Knowledge that Transforms

HE main objective of this paper is to suggest an econometric technique for identifying and testing the existence and structure of the spillover effects in the wage determination process. The main notion underlying the socalled spillovers hypothesis is that the wage settlements achieved by the ill' industry's bargaining units reflect not only the traditional labor and product market forces affecting the it' industry but also the wage settlements achieved by the jth industry's bargaining units. The terminology used to describe this process has varied considerably-pattern wage-adjustment, key-bargains, spillovers, imitation and dynamic market-interdependence.1 The research strategy underlying this paper is to decompose the wage changes into two components -a deterministic component that can be explained in terms of exogenous labor and product market forces, industry specific or economy-wide, and a residual component. Does this residual wage component stand for purely random effects or does it represent some spillover effects? One can study the covariance structure of the residual wage vector to make inferences about the dynamic interdependence in wage movements between different industries. The plan of the paper is as follows: In section I a brief review of the literature is presented. In section II the underlying theoretical model is explained and the estimates of the empirical model are presented. In section III some simulation experiments of the spillover structure are presented. In section IV the conclusions and policy implications are discussed. 1. Literature Review

ECONOMISTS have recently begun to assess the quantitative effect of on corporate rates of return. Several studies have analyzed the relation between the profitability of large firms and the amounts of faced by large firms.' But none has yet considered that the profitability of large firms might be related to the amounts of faced by smaller, fringe firms within their respective industries. Most occurs on a small scale and entrepreneurs are likely to estimate the of entering an industry on the basis of the performance of existing small firms. If these small firms fail or are unable to consistently earn a normal rate of return, potential competitors may regard as risky. The greater is this risk, ceteris paribus, the less likely new firms will be to enter. In this sense, the faced by small firms acts as an barrier enabling large firms to earn excess profits without attracting new competitors. Alternatively, can be thought of as the vehicle through which barriers work. For example, where significant scale economies exist, the of entering the industry is likely to be high and, therefore, excess profits of existing large firms will be protected from new competition. Many of the characteristics that traditionally have been termed barriers can be interpreted as factors increasing the amount of faced by potential competitors. Section I of this paper develops a measure of entry risk and section II incorporates it into a simple theoretical model of large-firm profit rates. Section III discusses the construction of variables and the selection of data for empirical estimation of the model, and section IV presents the statistical results. Finally, in section V, a model explaining on the basis of such conventional barriers as economies of scale, advertising, and research and development expenditures is developed and tested. The conclusions and implications of the study are summarized in section VI.

The task is an analysis of the traditional labor supply model using several competing methodologies. The approach is a step-by-step exploration of alternative labor supply estimating equations that attempts to identify the independent (marginal) effect of each particular change in the form of these equations. By systematically exploring what difference each of these changes makes to the parameter estimates, one can isolate which factors strongly affect estimated response parameters. Results from existing research can then be evaluated within a larger context, and future research can concentrate on resolving those methodological issues that do make a difference.

Daniel T. Dick, Marshall H. Medoff, Filtering by Race and Education in the U.S. Manufacturing Sector: Constant-Ratio Elasticity of Substitution Evidence, The Review of Economics and Statistics, Vol. 58, No. 2 (May, 1976), pp. 148-155

A DVERTISING is a subject that has held considerable interest for economists, especially with respect to the relationship between and market structure. It is commonly alleged that is inconsistent with perfect competition, but is to be found where there is monopolistic competition (however, see -Telser, 1964). A correlation between and market structure will be expected under two hypotheses which differ with respect to the direction of causality: (1) can act as a barrier to entry which allows market power to be developed; and, (2) is found under monopolistic competition as a result of optimization in light of the demand curves faced by market participants. This paper considers the second hypothesis; specifically, that as market structure in SMSAs deviates further from perfect competition, savings and loan associations (SLAs) make greater use of to attract deposits. The savings and loan industry provides a good data set since entry is determined by state and federal regulatory bodies. Thus, market structure can be taken as a given variable to which SLAs react rather than as an endogenous variable that may have been influenced by advertising. We can, therefore, specifically test the hypothesis that market structure affects in the absence of a simultaneous relationship in which affects market structure. In addition, the concept of market structure is expanded to consider the effect that branching by SLAs may have on advertising. The hypothesis that market structure is a causal factor in the decision can be derived from a model of the determination of the optimal price and expenditure for a monopolist, given that the demand for the product is a function of as well as price. The ratio of expenditures to sales revenue (referred to as advertising intensity) depends on the price elasticity of demand and the marginal value product of advertising.' This result may be expressed as,