Selling More Labor


BETWEEN 1900 and 1914, the United States received an average of nearly half a million immigrants a year. Two thirds of these immigrants were workers. American industry absorbed these new workers as a matter of course. There were occasional years of depression when employment shrank, but no one questioned the ability of industry to provide more and more jobs and to use more and more labor; no one suggested that there might be a limit to the number of jobs that industry could provide or the amount of labor that could be sold. In fact, when the immigration restriction act was passed, shortly after the war, the chief concern of employers was where they could now obtain sufficient men. Recently, in his Baltimore speech, President Roosevelt suggested a programme for reducing the number of unemployed. His programme had two principal parts. He proposed to limit the span of working life to the years between eighteen and sixty-five, and he advocated a reduction in the weekly working hours — a reduction which would be compensated by an increase in wage rates.

Mr. Roosevelt’s Baltimore speech vividly illustrates the reversal of popular opinion on the subject of employment during the last twenty years. For the views that he expressed in Baltimore are not his alone — they are the opinions which are dominant in America to-day. In a recent poll by the American Institute of Public Opinion, 76 per cent of the voters favored shortening the hours of labor in business and industry for the purpose of reducing unemployment, and 84 per cent said that the reduction of weekly hours should not be permitted to reduce weekly earnings. Just as thirty years ago it was taken for granted that the amount of employment is not limited and that industry could readily absorb a large influx of labor, so to-day many people seem to believe that the only way to give work to the unemployed is to take it away from the employed. To approach the problem of unemployment in this way is, of course, to confess defeat in advance. But such an approach is unnecessary. Boiled down to its simplest terms, the problem of unemployment is essentially one of selling more labor. Surely in a world where human wants arc virtually unlimited it must always be possible to discover a market for more labor. The problem is to discover the conditions that are favorable to the sale of labor and to put these into effect until unemployment has been cut down to the irreducible minimum.


No one knows how many unemployed there are in the United States. Of the several estimates the best is that of Mr. Robert Nathan, formerly Consultant on Unemployment Statistics of the President’s Committee on Economic Security. Mr. Nathan, after careful and elaborate study, put the number of unemployed in October 1935 at 10,600,000. Since that time there has been a moderate gain in employment. I am not convinced, however, that Mr. Nathan’s figures make adequate allowance for the substantial growth in employment produced by the starting of many thousands of small enterprises and the rise in self-employment. Between 1930 and 1935, for example, the number of farms in the United States increased by 500,000. Furthermore, the depression led many thousands of persons who could not obtain jobs working for others to go into business for themselves as taxi drivers, truck drivers, filling-station operators, and in other ways. In addition, during the period of revival, many new enterprises have been started, particularly in the consumer-goods industries. The actual volume of unemployment in the United States to-day is, I suspect, less than 9,000,000. This is a rough figure and it is somewhat less than most estimates. It is considerably above the much-publicized estimate of the New York Sun. The Sun’s estimate, however, does not pretend to be complete and there are serious objections to the sample upon which it is based and the methods by which it is derived. Apparently the Sun would have us believe that unemployment is scarcely any greater now than at the time of the unemployment census of 1930, although building construction is far less, railroad employment 500,000 less, industrial production no greater, and the job-seeking population about 2,500,000 larger.

A certain amount of unemployment is probably unavoidable. It is always off-season in some industries. Furthermore, men are always changing jobs for various reasons and enterprises are always going out of business. When one considers that there are approximately 38,000,000 persons in the United States who make their living by working for others, it is not extravagant to guess that the amount of ‘ normal ’ unemployment is in the neighborhood of 2,000,000. This represents the amount of maladjustment which is present in industry even during good times. The figure may at first seem large, but it represents an employment-giving efficiency of well over 94 per cent — and, as human institutions go, 94 per cent efficiency is high. If the present volume of unemployment is as much as 9,000,000, and if the ‘normal’ amount is about 2,000,000, the problem of absorbing the unemployed is one of finding jobs for about 7,000,000 men and women, plus about 500,000 more who represent the annual net increase in the number of job seekers.

These figures suggest that the reabsorption problem is considerably less serious than would appear from most of the current estimates of unemployment. In one important respect, however, the problem is greater than these estimates indicate. There are many thousands of persons engaged in agriculture at a very low level of income who should be transferred to jobs in non-agricultural industries. Their precise number is uncertain, but it may be well over a million. At any rate, these persons should be counted in estimating the number of additional jobs that are needed in the non-agricultural industries.


A considerable reduction in unemployment has already occurred. The peak of unemployment was reached in March 1933. Since that time the number of unemployed has dropped by over 5,000,000. At first sight this is an encouraging figure, because another gain of like amount would bring the volume of unemployment within about 2,000,000 of normal. The discouraging fact is that nearly three fourths of the reduction in unemployment occurred between March and October 1933. Since then the Federal Reserve Board’s index of industrial production has gained 34.6 per cent, but the volume of unemployment has dropped by only about 15 per cent. In the manufacturing industries, a rise in output of nearly 38 per cent between October 1933 and May 1936 has raised the index of factory employment by only one fourteenth. As the index has a downward bias during periods of revival, the actual gain in factory employment is somewhat more.

This lag of employment behind production is not confined to the United States; it is a world-wide phenomenon. Between June 1932 and June 1935, industrial production increased 24 per cent in the United Kingdom, but industrial employment gained only 8.6 per cent; in Germany, industrial production increased 77.3 per cent and employment 24.2 per cent; in Canada, industrial production increased 30 per cent, employment 6.8 per cent.

One reason why employment has failed to gain as rapidly as production is that it did not drop as rapidly. In fact, as late as March 1936 factory employment was virtually as high in relation to factory production as it was in 1929. In March 1936, factory output was 81.3 per cent of the 1929 level and factory employment was 80 per cent. A similar situation has prevailed in Great Britain, where in 1934 both factory employment and industrial production were practically up to the 1929 levels, and in Germany, where factory output was 80 per cent and employment 85 per cent of the 1929 level.

The lag of employment behind production on both downturns and upturns is partly explained by the fact that certain types of labor are needed in almost the same numbers regardless of whether output is large or small. In the main, however, it is a result of the fact that employers meet drops in production to some extent by reducing hours rather than by laying off men. This policy of spreading work in order to avoid dropping men has been pursued more widely during the current depression than ever before. Quite naturally the pickup in production has been met partly by increasing hours rather than by hiring men. In September 1934, for example, the average working week in factories in the United States was only 34.4 hours. By April 1936, it had increased to 38.7 hours, a gain of almost 12 per cent. During the same period factory employment increased 12 per cent. Half of the increase in man-hours during this period was accomplished by lengthening the working week, and half through an increase in employment. The most recent figures suggest that many employers, especially the smaller ones, are overdoing the increasing of working hours.

But although factory employment is as high relative to 1929 as is factory production, it is true that employers to-day are using fewer man-hours of labor to obtain a given volume of output. To-day the average working week in factories is over ten hours less than in 1929 and output per man-hour is about a fifth greater. The rise in output per man-hour has been just about sufficient to offset the effect of the shorter working week upon employment. Many persons attribute the rapid gain in output per man-hour to technological progress and express the fear that the number of jobs is being permanently decreased. It would be encouraging if the gain did represent improved efficiency, because only in this way can the standard of living be raised. Unfortunately, however, a large part of the gain is explained by other causes.

In the flush days of 1929 many enterprises felt rich enough to use many men in experiment, research, expansion, and other ways which had no immediate effect upon plant output. The quantity of labor purchased for these purposes to-day is much less. Furthermore, the gain in output per man-hour is partly a result of the efforts of the government to reduce unemployment. No nation has made a bolder or more heroic effort to increase employment than has the United States during the last three years, and yet the effort has yielded disappointing results. The procedure which has been followed has been that of raising wages in the hope that this would increase consumer purchasing power and provide markets for more production. This policy was the essence of the NRA. Stripped of all phraseology about consumer purchasing power, the procedure simply amounted to an attempt to sell more labor by raising the price of labor!

Now it is conceivable that under favorable circumstances a general increase in the price of labor might precipitate an upward spiral in commodity prices and so increase the sales of labor by making it relatively cheap, but the chance that a rise in costs would generate an inflationary spiral must be regarded as remote. In any event, our common sense should have warned us that raising the price is not likely to increase the sales of any article and that there is no reason to expect labor to be different in this respect from all other articles. It is not likely to be the one and only thing which can be sold in greater volume by increasing the price.

The government, however, did attempt to increase the sales of labor by raising its price, with the result that during the last three years the price of factory labor has advanced by 25 per cent and is now 3 per cent above 1929. The prices of the commodities which factory labor produces, however, are still about 15 per cent below the levels of 1929. With labor expensive and the products it makes still low in price, employers can afford to hire workers only under those conditions where their productivity is high. Hence the high productivity of labor represents, not merely mechanical and managerial progress, but partly the fact that under prevailing price relationships employers can afford to use labor only where it is much more productive than in 1929.

Part of the gain in output per manhour, however, must be attributed to better equipment and better methods of production. Such changes, of course, are always being made, but the depression and later the rapid rise in the price of labor have compelled employers to search harder than ever for ways of getting on with fewer men. After a hundred and fifty years of industrial revolution and of rapidly growing employment, it should be fairly evident to everyone that labor-saving methods do not necessarily reduce the total volume of employment. Nevertheless they may. Whether they do or not depends in the main upon (1) whether the savings are passed on to the public in the form of lower prices, or (2) whether, if the savings are not passed on, they are promptly spent, by enterprises for inventories or new equipment or are disbursed in the form of dividends to persons who promptly spend them. There is some evidence that during the last several years savings from laborsaving methods have been only partially passed on in the form of lower prices, investment, or dividend disbursements, and that to a substantial extent they have been used to build up the cash resources of the enterprises making them. Consequently, they may have retarded the gain in employment.


What can be done to increase employment? Mr. Roosevelt says shorten hours and raise wages. There is a strong case for not raising working hours above forty a week, but the experiment of attempting to make jobs by shortening hours and raising wages has already been tried with disappointing results, and the reason is plain to see.

Whether we like to admit it or not, there is a connection between the wages which employers must pay and the amount of labor which they can afford to hire. When wage rates rise, employers find it uneconomical to keep some men and they also find it easier to discover profitable ways of displacing labor. The rapid rise in the price of labor during the last three years has undoubtedly been a major influence limiting the gains in employment, and further rises in this price would retard still more the absorption of the unemployed.

Suppose that, instead of raising wages and reducing hours, we did almost exactly the opposite. Suppose that we reduced the wages and salaries of everyone who now has a job, on condition that employers use the money thus saved to hire more men. One of my trade-union friends, who does not like the idea, suggests that we begin by cutting the wages of economists. That is fair enough, provided, of course, that the colleges and universities use the savings to hire more economists! The money incomes of present job holders would be down 10 per cent, but the number employed would be up by one ninth and about half of the persons now out of work would have jobs. The output of goods would rise, but probably not quite so much as employment. In fact, the output available for consumption might increase by substantially less than the volume of employment.

Let us assume that it increased only 7 per cent — the proportion would probably be more. As the total demand for goods would be unchanged, employers would have to reduce prices in order to sell the larger output. This is why many of them would object to the plan. In order to sell 7 per cent more output, prices would have to drop on the average 6.5 per cent. The drop in the purchasing power of the incomes of those of us who now have jobs would be about 3.7 per cent. The community as a whole, however, would be better off because its standard of living would rise by the increase in the output of goods — by about 7 per cent. This gain in the standard of living would be accomplished principally by increasing the incomes of those who are now unemployed. But even those of us who now have jobs might not be worse off, for part of our present incomes are taken from us in the form of taxes and contributions to support the unemployed.

As employment increased, taxes and contributions for relief would drop and they might drop sufficiently so that even the present job holders would be slightly better off than now. At any rate, we are safe in concluding that the net loss to present job holders would be infinitesimal. It would seem a small price to pay for reducing unemployment by nearly one half.


This suggestion is crude and oversimplified and it fails to take account of the different effects of an increase in output upon different industries. It serves, however, to illustrate the tragic conflict of interest which exists between the unemployed, on the one hand, and business enterprises and the employed on the other hand. Both business and labor are interested in preserving prices and wages against cuts and are opposed to an expansion of production and employment which would threaten the existing structure of prices and wages. Hence my proposal, although sound economics, is obviously had politics — no politician of any party would dare advocate it. Politicians much prefer Alice-in-Wonderland schemes, such as trying to sell more labor by raising the price. Let us see, therefore, whether there are not other ways by which employment might be increased.

Most consumer-goods industries (except residential building) are already operating at about the 1928 rate. Absorbing the unemployed, therefore, is largely a matter of stimulating the capital-goods and the construction industries. The problem of stimulating these industries falls into two principal parts — stimulating home building by individuals and speculative builders, and stimulating expenditures on plant and equipment by business enterprises.

For six years residential construction has been far below the needs of the country. Naturally it helps to sell a man a house when he can be shown that the monthly payments will be little more than he must now pay in rent. Consequently it is important to keep down the cost of construction, which means the prices of building labor, building materials, and mortgage money. But the demand for new housing depends upon other things besides the relative costs of renting and owning — especially upon the ability of prospective buyers to make the necessary down payment and their willingness to go into debt for the remainder of the price. After a long and severe depression, time is required before people get their debts paid and sufficient savings accumulated so that they feel able to afford a down payment on a house. And naturally men will refuse to assume large debts until they feel some confidence in their economic future.

As times improve and as men look forward to greater security of employment, more and more persons will become willing to go into debt for the sake of owning a home. For some years to come, however, the demand for new housing is likely to be limited by the fact that for the average wage earner buying real estate on margin is an unwise and risky speculation. The reason is not primarily that houses are too expensive in relation to his wages, but that his employment is too insecure. Consequently the best way to stimulate residential building on a broad scale is to give workmen greater security of employment. The depression has led to the rapid spread of the seniority rule. Under this rule, layoffs in a given occupational group are based on length of service, the junior men being dropped first. There are many pros and cons to the rule, but in most industries it does have the important advantage of assuring reasonably steady work to at least the senior half of the working force. The extension of this rule to additional plants is probably the most important single step which could be taken to increase the demand for residential building.


The most immediate prospect of increasing employment is by increasing the expenditures of business enterprises upon plant and equipment. The engineers say that the plants of the country are using much equipment which could be replaced at a saving. The amount of such equipment is undoubtedly less than the engineers believe, but it is considerable. Furthermore, there are many plants which might profitably be enlarged and new ones which might profitably be built. In this connection it is important to observe that the gains in the efficiency of labor and management during the last several years mean that the existing plants of American industry can be operated with a smaller force than was required in 1929. Hence, in order to provide plants for the existing labor force to man and machines for it to operate, the United States needs more plants and more machines than it now has. It may seem paradoxical to assert that the United States has a shortage of plant at a time when many existing concerns are not operating at capacity. Nevertheless that is true, for during the last few years, while labor economies have been effected and the number of workers has been growing, the industrial plant of the country has scarcely been enlarged. Here is the crux of the unemployment problem. Why do not enterprises make the replacements, improvements, and extensions which they profitably could? If they could be induced to make them, a large part of the present unemployment would melt away. And the very process of putting men to work improving, enlarging, and building plants would provide larger consumer incomes and hence larger markets for industry.

Many concerns do not make replacements and extensions because they do not have the money. A leader in the furniture industry recently wrote me: ‘Probably more than 75 per cent of the machinery of furniture manufacturers north of the Ohio River and west of the Mississippi could profitably be replaced. Much of it is obsolete and actually preventing lower costs of production. If we had the money with which to buy machinery, I am sure much would be bought, but the money is not there, and, ironically enough, it is not likely to be there unless we meet the apparently impossible condition of getting it through profits.’

The substance of this statement applies to many railroads and to a wide variety of other enterprises which were almost ruined by the depression and which are only gradually getting on their feet. Many of these concerns have become burdened with heavy short-term bank loans which they are gradually paying off. Obviously these enterprises can contribute little or nothing to recovery in its early stages. Only after recovery has helped them increase their profits and reduce their bank loans will they in turn help recovery by entering the market for new equipment and for labor to make improvements and extensions.

But these concerns, though still numerous, are not typical of business as a whole. Most enterprises are well supplied with cash — in fact, the bank deposits of business enterprises as a whole are greater now than in 1929. Why do not concerns which have plenty of cash convert more of it into up-to-date plant and equipment? True, the outlay for plant and equipment is rising — last year it was about 30 per cent above 1934 and this year it is likely to gain even more than last. Nevertheless, the determination of managers to keep their enterprises highly liquid seems to be the dominant factor in the business situation today and the immediate cause for the failure of employment to increase more rapidly.

The expression which business men almost invariably use in explaining their reluctance to make expenditures is ‘uncertainty.’ But uncertainty is always present, because the future is always a mystery. What is peculiar about the uncertainty that confronts business men to-day? One thing is the artificial nature of the recovery. They are encouraged by the recovery that has been achieved thus far, but they are not induced by it to spend very freely because they are not yet convinced that it is ‘real.’ It is still being financed by an enormous government deficit — a deficit which grows rather than diminishes as business improves. Next year, for example, government revenues are estimated at about 75 per cent above 1933-1934, but the deficit will be substantially larger than three years ago.

Many people are led by the large and persistent deficit to fear inflation, Strange as it may seem, the average business man has the opposite fear — he fears deflation more than he does inflation. He admits that inflation is an ultimate possibility, but he is not led by this possibility to buy commodities. On the contrary, in so far as the prospect of inflation affects him at all, it encourages him to maintain a strong cash position, to be ready to buy commodities in case inflation actually occurs.

The immediate concern of the business man, however, is with what will happen when the government at last puts its finances in order. For a period of about two years the large expenditures of the present Administration were an exceedingly useful stimulant. Had the Administration permitted its deficit to drop gradually as recovery progressed and as revenues increased, the government’s expenditures would still be a stimulus. The persistence of the large deficit, however, creates the danger that the government will not get around to reforming its finances until just about the time that business is ready for a recession anyway. If government expenditures are then substantially cut and if taxes are substantially increased, will business go into a severe slump? This is the question which lately has begun more and more to worry the business man, and because no one can tell him the answer to it he is disposed to hold his cash and to be conservative about making commitments.

No less important than the uncertainty created by the government’s financial policy is the business man’s concern over future relations between the government and business. The disillusionment and discontent produced by the long and severe depression have led the government to undertake a drastic redefinition of the rights and obligations of property and of the relations between government and industry — in fact, the status of property has been changed more drastically than at any time since the abolition of slavery. Whether one approves or disapproves of these changes, one must concede that the process of making them is discouraging to enterprise. Even after the changes have been completed, business men are likely to be cautious about making commitments until they have acquired some experience under the new rules. At the present moment, however, business men are not sure that the process of redefining the status of property and the relations of government and industry has for the time being been completed — in fact, most of them believe that it has not.

The long and severe depression has not only led to reforms; it has also caused governments all over the world to make the restoration of production and employment a matter of public policy. This means that governments the world over have been assuming a new function. They have been acting as general managers of the economic system. No one has particularly intended it, no one has particularly planned or desired it — governments have simply been compelled by force of circumstances to assume the new responsibility. Had our government been more astute and skillful as an economic manager, it might have been able to offset the retarding effect of its reform policies upon recovery. The Administration, however, has made the fatal mistake of not distinguishing between the results which can be obtained through fear and compulsion and those which can be obtained only by creating confidence and inducing coöperation. It has sought to compel rather than to induce a revival of business.

This is well illustrated by its efforts during the last three years to make business enterprises increase their expenditures. The government began by using currency depreciation as a method of scaring managers into spending. This produced a temporary buying spurt in the summer of 1933, but; after inventories became fairly large the spurt stopped. Almost simultaneously the government tried to make industry increase its expenditures by compelling it to pay higher wages. A considerable increase in payrolls followed. Business managers, however, considered it wise to keep their enterprises liquid, and as wages rose they hunted for ways to limit their purchases of labor. This spring a tax bill was introduced for the purpose of compelling enterprises to disburse as dividends a large part of the money which they were not willing to invest. The Administration’s policy has tended to create a sort of contest between the government and business — the government attempting to compel enterprises to spend and business managers attempting to protect the cash position of their enterprises.

It is illuminating to compare the experience of the government as an economic manager with the experience of employers in managing labor. For many years plant managements attempted to get results from workers primarily by using drive methods. They found that these methods would produce results up to a certain point, but they also found that there were limits to the effectiveness of fear and compulsion. It took plant managements many years to make this discovery and many of them have still to make it, but during the last twenty years the literature of personnel management has consisted largely of discussions of how to get results, not by driving, but by inducing coöperation. The restoration of production and employment depends upon greater willingness on the part of business men to take chances. Business men cannot be scared or driven into taking many chances. They can, however, be induced to take them by favorable circumstances. Just as plant managements have had to learn the difference between compelling obedience and winning coöperation, so the government will not be a successful economic manager, at least during periods of depression, until it learns how to create conditions which make business men more willing to take chances. This does not preclude the institution of reforms during depressions, but it does mean that the more ambitious and sweeping the reforms, the more carefully and vigorously the government must at the same time encourage enterprise.


The employment problem of the moment seems to boil down to this: There are many expenditures which business enterprises could profitably make and which would go far to reduce unemployment, but which managers do not make because they consider it wise to maintain a strong cash position. In order to encourage industry to buy more labor, the following policies are needed: —

1. To avoid for the time being general increases in the price of labor.

2. To reduce the fear of drastic deflation on the part of business managers by making moderate cuts in government expenditures and permitting the increase in revenues still further to reduce the deficit. The government is here confronted with a delicate and important problem — of reducing its deficit so gradually as not to sacrifice the gains in production and employment which have thus far been achieved. In order to accomplish this, the curtailment in government expenditures must be accompanied by policies designed to encourage enterprise.

3. To abandon the policy of attempting to make business spend by the use of fear and compulsion and adopt as an objective of public policy the fostering of confidence and the creation of a willingness among business men to take chances,

4. In order to make a given increase in the demand for labor yield in the main new jobs rather than longer working hours, to seek the coöperation of industry in avoiding increases in the normal working week above forty hours until unemployment has been substantially reduced.

Such a programme, it should be noted, would not involve abandonment by the government of its objective of ultimately raising wages, for the simple reason that a substantial expansion of business could not go very far before competition among employers for men began to bid up the price of labor. It would not involve the abandonment of reforms which New Dealers hold important, such as the spreading of collective bargaining. The growth of collective bargaining requires above everything else a substantial increase in employment, because labor organizations cannot make and hold large gains except on a rising labor market.

Finally, a programme of this sort is the only real solution of the most difficult parts of the employment problem — finding jobs for the million and a half unemployed who are fifty-five years of age or over, accelerating the movement of people out of stranded areas where industry is not likely to revive in substantial degree, and drawing a part of the workers out of agriculture into other branches of industry.

The achievement of these results requires above everything else a substantial increase in the demand for labor, and this increase can be fostered only by policies which encourage business men to take chances and to increase production rather than to hold cash.