U NTIL the mid-1950's, the traditional view on the consumption of housing had been that the elasticity of housing consumption with respect to current income was less than unity.' However, recent literature on the theory of consumption function [15] [4] has pointed out that, in relating income to consumption, the concept of income should be that of permanent income rather than current or measured income, and that the use of measured income imparts a downward bias in estimating the effect of permanent income on consumption. Since housing consumption in particular may also be affected by the long-run prospect of income rather than by a single-year measured income, the permanent income elasticity may be higher than the previous estimates of the measured income elasticity of housing. The change in the concept of income encouraged a number of economists to test the effect of permanent income on housing consumption. Maisel and Winnick [14] tested the permanent income hypothesis of housing consumption by utilizing the I950 Survey of Consumer Expenditures. Contrary to the permanent income hypothesis, however, they reported that housing consumption was no more responsive to changes in permanent income than to changes in measured current income. Subsequently Margaret Reid [19] conducted an extensive cross-sectional study by using housing and income data obtained primarily from the 1950 Housing Census. Differing from Maisel and Winnick, Reid found that the demand for housing was indeed more responsive to changes in permanent income. Moreover, Reid's estimates of the permanent income elasticity of housing were substantially greater than one, and in fact ranged from 1.5 to 2. On the other hand, the author's recent time-series analyses [10] [11] indicated that while housing demand was more responsive to changes in permanent income, the permanent income elasticity of housing was still less than unity. The purpose of this paper is to obtain the cross-sectional estimates of permanent income elasticity on the basis of the 1960-1961-1962 reinterview Surveys of Consumer Finances. This study has three distinct features. First, this study differs from other housing studies in that it uses the instrumental variable method along the lines suggested by Livitan [12]. Livitan has recently shown that, for any year of analysis, the use of a lagged or future measured income as an instrumental variable yields a powerful test of the permanent income hypothesis. Second, since this study utilizes threeyear reinterview data, as a result both one-year and two-year lagged or future incomes can be used as instrumental variables. Due to the lack of data, Livitan [12] could not experiment with a two-year lagged or future income as an instrumental variable in his analysis of total consumption. Both Livitan [13] and Friedman [5], however, agreed that it is highly desirable to use a two-year lagged or future income as an instrumental variable. Finally, this paper develops and uses an extended version of Livitan's instrumental variable method in order to take account of the case in which a lagged or future income may not be a perfect instrumental variable for permanent income. The cross-sectional estimates of permanent income elasticity so estimated are substantially * The author is Professor of Economics at the University of Wisconsin-Milwaukee and a member of the Social Systems Research Institute at the University of Wisconsin, Madison. He is grateful to Melvin Lurie and Keith Phillips for their helpful comments on the initial draft of this paper. The author, however, remains responsible for the views expressed in this paper. This study was originally supported by the National Science Foundation under grant GS-631 and later by the Graduate School of the University of Wisconsin-Milwaukee. The data used in the study originally came from the Surveys of Consumer Finances conducted by the Survey Research Center, University of Michigan, in cooperation with the Board of the Governors of the Federal Reserve System. The author is indebted to the Social Systems Research Institute for making available the data coded on magnetic tapes. 1 For a brief account of the early history of housing studies, see the author's paper [10, pp. 82-83].

H OUSING demand is a major factor in determining national income via private residential capital formation. Moreover, it fluctuates so widely, in many cases independently, in comparison with the demand for other consumer goods that it has been the focus of considerable attention on the part of economists. Yet there are markedly different opinions about the basic relationship of housing demand to changes in income or prices. As early as 1857, Engel made the first classic study of the relationship of family expenditures to income based on budget data; among other things, the percentages of housing to total expenditure were roughly estimated for three different socioeconomic groups. Later, Wright interpreted these estimates to mean that housing expenditure for lodging or rent takes a constant percentage of income at all levels of income. This is known as one of Engel's laws of consumption. However, Schwabe in 1868 presented empirical evidence that the percentage of income spent on rent falls as income rises.1 On the other hand, Marshall in his theoretical analysis, viewed housing as a means of obtaining social distinction as well as shelter, and said that where the condition of society is healthy, and there is no check to general prosperity, there seems always to be an elastic demand for house room, on account of both the real conveniences and the social distinction which it affords. 2 With these views Marshall seems to distinguish himself from his predecessors by stating that the demand for housing is rather elastic with respect to income. In recent years further controversy based on various empirical evidences has arisen. These dissimilar findings have led to radically different implications for important issues in the field of housing such as residential capital formation and the incidence of property taxes. Morton arrived at an income elasticity of demand of around 0.6 for the value of a house purchased, using cross-section data. Combining his estimate with the view that the property tax rate is constant over the different levels of property value, Morton concluded that the property tax was regressive.3 Winnick similarly derived an income elasticity of demand for the value of a house purchased of about 0.5, and noted a downward trend in per capita residential housing stock since 1900 (observed by Grebler, Blank, and Winnick) when real income was rising. Combining these with the assumption of a low price elasticity for housing, Winnick concluded that there had been a downward shift in consumer preferences for residential capital formation since around 1900. Recently Muth presented an intensive study 6 of stock demand elasticities for housing in which he rejected Morton's conclusion, stating that the income elasticity of demand for housing is about unity; i.e., more elastic than what Morton estimated. The price elasticity for housing stock estimated by Muth also exceeded unity, differing substantially from Winnick's contention. In view of high elasticities with respect to both income and price Muth cast doubt on

OST research in housing demand has M focused on the explanation of aggregate behavior. Studies have tended to concentrate on statistical analysis, emphasizing construction of econometric models of aggregate housing demand. Research by Klein, Mattila, and Muth, among others, has provided considerable insight into the determinants of aggregate housing demand.1 However, it is being increasingly recognized that many relevant differences among individual households -for example, age of head, marital status, occupation of head, community type -ought to be considered in conjunction with the conventional economic variables, in studying the demand for housing. Moreover, a model built upon individual decision units has been developed in recent years to simulate the socio-economic system of the United States.2 This type of model also requires extensive knowledge of the behavior of individual decision-making units on housing expenditures. An analysis of crosssection data can provide this type of information about the individual demand for housing, but only fragmentary work has been done along these lines to date. In this paper, a cross-section analysis using economic and demographic characteristics is carried out to investigate the determinants of several aspects of the demand for housing. It attempts to answer the following questions: What are the factors affecting the decision to buy a house and the amount spent on purchase of a house? Given the purchase of a house, what are the factors influencing the decision to incur mortgage debt and the amount of mortgage debt incurred? In presenting the results, comparisons are made with empirical findings of earlier studies. This study differs mainly from other research in this area in that it treats house purchase and new mortgage incurment as jointly determined, and that investigation is undertaken of the marginal and conditional distributions of these aspects of behavior.

In a recent paper published in this Review, V. K. Chettv developed an interesting method of estimating the substitution parameters between liquid assets and found that commercial bank time and savings deposits, mutual savings bank deposits, and savings and loan association shares (T,MS, and SL hereafter) rank in the descending order of imnportance as near-moneys. While Chetty suggested inclusion of these nearmoneys in the definition of money, he argued that since T are closer substitutes for money than are SL, Milton Friedman's definition of money including T but not SL is also justified. Indeed, Friedman and Anna Schwartz (1970, p. 188) subsequently claimed that his definition of money is confirmed by Chetty's results. Chetty's finding, however, depends critically on the incorporation in his analysis of pre-1951 observations which do not reflect an important institutional change that occured in 1951 affecting the substitution parameters. Omitting such observations but using the same method, this paper will show that SL are closer substitutes for money than are T. Although this result does not affect the basic methodological contribution made by Chetty, it reverses his empirical finding. While I will not argue for including SL in the definition of money for the reasons discussed later, the present finding has an important implication for rejecting Friedman's concept of money now widely accepted in monetary analyses. Chetty employed 1945-66 annual timeseries observations for his analysis, but the analvsis should have been confined to a period after 1950. In the fall of 1950, the insurance provision of the Federal Savings and Loan Insurance Corporation was made more liberal than before in the event of default of an insured savings and loan association. As noted by Friedman and Schwartz (1963, p. 669), this provision in fact became identical to that governing the Federal Deposit Insurance Corporation. It is, therefore, quite reasonable to expect that the nearness of SL to money has increased since 1951. Indeed, there is evidence (see G. K. Kardouche, Lee (1966)) showing that the substitutability of SL for money has shifted due to the institutional change cited here. In addition, the post-1950 data pertain to the period of revival of monetary policy since the 1951 Accord and the analysis of substitution effects in a policy context should be directed to such data. With the 1951-66 data, Chetty's estimating equations are recomputed. Since there was evidence of autocorrelation among the least squares residuals, the equations are reestimated to remove autocorrelation by using Phoebus DhrvNmes' method with the following results.: 1 (1) log T = .0180 38.81 log (.061) (2.94) 1 + rT