E. RETIREE HEALTH BENEFITS

Enrollment in employer sponsored group health plans continues to be an important benefit available to retirees. Health benefits received from a former employer ranged from short-term post-retirement coverage (prescribed by the Consolidated Omnibus Budget Reconciliation Act (COBRA) provisions) to lifetime coverage under the employer's group health plan. The September 1994 CPS supplement collected data on both the availability of these benefits and the extent to which retirees enrolled in the plans. These questions provided updated information that is comparable to the retiree health benefits data collected in the August 1988 CPS supplement.

The two data sets provide the basis for analysis of the current status of employees as well as changes in access to and enrollment in health plans by retirees over the age of 55. This analysis indicates that there has been a significant erosion in the availability of these benefits to the retiree population, an increase in the costs born by retirees, and an associated decline in the propensity to enroll in the plans when offered.

Coverage Status. In 1994, approximately one third of the retiree population was covered by a former employer's group health plan. Lifetime employer sponsored health coverage or the option to retain such coverage was provided to approximately 30% of the retiree population, about 7.1 million of the 23.4 million retired persons over the age of 55. As discussed in Section A , these benefits are most often provided in conjunction with a pension but in some instances as the only retirement benefit.

The data indicate a significant drop-off in employer provided health coverage at the point of retirement and then a lesser decline in coverage that has occurred during the retirement years. Sixty five percent of retirees over 55 in 1994 reported that they received health benefits through their last job. Only two thirds of these (42%) continued this coverage for any period during their retirement and only about one half (34%) reported that they were currently enrolled in an employer sponsored plan. While the rates of coverage are slightly higher for younger retirees (those from 55 to 64) there is not a substantial difference between age groups.

There is a significant difference between public and private sector employees in both their overall coverage rates and in the magnitude of the changes in coverage during retirement. Public sector employees have much higher rates of coverage while working and experience less erosion of coverage. In 1994, 80% of public sector workers were covered by their employer before retiring with about two thirds (54%) reporting current coverage. Private sector retirees reported a pre-retirement coverage rate of only 60% and fewer than one half of these (27%) indicated that they were currently receiving benefits from a previous employer. Even this lower current coverage rate may be a high estimate. (see Technical Appendix)

Type and Cost of Coverage. Of the 8.0 million retirees who were covered under a former employer's plan in 1994, almost 4.9 million (or 61%) were paying some or all of the premium costs. The median annual premium cost to the retiree depended on the type of coverage selected and on the share of the employer's payment of costs.

In 1994, the median annual premium cost to the retiree for post-retirement employer group health plan coverage was $960. That median cost was $1,200 for family coverage and $684 for individual coverage. Younger retirees (ages 55 to 64) who paid for their coverage continuation into retirement were likely to pay more than older retirees (65 and over) for it: a median annual cost of $1,200 compared to $888 paid by older retirees. The difference in cost to each age cohort is most apparent for individual coverage.

A majority (53%) of covered retirees had family coverage. The median retiree cost for family coverage ranged from $1,104 if the employer shared some of the coverage costs, to $2,016 if the employer shared none and the retiree had to pay all of the costs. Costs for individual coverage ranged from $516 to $1,320.

Covered public sector retirees were more likely than private sector retirees to be paying for the coverage, but less likely to be paying for all of the premium costs. They paid higher median annual premium cost overall and for family coverage, but lower median annual premium cost for individual coverage. In contrast to the private sector covered retirees, the majority of public sector covered retirees had individual coverage rather than family coverage.

Reasons For Discontinuation of Employment-based Coverage. The lower levels of health benefits received by retirees relative to those available to active workers is associated with changes in both the supply (employer offers) and demand (workers' acceptance) of employment-based health coverage.

Some employers who offer health benefits to active workers do not extend them to retirees, or not at the same pre-retirement cost to the employee. Some workers who accept participation in an employer's health plan before retirement drop their coverage when they retire for cost considerations. Reasons for this include eligibility for Medicare and/or access to a cheaper source of coverage and a willingness to go uninsured for a period prior to Medicare eligibility.

Other forms of coverage available to retirees may only be short term. COBRA coverage and other retiree health plans designed to bridge the gap between retirement age and Medicare entitlement age are of relatively short duration. Workers who pick up that coverage early on in their retirements eventually exhaust their eligibility period. For all retirees, Medicare coverage becomes available when they reach age 65 and it replaces or supplements employer plans as the principal source of post-retirement health coverage.

The drop offs in coverage are attributable to all of these factors. While many survey respondents gave unspecified "other" reasons and some respondents named multiple reasons for not having kept their employment-based health coverage, simplifying assumptions were made to translate all retirees' responses into single mutually exclusive reasons to provide an overview of the relative importance of factors affecting coverage changes. (See Technical Appendix)

A relatively small percent of retirees in 1994 (15%) stated that health coverage was not available to retirees. This low percentage is probably due largely to the fact that COBRA continuation mandates availability for 18 months to all previously insured retirees except those at the smallest firms. Combining these with respondents stating they were ineligible or that their eligibility eventually expired indicates that about 36% of retirees in 1994 were not eligible to enroll in an employer's plan after retirement.

Twenty-five percent (25%) of the retirees who discontinued their employment-based health coverage when they retired dropped it for an alternative source of coverage (Medicare or another plan) or declared they "did not want/need" the coverage. "Too expensive" was the reason most often given, by 27% of retirees.

Among retirees who initially continued their employment-based health coverage into retirement but later dropped it, many more "other" reasons were given for discontinuation of coverage (by as many as 21% of them). However, as many as 25% named expiration of the eligibility period as the reason for discontinuing coverage, which implies a significant number of retirees indicated they would still be covered had they been offered lifetime coverage.

Private sector retirees who discontinued employment-based health coverage when they retired were less likely than public sector retirees to have done so voluntarily: were less likely to be covered by another plan, and more likely to find themselves ineligible for post-retirement coverage under their employers' group plans. Private sector retirees who initially continued coverage but later dropped it were also more likely than public sector retirees to drop it due to expiration of the eligibility period, and to have found its continuation "too expensive".

Changes from prior years. Between 1988 and 1994, the population of retirees over the age of 55 increased 27%, from 18.5 million to 23.4 million. This growth rate was higher in the private sector (33%) than in the public sector (28%).

Rates of retiree coverage under employer sponsored health plans, however, declined substantially in this period. The decline is evident in all aspects of the availability of coverage. A smaller percentage of the retiree population was covered at the time of retirement, after retirement, currently, and for life in 1994 than was reported in the 1988 survey. The decline is observable in both the public and private sector in roughly a proportional manner.

In 1988, 44% of retirees reported that they were currently covered by their previous employer. This represents a coverage level that is ten percentage points higher than that reported in the 1994 survey. Alternatively stated, there was a decline of 23% in the coverage rate of retirees over the six year period.

Although this decline is associated with changes in the overall rate of employer sponsored health coverage that have been widely reported for this period, the magnitude of the change appears to be predominantly due to a decline in the extent to which retirees are electing to maintain benefits. The rate of pre-retirement coverage declined by only 4 percentage points (from 69% to 65%) while the rates of those with coverage at some point after retirement declined 8 percentage points (50% to 42%) and those with current coverage by 10 percentage points. These changes occurred in approximately the same proportion for both public and private sector workers.

Changes in Cost to Retirees. In general employers were less likely in 1994 than in 1988 to pay all of the costs for retiree coverage. The one exception was that public sector employers were more likely in 1994 than in 1988 to pay all costs for family coverage. However, employers were more likely to share in the payment of costs with the retiree. Employers paid all costs for 42% of the covered retirees in 1988, but only for 37% of the covered retirees in 1994. Employers paid some of the costs for 33% of retirees in 1988, and for 42% of the retirees in 1994. This latter effect is likely to be substantially a result of retirees without some cost sharing provisions dropping their coverage, thereby raising the percentage with cost sharing without much change in the overall incidence of the offer of cost sharing.

Retirees required to contribute to the premium cost of their coverage saw that cost (median annual premium) go up 10% between 1988 and 1994, from $874 to $960 (in 1994 dollars). The cost for family coverage increased from $979 to $1,200, while the cost for individual coverage declined somewhat from $753 to $684. This apparent paradox is resolved by taking into account that (relatively) more employers were contributing to the payment of costs in 1994 than in 1988, so that (relatively) fewer retirees had to pay the full premium cost for individual coverage; by increasing the numbers of retirees paying partial premium cost, the median premium cost for individual coverage was lower in 1994 than in 1988.

Changes in Reasons for Discontinuation of Coverage. The questions about reasons for discontinuing health coverage were expanded in the 1994 survey. This makes a precise comparison across the period difficult
(see Technical Appendix) . The limited comparison that can be done indicates that the cost of post-retirement coverage became a more important factor, and expiration of eligibility a less important factor, in retirees' decisions to continue employer health coverage after retirement. In 1988, 21% indicated that they had dropped their coverage because it was too expensive compared to the 27% who provided this reason in 1994. Expiration of eligibility declined from 32% to 17% over the period, although other reasons for ineligibility were named in 1994 which made a total of 36% retirees not eligible for continuation of coverage after retirement.

The same percentage (5%) of retirees in 1988 and in 1994 declined the employer offered post-retirement coverage ("did not want/need coverage"). A lower percentage in 1994 discontinued employer coverage for Medicare coverage. Access to coverage under another plan was an alternative for relatively more retirees in 1994 (15%) than in 1988 (9%).

Characteristics Associated With Coverage of Private Sector Retirees. Coverage of private sector retirees under a former employer's group health plan after retirement is strongly associated with certain demographic and economic characteristics of the retirees. In general, the same variation of health coverage status with gender, race, education, income, occupation, industry, firm size, and collective bargaining status, which was observed with respect to workers before retirement, continues after retirement.

Demographic Variables. Among retirees, as among active workers, men are much more likely than women to be covered under an employer's group health plan. In 1994, more than half of the retired men but just over a third of the retired women who were covered just before they retired were still covered by a former employer's plan. If only lifetime coverage is considered, the gap is even greater with 33% of the retired men and 14% of the retired women recipients of this benefit in 1994.

Race also has a significant association with access to coverage. White retirees start retirement with higher coverage levels, are more likely to have continued coverage after retirement, and are more likely to have kept their former employer-sponsored coverage: 24% of white retirees could continue that coverage for life, versus 15% of black retirees, and 14% of retirees of all other races.

Coverage rates vary directly with education level. Retirees with "less than 12 full years" had the lowest coverage rate at retirement (52% covered) and remained the worst covered by an employer plan after retirement (16% with lifetime coverage). At the other end of the distribution, retirees with a "master's degree or higher" had the highest coverage at retirement (75% covered) and remained the best covered by employer plans after retirement (43% with lifetime coverage).

The association of coverage rate with age, and with age at retirement is more complex: one relationship holds for age brackets under age 65, and the opposite relationship for age brackets over age 65.

Up to age 65, the older the retiree cohort, the higher its coverage rate under former employers' health plans; beyond age 65, the older the retiree cohort, the lower its coverage rate. Employment-based coverage is, however, more likely to be lifetime coverage for older retirees than for younger retirees.

Older retirees have Medicare coverage and any employer provided coverage they continue is more likely to be lifetime coverage that supplements Medicare than short-duration COBRA coverage. Younger retirees, on the other hand, rely on employer provided coverage continuation in relatively larger numbers because they are typically not yet eligible for Medicare.

For private sector retirees who retired before age 65, the closer the age at retirement to the age of Medicare entitlement, the higher was the cohort's coverage rate under employer plans. On the other hand, beyond age 65, coverage rates are lower, the older the age at retirement.

While employment coverage continuation can be an inducement to early retirement, those who retire before age 55, and as early as at age 40, are likely not to have met minimum age/service eligibility requirements for coverage continuation at the time of retirement. This would account for the association of the lowest coverage rates with the cohort with the lowest age at retirement (ages 40-54). On the other hand, once they reached age 65, retirees were less likely to rely on employment health benefits than on Medicare for post-retirement health coverage.

Economic Variables. Low incomes are associated with low coverage rates and high incomes with higher coverage rates: more than half of retirees with pre-retirement annual earnings of $30,000 or more were currently covered by an employer plan; fewer than one in every ten retirees who earned less than $15,000 a year had that coverage.

Large firms are associated with high coverage rates and small firms with low coverage rates: 46% of retirees who worked for firms of 1,000 or more employees had employer supplied health coverage in 1994, and 40% had lifetime coverage; 22% of retirees who worked for firms of less than 100 employees had employer supplied coverage in 1994, and 7% had lifetime coverage.

Forty one percent (41%) of retirees covered by union contract had employer provided post-retirement coverage in 1994, and 36% had lifetime coverage. Only 19% of retirees not covered by union contract had that employer coverage, and only 17% had lifetime coverage.

The 17.4 million private sector retirees in 1994 joined the retiree ranks over a period of more than 20 years. The more recent their retirement year, the more likely they were to have had employer supplied coverage before retirement, to have continued that coverage after retirement, and to be currently (in 1994) covered under a former employer's health plan.

The same trend is evident for lifetime health coverage of retirees, but only up to retirees joining the ranks in the years 1985-1989; the most recent retirees (those who retired between 1990 and 1994) had lower lifetime health coverage rates than the previously most recent retirees (those who retired between 1985 and 1989).

This last observation indicates that recent retirees may have more access to employer offers of coverage continuation, but fewer of these offers are for lifetime coverage. They are more likely to be continuing COBRA coverage than more experienced retirees (those retired more than 5 years) who would have already exhausted any short-duration coverage continuation they might have picked up immediately following retirement. In addition, the most recent retirees may comprise relatively fewer older retirees than the more experienced retirees.

Retiree employment-based coverage rates varied widely across industries: 11% of retirees from the retail industry had lifetime health coverage under a former employer's plan (industry with lowest coverage rate); while 64% of retirees from the communications industry had it (industry with the highest coverage rate).

The same coverage rates also varied across occupations: it ranged from 9% of retirees in service occupations to 33% of retirees in professional specialty occupations.