Knowledge that Transforms

The response of the economy to two major -- although in important respects offsetting -- tax reforms has been much smaller than ardent supply-side revolutionaries expected, thus suggesting that a reassessment of the grounds for revolt is in order. This paper offers such a reassessment by first discussing how the evidence from the tax reforms of 1981 and 1986 reflects on our understanding of the response to taxation -- with particular reference to savings and capital gains realizations. I then reconstruct a 1992 view about how taxes affect behavior. A unifying theme is that the tax system does much more than alter the relative prices of real variables -- it also provides incentives to misreport income, restructure financial claims, time transactions, change the legal form of organization, and so on. For this reason, observed low tax elasticities of real variables may be due to either low elasticities of substitution or the fact that tax policy changes opportunity sets in complex ways. Disentangling these explanations requires an emphasis on the transaction-based nature of the tax system and the administration and enforcement of tax laws.

In the popular press and among policymakers the effect of no-fault divorce laws on divorce rates remains an issue (see e.g., New York Times, 23 July 1991). In my 1986 article in this Review, I show that the adoption of one form of no-fault divorceunilateral divorce-does not lead to a significant increase in divorce rates.' The theoretical model that is consistent with this empirical evidence would imply that the new law does not decrease the costs of divorce. It merely redefines which party has the right to terminate the marriage.

Two recent articles in this Review by H. Elizabeth Peters (1986) and William R. Johnson and Jonathan Skinner (1986) address an anomaly recently observed by labor economists. James P. Smith and Michael P. Ward (1985) have noted that real wage growth explains most of the increase in the female labor supply between 1950 and 1980. However, in the period after 1970, the female-labor-supply growth rate rose, while the real-wage growth rate fell. Robert T. Michael (1985) provides a partial explanation by identifying the effect of divorce on the labor-market decision of married women. In that article, a lagged relationship is observed between increases in the divorce rate and increases in the labor-force participation rate (LFPR) of married women with young children during the post-World War II period. Since the divorce rate rose in the period after 1960, it could explain part of the increase in the growth rate of the LFPR of married women observed after 1970. The articles by Peters and by Johnson and Skinner provide additional insights into this anomaly. They recognize that the introduction of no-fault grounds for divorce after 1970 may have contributed to the increase in the labor supply of married women after that date.' The no-fault grounds for divorce created what has been called The shift in female labor supply also could explain the slower real wage growth. Using 1979 data, Peters concludes that residence in a unilateral-divorce state had a statistically significant positive influence on whether a married woman was in the labor force. Johnson and Skinner use 1972 data to conclude that being in a unilateral-divorce state in that year had a negative effect on the labor supply of married women. Unilateral divorce was introduced in the United States in 1970, when California adopted its no-fault divorce statute. Therefore, Peters's data are probably more useful for testing the impact of unilateral divorce laws on married women's labor-force participation. In her article, the lack of compensation for marriage-specific investment at divorce creates an incentive for married women to increase their more general market capital during marriage by entering the labor force.2 A lack of compensation for marriage-specific investment implies that women are not being compensated for being good housewives and mothers. Given the range of commodities produced by housewives and mothers, marriage-specific investment should not necessarily vary with characteristics of women such as age, race, or education. If the impact of unilateral divorce on the LFPR of married women is due to the lack of compensation for mar*Regents' Professor of Management, Robert 0. Anderson Schools of Management, University of New Mexico, Albuquerque, NM 87131. I thank Michele Blazek, Steve Cauley, Dwight Grant, Ron Johnson, Rod Lievano, Sam Peltzman, John Schatzberg, David Weeks, Nelson Woodard, Douglas Young, and two anonymous referees for their valuable comments. The Anderson Schools of Management provided research funding for this project. IBefore 1970, most states had grounds for divorce that were based on fault, such as adultery or mental cruelty. Between 1970 and 1985, all the American jurisdictions enacted some form of no-fault divorce. The no-fault divorce statutes generally make either irretrievable breakdown or incompatibility the sole ground for divorce or add one of them to the fault grounds. While no-fault divorce is often called unilateral divorce, that is not technically correct, since some states require mutual consent to the no-fault grounds (see Doris J. Freed and Timothy B. Walker, 1990). 2While it was recognized by Peters (p. 443) that women who invest in marriage may face lower wages in the future, based on Gary S. Becker (1981 p. 15), marriage-specific investment has value based on its ahilitv to increase houiehold nrodiution

Efficiency measurement has become a very popular field in applied economics in recent years, and with this interest there has been a large intellectual investment in refining the empirical methods available to researchers in the area.' In this paper, we relate these developments to Harvey Leibenstein's original 1966 insight into the psychological ideas underlying the notion that economic agents may not achieve maximal efficiency in their productive decisions and behavior. Of course, it is always possible to argue that apparent inefficiency only arises from a failure of the observer to realize what it is that is being maximized. However, we shall evade this easy escape route into nonfalsifiable hypothesizing and instead shall take at face value the fact that too many empirical studies have come up with substantial measures of inefficiency for us to ignore its importance for normative economics.

Futures prices were well above spot prices for most commodities during most of the Great Depression; evidently the spectacular declines in agricultural prices caught many people by surprise. Based on the historical correlations between commodity prices and consumer prices, commodity markets anticipated stable consumer prices during the first year of the Great Depression. The dramatic drop in nominal Treasury bill yields, thus, should be read as a drop in ex ante real rates. Later in the Great Depression, markets anticipated deflation, but not as severe as actually occurred.