Knowledge that Transforms

The very poor performance of the stock market has been one of the major economic puzzles of the 1970's. The value of common stock has fallen significantly in relation to the price of final goods, the replacement value of the capital stock, and the value of pretax equity earnings. This fall in real share prices has raised the cost of capital to firms and has thereby reduced the incentive to invest in plant and equipment. Although no single factor is likely to have been responsible for this unusual performance of share prices, I believe that the sharp rise in during the past fifteen years has been one of the significant causes. Of course, should have no effect on real share values in an economy in which there are no taxes or other imperfections and in which portfolio investors correctly distinguish real and nominal magnitudes.' But the U.S. economy does have substantial taxes that are assessed on the basis of nominal (rather than real) capital income.2 An increase in the rate of raises the effective tax rate on equity earnings relative to the tax rate on other types of investment income. Individuals and financial institutions will therefore hold the existing stock of equity capital only at a lower real price. In Inflation and the Stock Market I presented a very simple model designed to capture the essential feature of this tax nonneutrality and its impact on share prices. The primary purpose of the analysis was to show how the overstatement of taxable profits caused by (because of historic cost depreciation and inventory accounting rules) could lead to lower real share prices even though reduced the real net-of-tax return on debt. This explanation stands in sharp contrast to the conventional view that lowers share prices because the (nominal) yield on debt rises. A second purpose of the analysis was to show how corporate stock could be a good hedge against inflation as long as the rate remained constant while being adversely affected by increases in the expected rate of inflation. And, finally, I wanted to indicate the importance of recognizing separately the roles of tax-exempt institutional investors and taxable individual investors. The analysis was definitely not intended to prove that must cause share prices to decline with existing U.S. tax rules. The model that I used is clearly far too simple in several ways to do more than illustrate a possible line of influence. The model assumes, among other things, that there are no retained earnings, no corporate debt finance, and no individual investment opportunities other than corporate stocks and government bonds.3 But the simplified model has the virtue of tractability and clarity that would be lost by adopting a more complex specification. In their comment, Irwin Friend and Joel Hasbrouck have presented a slightly different model of asset demand in an economy with taxes and inflation. I think that this alternative model is a useful complement to my own analysis. Moreover, as I shall explain in this reply, their model has the same implications as mine about the effect of on share prices. *President, National Bureau of Economic Research, and professor of economics, Harvard University. The study discussed in this note is part of the NBER Study of Capital Formation. The views expressed here are my own and not those of the NBER. 'Franco Modigliani and Richard Cohn have argued that many investors do not correctly evaluate either real profits or the relevant discount rate because they do not distinguish between real and nominal interest rates. 2This includes the use of historic cost depreciation, artificial inventory profits based on FIFO accounting, nominal interest income and expenses, and nominal capital gains. 3These features are included in a later model (see my 1980b article) designed to explain more of the complexities of the tax-inflation interaction.

The automotive emissions program is far from perfect. It is too inflexible and too costly; it is long on command-and-control and short on incentives. But, as this paper indicates, the program is achieving substantial reductions in pollutant emissions. The real question, then, is not whether the program is achieving emissions reductions but whether the benefits of those reductions are worth the costs and how the program might be redesigned so as to achieve those benefits at lower costs. 4 references.

Both with respect to inflation and productivity, macroeconomic performance in the last decade or so has been disappointing. Promises of policymakers to improve things have been frequently thwarted, not only in terms of inflation and growth, but also in terms of achieving targets for such summary measures of policy as the budget deficit or the money supply. Moreover, critics have charged that policy failures are to blame for many of our economic difficulties. The net effect has been to raise the practical question of whether the political process as now constituted can produce good economic policies. The last decade or so has also been marked, more than coincidentally, by substantial professional criticism of conventional neoKeynesian macro models and the theory of policy associated with them. One of the major punch lines of this research is that activist macroeconomic policy is misguided-indeed, at best it is seen as ineffective and at worst highly counterproductive. As a consequence, the so-called new classical economists argue that policymakers should avoid activism or and instead be guided by simple rules. The issue of rules vs. discretion, has, of course, been a source of long-standing professional debate and, while the new classical economists have provided some additional intellectual ammunition for rules, this hardly serves to explain the apparent popular resurgence in advocacy of rules. Recent interest in rules is reflected in the emphasis on monetary targeting, in proposed legislation-even constitutional amendments-for controlling the federal budget, and in attempts to untarnish the gold standard. This paper reviews the current state of the rules vs. discretion debate. I. The Nature of Policy: Some Preliminaries

My assignment here is to assess the implications of recent theoretical work in industrial economics for antitrust in the United States. I don't have space enough to present a comprehensive survey of that work, nor even to catalog all recent developments with apparent antitrust implications. I attempt instead to describe the general character of those implications, limiting myself to a few illustrative specifics. Industrial economics affects antitrust policy in three different ways. First, it is used in positive analysis aimed at determining whether or not current law has been violated in specific cases and at assessing damages due injured parties. Second, it should be used in evaluating the desirability of relief that might be imposed in particular cases in order to alter structure or conduct if a violation is found. Finally, the tools and results of industrial economics are important inputs in the formulation of general rules of law. I argue here that the new industrial economics can contribute a lot to the positive analysis of individual cases, but it has much less to say about the desirability of particular relief or of general rules of law. A final section briefly examines some implications of this situation.

The Arrow-Debreu analysis of decentralized resource allocation in a Walrasian economy assumes constant or decreasing returns to scale in production. Recently, several authors have extended this analysis to economies with a public monopoly, that is, a firm with increasing returns to scale. In this literature, the salient feature is the characterization of increasing returns to scale technologies as nonconvex production sets, so that under this definition both single and multiproduct firms may exhibit increasing returns. Here, our intended model is an economy with a competitive sector consisting of households and firms with convex technologies, and a public sector consisting of firms with nonconvex technologies. A special case is a single multiproduct firm which produces products for regulated markets (with a nonconvex technology) and produces products for unregulated markets (with a convex technology), for example, ATT firms with constant or decreasing returns are maximizing profits; the public monopoly is pricing at marginal cost, where potential losses are covered by the lump sum taxes; and all markets clear. An average cost-pricing equilibrium is a family of consumption plans, production plans and prices such that households are maximizing utility subject to their budget constraint; firms with constant or decreasing returns are maximizing profits; the public monopoly is pricing at average cost, that is, breaking even or making zero profits; and all markets clear. Unfortunately, all of the extant proofs of existence of a MCP or an ACP equilibrium are somewhat technical in nature and lack the transparency of counting equations and unknowns which many economists accept as an intuitive, if not formally correct, proof of existence. In view of this, one of the purposes of this paper is to demonstrate the existence of a MCP and an A CP equilibrium in a simple economy with increasing returns, where the equilibrium notions are characterized by systems of behavioral equations and market-clearing conditions. We give both an intuitive proof of existence by counting equations and unknowns, and a formal argument that these systems of equations have a solution by use of a simple fixed-point argument. In addition, we review several of the standard partial equilibrium prescriptions for the regulation of a public monopoly and show that in a general equilibrium model they can be interpreted as MCP or A CP equilibria.

In our 1980 paper we tested the joint hypothesis that prices are determined by the mean-variance (MV) capital asset pricing model (CAPM) and that beliefs are stationary. By focusing on the Invariance Law of Prices we avoided the questionable practice of estimating ex ante expectations with ex post returns. Moreover, we circumvented the need to identify the true market portfolio and hence avoided the ambiguity, noted by Richard Roll (1977), in the traditional security market line (SML) tests of the same joint hypothesis. However, Stuart Turnbull and Ralph Winter (T-W) and Richard Sweeney point to a further inconsistency in the joint hypothesis, that they believe can be removed by relaxing the stationarity assumption. This new concern is fundamental in that it applies to all empirical tests which assume stationarity of the return distribution, whether they are simply tests of the CAPM or tests employing the CAPM. The concern would apply a fortiori to tests that assume stationary betas as well. Both comments also suggest that the ad hoc addition of a random error term to our Invariance Law equation and the subsequent statistical tests of it are unnecessary. We first address these two criticisms and then address some further criticisms raised separately by T-W and Sweeney.