Committee on Economic Security (CES)

Volume I. Unemployment Compensation
Other Staff Reports
General Discussions

This overview of unemployment insurance by Professor Witte is the best and most-accessible summary of the issues involved in developing the U.S. system of unemployment compensation during the mid-1930s.


By Edwin E. Witte,
Executive Director of the President's Committee on Economic Security

February, 1935

On the verge of the nation wide adoption of unemployment insurance, there is a very widespread lack of definite understanding of the issues and objectives involved. Unemployment insurance has been in operation in England since 1911 and more recently has been adopted in most other Western European countries. In this country it has been the subject of much controversy for more than a decade, but outside of some twenty voluntary plans and the Wisconsin law, which came into operation only last July, there has been no actual experience with this institution. Within a limited circle, there is intense debate over details; outside of this small group, only a vague concept of what the implications are.

Place of Unemployment Compensation in a Program for Economic Security

For a better understanding of the problem, it seems appropriate to begin with the place of unemployment compensation in a comprehensive program for economic security. On this, as on all other questions relating to unemployment compensation, there is much disagreement, but there is little dispute over facts.

The present popularity of unemployment insurance is, to a considerable extent, due to the prevailing notion that it is a substitute for relief. This is a very erroneous impression. In no country has unemployment insurance replaced relief. In England only  half of the unemployed insured workers are now receiving unemployment insurance benefits. The others have exhausted their benefit rights and are being supported by public assistance grants or local poor relief. In Germany only one- eighth of the unemployed insured workers, according to the latest available figures, are in receipt of standard unemployment insurance benefits. Every plan for unemployment insurance in this country except the Lundeen Bill (which is so vague upon this point, as upon most others, that it is impossible to say what it does provide) applies only to workers who become unemployed after the system has been in operation for some time--usually two years after collection of contributions is begun. Unemployment insurance has little direct relation to the present depression. It will not lessen the relief load for several years, nor give people now on relief any increased allowances.

Every plan (except again possibly the Lundeen Bill, which is essentially a relief rather than an unemployment insurance measure) also strictly limits the periods during which benefits can be paid to unemployed workers. In England the benefits are flat sums varying with the number of dependents, but range between 30% and 40% of the average earnings of the industrial workers. These benefits are payable for a maximum period of 26 weeks, which is increased to a full year if the unemployed insured worker has not drawn any benefits for five years previously. In Germany standard benefits are paid for only six weeks, after which a means test is applied to determine eligibility for extended benefits. Actuaries of the Committee on Economic Security have estimated the probable operation of a national system of unemployment insurance in this country. Benefits of 50% of wages and not exceeding $15 per week, with a four week’s waiting period, can be paid for a maximum of only 16 weeks, if the contribution rate is 3% of payroll; for 22 weeks on a 4% contribution rate; and for 39 weeks on a 5% rate;--and these figures assume that the system should not be expected to remain entirely self- supporting during a prolonged depression.

During the present depression all countries with unemployment insurance systems increased contribution rates and decreased benefits and, in addition, most of them made loans or outright grants to the unemployment insurance funds. Where, as in Germany and Italy, they did not do so, the benefits were reduced to a point where little remains of unemployment insurance except the machinery. In all unemployment insurance bills in this country, authority is given for the reduction of benefits when the funds run low, and such reductions are inevitable in prolonged depressions unless the government comes to the rescue.

Unemployment insurance, then, should be regarded merely as a front line of defense against the hazard of unemployment. It is of value principally during the period just after the loss of a job when the worker has a reasonable expectation of returning to his old line of work within a short time. At this stage, the unemployed worker is, naturally, very reluctant to accept unaccustomed work which will remove him from immediate consideration for reemployment when hoped for new orders come in. The great majority of all workers who are laid off go back to work within the short periods during which unemployment compensation can be paid, but inevitably some workmen and more of them in depression than in normal times will still be without work when benefit rights are exhausted. For such workmen, something beyond unemployment compensation is needed.  That something more may be direct relief or work relief, both of which differ fundamentally from unemployment compensation in that they are based upon need rather than a contractual right.

Unemployment compensation and relief can be combined, but, as shown by England's experience from 1924 to 1931, the results are almost sure to be disastrous. Unemployment compensation works best when it is given as a matter of right in such limited amounts and for such limited periods as the receipts and the reserves warrant. These amounts and periods can be increased with higher contribution rates, but cannot possibly be made to cover all unemployment. Unemployment compensation is indispensable in a program for a reasonable degree of economic security, but should be conceived of as merely one of several measures needed for protection against the hazard of unemployment. It is complementary to, and not a substitute for, public assistance and public employment.

Federal Versus State Administration

Withal, unemployment compensation is indisputably worth while. Had a nation wide system of unemployment insurance with a 3% contribution rate been in operation in this country from 1922 on, above $10,000,000,000 would have been collected by 1933 and disbursed to unemployed workmen, mainly during the first years of the depression. Such disbursements would have had great value in maintaining purchasing power at a critical period, preserving morale, and lessening relief. Although not well adapted to many groups of workers, it is peculiarly valuable to the largest element in our entire population, the industrial workers who are ordinarily steadily employed.

The present is, moreover, clearly the opportune time for a real beginning with unemployment compensation. We are now, fortunately, on the up­swing of the business cycle. While there is still a terrifying amount of unemployment, the people who have jobs are no longer losing them in abnormal numbers. From the point of view of building up reserves, which is essential to a good start, no more favorable period will probably present itself for a long time. Politically, also, this is the time for action. Not only do we now have a national Administration which is earnestly desirous of establishing safeguards against the major hazards and vicissitudes of life, but public opinion is well prepared for such measures. When unemployment is less prevalent, it will be much more difficult to secure action on unemployment compensation. In the depression 1920 -21, President Harding named a commission to study methods of preventing unemployment and mitigating its evils. That commission made an excellent report, but not until the depression was over. There is good reason for believing that if the recommendations of this commission had been adopted some of the worst mistakes made in the present depression would have been avoided. As it was, the commission's work resulted in precisely nothing, because the favorable time for action had passed.

Conceding that unemployment compensation should be established on a nation wide basis now, what sort of an unemployment compensation system should be adopted? Ought we to have federal or state administration?

Prior to the New Deal, no advocate of unemployment insurance would have hesitated to answer: “Under the Constitution unemployment compensation must necessarily be established along state lines." The Democratic national platform of 1932 specifically committed the dominant party to “unemployment insurance and old age pensions through state legislation".

The New Deal has altered all this in the minds of some sincere advocates of unemployment insurance. While some of the support for a federally administered system of unemployment compensation represents merely the old game of opponents playing federal and state action against each other in the hope of defeating both, there can be no doubt of the sincerity of some of the recent converts to a federal system. They are people who have become dis­heartened because it is so very difficult to get even one state to act, let alone forty- eight states. The prestige of state governments is at a low point, while the present national Administration commands utmost confidence. Further, it is undeniable that a uniform, federally administered system would have great advantages both to corporations operating in several states and to employees whose work takes them into different states. The difficulty of providing unemployment compensation for such workmen under state systems of administration is the strongest argument for an exclusively federal system.

But there are also strong arguments on the other side. To begin with, despite the New Deal legislation, there is grave doubt whether the Federal government has authority to establish a national system of unemployment compensation. There is no constitutional basis for such action except the taxing power and the welfare clause. The latter has never been regarded by the Supreme Court as conferring any powers on the Congress not granted specifically, and the former, while broad, cannot be employed where the Court deems the tax a mere subterfuge for unauthorized regulation. Should the Agricultural Adjustment Act and the National Industrial Recovery Act be sustained, the Court's decisions may be brand enough to warrant setting up a federally administered unemployment insurance system (provided that they do not hinge upon the emergency character of these acts), but at present the only safe course is still state administration.

To many who have had experience with both federal and state administrations, moveover, the assumed vast superiority of federal over state action seems debatable. Federal personnel is on the whole undoubtedly superior and Federal salaries are distinctly higher. On the other hand, the very size of the Federal administrative machine is a great handicap. Its vast amount of red tape and its extreme centralization often paralyses action. There is probably no less friction between Federal Bureaus and departments as between states. The 73rd Congress was more ready to pass forward looking legislation than any state legislature has ever been, but that was probably due more to the extreme emergency confronting the nation than to the normal superiority of members of Congress over state legislators. In this connection it is not amiss to call attention to the fact that Congress did not enact a workmen's compensation act for the Government 's own employees or for the District of Columbia until ten years after all progressive states had taken such action.

What sort of an unemployment compensation act would emerge from the present Congress were it to attempt to write a complete law is entirely a matter of conjecture. The truth is that conditions are so diverse in this great country, that no single act could meet the situation unless it delegated broad discretionary powers to an administrative agency, and the present Congress has indicated that it is in no mood to follow the example set by the preceding Congress in this respect. In all probability a Federal unemployment compensation act would prove very disappointing to the more progressive industrial states, if for no other reason than that the non industrial states have the largest number of Senators.

These and still other reasons seem to throw the balance against an exclusively Federal system. An exclusively state system is even more out of the question. States cannot go it alone in this respect. The failure to secure state unemployment compensation laws is conclusive proof of this. While the subject has been agitated for years, many bills have been introduced in practically all legislatures, and numerous interim committees and commissions have reported favorably, only one state has passed an unemployment compensation law, and that, quite naturally, is a very meagre law. States will not enact unemployment compensation laws if it means placing their industries at a disadvantage in interstate competition. To make it possible for the states to act, the Federal government must protect them from the unfair advantage enjoyed by states that do not act. That can be accomplished only through a uniform tax on employers in all states, with an offset or refund of the tax to states which have set up unemployment compensation systems.

The President has characterized the desirable relationship of the Federal and state governments with respect to unemployment compensation as “a cooperative Federal- state system”--a New Deal in government, in which there is not a conflict over the respective rights of the Federal and state governments, but a joint attack on the grave problems of insecurity. In such a cooperative Federal -state system it is clearly the function of the Federal government to make it possible for the states to act, to encourage them to do so, and to help them with their problems. It alone can equalize competitive costs, and it alone can invest and liquidate reserve funds in such a manner that they will promote industrial stability, rather than the reverse. It must also assume responsibility for the compilation of nation wide statistics of unemployment and give guidance to the states in the difficult problems of administration. Beyond that, wide latitude to the states in meeting their peculiar conditions seems desirable. By this means both uniformity, where it is essential, and diversity, where necessary, can be secured.

The Committee on Economic Security, its staff, and the two major advisory groups, the Technical Board arid the Advisory Council, gave much attention to two other functions which the Federal government may very properly perform in a cooperative Federal- state system of unemployment compensation. These are reinsurance and transfer of compensation rights where workers move from one state to another. The desirability of such arrangements is at once obvious, but the difficulties in working them out practically are so great that every group which gave them consideration concluded unanimously that much further study will be necessary before anything along this line can be included, in the Federal Act. Such study should be one of the major duties of the Federal administrative agency.

Tax Offset Versus Tax Refund

Accepting the view that a cooperative Federal -state system of unemployment compensation is best and safest under existing conditions, considerable controversy has developed, within the circle of people who have been employed or consulted by the President's Committee on Economic Security, over the type of Federal law which should be enacted. This is the controversy which within this inner circle has been referred to as the "Wagner Lewis versus the subsidy plan."

The "Wagner -Lewis" plan was incorporated in the unemployment insurance bill endorsed by the President in the 73rd Congress and is also included in the Administrations Economic Security Bill in the present session. It provides for the levy of a payroll tax on all employers subject to the act, with a credit (in the present bill up to 90% of the Federal tax) for the contributions paid by any employer during the taxable year to the proper state authorities under the provisions of state unemployment compensation acts. When first announced it was hailed by all advocates of unemployment insurance (to quote Professor Paul H. Douglas) as "a truly brilliant method" and as representing "federalism at its best". Now, however, many of its former champions, along with the opponents of unemployment insurance, are very critical of this method.

The explanation of this change in opinion is the suggestion in the interim of a new method for inducing the states to enact unemployment insurance laws. This method has been called the “subsidy" or "grant- in- aid" plan, but is more accurately described as a tax return plan. It proposes that the Federal government shall levy a payroll tax on all employers throughout the country and then return to states which enact approved unemployment insurance laws all or a stated percentage of the payroll taxes collected from the employers within these states.

This is not a subsidy or grant- in -aid as these terms have been understood heretofore. No prior subsidy has ever been based upon the taxes collected from the state to which the grant- in- aid is made. Rather than a subsidy this is an earmarked tax which is returned to the states from which collected. The only essential difference between, this plan and the "Wagner -Lewis” proposal lies in the collection of the payroll tax. Under the present Wagner -Doughton -Lewis bill the states collect the contributions required from employers for unemployment insurance purposes and than deposit the moneys collected in the United States Treasury, to the credit of the state. Under the “subsidy” plan the payroll tax will be collected by the Federal government and the moneys from a given state credited to the account of that state in the United States Treasury.  Aside from this difference in the way in which the moneys credited to each state in its unemployment insurance account reach the United States Treasury, there is nothing that can be done under the “subsidy" system that cannot be done under the Wagner -Lewis plan, or vice versa.

The advocates of the "subsidy" plan have advanced many reasons for this preference.  Many of them believe that it is but a stepping stone to a national system of unemployment insurance; in fact, one of its leading advocates always refers to it as a "national plan.” Since under this plan as well as the other each state will have its own unemployment insurance system, it clearly does not eliminate the greatest weakness of state systems, the difficulties arising out of the shifting of workmen from one state to another. Under this plan, each state will have its own law and its separate fund, which presumably would lead to the same resistance against the subsequent establishment of a national system as might arise under the “Wagner- Doughton- Lewis” bill.

Many advocates of the "subsidy" plan believe that it lends itself more readily to the insertion of numerous standards in the Federal act with which the state laws will have to comply, and that this will bring about a desirable uniformity in these laws. Whether or not this is a valid argument is certainly debatable. Keen students of Federal legislation have noted the very strong feeling in this country that the Federal government should give aid to the states without trying to dictate to them. This sentiment seems to be particularly strong in the present Congress, and it is not at all improbable that a bill based on the "subsidy" plan would wind up with fewer standards than are included in the Wagner- Doughton- Lewis bill.

There also are arguments that the "subsidy" plan is sounder constitutionally. As neither the “Wagner- Lewis” or the “subsidy" plan has come before the courts, this argument is clearly speculative. The "Wagner -Lewis” device has in its favor a direct precedent in the unanimous decision of the United States Supreme Court upholding the provision in the Federal estates tax law allowing credit up to 80% of the Federal tax for the amount paid to the states under their inheritance tax laws. On the other hand, it has been held that a taxpayer cannot question a subsidy payment made out of general revenues. Both methods of securing state action give rise to the same question, whether the payroll tax is in fact a tax or a disguised method of regulation. Both are probably valid, and attorneys differ as to which is safest.

This leaves the more direct method of getting the unemployment compensation funds into the United States Treasury as the chief advantage of the “subsidy” system. There appear to be no constitutional obstacles in any state to the deposit of moneys collected for unemployment compensation purposes in the United States Treasury, but it must be conceded that the Wagner -Doughton-Lewis bill requires two operations to got these moneys into the Treasury and the "subsidy" plan only one.

The strongest arguments on the other side are that the "subsidy" plan would leave the states entirely dependent upon the Federal grants, since their unemployment compensation laws would not include any revenue raising features. There would, thus, be danger that the entire structure of unemployment insurance in this country might be wrecked through an adverse court decision or the failure of Congress to renew its annual appropriation for this purpose.

Another consideration which carried great weight with Administration and Congressional leaders in their decision in favor of the "Wagner- Lewis" rather than the "subsidy" plan was the use of the term "subsidy" to designate the tax return method. While it does not contemplate a subsidy from general revenues, there is danger that the Federal grants will come to be regarded as such and that there will be great pressure upon the Federal Government to come to the rescue of states when they fall into difficulties. If the “subsidy" plan is accompanied by requirements for high uniform minimum benefits throughout the country, as most of its advocates urge, then many states will certainly get into serious difficulties. The "subsidy" plan, consequently, may turn out to be a most dangerous device from the point of view of the United States Treasury. It presents the dangers of a system under which the United States government may be called upon to underwrite the state funds, without having any control over them.

Quite frankly, these arguments against the "subsidy" plan are matters of judgment and opinion as are the arguments in its favor. The truth seems to be that the differences between those methods of bringing about state action have been exaggerated. Either method will accomplish the purpose of removing the disadvantages to which states at present must subject their employers if they enact unemployment compensation laws.

At this point, it should be noted that a subsidy of the usual kind will not accomplish this cost essential purpose. Suggestions for such a subsidy from general tax revenues have been made from time to time and are now being renewed. Such a subsidy, if of a sizeable amount, would, it is true, operate as an inducement to the state legislatures to enact unemployment insurance laws, but would still have the employers in states which have no such laws. It is only through a uniform payroll tax that this greatest obstacle to state action can be removed.

Source of Funds

The suggestion that a subsidy from general revenues should be used to induce the states to enact unemployment compensation laws comes at this time from members of Congress who object to a payroll assessment against employers as a sales tax. This is also the argument against the pending Economic Security Bill used by advocates of the Lundeen Bill, which has a surprisingly large support despite its extreme vagueness. According to this view, the costs of all unemployment compensation should be met from income and inheritance taxes, rather than from contributions of employers and employees.

This argument has a strong appeal to liberals who do not like sales taxes. But is a contribution by employers to an unemployment insurance fund really a sales tax? In what essential does it differ from accident (workmen’s) compensation? Certainly, no one proposes that accident compensation shall be paid from income or inheritance taxes, any more than that the wage bill shall be so paid. Instead, accident compensation is regarded as a proper charge to be included in the cost of production, to be passed on to the consumers. Unemployment compensation, likewise, should be so regarded, particularly when it is limited to the period following the loss of a job while the employee has a good chance of returning to work soon and often is encouraged to wait for his old job by the employer himself.

Many, perhaps most, advocates of unemployment insurance who do not sympathize with the view that unemployment compensation should be paid for entirely from income and inheritance taxes, nevertheless believe that the government should make large contributions to the unemployment insurance funds. In England there is a tri party arrangement, the government, the employers and the employees all contributing equally. A similar arrangement is strongly urged by many of the dyed- in- the -wool advocates of social insurance in this country.

If unemployment insurance is considered a form of relief, there is much to be said for government contributions. If it is set up merely as a first line of defense, the argument for government contributions becomes much less strong. Under such a concept, government contributions should be used to provide work or public assistance to those whose unemployment benefit rights have been exhausted or who are excluded from the system entirely, rather than to increase unemployment compensation benefits. Particularly at this time when 10,000,000 workers are without employment, there is every reason to urge that whatever funds can be provided from general revenues should be used to give work and assistance to the unemployed, rather than to be paid into reserve funds from which benefits will be paid some years from now to those who may be out of work at that time.

This is the position taken by the Committee on Economic Security and the President. The pending Economic Security Bill is a companion measure to the four billion dollar appropriation for public work. While the Economic Security Bill carries no government contribution to unemployment compensation, the Administration program contemplates the largest contribution ever made from general revenues toward meeting the costs of unemployment. In contrast with the proposed four billion dollar appropriation for public work, plus eight hundred million more for relief prior to July 1, 1935, the British government contributed and loaned to the unemployment insurance fund a total of 355 million pounds in the fourteen years ending in March 1934, considerably less than one- half the contemplated expenditure from general revenues in this country in one year to provide work for the unemployed. Certainly, the American government cannot fairly be charged with being penurious if it prefers to use its funds to aid those now without employment, rather then to contribute to reserve funds to pay unemployment compensation in the future.

Apart from the government, the possible sources of funds for unemployment compensation are the employers and the employees. Except for those who hold that unemployment compensation should be paid for entirely from income and inheritance taxes, there are none who would excuse the employers from contributing. With reference to employer contributions there is no serious question other than how high the contribution rate should go.

On first thought it might appear that it makes little difference how high a rate employers are compelled to pay, so long as all must pay the same rate. By and large, the employers shift their contributions to the consumers; but they cannot always do so. A high rate may seriously curtail consumption and also has some tendency to encourage the replacement of labor by machinery. At this time some consideration may very properly be given to the fact that recovery is still incomplete and that industry cannot stand as high a contribution rate as in more normal times. In England the employer contributions average about 1 ½% of payroll; in Germany, about 3%. In this country, the Wisconsin act provides for a 2% contribution. The Wagner -Lewis bill as introduced in the Seventy- third Congress provided for a 5% rate, but, as revised by the Ways and Means Committee, for only a 3% rate. This is the highest employer contribution rate proposed in any state bill. How high the rate shall be, however, is largely a matter of judgment, to be decided by practical considerations, rather than abstract theories.

Regarding employee contributions, the main issue is whether or not employees should be required to contribute at all. Most of those who advocate employee contribution believe that it is immaterial how much the employees contribute so long as they are required to contribute some part of the costs. Those who hold this view argue that if the employees contribute they will have a greater interest in conserving the reserve funds and also in resisting unreasonable increases in benefits, since, presumably, increases in the contribution rates of employers would be paralleled by increases in the contributions of employees. Further, it is claimed that employee contributions are needed to assure the employees a real voice in the administration of unemployment compensation.

On the other hand it is urged that employee contributions are a tax on wages, imposed at a flat rate and without exemption even of the lowest paid wage earners, whose income is often insufficient. Moreover, such a tax on wages unlike the payroll tax on employers cannot be shifted to the consumers. In fact, the wage earners, as consumers, pay part of the tax imposed on employers and, with employee contributions, would have to pay still another tax.

The Committee on Economic Security and the pending Administration bill leave the question of employee contributions entirely to the state. No employee contributions are required in the Federal act, but the states are left free to add employee contributions, if they wish. Through employee contributions the benefits that can be paid can be increased. There may also be something in the argument that employee contributions will give labor a better claim to participation in the administration of the system and an interest in conserving the funds. On the other hand, employee contributions certainly are a tax on those least able to pay and represent a departure from traditional principles in this respect.


Who should be brought under unemployment compensation? This is a practical question to which no dogmatic answer can be given. The natural course for people who believe in unemployment compensation will be to include everyone who can possibly be included. Administrative difficulties, and above all, the obvious fact that if some groups of employees draw an undue share of the total benefits others will receive less than they should, are, however, countervailing considerations which render it desirable not to attempt maximum possible coverage, at least at the outset.

In European countries there are no numerical limitations in the social insurance laws such as are included in all but one of the American workmen's compensation acts. Such numerical limitations are illogical but are certain to be urged on both traditional and practical grounds. They greatly reduce the number of employees included within the scope of the law. More than one half of all manufacturing establishments have five or less employees, but employ only three per cent of employees engaged in manufacturing; eighty- five per cent of all retail establishments fall in this class but have only twenty-five per cent of the employees in this industry.

Traditionally in this country, agriculture, domestic service, and employment "not in the usual course of the trade, business, or profession of the employer" are exempted not only from the workmen's compensation laws but from practically all labor laws. Agriculture has been exempted from the British unemployment insurance act until now, when, finally, a special compensation scheme for agricultural workers is under consideration. Inclusion of these groups of workers is difficult administratively and, for short time and intermittent workers, is of relatively little value, since they seldom accumulate any substantial benefit rights. Unquestionably, it is desirable to include commercial agriculture and large domestic establishments, but as realists we must recognize that legislative bodies will probably prefer the traditional complete exemption.

In the Wisconsin law, regular part time workers are excluded. This was an exemption strongly opposed by the champions of this legislation and is without any justification; in fact, a difficult administrative problem is created through exclusion, while none arises through inclusion. Part time workers should, of course, be compensated only on the basis of the wage loss they actually sustain, rather than on the basis of full time work; but to deny them any protection is both unjust and troublesome.

With regard to seasonal industries, also, there is sure to be considerable demand for exemption. As British experience has demonstrated, there is danger that employees in such industries will draw a very undue share of the total money available for unemployment compensation unless the special restriction is introduced that only unemployment occurring within the actual season of work can be compensated. With such a provision, however, there is no good reason for excluding seasonal industries.

Quite often religious, charitable, and educational institutions are likewise exempted. Aside from tradition there is no good reason for such an exemption, and it seems little short of hypocritical that organizations which have waged the enactment of social legislation, as have many of these groups, should themselves ask to be excluded. To their credit be it said that but few such organizations are asking for an exemption at this time.

In any Federal law the employees of the states and their political subdivisions must necessarily be excluded, since the Federal government cannot tax the states. It is utterly illogical that governments should impose obligations on private employers that they will not themselves assume, but the inclusion of any public employees must necessarily be left to the states.


The benefits which can be paid, manifestly, depend primarily upon the rates of contribution. The higher the contribution rates, the more liberal the benefits can be. There are some important questions regarding benefits, however, which are independent of the contribution rate.

A fundamental question is whether the benefits should be designed to cover primarily short time or depressional unemployment. Either of these purposes is quite logical. If the former is deemed the major objective, a shorter waiting period and a longer maximum duration of benefits are warranted than if the system is primarily designed for depressional unemployment. In either case, the benefit rate should be high enough to sustain the unemployed worker and his family while benefits are being paid. In this country, benefits of less than fifty per cent of the wage are not likely to be satisfactory, but an absolute maximum limitation of possibly $15 per week is justified. All states also needs a minimum benefit, which obviously should vary from state to state, to conform with differences in costs of living and economic standards.

Available data indicates very great differences in the past unemployment rates of the several states, the range in the average rates for the years 1930 to 1934 being from a minimum of seventeen per cent to a maximum of above thirty- four per cent. It may be that over the entire business cycle these extremes would be less pronounced, but it would seem clear that benefit rates cannot be the same everywhere, unless the contribution rates are higher in the states with greater unemployment or the government is willing to underwrite the funds. The most comprehensive calculations of the benefits which can be paid are those which were made by the statistical staff of the Committee on Economic Security, based on the average unemployment throughout the country in the twenties and in the depression period. Most states would seem to be reasonably safe if they base their benefit schedules on the experience throughout the country in the years 1922 to 1931, as indicated in the Report of the Committee on Economic Security; but states with very unstable industries should either increase their contribution rates or reduce their maximum benefit periods. While uniformity in benefit rates throughout the country is desirable from the workers’ point of view, any attempt to impose such uniformity from Washington will almost surely result in the insolvency of many state funds.

Benefits for partial unemployment are a special problem which the authors of any unemployment insurance legislation must face. Such partial unemployment results most frequently from a policy of spreading the work, through reducing the hours of labor when orders diminish. This policy has always been advocated by organized labor, but only within limits. It involves a wage loss for which many workmen will feel that they should be compensated. To compensate all partial unemployment, however, may result in encouraging industries to maintain a large surplus labor supply, through using unemployment compensation to supplement inadequate wages. This has occurred in England and should be avoided in this country.  At the other extreme a policy may be adopted under which no compensation is paid for partial unemployment until the earnings of the workers are reduced to less than the amount they can receive as compensation for total unemployment. This is objected to on the grounds that it gives the workers nothing more if they work than if they do not do so. To overcome this objection and yet not throw the doors wide open to such abuses as have developed in England, it has been suggested that partial unemployment be compensated only to the extent that the wage loss exceeds a stated percentage which is slightly higher than the percentage of the wage paid as compensation for total unemployment (i.e., if the compensation rate is 50 per cent of the average wage, only partial unemployment causing a wage loss in excess of say 40 per cent is compensated).

Pooled Funds Versus Industry and Plant Accounts

The most heated controversy which has developed in this entire field concerns individual industry versus plant accounts. In European unemployment insurance laws all contributions are placed in a central pooled fund, from which benefits are paid without regard to the employer for whom the unemployed person last worked. (It is to be noted, however, that in England the banking and insurance industries have separate funds and that in Belgium, while there is a central pooled fund, the contribution rates vary by industries in accordance with their respective unemployment rates.) In this country, most, but by no means all, of the advocates of unemployment compensation strongly favor pooled funds on the European model; many of them refuse to recognize plant funds or accounts as being insurance at all. On the other hand, most employers who know anything about the subject strongly favor separate plant funds, or "unemployment reserves" as distinguished from "unemployment insurance."

The employers' preference for unemployment reserves is based mainly upon two considerations: (1) their desire to deal directly with their employees, and (2) their strong feeling that they should not be asked to pay for some other employer’s unemployment. The principal argument advanced for permitting employers to have such separate accounts is that when they must pay for their own unemployment they will keep unemployment down to the unavoidable minimum.

Organized labor, at first uncertain as to its preference, is now definitely committed in favor of pooled unemployment insurance funds. Its objections to separate plant funds are also two fold: (1) a belief that they are essentially "a company union device," in as much as they tie the employees closer to the company, and (2) a fear that many workmen will, under such a system, fail to receive benefits when unemployed, because the fund of their particular employer will have been exhausted.

It is difficult to appraise impartially these opposing claims because there has been little experience to test their validity. That the separate plant account system should furnish some incentive to reduce unemployment would seem unanswerable, but how effective this will prove, no one can be certain. How much employers can do to stabilize their employment will vary in different industries and, manifestly, many causes of unemployment are beyond the control of the individual employer. The relatively small number of employers in this country who have voluntarily set up unemployment reserve systems, on the whole, have not done anything so very striking in stabilizing employment, although some of them have a fine record in this respect. There is much reason for the view that what employers can do most effectively to keep down their unemployment compensation costs under a plant account system is not stabilization of their employment but distribution of available work among all of their employees, when slack times come. But upon this point also the evidence from actual experience is scanty.

Opponents of individual industry and plant accounts say that this is a system which attempts to differentiate between good and bad employers and to reward those who have a stable employment record. Besides this, however, there also is the factor that the risk of unemployment differs greatly between industries. Industries like building construction, coal mining and the capital goods industries generally have much more unemployment than, for instance, retail trade, public utilities, and the consumer goods industries. In all other lines of insurance there is some attempt to adjust the rates to the risk and this seems but equitable also in unemployment insurance. Such an adaptation of the rates to the inherent risk is possible under a pooled state fund system but not automatic, as under a system in which each employer bears the costs of his own unemployment. That a plant account is of advantage to employers with better than average stability of employment is obvious. But what of the employee? Clearly an individual account system is disadvantageous to the employee when he leaves one establishment to enter the employ of another unless he can take his unused benefit rights with him, which is difficult to work out. The most serious question of all, however, concerns the danger of the inadequacy of many plant funds. A single establishment cannot have the advantages of average exposure, which is fundamental in all true insurance. Under a system such as has been set up in Wisconsin, where each employer, however small, has his own account, it seems certain that many of these accounts will be exhausted when their employees most need protection. Of course, pooled state funds may also be exhausted, and if such funds become insolvent all employees in the state will suffer, while under the plant account system some employers, at least, will be able to fully meet all obligations. If plant accounts are allowed only to employers who furnish adequate security to guarantee payment of all promised benefits, all arguments directed against such individual accounts on the score of the inadequacy of the protection afforded would seem to be completely answered. There would still remain, however, the fact that the employers permitted to have suc. accounts would probably be the ones having less than average unemployment and their withdrawal from the central pooled fund would lessen its chances of remaining solvent.

Unemployment reserves, as distinguished from unemployment insurance, were taken up by advocates of unemployment compensation when the British unemployment insurance system was generally believed in this country to be a failure. Now that the British system is regarded as a great success, sentiment has swung very strongly in favor of pooled state funds, although most employers continue to favor reserves systems. The question has become whether any state should be permitted to establish a plant account system of unemployment compensation or allow any establishments or industries to contract out of the central pooled funds.

The Wagner -Lewis bill of the 73rd Congress permitted states complete free choice in these matters. The Committee on Economic Security and the pending Economic Security Bill permit states to set up an individual plant account system of the Wisconsin type, or to allow "contracting out" from pooled funds, or variations in the rates of contributions to pooled funds, but only under drastic restrictions. On all such cases the employers given such privileges must pay at least 1% on their payroll into the pooled state fund and they cannot reduce the contributions credited to their own accounts until they have built up reserves of at least 15% of their payroll (which few of them will be able to do in short of ten years). These provisions are of a compromise character and have been assailed, on the one hand, because they do not definitely outlaw plant funds and, on the other, because the restrictions are so drastic that employers will lack all incentive to reduce unemployment.

Guaranteed Employment

Raising some of the same questions which are involved in plant accounts, are the contractual arrangements between employers and employees known as “guaranteed employment." In such arrangements employers contract to furnish their employees a specified number of weeks of employment during the year and can legally be held for payment of the full wages for the guaranteed amount of work. Such contracts are quite common in England and have been experimented within a number of voluntary unemployment compensation plans in this country and in the Wisconsin law. Persons interested in continuing these experiments demand that the Federal act recognize such guaranteed employment contracts and exempt the employers from making any contributions to central unemployment compensation funds for any employees thus protected.

Guaranteed employment is clearly of advantage to the employee if they are guaranteed sufficient work to give them larger earnings than they would probably get without such a guarantee. Guaranteed employment, however, could also be used as a device to cheat employees out of their unemployment compensation rights. It is only when the guarantee is required to cover most of the possible working period and some provision is made for compensation to employees whose contracts are not renewed and who lose their jobs, that adequate safeguards are provided to make certain that guaranteed employment will be truly to the benefit of the employees.


The final major question arising in unemployment compensation concerns administration. On this subject there is no clash of interests between employers and employes, but a problem of keeping the administration as far removed from politics as possible.

The actual administration must be placed in the local employment offices.  Only involuntary unemployment is to be compensated. World experience has demonstrated that the best way to test willingness to work is through requiring the unemployed workman to register at the nearest employment office and to keep constantly in touch with it. It is through the employment offices everywhere that unemployment compensation claims are passed upon in the first instance and benefits paid. To this end a nation wide, efficient public employment office service, as contemplated in the Wagner- Peyser Act is an absolute essential.

Besides the local employment offices there must be a central state office through which contributions are collected and at which records are kept. A procedure for the settlement of disputed claims is also essential. Very helpful should prove advisory committees representative of employers and employees.

On a cooperative federal state system, there also must be a federal administrative agency. This federal agency must pass upon whether the state laws meet the prescribed requirements to entitle employers in these states to an offset against the federal payroll tax. It should also collect statistics upon unemployment experience throughout the country, so that rates may be established on a more scientific basis. It should assist the states in setting up unemployment compensation systems and in developing methods for dealing with such difficult problems as interstate transfer of employees and reinsurance.

Regarding the federal administrative agency the only serious question of policy is whether this should be an independent agency or attached to the Department of Labor. Every argument would seem to favor the latter arrangement. Not only is there now strong sentiment against the creation of any additional permanent independent federal agencies, but the administration of unemployment compensation is so inextricably interwoven with that of the employment offices that the United States Employment Service would have to be transferred from the Department of Labor to this agency. As the Department of Labor is the smallest of the cabinet departments, any such step would be most illogical, moreover, unemployment compensation is, clearly, a labor problem.

Back to Volume One Table of Contents