allowing review of the insurance available to older workers.

In performing the analysis, life insurance benefits were projected under the provisions of each insurance plan for employees at various assumed annual salary levels and lengths of service. Benefits were computed for an employee in mid-career (for example, age 40) and for older employees.

The same assumptions were applied to all three occupational groups studied, even though some of the salary levels would not be widely applicable in each group. That is, it is not likely that many production and service workers had a salary as high as $55,000, nor is it likely that many professional and administrative workers had a salary as low as $15,000 or $20,000, in 1988. Because benefit formulas may be designed for a specific group of workers having a known range of earnings, benefits shown at these unlikely earnings levels may not be meaningful. Hence, in examining the results of this analysis, one should focus on benefits at earnings levels that are appropriate for a particular occupational group.

Benefit levels

Table 1 shows the average life insurance amounts at the length-of-service and salary levels studied. In each occupational group, the benefit amount increased only slightly with service, yet rose significantly as salary increased. This is expected, as plans frequently base benefits on earnings and rarely on length of service.3 White-collar workers had the greater average benefit available at all salary levels, with the disparity widening with increasing annual salary. Thus, at $15,000, white-collar benefits were 44 percent higher than blue-collar benefits, while at $35,000, they were 55 percent higher.

Average life insurance amounts for whitecollar workers were more sensitive to salary changes than were those for blue-collar workers. For example, when salaries of white-collar workers increased 80 percent, from $25,000 to $45,000, average insurance benefits increased 60 percent. For blue-collar workers, the increase was 50 percent over the same salary range. The analysis for blue-collar workers in the upper salary ranges, though, may be skewed due to the aforementioned assumptions regarding the inapplicability of higher earnings to this occupational group. Over the lower applicable salary range of $15,000 to $25,000, when salary increased 67 percent, insurance increased 44 percent. In any event, one would expect greater sensitivity of white-collar workers' in

surance to salary changes because in 1988 nearly 80 percent of the white-collar participants in medium-sized and large firms had life insurance tied to earnings, compared with 50 percent of the blue-collar participants.

With life insurance benefits expressed as a percent of employees' annual salaries, average benefits for white-collar participants were always greater than annual salary, while for bluecollar participants that was true only at the lower salary levels. The following tabulation presents projected life insurance benefits as a percent of annual salary at 10 years of service:

Participants $15,000 $25,000 $55,000

$50,000-$59,999

3

20

Annual salary

As shown in table 2, dollar amounts of protection at any one salary level varied widely among the individual life insurance plans in the survey. Nevertheless, clusterings are apparent, reflecting the prominence of plans paying benefits equal to the annual salary or flat amounts such as $5,000, $10,000, and $20,000.

Life insurance for older workers

The Age Discrimination in Employment Act prohibits employers from discriminating against any person with respect to hiring, compensation, or privileges of employment based on the person's age. Originally, the Act protected individuals between ages 40 and 65, but as amended, it now applies to all employees 40 years of age or older.

One effect of the Age Discrimination in Employment Act is to ban mandatory retirement. Because of this, employees may choose to continue working past typical retirement age. For such employees, the cost of employersponsored life insurance may continue to increase, as the life expectancy of older workers declines. To compensate for this added cost, many employers have reduced the amount of life insurance protection afforded these workers. 5

Life insurance provisions for older workers varied widely in medium-sized and large private firms. In 1988, plans covering 56 percent of full-time participants imposed benefit reductions for older workers. The amount of insurance was first reduced at age 65 in plans covering 57 percent of those participants with age-related reductions, at age 70 for 32 percent,

1 Less than 0.5 percent.

Coverage for employees ages 65, 70, and 75 with 10 and 30 years of service is shown in table 3. As in table 1, there is little variation in benefit amounts based on length of service, and benefits still increase as salary increases. More significant is a 12- to 14-percent drop in protection at age 65 from comparable pay and service amounts unreduced by age provisions.7

As table 3 shows, the decline in benefits was most prominent after age 65, particularly between ages 65 and 70. Over this 5-year span, insurance amounts dropped 22 to 25 percent, depending on length of service and salary; between ages 70 and 75, the decline was 5 to 7 percent.

Table 4 presents the distribution of life insurance benefit amounts for older workers at the $15,000 and $35,000 salary levels. Prior to age-based reductions in coverage, 15 percent of participants at the $15,000 salary level had life insurance coverage of less than $10,000 (table 2). At age 65, however, 25 percent of plan participants had coverage of less than $10,000. The percent of employees who had less than $10,000 coverage continued to increase to 43 percent at age 70 and 48 percent at age 75.

At the $35,000 salary level, the percent of plan participants with less than $10,000 of coverage is lower than at the $15,000 level and does not rise as sharply as age increases. Only 13 percent of employees received these low benefits prior to age-related reductions, the figure

Footnotes

Excluded from coverage in the survey are benefits for executive management, part-time, seasonal, and temporary employees, as well as for employees who are on regular travel assignments (such as airplane crews and long-distance truckdrivers). In addition to life insurance, the survey examines the incidence and detailed characteristics of health care, short- and long-term disability insurance, retirement, and capital accumulation plans, and a number of paid and unpaid time-off items. It also reports on eligibility for a variety of other benefits. Key findings of the 1988 survey are in Employee Benefits in Medium and Large Firms, 1988, Bulletin 2336 (Bureau of Labor Statistics, 1989).

2 Provisions for maximum amounts of insurance, designed to limit benefits that are tied to earnings, are more common than provisions for minimums. Formulas providing benefits expressed as multiples of earnings (such as one or two times annual salary) commonly stipulate rounding rules; insurance amounts are most often rounded to the next higher thousand dollars.

3 In 1988, 58 percent of life insurance participants in medium-sized and large firms were provided with a basic benefit expressed as a multiple of their earnings, and an additional 7 percent derived their benefit from a graduated schedule based on earnings. Of the remaining participants, 31 percent were provided with a flat benefit amount and 3 percent with a flat benefit based on service.

4 Data from the Bureau's Employment Cost Index show average hourly wages and salaries of $11.84 for white-collar occupations in 1988, compared with $9.59 for blue-collar occupations. See Employment Cost Indexes and Levels, 1975-1988, Bulletin 2319 (Bureau of Labor Statistics, 1988), p. 48.

5 Prior to June 23, 1989, reductions in life insurance benefits for older workers were governed by guidelines established in the U.S. Department of Labor's 1979 interpretive bulletin (29 CFR 860.120). These guidelines allowed benefit reductions if justified by increased costs. On June 23, 1989, the Supreme Court, in Public Employees Retirement System of Ohio v. Betts, ruled that the Department of Labor's cost-justification guidelines were invalid. Data in this article reflect life insurance plan provisions in effect prior to this ruling.

6 For further information on age-related reductions in life insurance, see Michael A. Miller, “Age-related reductions in workers' life insurance,” Monthly Labor Review, September 1985, pp. 29–34.

with age, the need for coverage may decrease. Frequently, children of older workers have reached maturity, completed their education, and become independent. Where available, accumulation of home equity and investments, as well as spouse benefits under a retirement plan, help meet a surviving spouse's needs.

Protection for older workers

The 1978 amendments to the Age Discrimination in Employment Act affected practices that discriminate against workers over age 65. Prior to the amendments, some employers reduced coverage for workers over age 65 to $1,000 or $2,000-enough to help pay burial expenses. Others discontinued life insurance coverage at age 65, or excluded from participation those employees hired after age 65. As enacted in 1967, the statute applied to workers between the ages of 40 and 65, and prohibited discrimination on the basis of age in such areas as hiring, job retention, compensation, and other "terms, conditions, or privileges of employment." Because benefit reductions typically did not begin until workers reached age 65, the 1967 statute had relatively little effect on employer-provided life insurance plans. The 1978 amendments, which raised the maximum protected age to 70,3 however, did influence insurance plan provisions.

The act does not require that employee benefit plans apply uniformly to all workers, regardless of age. Differential treatment is permissible under a "bona fide employee benefit plan such as a retirement, pension, or insurance plan, which is not a subterfuge to evade the purposes of the Act. . . ." As interpreted by the U.S. Department of Labor in May 1979, this provision of the statute permits age-based reductions in employee benefits where such reductions can be justified by significant cost considerations. A benefit plan complies with the statute when costs incurred on behalf of an older worker equal those made on behalf of a younger worker, even though the older worker may receive a lesser amount of insurance coverage.5

Cost factors may be considered on a "benefit-by-benefit" or on a "benefit package" basis. Under the former, reductions in life insurance benefits do not violate the act (even if reductions start before workers reach age 65) provided that the reduction for an employee of a given age is no greater than that justified by the increased cost of covering employees of that age bracket. Cost comparisons may be made on the basis of age brackets of up to 5 years. Total denial of life insurance coverage on the basis of age, however, cannot be justified under a benefit-by-benefit analysis. (Life insurance coverage can legally cease when an employee reaches age 70 or upon separation from service, whichever occurs first.) It is, therefore, possible under the act to reduce life insurance coverage each year beginning at age 65 by a stated percentage of benefits or to make a larger one-time percentage benefit reduction at age 65 and maintain the resulting benefit level until age 70.

Where a benefit package approach is used, an employer has greater flexibility in adjusting individual benefits, "so long as the overall result is no lesser cost to the employer and no less favorable benefits for employees." Under this approach, life insurance could be reduced by an amount greater than warranted by cost considerations, or even discontinued, but only if an offsetting benefit is made available to the employees affected.

Basic coverage formulas

In 1984, employers of medium and large firms provided group life insurance to nearly all employees (96 percent); for four-fifths of the workers, the employer paid the full cost of basic coverage. (Although a small minority of employees also had supplemental plans which were paid for, at least in part, by the employer, data presented here refer to basic plans only.)

The amount of coverage for about two-thirds of the participants was based on earnings, generally determined as a multiple of the employee's annual wages or salary. As shown below, this method was most prevalent among professionaladministrative and technical-clerical employees, while flat amounts of life insurance applied to half of the production workers.

In most cases, the life insurance plans did not specify a dollar ceiling on benefits resulting from these multiple-ofearnings formulas. Where ceilings existed, they were usually $100,000 or greater and relevant mostly to those workers earning well over $50,000 a year.

Most of the remaining participants (one-third) received a flat amount of life insurance typically ranging from $5,000 to $15,000, regardless of earnings. Professional employees were more likely than the other two groups to be covered for $20,000 or more. Following are the amounts of life