Industrial Organization Literature Research Articles

WHO BIDS THE MOST FOR MARKET POWER?

In the traditional industrial organization literature market structure is an exogenous variable. However, sometimes market structure is a matter of choice. Firms can choose to operate as monopolists, but only if they pay for this right. When market structure is chosen a natural question is, “What types of firms will pay the price to operate as monopolists, and how will they differ from their competitive counterparts?” This paper develops a model which addresses this question and arrives at results that are novel when compared to the results of the traditional structure‐conduct‐performance paradigm.

Entry Barriers and Economic Welfare

The relationship between economic welfare and the number of firms in quasi-Cournot market is examined. In the first place, we presuppose the existence of strong (first-best) government that can enforce the marginal-cost principle to the firms along with regulating the number of firms. It is shown that there exist excessive number of firms at the free-entry quasiCournot equilibrium vis-a-vis the first-best welfare maximizing number of firms. The thrust of this result essentially survives even if we replace utopian first-best government by secondbest government that leaves the firms to pursue their respective profit maximization freely and engages solely in regulating the number of firms. It can be shown that the entry prevails again in this second-best world. Recent studies in the theoretical industrial organization literature have uncovered several instances which cast serious doubt on the reasonableness of a widespread belief that increasing competition will increase welfare (Stiglitz (1981), p. 184).' It has been shown that there are cases, which are not altogether unreasonable, where social welfare will be increased by strengthening, rather than weakening, the protection of incumbent firms from the threat of potential entry. This is in sharp contrast with the traditional belief.2 What is not known, however, is how robust these pathologies in fact are. We intend to settle this problem by proving two excess entry theorems in the quasi-Cournot (parametric conjectural variations) homogeneous oligopoly model (Seade (1980a, b)). In the first place, we presuppose the existence of strong (first-best) government that could costlessly enforce the marginal-cost pricing principle by firms in an oligopolistic market and regulate entry in pursuit of first-best social welfare optimization. It is shown that the number of firms at the free-entry equilibrium exceeds the first-best welfare optimizing number of firms. The result has already been noted by von Weizsacker (1980a, b) and others in terms of numerical examples. Our Theorem 1 asserts that this phenomenon always holds true for the family of quasi-Cournot models at hand. An implication of this result is that the existence of entry barriers which protect incumbent firms from potential competitors is not necessarily welfare decreasing in sharp contrast with the traditional belief. A problem remains with Theorem 1, however, in that an omnipotent first-best government does not exist in reality, and the first-best ideal is unrealizable in the actual economy. Therefore, even if the intervention by first-best government with view to restricting the number of firms within an industry may be welfare-improving, it does

When does marginal cost pricing lead to Pareto efficiency?

It is well-known that Pareto-efficiency requires, at least in many cases, that prices be set equal to marginal costs. For this reason economists often argue in favor of the marginal cost pricing rule as long as no overly large deficit results. If strongly increasing returns to scale cause a deficit which is considered too large, the marginal cost pricing rule is appropriately modified. To cover the deficit partly or fully, necessary conditions for optimal departures from marginal cost pricing are derived and their implementation is recommended. Normative considerations of this kind lead to what is called Ramsey pricing or pricing ~t la Boiteux. The normative character of Boiteux-Ramsey pricing is the starting point of a large part of the industrial organization literature and the basis for many specific policy recommendations, a typical example being proposals for the structure of telephone rates. Even if one is aware of the fact that quite often first best optimality requires marginal cost pricing and second best optimality requires Boiteux-Ramsey pricing, a further thought leads to the question of what the final outcome will actually be if such pricing rules ate adopted. Leaving distributional aspects aside, although they can be of great relevance, and focusing on efficiency only, o n e

Single-equation, multiple-regression methodology: Is it an appropriate methodology for the estimation of the structure-performance relationship in banking?

Tests of the structure-performance paradigm of the industrial organization literature have been carried out almost exclusively using a single-equation, multiple-regression methodology. The purpose of this paper is to suggest that where the firms being considered are multiple product in nature and may pursue objectives in addition to maximizing the value of the firm, such a methodology may be inappropriate. The results presented in this paper suggest that the absence of a consistently strong, positive, and statistically significant relationship between market concentration and bank profitability may be traced in part to such an inappropriate methodology.

Using Homothetic Composed Error Frontiers to Measure Water Utility Efficiency

Efficient delivery of public utility services is an issue which has been widely studied in the industrial organization and public finance literatures. Recent studies have generally focused on whether public or private water utilities are the most efficient. Other related issues of equal interest are how to measure efficiency and how to set standards of adequate performance. This study advances the literature in several important ways. One technical contribution is the use of a dual output production function in order to explicitly account for the separable production and distribution roles of water systems. The approach adopted here permits a test of the relationship between the two outputs and the finding is that the two roles must be separated in a cost or production function analysis. Most other studies focus only on the production of water, though Feigenbaum and Teeples [7] used a hedonic function to incorporate other characteristics of water provision in the output measure. The production function framework employed here allows a test of whether each output is produced under linearly homogeneous, nonlinearly homogeneous, or homothetic conditions. The closed form structure adopted in this study differs from the approximate form which was the basis for Feigenbaum and Teeples' work. A composed error framework is used to estimate frontiers for the production relationships. This requires a synthesis of the homothetic function [19] and composed error [1] literatures and an extension of these to the dual output case. The application of frontier analysis is a key to conducting efficiency tests. Previous work has not permitted tests of efficiency, but tests of whether the cost structures differ between groups of firms such as rural versus urban or public versus private firms. Efficiency must be assessed by judging either the actual cost or production performance against the corresponding potential for an individual firm. Actual performance relative to potential can then be used to test efficiencies across classes of utility firms. The paper is composed of five sections including the introduction. The next section

Factors affecting property tax reliance: Additional evidence

This paper contributes to the understanding of the variation in property tax reliance. Factors thought to underlie the relative use of property taxes in financing local public services are examined. Because of the importance of external debt and tax limits in determining both tax policies and local government organization, particular attention is paid to local government structure and the impact on tax composition. An index from the industrial organization literature is used to study the effects of local government fragmentation. The empirical results are consistent with expectations and indicate that external tax limits may increase property tax reliance by providing incentives to change the structure of local government.

A theory of conscious parallelism

In this paper the ‘rationality’ of conscious parallelism is studied within the context of a dynamic oligopoly model. The doctrine of conscious parallelism is modelled as the outcome of a signalling game in which the rules of response are specified axiomatically. This will result in a unique solution to the oligopoly problem that is based on firms' need to have consistent expectations that subsequently generate collusive behaviour. This model can then be used to more precisely define the conditions that facilitate or limit oligopolistic coordination, as found in the traditional industrial organization literature.

To Brand or Not to Brand? A Theoretical and Empirical Question

A firm's brand name is a valuable asset that can greatly enhance the demand for its products. This fact is widely recognized in both the economics and marketing literatures. Surprisingly, though, little rigorous analysis has been devoted by either of these literatures to the question of when a firm will identify one of its new products with an established brand name. It is on this question that our research is focused. The economics literature in this general area has focused on issues that are related to but quite distinct from the issue we examine. The traditional industrial organization literature discusses and attempts to capture empirically the relationships among advertising, brand names, and market power. (For some representative work We develop and test a model that predicts (1) whether a multiproduct firm will brand a new product with the established company name and (2) where the new product will be located in relation to the reputation estabished by the firm in other markets. Hotelling's (1929) model of spatial location serves as the basis for our theoretical investigation. However, we extend Hotelling's model to the case of experience goods whose qualities can only be known with certainty after the product is purchased and consumed. The logic that underlies our model is tested on a sample of firms from the U.S. liquor industry. We employ a mixed logit approach to capture the simultaneous decisions that are made by the firms in our sample. * The views expressed in this paper are our own, and not necessarily those of the Bell System. We are grateful to 0. Gur-Arie, the Michigan Liquor Control Commission, Peter Ochshorn, and Srinivasan Balakrishnan for valuable assistance with data collection and programming, and to the National Science Foundation for support through grant IST 8315690. The helpful comments of Roger Klein, Pablo Spiller, and an anonymous referee are also gratefully acknowledged. Any errors are, of course, our responsibility.

Elasticities and rent control revisited

Elasticity has played an important role in 3A many economic analyses. In order to mine 3R the policy implications, especially in the industrial organization literature, many economists have proposed rules involving price elasticities. Among them are Ramsey [1927], Kuhn [1970], Baumol and Bradford [1970], and Blinder [1980]. In a recent issue of this journal, Denton Marks [1983], based on the elasticity formula, demonstrated that rent control beand comes more attractive as the price elasticity of demand rises. Unfortunately, it can be shown that his result is extremely questionR = able and in most cases would be incorrect. Given the continued interest in modeling such policy implications, this paper points out that the impact analysis based on elasticities must be done with a great deal of care. Following Marks' notation, the fact that rent control is preferable can be shown by following derivatives:

Horizontal mergers and collusive behavior

In the Industrial Organization literature, it is generally felt that mergers hurt consumers; not only because of the increased industrial concentration they effect, but also because collusion becomes more likely. In this paper we show that, at least in one important case, this intuition is misguided. If a tacitly collusive agreement enforced by trigger strategies is not initially sustainable, mergers will tend to reduce the chance that it becomes sustainable in the future. This is so because the threat point implicit in the agreement becomes more favorable for outsiders.

Assessing Industry Concentration: The Case of Aluminum

The article examines the use and measurement of certain concepts from the literature of industrial organization when applied to the study of international industries. It is argued that considerable care needs to be taken in defining markets and in evaluating information concerning concentration. This issue is examined in the context of the international bauxite-aluminum industry. The factors leading to overstating and understating levels of concentration are discussed.

Investor Preference for Large Firms: New Evidence on Economies of Size

IN the last several decades, industrial organization literature has been replete with studies demonstrating a systematic relationship between accounting rates of return and firm size in the direction of greater (and more stable) accounting profits for larger firms.' These studies have been subject to criticism for an alarming number of statistical and conceptual problems, the most important of which alleges a serious bias in accounting profit measures as proxies for economic profit.2 The criticisms have resulted in a stalemate in the interpretation of empirical evidence on firm size and performance. Nevertheless, there is continued interest in the area, ostensibly resulting from policy concerns with large firm advantages in both product and capital markets. This paper provides new evidence concerning the industrial organization literature controversy of the connection between firm size and performance. Stock market data are used to examine the relationship between capital market returns, risk, and firm size. The findings suggest that even if there were no advantages of large size in the product market (equal rates of operating profit), large firms would still have a substantial capital-raising advantage over small firms. This advantage is unrelated to the widely recognized floatation cost economies. Even after controlling for nondiversifiable risk using the Capital Asset Pricing Model (CAPM), investors still demonstrate a statistically significant and economically important preference for the securities of larger firms. Investors apparently place a premium on large firms which is not attributable to large firms' observed lower risk as captured by the widely used measure of nondiversifiable risk (beta risk). This finding, a fresh contribution to the

Political Competition and Advertising as a Barrier to Entry

In the industrial organization literature there are few issues as controversial as the relationship between advertising and competition. The protagonists in this debate are those who feel that advertising improves resource allocation and those who argue that advertising is an anticompetitive device that promotes distortions in the efficient allocation of resources. One of the principal arguments that advertising is anticompetitive is based on the view that advertising can be used by established firms to erect barriers to entry into an industry. While this hypothesis has been studied extensively no consensus has been reached on its validity. This paper presents an alternative and novel test of this hypothesis based on a data source previously untapped by researchers in industrial organization. Economic theory predicts that the presence of monopoly power in an economy will distort the efficient allocation of resources. While entry of other firms should eventually eliminate the monopoly (or cartel), the entry process may itself be slowed by the defensive actions of the established firm(s). Firms earning excess profits have an incentive to take actions that deter entry. As Posner [16] has noted, resource distortions generated by efforts to protect monopoly may exceed the welfare loss of an uncontested monopoly. One technique that established firms could possibly use to reduce entry is advertising. Authors such as Bain [2], Comanor and Wilson [5], and Porter [15] have contended that in some industries established firms have a significant advantage in advertising effectiveness over potential entrants. Therefore, they argued that for some types of products, where advertising is an important method of product promotion, advertising can create barriers to entry. Established firms may be able to use advertising to create a barrier to entry because of a fundamental difference between an estalished firm and a potential entrant. Consumers are more familiar with an existing brand than a new brand of a product. This may convey a marketing advantage that enables the existing firms to formulate advertising campaigns that reduce the effectiveness of potential entrants' advertising in stimulating sales. A significant reduction in the ability of entrants to attract customers with advertising can

Efficiency, Market Power, and Profitability in American Manufacturing

A significant, positive association between concentration and profitability is perhaps the most widely reported statistical relationship to be found in the literature of industrial organization [25]. According to conventional oligopoly theory, this reflects the ability of the leading firms in concentrated markets to collude tacitly or explicitly. Some economists, most notably Harold Demsetz [8;9] have challenged the conventional view. Demsetz argues that it is the superior efficiency of large firms which results in both high concentration and high profits. He supports this argument with evidence, replicated by Round [20] and Carter [4], that concentration increases the profits of large firms but not small firms. If the firms in an industry are equally efficient, effective collusion should raise the profits of small and large firms alike. The fact that it doesn't strongly suggests that efficiency, concentration, and profits are interrelated in American manufacturing. There are two bits of evidence which indicate the large-firm efficiency and collusive behavior coexist in much of American manufacturing. The first is the profitability studies of Imel and Helmberger [11], Gale [10], and Shepherd [21]. These studies suggest that, considered together, market share and concentration are both significantly positively related to profitability. If, as argued by McGee [15] and Weiss [25], market share is capturing the effect of lower costs (and/or superior products), concentration must be capturing the effect of collusive behavior. The second is Carter's [4] refinement of Demsetz's methodology. Carter regressed price-cost margins on concentration, capital intensity, and advertising intensity for large and small firms separately. He reports a positive concentration coefficient for large firms which is greater than that for small firms and which is significantly associated with large-firm profitability. The difference in the estimated coefficients may be taken as evidence of large-firm efficiency and the significance of the large-firm coefficient as evidence of collusive behavior. Whether the concentration-profits relationship derives largely from efficiency, as Demsetz believes, or largely from collusive behavior, as the conventional view would have it, is a more difficult empirical question. The answer has important implications for merger policies, remedies pertaining to tacitly collusive oligopolies, and legislative proposals

The Structural Stability of the Concentration-Performance Relationship in Food Manufacturing

The literature of industrial organization is replete with analyses of the relationship between seller concentration and market performance. Most researchers have hypothesized a continuous linear relationship between profitability and concentration and have estimated that relationship accordingly.

A cost function approach to the estimation of minimum efficient scale, returns to scale, and suboptimal capacity: With an application to Canadian manufacturing

This paper presents a new method for utilizing the statistical cost technique to measure minimum efficient scale (MES), returns to scale and suboptimal capacity. An application of the duality theory between cost and homothetic production functions leads to justification for ignoring poor quality or unavailable capital data and the pooling of several years observations to improve the efficiency of the estimates. The methodology is applied to 91 four-digit Canadian manufacturing industries to obtain estimates of MES, returns to scale, and suboptimal capacity. For a subsample of industries, we demonstrate that the cost function estimates of MES and returns to scale are more closely related to engineering estimates than are the ad hoc estimates usually found in the industrial organization literature.

A Model of Duopoly Suggesting a Theory of Entry Barriers

The descriptive and empirical literature in Industrial Organization abounds with reference to economies of scale, product differentiation, and absolute advantages in cost or demand enjoyed by some firms over others. In particular, they feature prominently in Bain's list of barriers to entry in. However, textbook models of oligopoly often abstract away from all these features, and assume a homogeneous product with equal constant costs for all firms in the industry. The real-world complexities are confined to informal remarks, suggesting vague modifications of the formal results. This happens even in such outstanding treatments as those of Feller and Scherer.

The Cost of Capital and the Market Power of Firms

RECENTLY, significant research has been conducted on the functioning of the capital market. One thrust of this research (e.g., Fama, 1970) has demonstrated that the capital market is highly efficient, i.e., the prices of securities at a point in time seem to reflect available information, and security prices seem to adjust quickly over time to new information. Another thrust (e.g., Sharpe, 1964; Lintner, 1965; Mossin, 1966; Jensen, 1972) has been the theoretical development and empirical testing of a specific model, the Capital Asset Pricing Model (CAPM), which precisely defines risk and return, gives an economic justification for diversification, and under specific assumptions draws the equilibrium conditions between risk and return in the capital market. Not withstanding serious econometric difficulties of estimation, these studies picture a highly efficient capital market in which capital funds are allocated based only upon risk and return considerations as determined by a rigorous evaluation of relevant information. The capital market depicted in this recent capital market literature would seem to differ from the capital market depicted in the literature of industrial organization economics. For example, Baumol (1967) and Hall and Weiss (1967) have argued that the major barriers to entry are not in the structure of output markets, but in the capital market. The argument is that to enter and compete effectively in many basic industries, such as automobiles, chemicals, etc., a large sum of capital is necessary, and the capital market will not allocate a large sum to a new entrant. Basically the capital market fails in its allocative function because investment opportunities, albeit opportunities with high profit potential, are lumpy. That is, investment opportunities cannot be financed in small discrete amounts by new entrants, but can only be financed in large amounts by existing firms insuring basic industries marked by large firms with high market shares. Both the capital market literature and the industrial organization literature are valuable reference points for those who would hope to understand the allocation of capital in the economy and the effects it has upon the condition of entry, level of price and level of output found in industrial markets. The general purpose of this paper is to begin a reconciliation of these literatures with respect to their disparate views of the capital market. Since capital market theory relates capital costs to risk, our specific purpose is to determine if the market power of firms, as measured by size and seller concentration, seems to reduce the riskiness of firms and therefore their capital costs. In section II the difference between book profits and capital costs is stated. In section III, with the aid of the Capital Asset Pricing Model, the relationship between capital costs and risk is presented. In section IV the data and sample of firms are described, and in section V the data analysis is presented, which does in fact suggest that the risk and capital costs of powerful firms are lower than for other firms. Conclusions are presented in section VI.

Size of Firm and Size of Plant

Most industrial organization literature treats plant size and firm size as independent variables. It is assumed that public policy can limit firm size, while still enjoying plants of optimal scale. It will be shown that the two variables are linked both empirically and theoretically. The 1972 Census of Manufactures was employed to examine the relationship between size of firm and size of plant. The average number of plants per leading firm was calculated for leading firms in four-digit industries. Estimates of minimum practical levels of concentration show that much lower levels of concentration are possible without having to break up existing plants. However, to actually do so in a growing economy would force future expansions to be of sub-optimal scale. A strong empirical relationship was found between size of firm and size of plant. In all industries the plants of the leading firms are of above-average size, and in all industries those in the smallest firms are of below-average size. In most cases the plants of the leading four firms were larger than those of the next four, and the greater the advantage of the top four firms in size of firm, the greater their advantage in average size of plant. Plant size and firm size appear empirically to be closely related.

Advertising, Concentration, and Price-Cost Margins

The industrial organization literature contains many empirical tests of hypothesized relationships among elements of industry structure and performance. Most of these studies used ordinary least-squares regression to estimate single-equation relationships. This approach is incorrect if the relationship being estimated is part of a simultaneous-equations system. This paper treats three important relationships in industrial organization as a single simultaneous-equations system. The results are consistent with those obtained from single-equation models and suggest that simultaneous-equations bias is not an important factor in the estimation of structure-performance relationships.