The Keynes-Hansen
'Demand for Labor' Notion:

A Prosperity-Depression Theory by Which Labor
Loses in Boom and in Slump

Harry Gunnison Brown

[Reprinted from the American Journal of Economics and Sociology,
Vol. 18, No. 2 (January, 1959), pp. 149-156]

ROBERT L. HEILBRONER relates that when Professor Alvin H. Hansen of Harvard University (who "behind his back ... was called 'the American Keynes' ") went "to Washington to testify in the monopoly investigations ... he turned the committee into a hushed private seminar," and the chairman told him the discussion was "getting so interesting" that "we are violating our rules on all sides."[1] This was because, as Heilbroner expressed it, a "great current which had carried the capitalist ship along in the past was petering out, and henceforth progress would have to be made without the aid of a constant, favoring, urgent stimulus." As to what the stimulus was: "No one would have been more surprised than Parson Malthus," for "it was population growth."

Here is the way Hansen himself expresses the matter in Business Cycles and National Income:[2]

In the Great Depression of the Thirties there occurred for the first time in American history a drastic decline in the absolute rate of population growth. Every previous depression had been buoyed up by the capital requirements associated with an ever larger increment of population. The decade of the Nineteen Thirties enjoyed no such stimulus. In the decade of the Nineteen Forties, however, there was a strong resurgence of population growth, and this in part accounts for the high level of capital requirements in the years following the Second World War. The accumulated backlog of capital needs which confronted the economy after 1945 was in some measure greater by reason of the large growth in population in the decade of the Forties. By the same token the decade of the Thirties suffered from a dearth of investment opportunities, partly by reason of the drastic decline in the rate of growth.

And a few sentences further on, Hansen says:

After the economy has become adjusted to a rate of growth of around 16 or 17 million per decade, a decline in the rate of growth to less than 9 million could not fail to chill the outlook for investment. On the other hand, as we have just noted, the remarkable and unexpected spurt of population in the decade of the Forties has raised expectations with respect to profit- able investment outlets.

In these passages, the idea that increase of population increases the demand for labor is implied rather than directly stated. It is implied in the assertion that increased population conduces to business activity as contrasted with depression. And it is implied in the assertion that even a reduced rate of increase of population tends towards depression, as compared or contrasted with a "spurt of population."

But in his earlier book, Fiscal Policy and Business Cycles, Dr. Hansen is much more forthright, expressing himself as follows:[3]

It has been argued that cessation of population growth should be favorable to employment, since the supply of new workers in the labor market would be reduced. But it is easy to show that population growth, if it occurs in a period of territorial expansion, raises the demand for labor more than it raises supply. Thus, the volume of expensive investment associated with the net addition of one worker involves capital outlays on a house, amounting to, say, $4,000, and outlays on plant and equipment amounting to an additional $4,000. Eight thousand dollars of investment represents a far greater effect on the demand for labor than the effect on supply of one additional man-year of labor.

Before commenting on the main idea in the above pronouncement, we might ask what important difference it makes, if any, in Hansen's conclusion, whether or not population growth "occurs in a period of territorial expansion." The argument as it is stated in the succeeding sentences seems to be completely independent of the qualification, and such that it should stand or fall, if it has any meaning, regardless of "territorial expansion."

As a preliminary to discussion of the contention about the "demand for labor," it will perhaps help to make the discussion more realistic in relation to contemporary wage rates, if we reckon the capital outlays on the house as $8,000 and on plant and equipment as $8,000. For the price level is today approximately double what it was when Hansen's book was published. Then the last sentence would read: "Sixteen thousand dollars of investment represents a far greater effect on the demand for labor than the effect on supply of one additional man-year of labor."


The "Demand for Labor" Notion

THE PHRASEOLOGY seems to indicate-for perhaps we should reckon wages for "one man-year of labor" as about $4,000 to $5,000-that the addition to "supply" of one man-year of labor makes an addition to "demand" of three or four times that-of three "man-years" or four "man-years" of labor. But then we may reasonably ask, I think, why Hansen should set one man-year of labor in added "supply" of labor, over against $16,000 of investment as added "demand" for labor. Why not assume, instead, one man-decade of labor, one man-month of labor, one man-day of labor, one man-hour of labor or one man-minute of labor? We can but wish that Professor Hansen had explained for us just why $16,000 (or $8,000 in terms of 1941 prices) of "investment" should be associated with, com- pared with or in some sense equated with one man-year of labor. Is it because one man-decade of labor (for example) might look like an excess of supply of labor over demand instead of vice versa?

Economists have many times insisted that demand is not merely desire but depends on purchasing power.[4] Why does not Hansen tell us precisely how "one additional man-year of labor" provides the purchasing power for a demand amounting to $16,000 (or $8,000 in 1941 prices) ? There is, too, no sign of understanding, in the quoted passage, of how capital comes into existence through saving. Those who wish to invest in the construction of capital must save, i. e., deny themselves consumable or "present" goods. What they might have spent for such present goods can then be spent for capital or for the construction of capital. There is here no increase in demand for goods in general but merely an increase for capital balanced by a decrease in demand for consumable goods. Of course, an increase in the volume of circulating medium may increase the demand-at current prices-for goods in general and may thus bring about a rise in the price level.

It may, indeed, be easy to say, but certainly is not "easy to show," that "population growth . . . raises the demand for labor more than it raises supply."

We might add that demand for labor is commonly supposed, by economists, to have some relation to the productivity of labor. (An employer will seldom knowingly agree to pay a worker more than he believes the worker will add to what is produced.) Hansen seems to write, here, as if demand for labor depended on the housing and machinery "needs" of the laborers!

There is a wealth of evidence to show that most human beings have enough unsatisfied wants so that, if for any reason they do not need or want goods of a particular kind, such as houses, they will buy other goods -- more and better clothing, motor boats, electric refrigerators, musical instruments, books and newspapers, more and better furniture, etc. Or they will enlarge and beautify the houses they have. Or they will spend more in educating their children. Or they will invest more in the purchase of productive capital.

Those who do not have any desire to spend money, if there are any such, will presumably not work to earn money, and the quantity of goods produced to sell will therefore be lessened. If the population becomes smaller, the volume of goods produced will presumably be smaller. In any case, the assumption that if and because men do not want more or larger houses, therefore they will probably spend less in any appreciable degree -- i.e., that they will have an appreciably greater tendency to hoard their money -- and therefore bring a substantial decrease of demand for goods in general, is utterly gratuitous. And in the absence of such an assumption, the entire argument has no significant relevancy.

If Hansen is to make a case for the view that a declining rate of growth in population generates unemployment because of some consequential disinclination to invest, he must show that that disinclination to invest is not balanced by a corresponding inclination to spend. In other words, he must show that there is an appreciably greater tendency to hoard. Without such an assumption of increased "liquidity preference" and, therefore, increased hoarding, the argument that depression and unemployment must ensue loses all its plausibility.

Furthermore, in the light of the facts antecedent to and leading into the Great Depression of the Nineteen Thirties, the assumption that the initiatory force was hoarding, is unjustified. There was Federal Reserve credit restriction, beginning as early as the spring of 1928. And this restriction was accentuated in 1929 despite a level of wholesale prices already lower in the early part of 1929, prior to the stock market crash, than in 1928.' There was a great decrease, in the early Nineteen Thirties, of the volume of circulating medium. The data assembled by Dr. Clark Warburton[6] seem to indicate that changes in the velocity of circulation "are typically sequential in time to deviations in the quantity of money from its normal upward trend and are in the same direction." Entry into the depression, says Warburton, "was led by failure of the money supply; after the shortage of money had made itself felt ... declining use of money was a powerful intensification factor deepening the depression." Why, then, should we assume that "liquidity preference"-or "hoarding," or declining velocity of money--itself stemming, supposedly, in large degree from a declining rate of growth of population, was a significant initiatory cause, or even in any degree an initiatory cause, of the depression and of unemployment?

Let us temporarily ignore, however, all these flaws in Hansen's reasoning and in his assertions, and accept provisionally, more or less on faith, his view that increasing population "raises the demand for labor more than it raises supply."

But to say that increased population thus increases demand for labor more than it increases supply, is to imply that decrease of population decreases "the demand for labor" more than it decreases supply. And, as we have seen, Professor Hansen seems to look with a jaundiced eye even on increase of population, whenever the increase is at a substantially decreasing rate. Thus, if population does not increase, and at a sufficiently rapid rate, so that there is a relatively great desire-or "need?"-for new housing, industrial plant and equipment, the resultant lack of stimulus to such investment may (in Hansen's view) so decrease demand for labor as to bring about serious unemployment.

On the other hand, if population increases so fast as to make possible (again, in Hansen's view) relatively full employment, this is just because of the scarcity relative to population, of housing, plant and equipment. Thus these employed workers are nevertheless not so well provided, per worker, with either capital or land. The productivity of their labor thus tends to be lower and their wages must, therefore, be relatively low. In other words, labor can't win in either case. Labor must be, in the one case, to a large extent jobless; and it must, in the other case, be relatively unproductive and accept relatively low wages!

Could it perhaps be that Hansen would deny this and contend that with his putative high "demand" for labor, workers would have higher wages at the very time they were ill provided with capital; might he contend, that is, that wages have no special relation to the productivity of labor! Although Hansen's treatment of this matter differs superficially from that of Keynes, particularly in Hansen's argument about "demand for labor" in relation to "supply of labor," there seems to be substantial similarity in their conclusions.

For Keynes contends7 that accumulation of wealth can be, and has been, so large as to bring the "marginal efficiency of capital" down more rapidly than the "reward required for parting with liquidity" can be brought down. And he contends that, "in conditions mainly of laissez faire," this "can interfere . . . with a reasonable level of employment." Indeed, he seems clearly to attribute depression and unemployment in Great Britain and the United States during the post World War I period, largely, if not entirely, to this. Thus, Keynes too is saying that when capital equipment is plentiful so that labor, being well supplied with capital, has high productivity and might reasonably expect, therefore, to earn high wages, it is likely to be in substantial degree unemployed. And this is supposed to be the consequence of a "liquidity preference" which inhibits investing for the low returns realizable when capital is so plentiful.

In Keynes' approach, returns on capital are so low as to discourage in- vestment, because there has come to be so large an accumulation of capital -- presumably in proportion to the number of workers. In Hansen's approach, returns on capital are so low as to "chill the outlook for investment," because the population-and, therefore, the number of workers -- has increased so little; in other words, the amount of capital in proportion to the number of workers, is great. With both Keynes and Hansen, the large amount of capital per worker, tends to bring about decreased investment and decreased employment.

On the other hand, in the Keynesian theory as well as in Hansen's, labor has a better chance for employment when capital is relatively scarce, i.e., when workers are less well provided with plant and equipment and when, therefore, the productivity of labor is relatively low. Keynes does not, indeed, comment meaninglessly on "man-years of labor." But he certainly takes the position that when "the marginal efficiency of capital" is high -- which is when capital is relatively scarce -- "liquidity preference" is less likely to manifest itself in an excess of hoarding; and there is less likely to be unemployment. When capital is scarce, would-be wage earners can have jobs. But when capital is scarce, workers must be less well equipped with capital and their productivity (in the terminology of economics, "marginal productivity") must be, other things equal, lower. Hence their real wages must be relatively low. In short, with plentiful capital and high productivity of labor, jobs must be scarce; while with scarcity of capital and low productivity of labor, though there may be jobs, real wages must be low. In the Keynesian philosophy -- as in Hansen's philosophy -- labor loses either way.


Tax Incentives for Saving and Investment

BUT THERE IS A WAY of dealing with the alleged independent and initiatory cause of depression envisaged by Keynes and Hansen --assuming it to be such a cause -- which neither of these economists has apparently thought of. It is a method which would, at the very worst, give us a reprieve from the evil fate they warn us of. And even if we suppose that it could not, of itself, assure us of perpetual freedom from business depression and un- employment, it would provide enough gain to our economy to be very much worth while.

Both Hansen and Keynes emphasize as an important causative factor in the initiating of depression, a general unwillingness to invest. Keynes refers specifically to the inhibitory effect of liquidity preference when large investments in capital have brought the "marginal efficiency of capital" to a low percentage, e.g., 2 or 2-1/2 per cent. Hansen, as we have seen, regards large increase of population as a stimulus to investment, and decrease or unusually slow increase of population as retarding investment. Hansen must be assumed, therefore, to have a low "marginal efficiency of capital" in mind, in the latter case, as the proximate cause of the lack of new investment, a lack which, in his thinking, brings business depression.

But the returns which motivate investors are the returns they anticipate will come to them. It is not the per cent "marginal efficiency of capital" in adding to output which concerns them, but the per cent which comes to them personally. In other words, they invest for what is left after the yield of capital has been tapped by the community or state for the public exchequer. When Hansen says that population has not increased enough to make additional capital seem worth constructing and when Keynes says that capital has increased so much that its "marginal efficiency" is too low to overcome "liquidity preference," they must both have in mind a sequential small yield to investors. And this percentage of yield would be much larger if capital were not taxed.

If, therefore, we were to untax capital and draw sufficient additional revenue to make up the loss, by heavier taxes on the geologically-produced and community-produced value of land, this would certainly provide a greater reward to those who save and invest in capital. If it is really true -- as both Keynes and Hansen contend-that the lack of an adequate gain on investment leads to business depression and unemployment, and if by such a change in tax policy we can decidedly increase that gain, what are the overriding arguments against our doing so?

On the theory that it could, just conceivably, come about in some later decade or generation or century, that the return on capital to investors -- even though untaxed -- would be so low as to greatly increase liquidity preference and thereby initiate depression, such depression would still not be inevitable. An appropriate monetary policy could both satisfy-satiate, if necessary-liquidity preference, and provide enough additional purchasing medium to maintain the demand for goods and labor.

The change in tax policy here suggested would yield definite and substantial benefits, even though not needed at all to give us a reprieve from any depression generated in the way or ways Hansen and Keynes describe. The heavier tax on community-produced land values would lessen the waste of holding good land out of use for speculation, as it has lessened such waste in parts of Australia where such a tax system is employed. Labor would be better supplied with land, the productivity of labor would be greater and real wages would be higher. With lower land rent, the cost of housing to tenants would be lower.

Both cogent theory and available statistical data from Australia indicate that the larger percentage of gain to investors in new capital would bring about more capital construction in the communities, states and nations where this tax policy was followed. Thus, labor in them would be better provided with capital as well as better provided with land. For this reason too, then, the productivity of labor would be greater and wages would be higher. Why should not followers of Hansen and Keynes join in urging this reform? On the basis of their explanations of how business depressions are or may be brought about, such a tax policy would be a definite help in preventing them -- or, at worst, delaying them. On the basis of their own hypotheses, it would offer threatened humanity at least a reprieve and perhaps a long-even an indefinitely long! -- reprieve. Why do they ignore it? Do some of them fear, perhaps, that to express approval of a land-value-tax policy might make them professionally declasse? Or has it really never occurred to any of them that the possibility of land-value taxation has any bearing whatever on the adequacy or the correctness of the Keynes-Hansen analysis?


  1. In The Worldly Philosophers, New York, Simon and Schuster, 1953, pp. 289-90.
  2. New York, W. W. Norton, 1951, p. 75.
  3. New York, W. W. Norton, 1941, p. 41, footnote.
  4. This and the next four paragraphs are taken, with only slight changes, from my Basic Principles of Economics, 3rd ed., Columbia, Mo. (Lucas Brothers), 1955, Vol. II, p. 179, beginning with footnote and following with text.
  5. Basic Principles of Economics, op. cit., Vol. II, pp. 160-3, especially 163.
  6. In "Monetary Velocity and Monetary Policy," Review of Economics and Statistics, 30 (November, 1948), especially p. 309. See also his "Bank Reserves and Business Fluctuations," Journal of the American Statistical Association, 3 (December, 1948). Even if one is not convinced by Warburton's data here cited, that declining velocity of circulating medium is "sequential" to "failure of the money supply," Hansen's view that a decreasing rate of population growth generates unemployment remains equally implausible. Thus, some may contend that, with many borrowers, an increased interest rate charged by banks could bring about a cautious slowing down of their expenditures for goods and labor even before it reduced their borrowing. Having intended to borrow a certain amount in October, such a potential borrower might, recognizing the "tightness" of credit, slow down his expenditures in September in anticipation of borrowing less in October, than he would borrow had the bank rate remained low. But on this assumption, too, it is bank policy, and not a declining rate of population growth, that has decreased the demand for goods and for labor.
  7. The General Theory of Employment, Interest and Money, New York, Harcourt, 1936, p. 219 and, for "definition of the rate of interest," p. 167.