Knowledge that Transforms

In a recent note in this Review, Arthur Hurter, Joseph Martinich, and Enrique Venta (hereafter Hurter et al.) consider the question of the conditions under which the location decision for a cost-minimizing firm can be treated separately from the decision governing the firm's desired input mix. The authors argue that if the firm's production function is homothetic, total costs can be minimized by first determining an optimal location for the facility, and then determining the optimal input mix at this location. The implication of this viewpoint is that facility location and facility design problems can be analyzed independently. Thus the traditional theory of the firms which abstracts from spatial elements, and the traditional Weberian theory of industrial location which does not consider the choice of input mix, remain independently valid without involving the possibility of suboptimal decision making on the part of the firm. My purpose here is to clarify the conditions under which this type of separability does obtain: in particular, I argue that the facility location problem can be solved independently of the input mix selection problem if and only if a firm's production function is of the fixed proportions type. This implies that homotheticity is neither necessary nor sufficient for separability of these decision problems.

We are quick to add, however, that the difference in estimated OJT across the two groups is not large in 1966. In fact, H(1966)ID=1 H(1966)|D=O = $0.55 (with a standard error of $0.67) so that the difference is relatively small and the standard error is quite large. The conclusion is then that for this early period, the measured wage differential underestimates the true differential, but the extent of that understatement is questionable. [p. 559]

Women, youth, and minorities are frequently cited as an almost one-word explanation for declining productivity growth in the U.S. labor force as measured at the macro level. While some economists have proposed theories to account for productivity slowdown which emphasize such things as declining capital-labor ratio (J. R. Norsworthy, Michael Harper, and Kent Kunze, 1979); energy constraint (Edward Hudson and Dale Jorgenson, 1978); and macro instability (Richard Nelson, 1980; Michael Mohr, 1980), many analysts have cited the changing composition of the labor force as a significant contributing factor. (See, for example, Edward Denison, 1979; Norsworthy et al.; Frank Gollop and Jorgenson, 1980.) These assertions are tantamount to suggesting that admitting certain groups into the labor force necessarily lowers productivity and should not go unchallenged for several reasons. Inferring negative productivity coefficients to particular groups is likely to reinforce statistical discrimination whereby employers use perceived characteristics of groups as information shortcuts for making decisions about individuals (Edmund Phelps, 1972). It also has considerable implication for future U.S. productivity trends as the demographic composition of the labor force continues to change, and for formulating appropriate labor policies. This paper examines labor market trends and productivity measurement methodology as it pertains to labor force composition, and concludes that whether women, youth, and minorities have contributed significantly to the decline in productivity growth, and are likely to do so in the future, depends on how productivity is defined and measured. Current definitions of productivity are essentially based on micro concepts. I argue that they may be misleading indicators of trends in economic efficiency at the macro level, especially under the conditions of social change and labor-supply growth which characterized the 1970's.

The scholarly debate over the nature and cause of the significant racial disparities in prison incarceration rates in the United States has taken on renewed intensity in recent years. Two sorts of activities have spurred the debate. On one hand, researchers such as Alfred Blumstein (1982), Jan Chaiken and Marcia Chaiken (1982), and Joan Petersilia (1983) have begun to use powerful analytic and conceptual tools to scrutinize the hypothesis that racism or racial discrimination exists in the criminal justice system, or that it is the cause of the racial disproportionality of our prisons. On the other hand, minority scholars and public opinion leaders have begun a very visible and vocal attack on the results of the conventional social science community. (See, for example, National Minority Advisory Council on Criminal Justice, 1980.) These activities have stimulated much discussion among public policymakers and legislators. Ranking black members of the U.S. Congress, for example, have gone on record by questioning social science research findings that purport to show that racial discrimination in certain aspects of the criminal justice system does not exist - or, at least, that its alleged existence is not a cause of the greater representation of blacks in the prisons or the criminal population. Economists have not been leaders or even active participants in this debate. This is surprising for several reasons. Many of the conventional tools of econometrics can be called upon to resolve some of the statistical issues in dispute; post-Beckerian models are likely to yield more than negligible benefits in sorting out the theoretical effects of punishment on criminal activities; and radical labor market paradigms may prove useful in examining the historical evolution of prisons and punishment in America. A brief overview of a number of different areas of research on race and punishment will illustrate the inherent potential as well as the unrealized promise of economic approaches.

(3) WA = R1Pne + 1 _ R A n R2U, where pe is the forward-looking expectation and f31 is the credibility coefficient (or degree of belief) attached to the model and the announced policy rule. Total expectation is the weighted sum of the forwardand backward-looking beliefs. In the case of the simplest monetaristrational expectations viewpoint, P, =1, and pe = m*, where m* is the announced monetary target (with the trend in velocity offset by the economy's natural rate of growth). Aggregate supply price will be