United States Congressional Serial Set

II. BENEFIT ASSUMPTIONS

WEEKLY BENEFIT AMOUNT

Several facts have led the Council to conclude that existing benefit levels are on the average too low for estimating future costs. The facts are:

1. The average weekly benefit amount is now only about 35 percent of the average weekly wage; in the second quarter of 1947 it was less than 30 percent in eight States.

2. Even the maximum weekly benefit amount now ranges among the States from 35 to 59 percent of the average weekly wage, with 31 States in the 35 to 45 percent interval.

3. In 1947 more than half the benefit payments (57 percent) were at the maximum weekly benefit amount payable under the State laws; in eight States the proportion limited by the maximum exceeded 70 percent.

4. Increases in the cost of living have so greatly reduced the purchasing power of benefits that the average weekly benefit of $19.28 in July 1948 was worth only $11.11 in terms of 1935-39 dollars.

5. Even the present maximum weekly benefit amount would meet only 56.2 to 69.4 percent of the nondeferrable costs of living (49 to 53 percent of a total budget for family requirements) for a family of 4 in the 22 cities surveyed in June 1947, and the range among all 34 cities studied was from 48.9 to 86.4 percent.1

In order to determine the proper minimum rate over the next 10 years, it seemed prudent, on the basis of these facts, to assume for estimating purposes a higher level of benefits than now prevails in most States. The Council therefore assumed two sets of benefit conditions. The first set of assumed conditions is about equivalent to the provisions in the States with the most liberal benefits. These conditions assume weekly benefits equal, on the average, to at least 50 percent of previous weekly earnings up to a maximum benefit of $25 a week and a uniform duration of 26 weeks.

The second set of benefit assumptions used by the Council provides for a somewhat higher level of benefits. The cost estimates are projected over a 10-year cycle and it is reasonable to assume that benefits will rise during this period as they have during the past 10 years. In this second set of conditions, the Council assumed weekly benefits equal, on the average, to 50 percent of previous weekly earnings calculated on wages up to $80 a week.

There are many sets of benefit conditions, of course, which would result in approximately the same costs and any one of them would do equally well for the purpose of these estimates. Instead of a flatrate of 50 percent of weekly earnings up to $80 a week, a State might use a formula which would permit claimants with less than average wages to receive somewhat more than 50 percent, and those with greater than average incomes to receive somewhat less. One such formula resulting in approximately the same costs as the above formula is 60 percent of the first $25 of weekly wages plus 40 percent of the next $55. One formula with dependents' allowances resulting in approximately the same costs as the above formulas is 60 percent of

See Unemployment Benefits, Wages, and Living Costs, Social Security Bulletin, April 1948, pp. 3-9. 83404-4914

The Council recommends the minimum rate of 1.2 percent for benefit purposes because this rate seems to be applicable to the majority of the States under both benefit assumptions and both the favorable and unfavorable economic assumptions. We might have suggested a higher rate that would have covered the costs of even the highest-cost State, but this approach was rejected because it would require many States to collect more than they needed for an adequate level of benefits. Similarly, the Council might have proposed a much lower rate that would have covered the costs only in the lowest-cost States; but this approach was rejected because it would not accomplish the Council's purpose of reducing interstate competition for lower contribution rates. With such a minimum rate, most States would still be in the position of having to decide whether they would provide more liberal benefits or reduce the contribution rate to the minimum. The Council believes that the rate of 1.2 percent will avoid interstate competition in contribution rates among most States, but again reiterates the fact that, under its cost assumptions, a few States will have to charge more than the minimum rate, and that all States, under the State-Federal system, must be responsible for providing adequate contribution rates and benefits in relation to their own experience.

These estimates do not undertake to indicate what unemployment insurance will cost in the individual States over the next 10 years or what rates particular States should charge. Much more detailed study on an individual State basis would be needed before conclusions of this type could be reached. The estimates for the individual States are rough calculations based on their past benefit experience (the war years, 1942-44, were not considered in these estimates), and future benefit experience in many States will probably differ from past experience. The estimates do, however, give a basis for establishing a national minimum rate; for this purpose it is not necessary that the costs in each State be accurately predicted as long as the general picture is reasonably correct.

I. ECONOMIC ASSUMPTIONS

de

Benefit costs for a specific unemployment insurance program pend primarily upon the economic conditions prevailing during the period under consideration.

In order to determine costs over a complete business cycle, the du ration of the cycle must be established. If estimates are projected for only 3 or 4 years ahead they cannot adequately take account of a relatively severe decline, with unemployment reaching 5, 8, or 10 million, and subsequent return to predepression levels of business activity. On the other hand, it would be impractical to plan the financial structure of an unemployment insurance program too many years ahead. In view of these considerations, therefore, variations in economic activity over a 10-year period were considered. Ten years was deemed long enough to encompass anticipated varia tions in economic activity but not too long for practical purposes of planning.

To estimate costs over a business cycle, three basic assumptions need be established: (a) a high level of employment at the beginning and end of the cycle; (b) employment declining in the early phase of the cycle and increasing in its later phase; and (c) the range in the volume

of unemployment. The precise shape of the pattern does not significantly affect the size of the estimates. The slope may be irregular and the trough shifted to the left or right without affecting costs. It is important only that there be peak levels of employment at the beginning and end of the cycle and a specified range of variation in unemployment over the period. Detailed differences during the course of a business cycle tend to average out over the cycle.

A. FAVORABLE PATTERN OF EMPLOYMENT

One set of cost estimates was based on the assumption that unemployment during the next 10 years would vary from 2 to 5 million as follows:

B. UNFAVORABLE PATTERN OF EMPLOYMENT

It is possible that estimated unemployment of 5 million at the trough of the business cycle might prove to be over-optimistic. Another set of estimates was therefore prepared based on the assumption that unemployment would range from 2 to 10 million during the course of the business cycle. In the 2 to 10 million cycle, unemployment was assumed to vary in the following manner:

Unemployment insurance, as it operates in all States, compensates the highest proportion of unemployed workers during peak levels of employment and the initial stages of an economic set-back. As the depression deepens, a growing proportion of unemployed workers exhaust their benefit rights and find it difficult or impossible to get new jobs. During the later stages of a depression, although the absolute number of unemployed may be large, the percentage of the unemployed receiving benefits is much smaller than in the early stages. A fairly rigid demarcation develops among the unemployed between

workers in the turn-over group who stand a good or reasonable chance of finding a job, and those in the hard-core group who have relatively little chance of reemployment during the depression.

The cost estimates under both economic patterns were based on the assumption that turn-over among covered workers during periods of peak employment would average 2 to 3 percent of covered employment per month. This turn-over pattern is indicated by data on initial claims and covered employment reported by the State employment security agencies.

The turn-over group consists in large part of workers out of a job because of frictional factors in the economy that are prevalent in both good times and bad. Even if the workers in the turn-over group had as good chances of finding employment during the depression as during peak business activity, however, the emergence of the hardcore in a depression with almost no chances of finding a job tends to reduce the hiring prospects of unemployed workers taken as a whole. As a result, turn-over tends to decline during a depression. This phenomenon was taken into account in the preparation of the cost estimates.

An even more unfavorable pattern than either of those assumed, with unemployment rising to as much as 13,000,000, would raise costs on the average by perhaps 5 to 10 percent. These higher costs would result mainly from the increased number of initial layoffs averaged over the 10-year period, but also to a lesser extent from the longer duration of compensated unemployment. It is significant, however, that even extreme assumptions for the volume of unemploy ment do not increase costs substantially. Since unemployment benefits are paid for a limited duration and since eligibility depends upon recent earnings, the effect of large-scale unemployment on the costs of the system is limited.

Some consideration was given to the possibility that employers might rotate jobs by hiring workers as they exhaust benefit rights and laying off others as they gain eligibility for benefits. If this type of share-the-work were widespread, it would increase costs considerably. Because of seniority rules and employment practices, however, the extent of this type of job rotation is likely to be slight. On the other hand, the more normal share-the-work practice of reducing the number of hours worked per week would tend to reduce benefit costs. The cost estimates were based on the assumption that these contrary tendencies would about cancel out and that share-the-work practices would not affect benefit costs.

D. LABOR FORCE

Under both economic patterns, the labor force was assumed to increase at an average of 600,000 a year over the 10 years. At present, the labor force is growing at a rate of more than a million a year. Such growth, however, is unusual during peacetime and is probably attributable to the prevailing boom conditions. As conditions become more stable, the growth in size of labor force will probably tend toward the long-run average of 1 percent per year. About 1.2 million people will probably reach working age each year, while slightly more than half a million will leave the labor market because of age, infirmity, marriage, or death. During the past 12 months, the labor force has been averaging about 62 million.