Online Library of Liberty

A collection of scholarly works about individual liberty and free markets. A project of Liberty Fund, Inc.

Advanced Search

Friedman on a Volunteer Army

Related Links:

Source: New Individualist Review, editor-in-chief Ralph Raico, introduction by Milton Friedman (Indianapolis: Liberty Fund, 1981). Chapter: MILTON FRIEDMAN, Why Not a Volunteer Army?

Copyright: The copyright to this publication is held by Liberty Fund, Inc. The New Individualist Review is prohibited for use in any publication, journal, or periodical without written consent of J. M. Cobb, J. M. S. Powell, or David Levy.

Fair Use: This material is put online to further the educational goals of Liberty Fund, Inc. Unless otherwise stated in the Copyright Information section above, this material may be used freely for educational and academic purposes. It may not be used in any way for profit.

Why Not a Volunteer Army?

OUR MILITARY FORCES currently require the services of only a minority of young men. At most, something like one-third will have seen military service by the time they reach twenty-six. This percentage is scheduled to decline still further as the youngsters born in the postwar baby boom come of age. Some method of “selective service” is inevitable. The present method is inequitable, wasteful, and inconsistent with a free society.

On this point there is wide agreement. Even most supporters of a draft like the present one regard it as at best a necessary evil; and representatives of all parts of the political spectrum have urged that conscription be abolished—including John Kenneth Galbraith and Barry Goldwater; the New Left and the Republican Ripon Society.

The disadvantages of our present system of compulsion and the advantages of a volunteer army are so widely recognized that we can deal with them very briefly. The more puzzling question is why we have continued to use compulsion. The answer is partly inertia—a carryover from a total war situation, when the case for a volunteer army is far weaker. But even more, the answer is the tyranny of the status quo. The natural tendency of an administrator of a large, complex, and ongoing activity is to regard the present method of administering it as the only feasible way to do so and to object strenuously that any proposed alternative is visionary and unfeasible—even though the same man, once the change is made and it becomes the existing method, will argue just as strenuously that it is the only feasible method.

This bureaucratic stand-pattism has been reinforced by a confusion between the apparent and the real cost of manning the armed forces by compulsion. The confusion has made it appear that a volunteer army would be much more expensive to the country and hence might not be feasible for fiscal reasons. In fact, the cost of a volunteer army, properly calculated, would almost surely be less than of a conscripted army. It is entirely feasible to maintain present levels of military power on a strictly volunteer basis.

The other disadvantages that have been attributed to a volunteer army are that it might be racially unbalanced, would not provide sufficient flexibility in size of forces, and would enhance the political danger of undue military influence. While the problems referred to are real, the first and third are in no way connected with the use of voluntary or compulsory means to recruit enlisted men and do not constitute valid arguments against abolishing the draft. The second has more merit but devices exist to provide moderate flexibility under a voluntary as under a compulsory system.

There is no reason why we cannot move to volunteer forces gradually—by making conditions of service more and more attractive until the whip of compulsion fades away. This, in my opinion, is the direction in which we should move, and the sooner the better.

A volunteer army would be manned by people who had chosen a military career rather than, at least partly, by reluctant conscripts anxious only to serve out their term. Aside from the effect on fighting spirit, this would produce a lower turnover in the armed services, saving precious man-hours that are now wasted in training or being trained. Also it would permit intensive training and a higher average level of skill for the men in service; and it would encourage the use of more and better equipment. A smaller, but more highly skilled, technically competent, and better armed force could provide the same or greater military strength.

A volunteer army would preserve the freedom of individuals to serve or not to serve. Or, put the other way, it would avoid the arbitrary power that now resides in draft boards to decide how a young man shall spend several of the most important years of his life—let alone whether his life shall be risked in warfare. An incidental advantage would be to raise the level and tone of political discussion.

A volunteer army would enhance also the freedom of those who now do not serve. Being conscripted has been used as a weapon—or thought by young men to be so used—to discourage freedom of speech, assembly, and protest. The freedom of young men to emigrate or to travel abroad has been limited by the need to get the permission of a draft board if the young man is not to put himself in the position of inadvertantly becoming a law-breaker.

ONE GOOD EXAMPLE of the effect on freedom of a volunteer army is that it would completely eliminate the tormenting and insoluble problem now posed by the conscientious objector—real or pretended.

A by-product of freedom to serve would be avoidance of the present arbitrary discrimination among different groups. A large faction of the poor are rejected on physical or mental grounds. The relatively well-to-do used to be in an especially good position to take advantage of the possibilities of deferment offered by continuing their schooling. Hence the draft bears disproportionately on the upper lower classes and the lower middle classes. The fraction of high-school graduates who serve is vastly higher than of either those who have gone to college or those who dropped out before finishing high school.

A volunteer army would permit young men, both those who serve and those who do not, to plan their schooling, their careers, their marriages, and their families in accordance with their own long-run interests. As it is, the uncertainty about the draft affects every decision they make and often leads them to behave differently than they otherwise would in the correct or mistaken belief that they will thereby reduce the chance of being drafted.

Substitution of a volunteer army (or of a lottery) for the present draft would permit colleges and universities to pursue their proper educational function, freed alike from the incubus of young men—probably numbering in the hundreds of thousands—who would prefer to be at work rather than in school but who now continue their schooling in the hope of avoiding the draft, and from controversy about issues strictly irrelevant to their educational function. We certainly need controversy in the universities—but about intellectual and educational issues, not whether to rank or not to rank students for their draft boards.

Similarly, the community at large would benefit from the reduction of unwise early marriages contracted at least partly under the whip of the draft and from the probably associated reduction in the birth rate. Industry and government would benefit from being able to hire young men on their merits, not their deferments.

So long as compulsion is retained, inequity, waste, and interference with freedom are inevitable. A lottery would only make the arbitrary element in the present system overt. Universal national service would only compound the evil—regimenting all young men, and perhaps women, to camouflage the regimentation of some.

If a very large fraction of the young men of the relevant age groups are required—or will be used whether required or not—in the military services, the advantages of a volunteer army become very small. It would still be technically possible to have a volunteer army, and there would still be some advantages, since it is doubtful that literally 100 per cent of the potential candidates will in fact be drawn into the army; but if nearly everyone who is physically capable will serve anyway, there is little room for free choice, the avoidance of uncertainty, and so on. To rely on volunteers under such conditions would then require very high pay in the armed services, and very high burdens on those who do not serve, in order to attract a sufficient number into the armed forces. This would involve serious political and administrative problems. To put it differently, and in terms that will become fully clear to non-economists only later, it might turn out under these special circumstances that the implicit tax of forced service is less bad than the alternative taxes that would have to be used to finance a volunteer army.

Hence, for a major war, a strong case can be made for compulsory service. And indeed, compulsory service has been introduced in the United States only under such conditions—in the Civil War, World War I, and World War II. It is hardly conceivable that it could have been introduced afresh in, say, 1950, if a system of compulsory service had not so recently been in full swing. As it was, the easiest thing to do when military needs for manpower rose was to reactivate the recent wartime technique.

Under conditions found at present, the number of persons who volunteer for armed service is inadequate to man the armed forces, and even so, many who volunteer do it only because they anticipate being drafted. The number of “true” volunteers is clearly much too small to man armed forces of our present size. This undoubted fact is repeatedly cited as evidence that a volunteer army is unfeasible.

It is evidence of no such thing. It is evidence rather that we are now grossly underpaying our armed forces. The starting pay for young men who enter the armed forces is now about $45 a week—including not only cash pay and allotments but also the value of clothing, food, housing, and other items furnished in kind. When the bulk of young men can command at least twice this sum in civilian jobs, it is little wonder that volunteers are so few. Indeed, it is somewhat surprising that there are as many as there are.

TO MAN THE ARMY with volunteers would require making conditions of service more attractive—not only higher pay but also better housing facilities and improved amenities in other respects. It will be replied that money is not the only factor young men consider in choosing their careers. This is certainly true—and equally certainly irrelevant. Adequate pay alone may not attract, but inadequate pay can certainly deter. Military service has many non-monetary attractions to young men—the chance to serve one’s country, adventure, travel, opportunities for training, and so on. Not the least of the advantages of a volunteer army is that the military would have to improve their personnel policies and pay more attention to meeting the needs of the enlisted men. They now need pay little attention to them, since they can fill their ranks with conscripts serving under compulsion. Indeed, it is a tribute to their humanitarianism—and the effectiveness of indirect pressure via political process—that service in the armed forces is not made even less attractive than it now is.

The personnel policies of the military have been repeatedly criticized, and, with no spur, repeatedly left unreformed. Imaginative policies designed to make the armed forces attractive to the kind of men the armed services would like to have (plus the elimination of compulsion which now makes military service synonomous with enforced incarceration) could change drastically the whole image that the armed services present to young men. The Air Force, because it has relied so heavily on real volunteers, perhaps comes closest to demonstrating what could be done.

The question how much more we would have to pay to attract sufficient volunteers has been analyzed intensively in the Department of Defense study of military recruitment. Based on a variety of evidence collected in that study, Walter Oi estimates in his paper1 that a starting pay (again including pay in kind as well as in cash) of something like $4,000 to $5,500 a year—about $80 to $100 a week—would suffice. This is surely not an unreasonable sum. Oi estimates that the total payroll cost (after allowing for the savings in turnover and men employed in training) would be around $3 billion to $4 billion a year for armed forces equivalent to 2.7 million men under present methods of recruitment, and not more than $8 billion a year for armed forces equivalent to the present higher number of men (around 3.1 or 3.2 million men). Based on the same evidence, the Defense Department has come up with estimates as high as $17.5 billion. Even the highest of these estimates is not in any way unfeasible in the context of total Federal government expenditures of more than $175 billion a year.

Whatever may be the exact figure, it is a highly misleading indication of the cost incurred in shifting from compulsion to a volunteer army. There are net advantages, not disadvantages, in offering volunteers conditions sufficiently attractive to recruit the number of young men required.

This is clearly true on the level of individual equity: the soldier no less than the rest of us is worth his hire. How can we justify paying him less than the amount for which he is willing to serve? How can we justify, that is, involuntary servitude except in times of the greatest national emergency? One of the great gains in the progress of civilization was the elimination of the power of the noble or the sovereign to exact compulsory servitude.

ON THE DIRECT budgetry level, the argument that a volunteer army would cost more simply involves a confusion of apparent with real cost. By this argument, the construction of the Great Pyramid with slave labor was a cheap project. The real cost of conscripting a soldier who would not voluntarily serve on present terms is not his pay and the cost of his keep: it is the amount for which he would be willing to serve. He is paying the difference. This is the extra cost to him and must be added to the cost borne by the rest of us. Compare, for example, the cost to a star professional football player and to an unemployed worker. Both might have the same attitudes toward the army and like—or dislike—a military career equally; but because the one has such better alternatives than the other, it would take a much higher sum to attract him. When he is forced to serve, we are in effect imposing on him a tax in kind equal in value to the difference between what it would take to attract him and the military pay he actually receives This implicit tax in kind should be added to the explicit taxes imposed on the rest of us to get the real cost of our armed forces.

If this is done, it will be seen at once that abandoning the draft would almost surely reduce the real cost—because the armed forces would then be manned by men for whom soldiering was the best available career, and who would hence require the lowest sums of money to induce them to serve. Abandoning the draft might raise the apparent money cost to the government, but only because it would substitute taxes on the general population in money for taxes on certain “selected” young men in kind.

There are also some important offsets even to the increase in apparent money cost. In addition to the lower turnover, already taken into account in the estimates cited, the higher average level of skill would permit further reductions in the size of the army, saving monetary cost to the government. Because manpower is cheap to the military, they now tend to waste it, using enlisted men for tasks that could be performed by civilians or machines, or eliminated entirely. Moreover, better pay at the time to volunteers might lessen the political appeal of veteran’s benefits that we now grant after the event. These now cost us over $6 billion a year, or one-third as much as current annual payroll costs for the active armed forces—and they will doubtless continue to rise under present conditions.

There are still other offsets. Colleges and universities would be saved the cost of housing, seating, and entertaining hundreds of thousands of young men. Total output of the community would be higher both because these men would be at work and because the young men who now go to work could be offered and could accept jobs requiring considerable training instead of having to take stop-gap jobs while awaiting a possible call to the service. Perhaps there are some effects in the opposite direction, but I have not been able to find any.

Whatever happens to the apparent monetary cost, the real cost of a volunteer army would almost surely be less than of the present system and it is not even clear that the apparent monetary cost would be higher—if it is correctly measured for the community as a whole. In any event, there can be little doubt that wholly volunteer forces of roughly the present size are entirely feasible on economic and fiscal grounds.

It has been argued that a military career would be so much more attractive to the poor than to the well-to-do that volunteer armed services would be staffed disproportionately by the poor. Since Negroes constitute a high proportion of the poor, it is further argued that volunteer armed forces would be largely black.

There is first a question of fact. This tendency is present today in exaggerated form—the present levels of pay are comparatively more attractive to blacks than the higher levels of pay in voluntary armed forces would be. Yet the fraction of persons in the armed forces who are black is roughly the same as in the population at large. It has been estimated that even if every qualified Negro who does not now serve were to serve, whites would still constitute a substantial majority of the armed forces. The military services require a wide variety of skills and offer varied opportunities. They have always appealed to people of varied classes and backgrounds and they will continue to do so. Particularly if pay and amenities were made more attractive, there is every reason to expect that they would draw from all segments of the community.

In part, this argument involves an invalid extrapolation from the present conscripted army to a volunteer army. Because we conscript, we pay salaries that are attractive only to the disadvantaged among us.

BEYOND THIS QUESTION of fact, there is the more basic question of principle. Clearly, it is a good thing to offer better alternatives to the currently disadvantaged. The argument to the contrary rests on a political judgment: that a high ratio of Negroes in the armed services would exacerbate racial tensions at home and provide, in the form of ex-soldiers, a militarily trained group to foment violence. Perhaps there is something to this. My own inclination is to regard it as the reddest of red herrings. Our government should discriminate neither in the civil nor in the military services. We must handle our domestic problems as best we can and not use them as an excuse for denying blacks opportunities in the military service.

One of the advantages cited for conscription is that it permits great flexibility in the size of the armed services. Let military needs suddenly increase, and draft calls can be rapidly stepped up, and conversely. This is a real advantage, but can easily be overvalued. Emergencies must be met with forces in being, however they are recruited. Many months now elapse between an increase in draft calls and the availability of additional trained men.

The key question is how much flexibility is required. Recruitment by voluntary means could provide considerable flexibility—at a cost. The way to do so would be to make pay and conditions of service more attractive than is required to recruit the number of men that it is anticipated will be needed. There would then be an excess of volunteers—queues would form. If the number of men required increased, the queues could be shortened, and conversely.

The change in scale involved in a shift from conditions like the present to a total war is a very different matter. If the military judgment is that, in such a contingency, there would be time and reason to expand the armed forces manyfold, either universal military training, to provide a trained reserve force, or stand-by provisions for conscription could be justified. Both are very different from the use of conscription to man the standing army in time of peace or brush-fire wars, or wars like that in Viet Nam which require only a minority of young men.

The flexibility provided by conscription has another side. It means that, at least for a time, the Administration and the military services can proceed fairly arbitrarily in committing U. S. forces. The voluntary method provides a continuing referendum of the public at large. The popularity or unpopularity of the activities for which the armed forces are used will clearly affect the ease of recruiting men. This is a consideration that will be regarded by some as an advantage of conscription, by others as a disadvantage.

THERE IS NO DOUBT that large armed forces plus the industrial complex required to support them constitute an ever-present threat to political freedom. Our free institutions would certainly be safer if the conditions of the world permitted us to maintain smaller armed forces.

This valid fear has been converted into an invalid argument against voluntary armed forces. They would constitute a professional army, it is said, that would lack contact with the populace and become an independent political force, whereas a conscripted army remains basically a citizen army. The fallacy in this argument is that such dangers come primarily from the senior officers, who are now and always have been a professional corps of volunteers. A few examples from history will show that the danger to political stability is largely unrelated to the method of recruiting enlisted men. Napoleon and Franco both rose to power at the head of conscripts. The recent military takeover in Argentina was by armed forces recruiting enlisted men by conscription. Britain and the U. S. have maintained freedom while relying primarily on volunteers, Switzerland and Sweden while using conscription. It is hard to find any relation historically between the method of recruiting enlisted men and the political threat from the armed forces.

However we recruit enlisted men, it is essential that we adopt practices that will guard against the political danger of creating a military corps with loyalties of its own and out of contact with the broader body politic. Fortunately, we have so far largely avoided this danger. The broad basis of recruitment to the military academies, by geography as well as social and economic factors, the ROTC programs in the colleges, the recruitment of officers from enlisted men, and similar measures, have all contributed to this result.

For the future, we need to follow policies that will foster recruitment into the officer corps from civilian activities—rather than primarily promotion from within. The military services no less than the civil service need and will benefit from in-and-outers. For the political gain, we should be willing to bear the higher financial costs involved in fairly high turnover and rather short average terms of service for officers. We should follow personnel policies that will continue to make at least a period of military service as an officer attractive to young men from many walks of life.

There is no way of avoiding the political danger altogether, but it can be minimized as readily with a volunteer as with a conscripted army.

Given the will, there is no reason why the transition to volunter armed forces cannot begin at once and proceed gradually by a process of trial and error. We do not need precise and accurate knowledge of pay and amenities that will be required. We need take no irreversible step.

OUT OF SIMPLE justice, we should raise the pay and improve the living conditions of enlisted men. If it were proposed explicity that a special income tax of 50 per cent be imposed on enlisted men in the armed services, there would be cries of outrage. Yet that is what our present pay scales plus conscription amount to. If we started rectifying this injustice, the number of “real” volunteers would increase, even while conscription continued. Experience would show how responsive the number of volunteers is to the terms offered and how much these terms would have to be improved to attract enough men. As the number of volunteers increased, the lash of compulsion could fade away.

This picture is overdrawn in one important respect. Unless it is clear that conscription is definitely to be abolished in a reasonably short time, the armed services will not have sufficient incentive to improve their recruitment and personnel policies. They will be tempted to procrastinate, relying on the crutch of conscription. The real survival strength of conscription is that it eases the life of the military high command. Hence, it would be highly desirable to have a definite termination date set for conscription.

The case for abolishing conscription and recruiting our armed forces by voluntary methods seems to me overwhelming. One of the greatest advances in human freedom was the commutation of taxes in kind to taxes in money. We have reverted to a barbarous custom. It is past time that we regain our heritage.

[* ] Milton Friedman is Paul Snowden Russell Distinguished Service Professor of Economics at the University of Chicago, and past President of the American Economics Association. This article is an amplified text of his paper presented to the Conference on the Draft held at the University of Chicago in December 1966.

[1 ] See his article, pp. 13-16, this issue.

Last modified April 10, 2014