Stephen F. LeRoy

Atheoretical macroeconometrics: A critique

Abstract The spurious nature of the restrictions used to identify many macroeconometric models has led some researchers to advocate a style of econometric inquiry that is less dependent on prior theoretical restrictions of the sort that were central to the approach of the Cowles Commission. This development, which we call atheoretical macroeconometrics, is summarized and evaluated in the current paper. It is contrasted with an updated version of the Cowles Commission approach. We conclude that while some of the exercises of atheoretical macroeconometrics are valid, those that have attracted the most attention and appear the most innovative-exogeneity testing, impulse response analysis and policy analysis using estimated vector autoregressions - are based on incorrect analysis.

Paradise Lost and Regained: Transportation Innovation, Income, and Residential Location

It was long characteristic of American cities that the rich lived on the edges while the poor lived in the centers. In the 197Os, however, that residential pattern began to change as rich households moved into many city centers, displacing the poor. In this paper, we offer au account both of why the earlier pattern prevailed until the 1970s and of why it is changing now. The key to our explanation is the altered role of the car as a means of commuting to work. When the car first became practical for commuting it gave the rich access to cheaper suburban land than was available to the poor. Accordingly, the rich lived in the suburbs and the poor in the city centers. The relative cost of commuting by car has fallen, however, and now almost everyone can afford it. The poor, therefore, now have access to suburban locations and, thus, the comparative advantage the rich once had in those locations has been eroded. Consequently, the rich are beginning to return to the city centers. Our explanation is based on a simple extension of the model developed by Alonso [l], Muth [15], and others. In the Alonso-Muth model employment is concentrated in a central business district (CBD), and workers commute to that CBD from a residential area surrounding it. Because the cost of commuting increases with the length of the commute, the unit rent of housing must fall with distance from the CBD. Households of different incomes will base their choice of where to live upon two opposing sources of benefit. First, since the rich use more housing than the poor, they benefit by living farther from the CBD where the rent of housing is lower. Second, however, the rich value their time more highly than the poor, so they benefit from living nearer the center of the city, where commuting times are shorter.

Knight on Risk and Uncertainty

It is argued that the received interpretation of Frank Knights_(1921) classic risk-uncertainty distinction-as concerning whether or not agents have subjective probabilities-constitutes a misreading of Knight. On the contrary, Knight shared the modern view that agents can be assumed always to act as if they have subjective probabilities. The authors document their contention that by uncert ainty Knight instead meant situations in which insurance markets collapse because of moral hazard or adverse selection. Knights discussion of market failure, a lthough always informal and in placesinaccurate, was in many respects a remarka ble anticipation of the modern literature. Copyright 1987 by University of Chicago Press.

Risk Aversion and the Dispersion of Asset Prices

Several recent papers have investigated the implications for asset price dispersion of conventional security valuation models. LeRoy and Porter (1981) used the present-value equation to demonstrate that the coefficient of dispersion of stock prices must be lower than that of dividends. At the same time Shiller (1979), employing the expectations hypothesis of the term structure of interest rates, derived upper bounds for the dispersion of long-term interest rates and for the one-period rate of return relative to the dispersion of short-term interest rates. Shiller subsequently extended the analysis to stock prices (1981). All of these papers, as well as those of Singleton (1980) and Huang (1981), conducted empirical tests of the implied variance bounds. For the most part, they found that the implications were flagrantly violated: stock prices and Several recent papers have concluded that asset price volatility is too great to be consistent with conventional security valuation models. We observe that the conventional models are based on risk neutrality or near risk neutrality and demonstrate that, in an exchange setting, risk aversion generally leads to greater dispersion of asset prices than does risk neutrality. It follows that volatility tests based on models assuming risk neutrality will generate downward-biased estimates of implied price variances to the extent that individuals are risk averse.

Equilibrium valuation of illiquid assets

Summary. We develop an equilibrium model of illiquid asset valuation based on search and matching. We propose several measures of illiquidity and show how these measures behave. We also show that the equilibrium amount of search may be less than, equal to or greater than the amount of search that is socially optimal. Finally, we show that excess returns on illiquid assets are fair games if returns are defined to include the appropriate shadow prices.

On the arbitrage pricing theory

The Arbitrage Pricing Theory relates the expected rates of return on a sequence of primitive securities to their factor exposures, suggesting that factor risk is of critical importance in asset pricing. However, we show that if the sequence of primitive returns is replaced by a sequence of returns on portfolios formed from the primitive securities, then the factor subspace is arbitrary. The implication is that the theorems relating expected returns to factor risk require substantial reinterpretation. Our reinterpretation consists of a demonstration that exact and approximate factor pricing do not constitute substantive characterizations of asset pricing. Instead, they are implications of the characterization of the returns space as a Hilbert space (exact factor pricing corresponds to the Riesz representation theorem and approximate factor pricing is a consequence of Bessels inequality).

Bubbles as payoffs at infinity

SummaryWe define rational bubbles to be securities with payoffs occurring in the infinitely distant future and investigate the behavior of bubbles values. We extend our analysis to a setting of uncertainty. In an infinite horizon arbitrage-free model of asset prices, we interpret the money market account as the value of a particular bubble; a similar interpretation holds for other assets related to the state-price deflator and to payoffs on bonds maturing in the distant future. We present three applications of this characterization of bubbles.

On Policy Regimes

Lucas’s (1976) “Econometric Policy Evaluation: A Critique” is widely regarded as the most influential paper in macroeconomics of the 1970s. It is usually read as making the simple point that the Keynesian econometric models of the 1960s were not well founded on microeconomic principles and in particular that Keynesian model builders played fast and loose with agents’ expectations (for example, Mankiw, 1990). On this conservative reading, Lucas was surely correct. Lucas, however, had more in mind than merely criticizing particular Keynesian macroeconomic models. He also argued that the entire mode of analyzing policy interventions employed in Keynesian policy analysis was in conflict with general equilibrium theory. Here Lucas was distinguishing the methodological question of how one analyzes policy using any macroeconomic model from the substantive question of whether Keynesian models are correct. That he considered the former problem the basic one is indicated by the fact that he contrasted the Theory of Economic Policy mode of policy analysis with his recommended mode within the context of an abstract functional representation of a generic macroeconomic model. Particular Keynesian and new classical models were considered only to develop the general argument.

Risk-aversion and the term structure of real interest rates

Abstract The term structure of real interest rates is studied in the context of a consumption-based general-equilibrium model. It is shown that the expectations hypothesis is approximately satisfied for low interest rate volatility. Otherwise the term premia are generally positive.

A monetarist model of inflation

Abstract We consider an overlapping generations model with new money introduced via autocorrelated lump-sum transfers. Hence, the own rate of return on money is fixed, implying that (dynamic) neutrality may fail due to inflation tax effects. We derive a condition, less restrictive than serial independence of monetary growth rates, which is necessary as well as sufficient for neutrality. We determine whether various properties of the joint distribution of monetary growth rates and inflation rates which are weaker than neutrality—for example, positive correlation between the two—can be expected to obtain in general settings.