W HETHER or not shortages of highly educated and trained personnel, such as engineers and scientists, existed or were serious in recent years has been the subject of widespread discussion and indeed controversy.1 Typically, this discussion focuses on the question of whether there ought to be more engineers and scientists, in terms of our competition with the Russians, for example. It is not surprising then that controversy flourishes, since the answer to this question depends ultimately on one's value judgments, with respect to market and nonmarket variables. But more fundamentally, much of the disagreement over the issue stems from the variety of meanings attached to the term and the lack of empirical tests. To clarify the matter, Blank and Stigler in their recent book2 define and provide an empirical test for the existence of shortage. This paper re-examines some of the empirical evidence they submit and supplements it with additional evidence, part of which extends the analysis to more recent date. Briefly, here is summary of the BlankStigler approach and findings.3 They begin with this definition of shortage: a shortage exists when the number of workers available (the supply) increases less rapidly than the number demanded at the salaries paid in the recent past. Then salaries will rise, and activities which once were performed by (say) engineers must now be performed by class of workers who are less well trained and less expensive.4 (Italics in original.) Since I929, and more particularly since I939, they find that the earnings position of engineers has deteriorated substantially relative to all earners and to other professional earners. Although they note slight upturn in the relative earnings position of engineers since the beginning of the Korean WVar, this reversal is characterized as a minor cross-current in tide. Thus, the evidence leads them to conclude that there has been no but rather an increasingly ample supply of engineers. Furthermore, they predict that the downward trend in relative earnings position will continue.5

M R. HICKMAN's article is a well balanced contribution, and I have only a few comments to make. When one speaks of autonomous shifts in consumption, one has to be careful to exclude shifts which are really induced. Certain shifts which can properly be classed as induced may not appear such at first thought. Thus there are clearly cyclically-induced shifts in the consumption function. These relate to changes in expectations caused by cyclical movements of investment and aggregate income. It may be possible eventually to establish a fairly standard pattern of this form of cyclical behavior, though doubtless the cyclically-induced shifts in the consumption function will vary more or less from cycle to cycle. Mr. Hickman himself implicitly refers to such cyclically-induced shifts in the first paragraph of his section II. Next it is important to weigh carefully contrived induced changes in consumption. These played an important part in the recovery of I954-55. They involved not only tax cuts, but also a deliberate program designed to push the expansionary role of consumer credit to the limit. Mr. Hickman also takes cognizance of this, but I believe not quite adequately. With respect to the relative importance of gross private investment and consumption in the downturn, I note that investment declined by $I0.4 billion from the second quarter of I953 to the fourth quarter of I953, while consumption declined by a mere $i.i billion all annual rates. Also with respect to the recovery, I note that from the second quarter of I954 to the fourth quarter of I954, gross private investment increased by $3.6 billion while consumption increased by only $4.4 billion. An increase in consumption of this magnitude in relation to the magnitude of the increase in investment is not at all out of line with typical cyclical behavior. And from the fourth quarter of I954 to the fourth quarter of I955 gross private investment increased by $I5.9 billion while consumption increased by only $22.0 billion -again quite in line with normal cycle behavior. A point is made of the fact that the of increase of consumption expenditures diminished during the first half of I953. This also is typical consumption behavior at the upper turning point, and in no way proves that consumption leads. In the I948-49 recession the declines in the rate of increase in the last three quarters of I948 were (in billions of dollars) 4.8, 2.8, and o.g. I am unable to find any peculiarly autonomous behavior of consumption in the I953-54 recession.

T HE recent report of the Economic Policy Commission of the American Bankers Association entitled A Plan for Member Bank Reserve Requirements is a remarkable document. The bankers are, in effect, asking Congress to hand them on a silver platter $9.8 billions of earning assets in place of an equivalent amount of unearning cash assets which they are now required to hold as reserves. The proposal is to count vault cash as part of the required reserves and to reduce the reserve requirements from the present levels (20 per cent for central reserve city banks in New York and Chicago, i8 per cent for reserve city banks in some 50 of the largest cities, and I2 per cent for smaller banks) to a uniform io per cent.' Of the $9.8 billion, $7.7 billion is accounted for by the reduction in reserve requirements and $2.I billion is accounted for by the inclusion of vault cash as part of required reserves. It is evident that only a very small part of the windfall would accrue to the smaller socalled country banks which hold about 38 per cent of the total assets of member banks. The proposal, if enacted into law, would conspicuously favor the large banks. American history is replete with government subsidies on a handsome scale. In many, possibly even in most cases, these subsidies from the railroad land grants to low-cost housing -can be justified from the standpoint of the general welfare. But no one will deny, I think, that there are few if any actions of government that demand a more conscientious assessment of general social benefits and costs. Subsidies, open or veiled, should continually be subjected to careful scrutiny. And this is especially true of subsidies which are veiled in mystery as is the case with the one here under consideration. The Commission says that a clear-cut understanding on the part of the public is highly important. Unfortunately, the report falls considerably short of this worthy aim. Still there is no need to feel alarmed. The Congress has evidenced in recent years a high degree of enlightenment with respect to monetary and banking matters and is not likely to act hastily on this proposal. World War II could have been financed entirely (apart from taxes and bond sales to the public) by the Federal Reserve Banks. This would have involved no subsidy to anybody. The war was indeed partly financed in this manner. The Federal Reserve Banks absorbed about $22 billion of new United States securities. The commercial banks, however, absorbed much more, about $69 billion. To enable them to acquire this huge volume of earning assets, they were supplied with the requisite reserves. This cost the banks not a cent. Some economists objected strongly to this procedure. They wanted the Federal Reserve to do all the bank financing in order to prevent the bestowal of a huge windfall of earning assets on the commercial banks. The policy pursued could, however, be justified. The volume of monetary transactions was rising by leaps and bounds under the rapidly growing war economy. This development involved huge increases in the cost of banking operations. War financing involved extensive banking services performed for the Treasury by the banks. Had the war bankfinancing been done exclusively by the Federal Reserve, the commercial banks would have had to be subsidized in some other manner, or else they would have been compelled to charge unbearably high service charges. The Economic Policy Commission deplores the fact that the Federal Reserve Banks had absorbed so high a proportion of the war issues. The commercial banks could have done the job with less use of Federal Reserve credit had the reserve requirements been reduced. Had this been done, nearly all of the asset windfalls would have fallen to the commercial banks and virtually none to the Federal Reserve Banks. The Commission now wishes to back The report suggests that this may be lowered or raised by the Federal Reserve Board within the range of 8 and I2 per cent.

seems to me much too severe and based in part on a misinterpretation or misunderstanding of the Ricardian doctrines. Knight as a full adherent of the post-I870 revolution in value and distribution theory, or departure from the Ricardian-classical to the utility and marginalist analysis, is I think half blind to the insights, of enduring value, contained in the former and preserved in the Marshallian synthesis. But I cannot, for lack of space, go further into this subject. My final advice to the reader must be: don't rely on this review but read the book!