.

The Brookings Institution has undertaken to study and report the distribution of income in relation to economic progress, and has published its findings and its conclusions in a series of books. These books may well turn out to be among the most important publications since the invention of printing.

Especially pertinent to the problems of the distribution of wealth is Spurgeon Bell's "Productivity, Wages and National Income" (The Brookings Institution, Washington, D. C., $3.00.) This book concerns itself with studying "how gains from increasing productivity are distributed among the groups participating in production and the consuming public; and the effects of the existing distribution upon employment and national income." It presents statistics from a number of different industries: manufacturing, mining, transportation, and light and power in particular. It examines "the relation between the expansion of productive efficiency and the income of .the wage earning population," in order to discover "whether there might perhaps be some deep-seated maladjustment in the economic organism which was serving continuously to restrain or retard the rate of economic expansion."

We need not, of course, go to statistics in order to learn that wages, considered as a share of the product, are decreasing and have .been decreasing for a long time. George announced clearly the law by which this contraction of labor's share must take place in such a society as ours. The extent to which there has been a concentration of economic power in the United States ia documented toy a number of monographs published by the Temporary National Economic Committee; in particular, Monograph No. 26, on "Economic Power and Political Pressures," reveals the extent to which economic forces seek to influence the American State (This monograph can fee had from the Superintendent of Documents in Washington for 25c.) Now, people desire economic power only in order that they may obtain more satisfactions; and to obtain satisfactions by means of power must be to obtain them at the expense ot someone's production. It follows necessarily, therefore, that as political and economic power become centralized, there must be a decline in the share of the producer; there must be a fall in wages.

So much for a priori reasoning. Plausible though it may be, deduction a priori is an object of intense popular suspicion. The common phrase, "it sounds swell in theory, but it won't work," reflects this skepticism. The attitude is natural enough; people have been fooled too often with what looked like air-tight logic and proved to be only wind. We want proof, not by spinning theories, but by actual test. Very properly, we want to be shown.

Is it true that wages, though they may increase as a quantity, tend to fall as a proportion? "Productivity, Wages and National Income" answers this question for us. First: though wages (according to the theory) may increase as a quantity, the record shows that they have not increased at all. Using 100 as an index, we find that in the period 1919 to 1938, inclusive, the average weekly earnings of industrial wage earners have fluctuated between the 1919 low of 87.6 and the 1929 high of 104.2. Note that in 1919 real weekly wages were less than in any subsequent year, less even than in 1933, when the wage index stood at 89.6. This shows the danger of assuming a necessary correlation between wages and "prosperity" for in 1919 there was moderately full employment. We need to know, not only how much the weekly wage was, tout also how many people weren't getting any wage at all.

After 1922 wages leveled off, and for nine years did not vary much from 100; the lowest figure is 98.8 in 1925, and 'the highest 104.2 in 1929. A period of four years in which there was a drop to the lower 90's is followed by the years 1936-38, during which the wage index fluctuates between 98.8 and 102.8. During the boom period of the twenties, the collapse of the depression, and the hypodermic economy of the later thirties, real weekly wages fluctuated less than 8% from the average.

Before we leave the subject of wage rates, we may take a look at annual earning figures. Here is where the great ups and downs of the boom and depression periods become apparent. In 1919 and in 1933 the weekly earning power of labor was about the same -- indeed, it was a trifle higher in 1933; but in 1919 unemployment was estimated at 8%, and in 1933 at 46%. If we take this unemployment into consideration we find that the 1933 wage was, not 89.6, but instead about 50; the estimated annual income of employed industrial workers was $915 in 1933, but if we include the zero incomes of the unemployed in figuring our average, the money wage per worker drops to $497.

During all this time, two things were happening; the productivity of labor was increasing, and employment was falling off. In manufacturing generally, unit wage cost declined by about one third, although real weekly wages remained about the same; this means that labor produced three units in place of two. but received none of the increase as wages. The change in the railroad industry was about the same. In mining, unit wage cost decreased by over 50%: and in mining there was a considerable drop in real weekly earnings as well: about 20%.

In one important field production increased even faster than productivity; this was the electric power industry, in which productivity increased to 220 and output to 240, both starting from 100 In 1924. But even here there was a 40% drop in the unit wage cost, so that increasing production did not lead to any material increase in employment. Dr. Bell observes that "Unit wage and salary cost fell considerably more than the price of service."

Every science is largely dependent upon measurements. Indeed, it is only by making measurements that our sciences can be reduced to a quantitative form. And in its task of measuring, science must solve two problems: what to measure, and how to measure it. No science has solved these problems to perfection. But in economics we have not even made a beginning toward a rational system of measurement. Why? We don't know what we want.

The practical difficulties in the way of finding valid units in which to measure economic variables are formidable enough. A perfect system would measure satisfactions; would enable us to draw a graph of a man's pleasure at beholding a blonde, a blonde's pleasure at beholding a steak. Such perfection is probably forever unattainable. Failing that, we try to measure the "national dividend" by car loadings, bank clearings, money wages, real wages, and so on. That there is some sort of correlation between these and actual satisfactions produced is undeniable; but in the absence of even approximate knowledge of the degree of correlation, to use them as measurements of production seems to be climbing very far out on the statistical limb.

Does it really matter whether real wages in 1933 were 89.6 or 98.6? Is it so important to know whether workingmen had as much to eat one year as another? Isn't the really important question, not "How much did he get?" but rather "How was the pile divided?"

Such a question must lead us to ask who were the sharers in the division. The Brookings Institution study does not go farther below the surface than is customary; it distinguishes between "wages" and "salaries," but makes no distinction among non-labor incomes, which are lumped as "earnings on capital." That the income from a unique mine such as the Climax deposit is vastly different from the income of an individual capitalist in such a competitive field as (say) dress manufacturing must be perfectly well known to every serious student of economics; but it has not influenced economic analysis to any great extent. Yet the fact remains that we must separate non-labor income into one part which may be regarded as a reward offered in a free market for valuable service, and another, resulting from privilege, monopoly, superior bargaining power, or some other exploitative advantage; and that until we have made such a separation we shall learn little from our statistics.

Dr. Bell recommends that industry seek a remedy by passing on to the consumer, in the form of lower prices, the savings which result from increased productive capacity. It is a vain counsel -- doubly vain; for farmers and manufacturers in highly competitive fields have no choice but to sell at the market, often at a loss; while asking those who enjoy a measure of protection from competition to waive their profits seems a little like asking the wolf to disgorge Little Red Riding Hood.

The economic machine exists for the purpose of producing wealth. If we consume wealth, someone has produced it If someone else has produced it, and we have not rendered a commensurate service in exchange, we are economic parasites. In nature, parasite and host often achieve an acceptable biological balance; but in economics it seems to be otherwise. Economic parasitism has increased -- for to what other cause can we attribute the decline in the portion of the wage earner? -- and apparently we are nearing the time when our parasites will overtax the host's capacity to support them. The growing army of bureaucrats In Washington swells the number alarmingly.

Must all perish together? We need not allow our pessimism to run away with us. "Productivity, Wages and National Income" is still one of the most important books of the present day. Despite the handicap of having to express its findings in units which have no measurable correlation with anything whatever in the real world, it supplies us with what is probably the best information yet obtainable. Its obvious sincerity, its thoroughness, its recognition of the existence of "some deep-seated maladjustment," are all happy auguries. For centuries men believed that the earth was the center of the universe, and invented horrible pretzel-curves to describe the motions of the planets; all this complication was swept away by a simple change of viewpoint. May we not hope for some similar revolution in economics?