Although Imai and Rubenstein are correct that our proof — that the Miller- Modiglianl (M-M) and Sharpe-Lintner-Mossin (S-L-M) capital asset pricing models are mutually consistent — is incomplete, their comments indicate some confusion about the relationships involved in the equilibrium pricing of assets in these models. Further, they seem to imply that Stiglitz's proof in terms of dollar returns is in some sense superior to a proof in terms of rates of return. This is erroneous. Accordingly, we shall further clarify the relationships inherent in the models and correct our presentation of the proof of the invariance of capital costs in the context of the S-L-M model.

Many people from various professions have long been interested in determining what factors influence common stock prices and the rate of return (or cost of equity capital) which investors expect to obtain from an investment in common stock. The response to this interest has been numerous articles, both theoretical and empirical. A problem of many recent empirical studies is the attempt to explain a complex situation with an oversimplified model which does not explain well interactions among the factors that affect share prices. The present paper does not claim to resolve all the issues that exist but it does attempt to broaden the perspective by developing a more comprehensive theoretical model.

“Computation of the Efficient Boundary in the E-S Portfolio Selection Model” William W. Hogan and James M. Warren Volume VII, No. 4, September 1972

Regression analysis is by far the most popular of all the statistical techniques employed in financial research. Almost without exception, the models employed have been additive in both the parameters and the variables (regressors).

Using the mean-variance model, Sharpe [5] and Lintner [4] have derived an equilibrium model for price determination under uncertainty. Jean [2] has tried to generalize this model so that other moments of the distribution will be taken into account. The purpose of this note is to show that unlike the Sharpe-Lintner model, Jean's results make no economic sense.

Unfortunately Professors Arditti and Levy (A-L) in their comment published in this issue of this journal did not realize that the determination of the investor optimum in my paper [1] was simultaneous with respect to the three parameters, the mean and the variance and the third moment of portfolio returns. When the nth moment was introduced, it was assumed that the investor chooses on the basis of all n parameters — the mean, second moment, third moment, etc. — through the nth moment.

The federal deposit insurance system of the United States was instituted as a result of severe financial crises in the United States and was intended to prevent the recurrence of some of the evils of such crises, specifically those resulting from bank deposit losses. Since the Federal Deposit Insurance Corporation (FDIC) was established in 1934, the United States has not experienced banking panics or harmful effects on the economy due to widespread destruction of bank deposits. As a result, there is a fairly widespread feeling that the present deposit insurance system has been a success. If prevention of panics and large-scale deposit destruction are the only criteria for success, this judgment may be justified, although, if the insurance system uses resources, it must be proved that panics and deposit losses would be unacceptably frequent or large in its absence.

“A Note on Geometric Mean Portfolio Selection and the Market Prices of Equities” Robert H. Litzenberger and A. P. Budd Volume VI, No. 5, December 1971

Results of the analysis imply that rates of return on the property-liability portfolios included in this study are less than rates of return on either the investment company portfolios or the Standard and Poor's industrials. However, it appears that these differences are likely due to different investment objectives of the different portfolios.

Gibson offers a proposal for partial reform of the present system of deposit insurance in the United States.

A. The responsibility for insurance should be divided between FDIC and the Federal Reserve. “Normal” bank failures should be handled by FDIC and “depression” failures by the Federal Reserve.

B. Insurance should be extended to all deposits regardless of size and should not be limited to those of $20,000 or less.

Considering Professors Monroe's and Trieschmann's own analysis and methodology, it seems that their conclusion and implications of the results are considerably reduced in scope from those presented in their paper. I would like to discuss this new approach within the broader scope of their work. I will conclude with several points which closely relate to their work but which are not, by their choice, a part of the present paper.

For as long as governments have intervened in economic affairs, private decision makers have searched for signs and indicators of ongoing or impending government actions. A century ago, anxious British bankers relied on changes in the bank rate to provide them with information on the intent of the Bank of England. Contemporary decision makers not only analyze changes in policy-tools monetary, fiscal, or debt policy actions—but they carefully scrutinize the words of ever more talkative economic policy makers. The State Department's legendary crew of “China watchers” could not analyze pronouncements by Chairman Mao any more carefully than money market bankers and security dealers analyze an economic report of the President or a speech by the Chairman of the Board of Governors of the Federal Reserve.

Aggregation of monies must be determined in accord with the theory using the aggregate. This theory must specify aggregation (1) eligibility rules and (2) weights. Four current methods are evaluated on this basis: Fisher's aggregation bases eligibility on significant quantities Vi of monies Mi and attaches quantity weights Vi/Vn (Vn being numeraire). Pesek-Saving aggregation merely substitutes social marginal value weights Vipi/Vnpn. Friedman-Schwartz aggregation has eligibility depend on good subsequent statistical performance and uses weights given by private marginal values (gross of subsidies). Chetty's aggregation has all assets eligible and weighted by cross-elasticities of demand.

I find a great deal in Kaufman's paper with which I am in agreement, and I think that setting out the two approaches, the neutralized variable approach versus the ideal indicator, is very useful. I have nothing to add to what Kaufman has said about the work on neutralized variables, even though I know there are a few people who do not consider this a closed issue.

The purpose of this research is to compare empirically the usefulness of two historical risk, measures to predict the probability of future loss of various magnitudes resulting from investment in different common stocks. The two risk measures are the mean absolute deviation and the standard deviation of the natural logarithms of monthly investment performance relatives.

The “market model” of capital asset pricing theory posits that the oneperiod return on an asset is a linear function of the one-period return on a “market factor” plus the effect of factors that are unique to that asset. The coefficients of the model, estimated using realized returns, can be used for predicting asset returns conditional on market returns, and the slope or “beta” coefficient provides an estimate of the asset's risk. Although the market model has been applied to a wide variety of capital market studies and is now being applied by practitioners for assessing asset risk, very little research has been undertaken to discover the determinants of the beta coefficient.