Knowledge that Transforms

SINCE 1954 nearly all vestiges of legal segregation of public schools in the South have been eliminated. But, at the same time, the suburbanization of households in Southern cities has contributed to the de facto resegregation of some schools. Just as the Tiebout hypothesis (1956) suggests that households locate according to their preferences regarding local public goods, there is some evidence to indicate that preferences for segregated schools may have contributed to this resegregation. Glantz and Delaney (1973) found that, by one measure, metropolitan residential segregation increased more during the 1960's in Southern metropolitan areas -where city schools were substantially desegregated than in the Northern metropolitan areas studied. And for some organizations, a primary reason for supporting recently ordered metropolitan desegregation plans which would effectively combine city and suburban school systems is the belief that desegregation of city schools alone merely contributed to white flight from the city.' But, because there are other factors which may also cause suburbanization of whites, such as employment decentralization and the growth of Negro ghettos, it is not clear whether school desegregation has had an independent effect on the demand for housing by households. This paper presents an analysis of housing prices to determine whether desegregation has an independent effect on the price paid by whites for housing. Since the supply of housing is relatively inelastic in the short run, an effect of this kind would support the hypothesis that desegregation affects households' demand for housing. While shifts in demand will result primarily in price effects in the short run, quantity changes and locational rearrangement will be most important in the long run. In order to determine if shifts in demand have accompanied desegregation, this paper will examine such price effects. The metropolitan area studied is Atlanta, Georgia, a Southern city which experienced school desegregation and apparent white flight during the 1960's. In 1960, a year before desegregation was begun, 37.2 % of the students in the city school system were Negro. By 1970 this figure was 67.1%. During the decade the proportion of Negro families in the city rose from 34.0% to 47.7%o. These changes reflect a number of different locational trends, one of the most important of which was a rapid growth in the city's Negro population. In order to assess the independent effect of school desegregation on housing demand, it is above all necessary to separate the effect of school racial composition from that of neighborhood racial composition, as well as other supply and demand factors affecting housing prices. The analysis presented in this paper uses data on housing prices and characteristics drawn from 1960 and 1970 census tract reports for Atlanta. The comparatively rapid end to de jure segregation in that city during the decade of the 1960's provides a unique opportunity to separate the effects of neighborhood and school integration. The empirical analysis supports the hypothesis that school desegregation does have a significant effect on housing prices, independent of neighborhood racial change. Section I discusses the use of housing prices in determining the effect of public service characteristics on housing demand. Sections II and III describe the data and the empirical findings. Section IV summarizes the analysis. Received for publication December 3, 1973. Revision acaccepted for publication July 30, 1974. * I am grateful to Professors Martin Feldstein, John Kain, Richard Freeman, and Gregory Ingram, members of the public finance and urban economics seminars at Harvard, and referees for this Review for comments on earlier drafts of this paper. Financial support was provided by the Ford Foundation. 1 See, for example, the testimony of William L. Taylor, Director of the Center for National Policy Review in hearings before the Senate Select Committee on Equal Educational Opportunity, November 30, 1971, p. 10475 or Junie Brown, City School Case: No End in Sight, Atlanta Journal-Constitution, December 31, 1972, pp. IA, 6A.

IN the area of international trade, there have been a number of studies designed to explain imports and exports disaggregated by commodities.' These studies, however, all concentrate on the trading patterns of industrial countries and there has been very little in the way of a systematic analysis of the imports and exports by types of commodities for developing countries. There is, of course, an obvious reason for ignoring these countries, and that has to do with the availability, or rather paucity, of adequate data; this data problem is most pronounced in the area of prices of disaggregated imports and exports. Although it is generally believed that the trading patterns of developing countries differ from those of the developed, an empirical examination would be necessary to determine the extent of such a difference. The purpose of this paper is to study the behavior of the imports of Venezuela at both the aggregated and disaggregated levels during the period 1953-1972. Data on imports at a disaggregated level have recently been made available for Venezuela for a nine-commodity breakdown;2 the data include the value and price for each of these categories on an annual basis, as well as the price of the domestic competing product. These nine categories cover some 80% of the total Venezuelan imports. The results of this study should provide information on the behavior of different types of import goods and on the nature and extent of the bias involved in estimates of elasticities based only on aggregate imports.

THEORISTS of the American labor movement long have argued over the causes of fluctuations in union membership and the prospects for future union growth. For a number of years, the hypothesis of Commons (1932) and Perlman (1966), which related union expansion to the business cycle, was the most widely accepted explanation of the fluctuations in union membership. the years, a number of critics such as Dunlop (1948), Shister (1953), and Bernstein (1954) challenged the business cycle hypothesis on the grounds that American unionism is too complex and diffuse a social phenomenon to be understood in such simple terms. They contended that a multicausal system (including the cycle) is necessary to account for the rise of trade unionism. Although union theorists have frequently provided statistical data that are consistent with their inferences concerning the major factors influencing union growth, prior to the study by Ashenfelter and Pencavel (1969) their hypotheses had not been tested on an empirical plane which could disprove the suggested causal relationship between such factors and union growth. Ashenfelter and Pencavel (A-P) used multiple regression analysis to estimate a single behavioral relationship, including social and political as well as economic variables, capable of explaining the growth of American trade unions membership in the period 1900-1960. Although the model of A-P apparently has identified the determinants of union growth, there is some question as to whether the model has equally identified the determinants of future union growth. In econometric time-series analysis, it is assumed that differential periods of time are homogeneous, except for differences in the explicit variables of the system that are measured, and for differences in random effects. Over long sweeps of time, this assumption may become very tenuous.' This issue lies at the heart of the recent debate among the so-called saturationists and the school concerning the future prospects of the rate of growth of the American labor movement.2 The saturationists argue that significant changes have occurred within the structure and composition of the American labor force which have caused the past determinants of union growth to be inoperable in the future.3 Proponents of the historical approach challenge the general validity of the influence of structural factors on the comparative propensity of workers to join unions. Taft (1963) contends that the saturationists argument assumes propensities and psychological attitudes which have not been proven. In fact, actual experience has shown these assumptions to be baseless, and not a scintilla of evidence has been presented to justify these conclusions. They maintain the labor movement increases its size in two ways at a modest pace over long spans of time and in sharp spurts at infrequent intervals, ... . and that the slow growth of unionism in the post World War II period easily can be fit into the theory.4 The debate between the saturationists and the school has not yet been resolved. The level of actual union membership has increased by 4,300,000 or 25.4% for the period 1953-1970, but the level of real union membership, as measured by the per cent of nonagricultural employment organized, has declined from 34.1 to 30.1. The recent success of the labor movement in organizing some difficult structural groups such as government employ-

RECENT publications have underscored the priority being given by government officials to policies promoting more rapid technological change in Canadian manufacturing industries.' While major emphasis is being placed on encouraging the indigenous commercial development of new production techniques and products, i.e., innovation, relatively little concern has been paid to the subsequent diffusion of process and product innovations. Since the potential benefits of any innovation will remain largely unexploited by slow subsequent rates of adoption, policies fostering rapid adoption of existing technology are complements to policies encouraging innovation. Furthermore, quick imitation is an effective (and frequently necessary) means of blunting competitive disadvantages created by rival innovation.2 In view of the high costs usually associated with innovation, and the relatively small size of the domestic economy, imitative-rather than innovative-excellence might be a more feasible objective for many Canadian industries. Although an implicit presumption exists that geographical, cultural and economic ties with the United States insure that best practice technology in Canada will be close to that in other countries, the limited empirical evidence is inconclusive. Hufbauer (1966) found that production of a new synthetic material first took place in Canada approximately fourteen years, on average, after its initial production by the innovating country. Globerman's study (1974) comparing the adoption patterns of special presses in the paper industries of Canada and several Western European countries found that the rate of diffusion was faster among European firms, although the presses were initially introduced at approximately the same time as in Canada. Studies by Ault (1973) and Baumann (1973) of diffusion in the iron and steel industry provide evidence that Canadian firms, on average, adopted new technology as quickly as major Western European companies, and in the early stages, more quickly than their U.S. counterparts. In the light of limited research on an important policy issue, this study attempts to provide additional evidence on two related questions: 1. Do firms in Canada adopt innovations as quickly as their foreign counterparts? 2. What important factors influence industrial diffusion patterns in Canada? Analysis is focused on the diffusion of a major manufacturing innovation, numerical control, within comparable sets of potential users in Canada and the United States.3

P REVIOUS studies, Houthakker and Taylor (1970), Phlips (1972), and Houthakker and Verleger (1973), on the demand for gasoline have concentrated on the private consumer demand at least with respect to model formulation, although it is not clear that the data used in these studies were so restricted. These models have been specified as single equations involving short-run dynamic linear adjustments. In terms of the regressions actually run, the main effect has been to introduce lagged values of the dependent variable. Very little attention has been paid to prices other than own price. Our approach in this paper is to formulate the model in terms of a simultaneous equation system, to pay particular attention to relative prices in addition to own price and income, and to separate the demand by households from that by commercial (truck) users.

A CCORDING to the expectations hypoth.t esis, the forward rates of interest implicit in the yield curve provide unbiased estimates of the market's expectations of future spot rates. Recently, the empirical validity of the expectations hypothesis has been tested by examining the conformity of movements in forward rates to the properties implied by the efficient market model.' latter model contends that, subject to the constraints imposed by the costs of obtaining and using information, expectations incorporate all publicly available information and prices (forward interest rates) always reflect these expectations. In this paper, the utilization of the efficient market model as a theoretical construct for testing explanations of the yield curve is extended in two directions. First, the long dormant question of the predictive accuracy of forward rates is reopened. Interest in this topic ebbed following the publication of Meiselman's work which presented a test of the expectations hypothesis that did not depend upon forecast accuracy for its statistical power. However, it appears that the analysis of forecast errors contains valuable information regarding the validity of the expectations hypothesis and the efficient market model.2 second point of departure from the existing literature is the performance of semistrong form tests of market efficiency. purpose of these tests is to determine whether the market employs all publicly available information in particular, estimates of income and liquidity variables in forming expectations of future rates. By comparison, previous investigations of the efficiency of the debt market have been limited to weak form tests which merely seek to determine whether the market correctly uses past interest rate data in forming expectations.3 Received for publication June 27, 1973. Revision accepted for publication June 10, 1974. * authors are indebted to Michael Adler, John Ciccolo, Jr., Stanley Diller, Lawrence Fisher, Kevin Hurley, Levis Kochin, Charles Nelson, Frank Schott, and Christopher Sims for helpful comments and criticism; and to Patricia Chick and Rona Stein for their excellent research assistance. Earlier versions of the paper were presented at the meetings of the American Statistical Association, Montreal, Canada, August 1972 and at seminars at Columbia University, Harvard University, the University of Chicago, and the University of Minnesota. Comments on those occasions also contributed to the paper. views expressed are those of the authors and do not necessarily reflect those of the individuals mentioned above, or the Federal Reserve Bank of New York. authors, of course, bear sole responsibility for all remaining errors. 1 See, for example, Richard Roll, Behavior of Interest Rates (New York: Basic Books, 1970), and Thomas J. Sargent, Rational Expectations and the Term Structure of Interest Journal of Money, Credit and Banking, IV, February 1972, pp. 74-97. Within the context of the expectations hypothesis, the efficient market model implies that forward rates of interest follow a martingale sequence. One important property of such sequences is that forward rates applicable to a particular point in time are not expected to change. original analytical work on the efficient market model is due to Samuelson and Mandelbrot. Paul A. Samuelson, Proof that Properly Anticipated Prices Fluctuate Randomly, Industrial Management Review, VI (Spring 1965), pp. 41-49 and Benoit Mandelbrot, Forecasts of Future Prices, Unbiased Markets and Martingale Models, Journal of Business, XXXIX (Special Supplement, January 1966), pp. 242-255. 2 Meiselman argued that the hypothesis would be corroborated if it were shown that expectations were revised systematically. David Meiselman, Term Structure of Interest Rates (Englewood Cliffs, New Jersey: PrenticeHall, 1962). Prior to the publication of this study the expectations hypothesis had been cast into disrepute because empirical researchers found that forward rates were inaccurate predictors of future spot rates. Among those who rejected the expectations hypothesis because forward rates had little predictive value were Hickman and Culbertson. By comparison, Macaulay-who found only limited evidence of predictive accuracy -argued that while successful forecasting might be rare, forecasting might still be attempted. See, W. Braddock Hickman, The Term Structure of Interest Rates: An Exploratory Analysis (New York: National Bureau of Economic Research, 1942), mimeo; John W. Culbertson, The Term Structure of Interest Quarterly Journal of Economics, November 1957, pp. 485-517; and Frederick R. Macaulay, Movement of Interest Rates, Bond Yields and Stock Prices in the United States Since 1859 (New York: National Bureau of Economic Resarch, 1938). 3 implications of this limitation in Sargent's work are discussed by Robert Shiller, Rational Expectations and the Term Structure of Interest Rates: A Comment, Journal of Money, Credit and Banking, V, August 1973, pp. 856-860.

ECONOMISTS have suggested that differential growth in regional economy in the United States depends substantially upon labor supply, if it is assumed that the demand for a region's exports and its supply of capital are perfectly elastic.' The spatial movement of labor force or migration has been a hot subject in studies concerned with differential growth in regional income and employment. Most of these studies are concerned with a gross migration, and they always agree that employment or income consideration dominates other factors in making locational decisions among migrants.2 However, it should be noted that it is the rate of net migration (i.e., the difference of in-migration and out-migration divided by population) that directly affects the rate of labor force growth and, consequently, regional growth. Recently, more and more people have been commenting on the paradoxes of affluence. Discontent with the quality of life in the United States seems to have increased proportionally with technological advancement and growth in material wealth. Environmental quality, individual equality, economic opportunity and status, and a host of other forces that combine to shape the quality of life of the individual, are now major considerations in any public policy decisions. The primary objective of this paper is to explore the relationships between the variations in net migration rates among states and the levels of quality of life measured in those states. We first present a production model of the quality of life and some empirical results. Then follows a theoretical model, in which the decision of a household head to migrate is treated so as to maximize his quality of life. Empirical tests of the hypothesized relationship described in the migration model are contained in section IV. Concluding remarks follow.

B AINis (1956) analysis of the sources of barriers to new competition suggests that scale economies to the plant or firm allow sellers to elevate price somewhat above average cost without attracting new rivals. Multivariate statistical analyses of the determinants of seller concentration and profits henceforth have included measures of scale economies among their independent variables. However, direct measures, based on engineering or survey evidence, have been available to only a few investigators (Bain, 1956; Eastman and Stykolt, 1967). Others have employed various statistical proxies for the missing firsthand evidence. This paper proposes a new way to form these surrogate measures, compares its properties with those of previous approaches, and reports its performance in two recent studies of allocative efficiency in manufacturing industries.

det xoo'xoo 1 ) (N n + 1)/(n-2) has the det xo'xo F-distribution with n 2 and N (n 1) degrees of freedom, assuming normality of the underlying distributions. If the statistic is (almost) zero, det xoo'xoo (almost) equals det xo'xo, and the multicollinearity among the columns of xo is (almost) wholly attributable to the columns of xoo. Each step in the proposed third stage is virtually a replica of the second stage, the only difference being that fewer variables are involved. Thus, to show why the proposed third stage is effective in detecting multicollinearity patterns, it suffices to refer to the justification of the second stage, in footnote 3. 4 In practice, calculation of a few, strategically chosen, principil minors should suffice.

IN the empirical literature on market interferences a variety of numerical techniques have been used to analyse competitive equilibria which are not directly observable. Harberger (1963, 1966), and Johnson and Mieskowski (1970), for instance, in their analyses of factor market distortions in the United States economy use a mixture of differential calculus and linearization assumptions to estimate efficiency losses and distributional impacts of particular market interferences. Acceptance seems to be implied in these procedures of the reliability of approximate methods for calculating unobservable equilibria. This issue of reliability forms the subject matter of the present paper.' A recent joint paper (1972) examined the robustness of Harberger's results (1966) when a competitive for the United States economy in the absence of distortionary taxation was calculated using a procedure for the computation of competitive equilibria due to Scarf (1967, 1969, 1973). The results for particular parameterizations suggested that the gain in simplicity of approximate methods may in some circumstances counterbalance the precision of more refined computational devices. These results, however, were obtained for a problem of small dimensionality and limited complexity, and the comparison between Harberger's results and true general solutions was made only on the basis of one summary statistic. In addition, the approximate solution device used by Harberger (1966) does not correspond to conventional notions of either or general analysis. It is thus of some importance that the comparison between general solutions and various forms of analysis be carried further before any conclusions on computational experience are used as a guide in other contexts. These issues are taken up here in the context of a-particular model which allows different forms of analysis to be used and compared to general solutions. The results presented are put forward as evidence on computational experience. This paper considers a general model of the United Kingdom economy used in recent work on an assessment of tax changes in the United Kingdom economy (1973). Using this model the gain2 to the United Kingdom from the abolition of the distortionary features of capital income taxation is calculated by various methods and compared to the general solution. Section II presents a characterization of competitive equilibria for an economy with taxation used in a recent paper by Shoven and Whalley (1973) which underlies the United Kingdom tax model. As no explicit statement of partial equilibrium analysis is to be found in the literature, two alternative characterizations of such procedures which are later applied to the model, are devel-