-specific factor—climate is used as the example. The effects of differences in tax rates and technology are determined in a model where the price of housing is endogenous. Compensation for an inferior climate occurs through regional differences in income levels or the price of housing, dependent on the manner in which climate affects production or consumption. The market distribution of households is found to be suboptimal in cases where utility is derived directly from the consumption of climate. The locational choice of households is determined by a variety of factors. Recent contributions concentrating on climatic variations are Graves [1979], Graves and Linneman [1979], and Izraeli [1973]. A problem with the migration literature has been the lack of specifi-cation of a model that contains a system of urbanized areas. The model should determine the equilibrium size of urban areas. The migration flow results from the adjustment of the stock of population to changes