D. RECEIPT OF PRIVATE PENSION ANNUITY BENEFITS

Although the proportion of retirees who are receiving an annuity type pension benefit has been decreasing, among those who received an annuity benefit in recent years, the amounts of these benefits have been steadily increasing.

For annuity recipients who first began receiving benefits in 1993 or 1994, median annual benefits were $8,400 and mean annual benefits were $12,107. This represents an increase from $5,570 and $7,770 in 1988-89.

The median annuity provides a replacement of about 27% of the income of retirees in the year before they retired.

There is, however, a great deal of variation in the distribution of the annuity benefits that are being received. Among new pensioners in 1993-94, men received median annual benefits of $9,600 and mean annual benefits of $14,654, while women received median annual benefits of $4,800 and mean annual benefits of $6,748.

This gap in median pension amounts received by men and women which had widened from 1978 to 1989, has narrowed in recent years. In 1988-89 the median annuity benefits received by women were 37% of those received by men. In 1993-94 the median of women's benefits for new pensioners had risen to 50% of those received by men.

While the gap in median benefits received by men and women has narrowed, the percentage of all new private annuity recipients who were women decreased from 35% in 1989 to 29% in 1994. The shift in the form of benefit payment from annuities to lump sums has disproportionately affected women, who are more likely than men to be employed in small firms. It is among small firms where the shift from DB to DC coverage has been greatest.

Variation in annuity benefits is also evident along a variety of other dimensions that generally parallel distributions of income and plan participation rates. Workers with less than a high school education receive benefit levels at about one third those of persons with a master's degree or higher level of educational attainment. Firm size and occupation are also significantly associated with the amount of the annuity.

Wage replacement rates. The extent to which the retirement income system is able to replace pre-retirement earnings is generally accepted as the best indicator of its adequacy. In general, a replacement rate of between 60% (for higher income) and 80% or more (for lower income) is assumed to be required to maintain a worker's standard of living in retirement.

The survey collected information on respondent's earnings in the year prior to retirement which permits replacement rates to be estimated.

For all private sector workers receiving annuity benefits in 1994 (which includes some who continue to work) the income from a private pension plan replaced a median of 26% of reported pre-retirement earnings and Social Security replaced 31% of earnings. The median total replacement rate for all of these persons was 61%. (Note: these rates are not additive because they are medians.) The overall rates include about 1 million mostly early retirees who are not receiving Social Security, primarily because they have not yet reached the age of 62. The combined replacement rates for those who are receiving Social Security benefits as well as a private pension is somewhat greater at 67%.

The rates of income replacement, however, are somewhat deceptive because Social Security and some private benefits have been adjusted for inflation while pre-retirement wages remain in nominal dollars. A more meaningful measure is the replacement rate when final year's earnings are expressed in 1994 dollars. This indicates the degree to which benefits provide purchasing power equivalent to pre-retirement levels.

When replacement rates are calculated using this measure of "purchasing power replacement" they are considerably lower. Adjusting reported pre-retirement earnings to current dollars yields an overall replacement rate of 41%, with private pensions providing 18% of "real" pre-retirement income. Considering only those who receive both Social Security benefits and a private pension, the comparable measure is slightly higher at 45%.

The level of these rates is partially a result of many older retirees who spent the majority of their working years in a period in which the private pension system was just emerging as a significant source of retirement income. A more accurate measure of the capacity of the system to maintain the living standards of future retirees is the level of benefits attained by the most recent group of retirees, the majority of whom have spent at least half of their career when the pension coverage rate among private sector workers has been essentially constant.

For persons receiving their initial benefits in 1993 or 1994, among the group of private sector workers who received both private pension annuities and Social Security benefits, the combined benefits replaced 58% of pre-retirement wages. For those receiving just Social Security benefits, these replaced about 40% of pre-retirement wages. The median rate of replacement for new recipients was the same as that reported by new recipients in 1989.

These averages, however, obscure a great deal of variation in the extent to which new benefit recipients are able to match their prior year's earnings. As indicated in the graph below only a little over 40% of recipients received initial combined benefits that replaced two thirds of pre-retirement wages. The bottom fifth of the distribution reported that they were able to replace less than one half of prior earnings.

Post-retirement benefit increases. The differences observed between nominal and "purchasing power" replacement rates are due to the fact that the majority of private pension annuity recipients are not receiving cost of living adjustments while Social Security benefits have been roughly indexed to the cost of living. Only about 30% of private pension recipients report that they have ever received a cost of living increase, and among those who have, the increases have averaged only about 2.1% per year. Thus for most pension recipients, the real value of total pension benefits and, therefore, their ability to replace the value of prior earnings will decline over time.

This can be observed by measuring the "real" replacement value of annuities which were initially received at various points in time. As shown in the graph below, pensioners who began receiving annuities in 1984 had initial median wage replacement rates of 24%. Today their median annuity benefits, even including post-retirement increases received by some, replace only 17% of pre-retirement wages. For pensioners who retired in 1979, the replacement rate of their annuities has declined from an initial average of 23% of their pre-retirement wages to 11% of their 1979 wages adjusted to 1994 dollars. The path of replacement rate erosion is effectively parallel for the different years of initial receipt.

Joint and survivor option. All private pension plans which provide benefits in the form of an annuity are required by law to provide a J&S annuity as the standard form of benefit for married recipients. A J&S benefit provides, in the event of death of the recipient, for benefits to be continued at a reduced rate to a surviving spouse. Recipients can opt out of this arrangement by written agreement of both the recipient and the recipient's spouse.

Overall, 54% of married pension recipients with annuity benefits selected J&S benefit, with 62% of men and 28% of women opting for a J&S. The percentage selecting a J&S was higher for more recent recipients, with 38% of recipients who began receiving benefits prior to 1975 choosing the J&S, compared to 62% of those retiring in 1993 or 1994. Among recipients who were divorced, only 7% reported selecting a J&S benefit compared to 59% of married recipients.

Survivors' benefits. Out of 13 million persons ages 40 and over who are widowed, 18% receive benefits based on a pension earned by a deceased spouse. The rate of receipt was 21% for women and 5% for men. Median annual benefit amounts received by survivors were $4,800 for women and $3,600 for men.