Recent financial literature has discussed how a creditor should determine its investigation and extension policy. Mehta [8,9] has developed a sequential process for credit extension, and others [1,2,4,7,10,12,14] have used credit-scoring functions to develop decisions rules. Instead of discussing the use of a particular system or the development of a new system, this paper shifts the focus to selection of the best of alternative systems. Different creditors face different profit-loss ratios on loans, business volume, and prior probabilities of good and bad customers. Furthermore, since the alternative systems have different initial costs, effectiveness, and investigation costs per application, no one system is optimal for all creditors. Finally, any credit-scoring alternative declines in effectiveness over time. Measurement of the overall effectiveness of a system requires that the optimal time between updating the system be known.

This brief paper will show that (a) a theoretical equilibrium state of the world exists in the absence of capital controls and trade barriers when prices for the same goods in different markets are equal, after translation at the spot exchange rate; (b) differences in rates of aggregate price change in different markets eventually cause offsetting exchange rate changes which restore condition (a); (c) returns on equivalent securities denominated in different currencies but covered in the forward market are almost instantaneously equalized; (d) the market's expected rate of change of the exchange rate equals, to a close approximation, the control-free interest rate differential between the two currencies; (e) in the absence of predictable exchange market intervention by central banks, the interest rate differential is the best possible forecaster of the future spot rate; and (f) the forward rate also provides the best forecast of the future spot rate. A final corollary identifies a relationship between inflation rates and international interest rate differentials.

A two period model of bank behavior is developed where the bank chooses, at the beginning of period one, optimal levels of loans, securities and debt. Given perfectly illiquid loans, perfectly liquid securities investments, and alternative assumptions about the liquidity and cost of debt, the bank chooses the endogenous variables as follows: (1) If the marginal cost of borrowing everywhere exceeds the marginal return to securities then: a) near the trough of the loan demand cycle marginal returns on securities and loans are equated and debt is zero; b) in the center of the loan demand cycle all deposits are committed to loans as their marginal return exceeds the marginal return on securities but is less than the marginal cost of debt; c) near the peak of the business cycle securities are zero and non-deposit liabilities are acquired until their marginal cost is equal to the marginal return on loans. (2) If the marginal cost of borrowing is, in some range, less than the marginal return to securities, then instead of possibility b) above, the bank will borrow to finance both loans and securities as a precaution against the prospect of future borrowing at a higher cost.

In this paper, a short-run partial-adjustment model of the demand for credit union shares was specified and estimated with time series data. The estimated results were used to derive long-run, equilibrium demand coefficients and elasticities. The main conclusions are that credit union shares are substitutes for deposits at savings and loan associations, time and savings deposits at commercial banks, and marketable bonds. Moreover, the implications of the statistical results are that credit union and savings and loan shares are more closely related to more liquid assets than to long-term assets. While real income was employed as a constraint variable, it was employed as a maintained hypothesis since the use of a wealth constraint led to perverse results. Also, some evidence was presented that the elasticities of the demand function for credit union shares are different from those of an aggregate savings deposits function. Thus, it is likely that an aggregate demand function will contain aggregation bias.

The U. S. National Bank was declared bankrupt in October of 1973. Just a few days short of a year later the Franklin National Bank failed. These were the largest bank failures ever experienced in the United States and they occurred against a backdrop of extensive bank regulatory operations first to prevent them and then to reduce the effect of these failures upon bank depositors and the banking public. The objective of this study was to determine the effects of these two large bank failures upon the structural relationship of risk and return of equity investments in other large commercial banks. Using historical hindsight, it can be seen that these large bank failures were isolated and noncumulative, but it should be remembered that the investors in bank equities at that time did not know that these failures would not precipitate other bank failures. This worry or apprehension.should have.been reflected in the equity prices of other large bank stocks. Did these failures substantially affect the perceived risk of failure in other large commercial banks? Did these failures result in a substantial increase in the required rate of return on bank equity? Under the efficient markets hypothesis it is assumed that all current prices reflect all knowledge and thus equity prices should adjust quickly to the presence of new information contained in a bank failure. Therefore, in order to determine the influence of these banking failures upon the structural relationships of risk and return of bank equities, the market model was used to relate the weekly holding period returns on 19 large commercial bank common shares to the weekly holding period returns on the S & P 500 index over the past four and one half years.

This study found that there was a lengthy period of structural stability when the realized returns on large bank stocks were compared with the realized returns on the market portfolio. Further, the study found that the U. S. National failure had no structural effect upon the relationships of risk cognizance and required return in that there were no resulting significiant changes in the parameters of the market model before and after the failure. On the other hand, the failure of the Franklin National had a significant structural effect upon risk cognizance. The perceived level of unsystematic risk significantly increased above that of a base period established before the Franklin failure. There was no structural change found in either the intercept term or the measure of systematic risk after the Franklin demise, only a change in unsystematic risk. The most significant change occurred just after the closing of the Franklin and subsided completely during the first and second quarter of 1975.

The rising cost of funds internationally is forcing multinational corporations to pay more attention to effective cash management on a global basis. However, the available literature is preoccupied with cash management in only one currency. This is a serious oversight given the heavy involvement of U.S. firms overseas. In 1970, for example, the ratio of foreign source earnings plus income from abroad (royalties, fees, service charges) to total U.S. corporate after-tax profits was over 25 percent [14]. If export and import activities were included, this statistic would be more impressive yet.

In application of portfolio selection algorithms [3,4] and in tests of the effectiveness of these approaches [1,2], it is sometimes useful to know, a priori, the size of the set of possible portfolios that may be encountered. Given a set of linear restrictions such as that worked by Frankfurter, Phillips, and Seagle [1,2], the set of possible portfolios is finite. This note presents a simple algorithm for determining the size of this set. Only two inputs are required:

1. The size of the universe of securities under study, and

2. A functional relationship which acts as a constraint on the weights.

The following is a heuristic algorithm without a rigorous, generalized proof.

Concern for the social objective of an efficient financial system led several researchers durnng the 1960's to study economies of scale in financial intermediation. These studies were used to draw inferences regarding an optimal structure of financial institutions. Today, there is added reason for the concern for the social cost of providing financial intermediary services. Since 1966, deposit rate controls have been extended to savings and loan associations as well as other thrift institutions whereas the Banking Act of 1933 had previously extended these controls to commercial banks.

In a past issue of this journal, Carl J. Norstrøm [3] developed a sufficiency condition for the existence of a unique nonnegative internal rate of return for an investment project. His method is appealing because it involves a simple computation using cumulative undiscounted cash flows and because it applies to a wide range of practical situations.

In a previous paper in the Journal of the American Statistical Association, Fielitz and Smith [16] present evidence in favor of asymmetric members of the stable family of distributions for describing stock price changes. Support for this contention is derived from an empirical examination of the daily closing prices of 200 stocks from the New York Stock Exchange from December 23, 1963, to November 29, 1968. In particular, the empirical examination of these securities indicates that outliers (especially in the short-tail (left-hand tail) of the distributions), kurtosis, and skewness are present in the distributions. These characteristics are consistent with the hypothesis that stock price changes belong to the asymmetric class of members of the stable family of distributions.

In [1] the authors proved the following proposition.

Proposition 1: If the positive random variables are exchangeable and linearly independent and if u(x) is strictly concave and satisfies

then the unique optimal choice is given by .

The purpose of this paper is to compare the performance of stocks and mutual savings and loan associations. Since the philosophies underlying stocks and mutuals supposedly differ, assertions are often made that operational differences also exist. The profit motive is presumed to dominate the operations of stock associations while mutual associations are said to be operated more in the interest of the general public. As the pressure to allow more conversions increases, the question of whether significant differences in operating characteristics exist becomes increasingly important.

In a recent article in this Journal, Amir Barnea [1] proposes a criterion for assessing the market-making efficiency of New York Stock Exchange specialists. The appealing aspects of Barnea's method are that it uses publicly available data (common stock prices) and operates on a variable of primary concern to investors, the variance of returns on common stock. The difficulty we see in applying his approach, however, is that Barnea's performance criterion can be sensitive to a number of factors in addition to any impact the specialist might have, and that effective specialist intervention might have either a positive or negative impact on the performance measure. Thus, his specialist ranking seems to be quite misleading, and his empirical findings appear to be amenable to a substantially different interpretation than that which he provides.

In a recent article in this journal [6], Clement G. Krouse and Wayne Y. Lee (hereafter K-L) presented a model of optimal equity financing of a corporation based on Pontryagin's maximum principle. In this note the basic assumption of a constant internal rate of return of the K-L model is relaxed. As a result, the financial implications of the K-L results remain essentially unchanged, but their applicability is extended considerably, and some undesirable solution characteristics are eliminated.

The focus of our study concerns the information effect of newly reported data. Our time frame of reference is the period after the data become available. The information we analyze is generated by a multivariate model which utilizes publicly available data. We concentrate on an extremely poor performing group of companies where the new information indicates a change in status from a going concern to a potential bankruptcy; i.e., the firms possess characteristics similar, to other firms which were bankrupt in the past. Our sample, however, is comprised only of firms which, in fact, did not fail.

In my paper utilized time-variance relationships to detect the impact of NYSE specialists on stock price changes. Schwartz and Whitcomb (SW) seem to agree that return variability and not average bid ask spreads is the appropriate measure on which performance of NYSE specialists should be evaluated. They raise, however, several objections regarding the use of time-variance relationships to evaluate specialists' performance. In particular they show that the proposed performance measure which is the average (per specialist unit) ratio of short term relative to long term return variance depends on the degree of autocorrelation in the return series. For first order serial correlation of returns they obtain,

where r is the performance measure (in terms of a single security),

T is the length of the differencing interval,

are returns variances for one day and T day intervals respectively,

ρ1,2 is first order serial correlation.

I welcome the oppostunity to second the analysis of Davies and Ronning [4]. The considerations they raise crop up repeatedly in various contexts of uncertainty analysis and prompt the following remarks dealing with various tonics described in ray title.