It has been argued recently that the size of the holdings of foreign exchange by commercial banks provides a better measure of the adequacy of international means of payments than the size of official reserves.' At first glance this appears obvious for it is these commercial holdings of foreign exchange which are used directly for financing international exchange while official reserves are used only to finance imbalances in countries' balance of payments which result from the maintenance of relatively exchange The argument becomes less clear, however, when one stops to question what is meant by the adequacy of international means of payments. Within a free market context, what does it mean to say that commercial holdings of foreign exchange are inadequate? The commercial interests involved clearly can not feel that their foreign exchange holdings are inadequate (apart from a desire to have higher wealth positions in general) for otherwise they would simply exchange domestic for foreign currency until their foreign currency holdings were no longer inadequate. In other words, from the point of view of commercial banks and traders, at any point in time would merely mean a temporary disequilibrium situation. traders on both sides of the market felt their foreign currency holdings to be inadequate then they would in effect merely swap currencies with one another (a practice now common between central banks). the size of the desired swaps did not match on each side of the market, then under flexible rates the price of the relatively scarce currency would be bid up until desired holdings equalled actual holdings, i.e., until foreign currency holdings were adequate. As Yeager has put it, If no authority concerned itself with gold and foreign exchange, and if private persons, firms and dealers such as banks, found their holdings inadequate, they would bid for additional amounts, thus depressing the home currency on the exchange market, stimulating exports relative to imports, and making available the quantity of foreign exchange desired at the new level of exchange rates. 2 Under a fixed rate system the increased demand for foreign currency would be reflected in official reserve losses. In either case, observed foreign currency holdings would always reflect desired or adequate holdings except for the effects of transitory disequilibrium. We could, however, meaningfully speak of inadequacy in terms of a discrepancy between desired and actual holdings if a free market does not exist. In other words, where exchange controls, etc. effectively prevent traders from satisfying their demands for foreign balances then we could unambiguously say that observed holdings were inadequate. As is brought out in Heller's figures,3 the rapid expansion of holdings of foreign currencies by banks in industrial Europe as postwar exchange controls were loosened suggests that there was considerable inadequacy at the beginning of the period. one accepts the argument put forward here that one can meaningfully speak of an inadequacy of commercial holdings of foreign exchange only where traders do not face free markets for foreign exchange, then inadequate commercial holdings of foreign exchange are themselves a reflection of an inadequacy of official reserves (at least from the point of view of the country in question). In other words, inadequacy of commercial holdings of foreign exchange is a reflection of impediments placed on the foreign exchange market which in turn reflect that the government of the country in question feels that its official reserve holdings are below their desired level, i.e., that they are inadequate. At first glance Heller's figures would seem to contradict this argument. Over the 1951 to 1966 period the global ratios of official reserves to imports and banks' foreign exchange holdings to imports show quite different trends, the former falling by almost one half while the latter almost tripled. Hence, Heller's conclusion that, while according to

Journal Article Optimal Savings under Uncertainty Get access Jerusalem D. Levhari, Jerusalem D. Levhari The Hebrew University Search for other works by this author on: Oxford Academic Google Scholar T. N. Srinivasan T. N. Srinivasan Indian Statistical Institute Search for other works by this author on: Oxford Academic Google Scholar The Review of Economic Studies, Volume 36, Issue 2, April 1969, Pages 153–163, https://doi.org/10.2307/2296834 Published: 01 April 1969 Article history Received: 30 April 1968 Revision received: 26 August 1968 Published: 01 April 1969

The best available indication of the present state of college level auditing education can be deduced by considering widely used textbooks. A recent survey of textbooks used by members of the American Association of Collegiate Schools of Business revealed that the five leading textbooks are used in over eighty-seven percent of, the schools. In all of these books predominant emphasis is placed on verification of financial statement categories and a majority of the chapters are directed to serial treatment of specific general ledger accounts. Within these chapters a great deal of detailed procedural advice is given. There are certain current trends in accounting education, in addition to a preference for conceptual material, that must be considered in shaping the content of the auditing course. There is not room enough for remedial courses in the present accounting curriculum. The auditing course should not be used as a review of accounting theory, which should have been learned in intermediate accounting, and there is not time to cover many features of internal control, which are more appropriately covered in the systems course.

A mixed model of regression with error components is proposed as one of possible interest for combining cross section and time series data. For known variances, it is shown that Aitken estimators and covariance estimators are in one sense asymptotically equivalent, even though the Aitken estimators are more efficient in small samples. Turning to unknown variance components, Zellner-type iterative estimators are compared with covariance estimators. Here, few small sample properties are obtained. However, it is shown that covariance and Zellner-type estimators have equivalent asymptotic distributions and equivalent limits of sequences of first and second order moments for weakly nonstochastic regressors. For the model analyzed, the theoretical results obtained, as well as ease of computation, tend to support traditional covariance estimators of the regression parameters. An additional interesting result presented in an appendix is that ordinary least squares estimates of the fl's (ignoring the error components) have unbounded asymptotic variances. On efficiency grounds, this argues rather strongly for some care in combining data from alternative sources in regression analysis.

I. Introduction, 688. — II. Restraints on the portfolio-management incentive for gold conversion, 689. — III. The adjustment-signaling incentive for gold conversion, 692. — IV. Summary and conclusions, 694.