Knowledge that Transforms

Bernard M. S. van Praag, Jan S. Spit, Huib van de Stadt, A Comparison Between the Food Ratio Poverty Line and the Leyden Poverty Line, The Review of Economics and Statistics, Vol. 64, No. 4 (Nov., 1982), pp. 691-694

Ronald R. Braeutigam, Andrew F. Daughety, Mark A. Turnquist, The Estimation of a Hybrid Cost Function for a Railroad Firm, The Review of Economics and Statistics, Vol. 64, No. 3 (Aug., 1982), pp. 394-404

JN the thirty years since Joe Bain's seminal I study (1951), many economists have attempted to determine whether or not there exists a critical level of market concentration at which a discontinuity occurs in the relation between industry concentration and profitability. In this paper we develop and test a more general model, based on organizational costs, which posits the existence of two critical levels of market concentration: one at which an existing cooperative equilibrium will break down as concentration declines, and one at which a cooperative equilibrium will become attainable as concentration increases. We show that the standard model of the critical level of concentration can be regarded as a special case of our more general model, and also that the more general model attains a statistically superior fit. In section II we briefly review some of the previous literature in this area. We develop our model in section III and specify and estimate it in sections IV and V. In sections VI and VII we summarize and interpret our results and present some policy implications of our findings.

The hypothesis of this paper is that crude oil, like any other unfinished commodity, is valued for the products derived from it; the purpose is to offer an empirical explanation for changes in the crude price charged by the members of OPEC. The model results show that the market-clearing prices reported to prevail for petroleum products on the principal petroleum spot market at Rotterdam are the primary determinants of changes in official crude prices. A systematic relationship between offical and spot prices is argued to have prevailed since 1974. An appendix clarifies five types of data required for the model. 13 references, 4 tables.

T HIS study develops an empirical model for the joint determination of the demand and supply for bilateral trade, in which both bilateral quantities and bilateral prices are endogenous. In theory each bilateral demand (supply) equation would contain income (output) variables together with all relevant bilateral prices. Given differences in the commodity composition of trade and given price discrimination, export prices of any country will differ over destination countries. In addition, transport costs and tariffs will drive a wedge between bilateral export and import prices. Thus, a bilateral demand and supply system for aggregate trade among n countries would require consideration of 2n2 price variables, an overwhelming task for empirical research. As a compromise between pure theory and practice, we use a two-tier construction in which total import demand and export supply are determined by aggregate real income (output) and relative aggregate prices, and then bilateral trade flows are determined by theory-based allocations featuring relative bilateral prices. This approach is more general than either specifying only bilateral demand behavior (assuming infinitely elastic supply at given prices) or including the supply side through stringent assumptions such as constant market shares.' For empirical application the model is extended to include proxies for nonprice competition and structural factors, a flowadjustment specification of dynamics, and a trend/cycle decomposition of income (output). The model is estimated from a panel of bilateral trade flows covering the United States, Japan, France, Germany, and United Kingdom for the years 1958-1974. The theoretical model is developed in section II, the empirical specification is presented in section III, and the empirical results are discussed in section IV.

The right conclusion could be that there is a strong connection between the spectral components with specified periods. A possible economic interpretation could identify a strong long-term (in the sense of the specified periods) connection between both series. But that would still not allow us to say anything concerning the size and direction of the time delay between the two series. For such an analysis we would have to, at least, consider the phase angles between the components. Because of the ambivalent meaning of the phase angle (in relation to the direction and number of periods) the final answer could come only from the time domain analysis. However, the fact that only spectral components with longer periods, i.e., smaller frequencies, show coherency suggests something else: In order to influence the balance of trade the changes in the terms of trade should be durable enough. Mathematical formalization of this concept of durability would therefore be helpful. I would propose that further research be done in this respect.

THE ease of substitution between labor and other factors of production is an important determinant of the elasticity of demand for labor and thus of the economic effects of unionism. All else the same, the greater the elasticity of substitution, the greater is the elasticity of derived demand and the greater the displacement of labor for a given union-induced wage increase. In sectors where this elasticity is large, unions are likely to be relatively weak and able to win only slight wage gains (Freeman and Medoff, 1981 and forthcoming) or, if they win large gains, will have to pay a high price in terms of lost jobs. In the sectors where the elasticity is small, unions may be able to extract a substantial wage premium at little cost in terms of employment. Moreover, as a simple general equilibrium model indicates, the elasticity of substitution between labor and other factors in the unionized sector is a key parameter in determining the impact of the IIunion wage effect on the earnings of nonunion workers and on the efficiency of the economy (Johnson and Mieskowski, 1970). Despite the importance of the elasticity of labor demand for an analysis of unionism, little attention has been given to the absolute and magnitude of this elasticity under collective bargaining. While there is some discussion of technological change and the substitution of capital and nonproduction workers for organized production workers in the institutional literature (e.g., Slichter, 1941; Slichter, Healy and Livernash, 1960; Bok and Dunlop, 1970), modern econometric work provides no estimates of the relevant parameters. As a result, Johnson and Mieskowski (1970), Rees (1963), Lewis (1964) and others have been forced to evaluate union effects with guesstimates of the elasticities of concern in union settings. This paper attempts to fill some of the gap in our knowledge by providing estimates for U.S. manufacturing of the constant output elasticity of demand for unionized and nonunionized production workers and of the elasticity of substitution between these workers and other inputs. The analysis concentrates on what we call the relative inelasticity hypothesis, which states that the demand for production workers will be more inelastic in the presence of a union for two reasons: the likelihood that unions have organized and survived in sectors with low elasticities, and the effects of various contract provisions on the ability of management to substitute other factors for production labor. The study is divided into four sections. Section I develops the rationale for the inelasticity hypothesis and describes the nature of the empirical analysis conducted to test its validity. The second section uses a 1972 state by 2-digit Standard Industrial Classification (SIC) industry data file for manufacturing to estimate the elasticity of substitution between production labor and both nonproduction labor and capital in the union and nonunion sectors of U.S. manufacturing industries. Section III provides estimates, based on a 1968-72 sample of manufacturing establishments, of the elasticity of substitution between production and nonproduction labor (the only two inputs for which information is available). The final section briefly summarizes the findings and discusses their implications for understanding the impact of trade unionism on the U.S. economy. To preview the ensuing discussion, our main conclusion is: Substitution between production labor and other inputs is generally lower in union Received for publication March 6, 1978. Revision accepted for publication August 17, 1981. * Both authors are with Harvard University and the National Bureau of Economic Research. Supported by U.S. Department of Labor Grant No. J-9-M-6-0094, National Science Foundation Grant No. APT77-16279, and the National Bureau of Economic Research (under its program of research on labor economics). We are especially grateful to Jane Mather for her invaluable assistance on this project and to Greg Bialecki, Charles Brown, Gary Chamberlain, Kathy Coons, Jon Fay, Martin Van Denburgh, and Lori Wilson for their significant contributions. The study has not been reviewed by the Board of Directors of the National Bureau.

N the last decade several papers have enriched the theory of pricing. Kamien and Schwartz (1971), Gaskins (1971), Baron (1973), and Flaherty (1980)' have made major contributions with dynamic and stochastic models. They predict important new implications for pricing and afford an opportunity for more refined tests of this behavior. We test these new implications and provide additional confirmation of pricing. Insights into how our tests add to the evidence can be seen by briefly examining the theories and the earlier tests. Bain's (1956) static model predicts that monopolists in markets with high barriers to entry will limit to forestall entry, rather than charge a short-run maximizing price which would encourage entry and lead to lower future profits. It also predicts that monopolists facing low barriers will not price, because their opportunity cost of short-run profits forgone to forestall entry is great. When this opportunity cost exceeds the savings from reduced entry, firms prefer the short-run maximizing price. Conversely, when barriers are high, the opportunity cost of forestalling entry is low and firms price. The new models predict intermediate results. By definition, as price exceeds the level at which entry is forestalled, the entry rate or its probability increases above zero. The new theories assume that when price is slightly above the entryforestalling level, a flood of entry is not induced but a gradual increase in its rate or probability occurs. Firms may thus price to regulate the entry rate or probability, not forestall entry. Firms select intermediate prices, setting the marginal entry cost equal to the marginal benefit of a higher price. They reflect Bain's conclusions in the sense that when barriers are low, monopolists may charge high prices, letting entry erode future profits. However, if entry barriers are at intermediate levels, a monopolist's price may be lower. At yet higher barriers optimal prices climb, but often remain above the entry forestalling price.2 These models predict that many industries with high concentration would initially have high profits, but that entry would lead to reduced concentration and profits over time. This accords with Bain's (1970) finding that high concentration tends to erode over time and Brozen's (1971) finding that high profits in initially highly concentrated industries tend to erode over time. Many empirical cross-sectional studies show that measured profit rates are positively correlated with measures of structure-entry barriers and concentration (Weiss, 1974). A related literature has examined entry rates. Harris (1973) and Orr (1974) show entry rates rising as pre-entry profits are higher and falling as barriers are higher. Both the methodologies and the interpretations of the profit rate studies have been challenged. Brozen (1969, 1971) argues that with correct specification, they disintegrate. However, his tests would reject pricing if profits eroded over time as the new theories predict. Demsetz (1973) offers another rebuttal. He notes that firms in an industry may experience efficiencies (i.e., scale economies, superior inputs, or superior foresight). Superior firms will be winners in the market, earning higher profits and expanding their market shares. Industries with such superior firms would then be characterized by high concentration and high profits. Industries without such superior firms should have neither high profit rates nor high concentraReceived for publication April 1, 1981. Revision accepted for publication January 27, 1982. * Cornell University and Oklahoma State University, respectively. We are indebted to Joe Bain, William Greene, N. Kiefer. J. D. Rea, R. Reynolds, and anonymous reviewers for useful comments. Related results appear in J. Shaanan (1979). 1 A related set of extensions have posited other approaches to entry deterrence (Spence, 1977; Salop, 1979; and Kirman and Masson, 1980). 2 This is consistent with Gaskins (1970) but not Gaskins (1971), which is a special case in which fringe firms may be driven out.

D RAMATIC changes in the division of labor compensation between wage and nonwage payments have occurred since World War II. Voluntary employer contributions to private pension and welfare funds have risen from 1.2% of wage and salary compensation in 1947 to 8.3% in 1979, with half the increase taking place since 1969 (see table 1). Major changes in the composition of the fringe benefit package have also taken place during this period. Employer contributions to group health insurance plans have risen from 24% of total fringe benefits in 1947 to 40%o in 1979. If this growth continues, group health insurance will replace pension and profit-sharing plans as the largest fringe benefit expenditure by employers. These trends in the compensation mix have long been noted by researchers and various factors have been offered as explanations, including the favorable tax treatment of fringe benefits, demographic changes such as increased life expectancy, and unionization (Rice, 1966; Ture, 1976; Mabry, 1973; Freeman, 1978). However, most explanations have not been subjected to rigorous empirical testing and the few results that have been published are both questionable and dated. For example, Rice (1966) estimated that changes in the tax rate structure had little influence on the relative increase in fringe benefits from World War II to 1963.' More recent studies (Schiller and Weiss, 1979, 1980) have examined a specific fringe benefit the private pension plan for its impact on turnover and its role as an 'equalizing difference, but these studies have not analyzed the determinants of the compensation mix. This paper discusses the theoretical determinants of the division of labor compensation between fringe benefits and cash income and empirically estimates their importance. It concentrates on the impact of taxation on the compensation mix for several reasons. First, the assumption that rising tax rates encourage individuals to substitute fringe benefits for cash income underlies estimates of the welfare cost of tax preferences (Browning, 1979). Second, proposals to fully tax employee fringe benefits under a comprehensive9' income tax have surfaced in recent years (Nolan, 1977; Clotfelter, 1979). Finally, current federal tax policy is calling for large tax rate reductions to increase supply incentives. Unlike the determinants of the compensation mix that change slowly over time, such as life expectancy, tax rate changes can occur rapidly through legislative action. Reducing tax rates might enlarge the tax base by increasing the fraction of labor income received as taxable cash wages and salaries. Section II outlines the tax treatment of various forms of labor compensation, and in section III other factors thought to influence the choice between wage and nonwage compensation are discussed. Empirical analyses of the compensation mix are presented in section IV. A brief summary and implications of the major findings for the future growth of fringe benefits follow in section V.