I. Introduction, 309. — II. The ambiguity in Hicks's concept in his partial analysis, 311.

HIS article is concerned with the deterT minants of consumer saving in the short run. Preoccupation with consumer saving springs from a desire to predict year-to-year changes in the over-all level of economic activity and to show the effect of consumers' actions on these year-to-year changes. As its central theme, the article focuses on and demonstrates the existence of a strong relationship between tangible consumer investment and consumer saving. Potentially, the use of this relationship, which has been hitherto overlooked, may enable economists to achieve better predictions of consumer saving. The analysis is restricted to the consumer sector only. (Thus, this article does not deal with the essentially entrepreneurial investments of consumers'whose occupation classification is self-employed businessman or farm operator.) Supporting data are introduced from the I94952 Surveys of Consumer Finances. Deflationary saving, or in net claims, defined as disposable money less total expenditures for goods and services, has been taken as our dependent variable. This variant of saving does not count as saving purchases of tangible assets such as houses or cars; only changes in claims to assets are counted as saving. We chose this variable because we are in-' terested in short-run fluctuations in the level of economic activity. Deflationary saving measures the extent to which the actions of economic units increase (the case of negative or decrease (the case of positive the circuit flow of The usual variants of saving which define saving as change in net do not measure directly additions to or subtractions from the flow and thus are less suitable for our purposes.' If the magnitudes and movements of the net worth variants of saving were highly correlated with deflationary saving, they could be used as proxies for it, but this is not the case.2 The question may legitimately be raised as to why we do not consider the use of Milton Friedman's3 (or saving) concept. The answer is that the permanent consumption concept is designed to eliminate just those transitory elements of consumption (or in which we are chiefly interested, namely those which account for most of the year-to-year variation in saving. For a similar reason we have employed measured income rather than income.

T HIS paper is concerned with factors associated with the purchases of used automobiles. In addition to reporting the results of an empirical investigation, the paper demonstrates a particularly useful method for analyzing survey data when information from more than one independent cross-sectional survey is utilized. The approach consists first of testing the relationship between used car purchases and socioeconomic variables, and second, of relating the residual variability in used car purchases to attitudinal and expectational variables. In the first stage, data from the I955, I956, and I957 Surveys of Consumer Finances (combined) are used, while the second stage makes use of the reinterview part of the I953 Survey.' Table i provides general background information about used car purchases by car owners, and about the frequency of multiple car ownership among used car purchasers (namely, of spending units who bought their used car as a second car). Almost a quarter of all car owners buy a used car in any one year, and about one fifth of the purchasers buy their used car as a second car. In examining the variations of these proportions among population sub-groups (with proper account given to sampling variations), an important pattern emerges. It shows a decline in the frequency of used car purchases and an increase in the frequency of multiple ownership among purchasers, with a rise in the socio-economic status of the group. This pattern is maintained within occupation, education, liquid asset (bank accounts and government bonds), and income groups. In addition, it appears that home owners are more frequent used car buyers (and multiple owners) than non-owners, and the proportion of multiple car owners among used car buyers rises with the increase in the number of income receivers in the spending unit. Individually, all these and other independent variables manifest a relationship to used car purchases. However, since they are not independent of each other, a multivariate analysis is required before significant relationships can be established.

T HE much-used term has never been clearly defined in the literature. It is generally agreed that it cannot mean an increase in cash holdings, for all cash (money) is held by somebody at all times. A common explanation is that hoarding is a decrease in velocity. As a matter of fact, a decrease in velocity seemingly explains most of the phenomena that are ordinarily thought to result from hoarding. There is a tendency among some economists, however, to assert that velocity is a meaningless ex post coefficient. On first thought, it seems possible to argue with those who hold this point of view. Consider the stock of coin and currency. It is held at all times, but it changes hands from time to time. If you count the number of times each unit changes hands, you have the operational basis for a seemingly satisfactory definition. When you consider demand deposits, however, difficulties arise because, in transfer of ownership, you cannot identify the unit transferred. This problem arises from the nature of a demand deposit dollar. It is not a thing; it is only an idea, accepted by practically everybody in a developed economy and dignified by an institution, the banking system, that records the idea on paper. But its essence is as evanescent as that of a poltergeist. After all, an accounting system could be devised to keep track of poltergeists; and, if people took this system seriously enough, they would talk of a stock of poltergeists, of its turnover,' and of increases and decreases in it. In other words, velocity is not a good concept to use in the definition of another term, for it does not itself possess a good operational definition. It is fairly clear, however, that the monetary concept is associated in the minds of the profession, on the one hand, with the notion of holding money and, on the other, with the notion of the flow of money through the economy. It it also associated with certain observable real phenomena, all of the sort described as deflationary. A satisfactory definition should rationalize this somewhat cloudy collection of ideas and, at the same time, demonstrate why each individual idea has some intellectual nexus, not necessarily completely sound, with it.

T HE intent of this study is to ascertain that linear combination of financial characteristics which large industrial corporations with low ratios of earnings per share to common stock price from those with high ratios.' The linear transformation of several variables into a single variate (z) permits the categorization of firms on the basis of whether the z values are greater or less than a predetermined mean value. The proposition which underlies the division of firms into high and low ratio groups is that, if allowance is made for the historical nature of earnings and for market imperfections, the earnings-to-stock-price ratio reflects the composite market valuation of such factors as financial risk and dividend policy.2 With this in mind, it is interesting to inquire whether certain basic measures can be used to differentiate successfully between the two classes. Discriminating variables, that is, financial characteristics chosen to reflect individual elements of risk and other factors which affect e/p ratios, include the ratio of dividends to earnings, the ratio of current assets to current liabilities, the rate of return on additional investment, the relative change in sales, and the comparative stability of the common stock price. The construction of the problem is designed to parallel the thinking of investors and/or financial executives. Despite the apparent continuity of risk gradations, firms tend to be grouped on the basis of low, medium, and high risk; e/p ratios (or their reciprocals) are often used as the initial stratification variable; and attention is customarily directed to the extreme classes. The evaluation of common stock and other corporate securities tends in ddition -to be carried out in terms of certain conventional ratios. Discriminant analysis, as employed here, is not intended as a substitute for multiple regression analysis. Given little knowledge as to the appropriate form and complexity of the general regression function, this approach nonetheless serves as a useful device for observing directly those characteristics which distinguish lowand high-risk categories. The relevant information obtained is large relative to the sample size. The derived relationships may in turn facilitate the formulation of multiple regression functions. The potential utility of the analysis which follows is at least threefold. First, procedures for the selection of underand overvalued stocks may be improved by the introduction of discriminant analysis. If the discriminating index suggests that a firm clearly belongs to one group while its e/p ratio indicates otherwise, some reason exists for believing the company's stock to be underor overpriced. Second, partial conclusions may be drawn as to the influence of changing stock market levels upon the importance of different factors which condition e/p ratios. Distributions of e/p ratios, exhibited in Table i for samples of large industrial firms, reflect (for example) a greater central tendency for the I952-55 period than for I948-5I. If the discriminating function based upon the I952-55 data fails to predict well for the earlier period, there is some presumption that weights of the individual variables have shifted. The index characteristic of discriminant analysis affords certain advantages in this respect. The discriminant function is applicable whatever the level of stock prices, provided the * The helpful assistance of W. W. Cooper and Carl Hensley, Carnegie Institute of Technology, and Charles Christenson, Harvard University, is acknowledged. The computations were performed in the computer center at Carnegie Institute of Technology. I The method employed is described in G. Tintner, Econometrics (New York, 1952), 96-I02. See also M. G. Kendall, The Advanced Theory of Statistics (London, I946), Vol. ii, 34I-48. By best discriminates is meant that the chance of erroneous classification is approximately minimal. 2Earnings-to-stock-price ratios are hereafter referred to as e/p ratios.

D URING the past decade, a considerable amount of information has become available about share ownership among various population groups. Following the first collection of data on this subject in the I947 Survey of Consumer Finances,' additional information was gathered in subsequent Surveys, in the I952 study of the Brookings Institution,2 and in the I956 Census of Shareowners conducted by the New York Stock Exchange.3 These studies were primarily concerned with the frequency of stock ownership. They showed that only one tenth of the nation's families own publicly-traded common and preferred stocks, and demonstrated considerable variability in ownership among different population groups. Some of the studies, notably the recent Surveys of Consumer Finances, also collected data about the approximate size of stockholdings, and these shed light on the concentration of stocks among different groups of stockholders. Finally, a recently completed reinterview study, conducted by the Survey Research Center, contained attitudinal information related to stock ownership. In the present paper, data from the Surveys of Consumer Finances will be used to investigate the economic and demographic factors associated with the frequency of stock ownership and with the amount of stocks owned. Subsequently, data from the Survey Research Center reinterview study will be used to study the association of stock ownership with some more dynamic attitudinal and expectational variables. In selecting the independent variables to be used in the investigation, primary attention was given to factors which are presumed to explain stock ownership by an individual. These are: his knowledge of the stock market and his familiarity with stocks as an investment outlet; his income and wealth, which reflect his financial ability to enter the stock market; his general personality, which may indicate his attitude toward entering the stock market; and his price expectations, as well as other economic and financial expectations. On the operational level, many of these considerations cannot be tested directly. They can only be approximated by variables about which information is easily gathered and is readily available. It is to be remembered, how, ever, that the surveys used here were not uniquely designed to study stock ownership. Therefore, they do not contain all of the variables which should, ideally, be used in the present analysis. The selection of variables was thus constrained by the availability of data. Following are the operational variables which were selected for the present analysis: A. Socio-Economic Variables: (i) income, (2 ) education, (3) liquid asset holdings, (4) age, (5) occupation, (6) region, and (7) size of place of residence. B. Attitudinal and Expectational Variables: ( i ) price expectations, (2 ) job preference, taken as an indication of security mindedness, (3) investment preference, (4) personal financial expectations, and (5) general economic expectations. Of the socio-economic variables, education, age, occupation, liquid assets, and income were found significantly related to the frequency of stock ownership. Only income and liquid assets, however, seem to be directly related to the amount of stocks owned. Education and age are perhaps associated with it indirectly, through their effect on income and liquid assets. With regard to the attitudinal variables, * The author gratefully acknowledges the help of Professors Katona, Morgan, Lansing, and Mueller of the Survey Research Center, University of Michigan, in the preparation of this study. 'The Survey of Consumer Finances is conducted annually by the Survey Research Center, University of Michigan, for the Board of Governors of the Federal Reserve System. The results of the survey are published in the Federal Reserve Bulletin. I am grateful to the Board of Governors for permission to use these data in this study. 2Lewis H. Kimmel, Share Ownership in the United States (Brookings Institution, Washington, I952). The study was sponsored by the New York Stock Exchange. Who Owns American Business?, I956 Census of Shareowners, prepared by the Department of Public Relations and Market Development, New York Stock Exchange.