I HAVE REFERRED to various union make-work and featherbed practices. These practices, and the public toleration of them, spring from the same fundamental fallacy as the fear of machines. This is the belief that a more efficient way of doing a thing destroys jobs, and its necessary corollary that a less efficient way of doing it creates them.
Allied to this fallacy is the belief that there is just a fixed amount of work to be done in the world, and that, if we cannot add to this work by thinking up more cumbersome ways of doing it, at least we can think of devices for spreading it around among as large a number of people as possible.
This error lies behind the minute subdivision of labor upon which unions insist. In the building trades in large cities the subdivision is notorious. Bricklayers are not allowed to use stones for a chimney: that is the special work of stonemasons. An electrician cannot rip out a board to fix a connection and put it back again: that is the special job, no matter how simple it may be, of the carpenters. A plumber will not remove or put back a tile incident to fixing a leak in the shower: that is the job of a tile-setter.
Furious “jurisdictional” strikes are fought among unions for the exclusive right to do certain types of borderline jobs. In a statement prepared by the American railroads for the Attorney-General’s Committee on Administrative Procedure, the roads gave innumerable examples in which the National Railroad Adjustment Board had decided that
each separate operation on the railroad, no matter how minute, such as talking over a telephone or spiking or unspiking a switch, is so far an exclusive property of a particular class of employee that if an employee of another class, in the course of his regular duties, performs such operations he must not only be paid an extra day’s wages for doing so, but at the same time the furloughed or unemployed members of the class held to be entitled to perform the operation must be paid a day’s wages for not having been called upon to perform it.
It is true that a few persons can profit at the expense of the rest of us from this minute arbitrary subdivision of labor— provided it happens in their case alone. But those who support it as a general practice fail to see that it always raises production costs; that it results on net balance in less work done and in fewer goods produced. The householder who is forced to employ two men to do the work of one has, it is true, given employment to one extra man. But he has just that much less money left over to spend on something that would employ somebody else. Because his bathroom leak has been repaired at double what it should have cost, he decides not to buy the new sweater he wanted. “Labor” is no better off, because a day’s employment of an unneeded tile-setter has meant a day’s disemployment of a sweater knitter or machine handler. The householder, however, is worse off. Instead of having a repaired shower and a sweater, he has the shower and no sweater. And if we count the sweater as part of the national wealth, the country is short one sweater. This symbolizes the net result of the effort to make extra work by arbitrary subdivision of labor.
But there are other schemes for “spreading the work,” often put forward by union spokesmen and legislators. The most frequent of these is the proposal to shorten the working week, usually by law. The belief that it would “spread the work” and “give more jobs” was one of the main reasons behind the inclusion of the penaltyovertime provision in the existing Federal Wage-Hour Law. The previous legislation in the states, forbidding the employment of women or minors for more, say, than forty-eight hours a week, was based on the conviction that longer hours were injurious to health and morale. Some of it was based on the belief that longer hours were harmful to efficiency. But the provision in the federal law, that an employer must pay a worker a 50 percent premium above his regular hourly rate of wages for all hours worked in any week above forty, was not based primarily on the belief that forty-five hours a week, say, was injurious either to health or efficiency. It was inserted partly in the hope of boosting the worker’s weekly income, and partly in the hope that, by discouraging the employer from taking on anyone regularly for more than forty hours a week, it would force him to employ additional workers instead. At the time of writing this, there are many schemes for “averting unemployment” by enacting a thirty-hour week or a four-day week.
What is the actual effect of such plans, whether enforced by individual unions or by legislation? It will clarify the problem if we consider two cases. The first is a reduction in the standard working week from forty hours to thirty without any change in the hourly rate of pay. The second is a reduction in the working week from forty hours to thirty, but with a sufficient increase in hourly wage rates to maintain the same weekly pay for the individual workers already employed.
Let us take the first case. We assume that the working week is cut from forty hours to thirty, with no change in hourly pay. If there is substantial unemployment when this plan is put into effect, the plan will no doubt provide additional jobs. We cannot assume that it will provide sufficient additional jobs, however, to maintain the same payrolls and the same number of man-hours as before, unless we make the unlikely assumptions that in each industry there has been exactly the same percentage of unemployment and that the new men and women employed are no less efficient at their special tasks on the average than those who had already been employed. But suppose we do make these assumptions. Suppose we do assume that the right number of additional workers of each skill is available, and that the new workers do not raise production costs. What will be the result of reducing the working week from forty hours to thirty (without any increase in hourly pay)?
Though more workers will be employed, each will be working fewer hours, and there will, therefore, be no net increase in man-hours. It is unlikely that there will be any significant increase in production. Total payrolls and “purchasing power” will be no larger. All that will have happened, even under the most favorable assumptions (which would seldom be realized) is that the workers previously employed will subsidize, in effect, the workers previously unemployed. For in order that the new workers will individually receive three-fourths as many dollars a week as the old workers used to receive, the old workers will themselves now individually receive only three-fourths as many dollars a week as previously. It is true that the old workers will now work fewer hours; but this purchase of more leisure at a high price is presumably not a decision they have made for its own sake: it is a sacrifice made to provide others with jobs.
The labor union leaders who demand shorter weeks to “spread the work” usually recognize this, and therefore they put the proposal forward in a form in which everyone is supposed to eat his cake and have it too. Reduce the working week from forty hours to thirty, they tell us, to provide more jobs; but compensate for the shorter week by increasing the hourly rate of pay by 33.33 percent. The workers employed, say, were previously getting an average of $226 a week for forty hours work; in order that they may still get $226 for only thirty hours work, the hourly rate of pay must be advanced to an average of more than $7.53.
What would be the consequences of such a plan? The first and most obvious consequence would be to raise costs of production. If we assume that the workers, when previously employed for forty hours, were getting less than the level of production costs, prices and profits made possible, then they could have got the hourly increase without reducing the length of the working week. They could, in other words, have worked the same number of hours and got their total weekly incomes increased by one-third, instead of merely getting, as they are under the new thirty-hour week, the same weekly income as before. But if under the forty-hour week, the workers were already getting as high a wage as the level of production costs and prices made possible (and the very unemployment they are trying to cure may be a sign that they were already getting even more than this), then the increase in production costs as a result of the 33.33 percent increase in hourly wage rates will be much greater than the existing state of prices, production and costs can stand.
The result of the higher wage rate, therefore, will be a much greater unemployment than before. The least efficient firms will be thrown out of business, and the least efficient workers will be thrown out of jobs. Production will be reduced all around the circle. Higher production costs and scarcer supplies will tend to raise prices, so that workers can buy less with the same dollar wages; on the other hand, the increased unemployment will shrink demand and hence tend to lower prices. What ultimately happens to the prices of goods will depend upon what monetary policies are then allowed. But if a policy of monetary inflation is pursued, to enable prices to rise so that the increased hourly wages can be paid, this will merely be a disguised way of reducing real wage rates, so that these will return, in terms of the amount of goods they can purchase, to the same real rate as before. The result would then be the same as if the working week had been reduced without an increase in hourly wage rates. And the results of that have already been discussed.
The spread-the-work schemes, in brief, rest on the same sort of illusion that we have been considering. The people who support such schemes think only of the employment they might provide for particular persons or groups; they do not stop to consider what their whole effect would be on everybody.
The spread-the-work schemes rest also, as we began by pointing out, on the false assumption that there is just a fixed amount of work to be done. There could be no greater fallacy. There is no limit to the amount of work to be done as long as any human need or wish that work could fill remains unsatisfied. In a modern exchange economy, the most work will be done when prices, costs and wages are in the best relations with each other. What these relations are we shall later consider.