Postretirement increases under private pension plans.
A lower rate of inflation--but continuing over a longer period--may have still greater effects on today's retirees, who often will receive pension benefits for 15 years or more. An inflation rate of 5 percent a year would, after 15 years, cut in half the purchasing power of the original pension benefit, and a 7.5-percent annual rate of price increase would result in a two-thirds reduction. Thus, even without "double-digit" inflation, the value of a fixed pension can be seriously eroded during retirement. To offset part of this loss, many employers grant pension increases to retirees or their beneficiaries.
Information on the extent of these postretirement pension increases is available from the Bureau's annual survey of the incidence and provisions of employee benefit plans. This survey is conducted in the United States--excluding Alaska and Hawaii--in private sector establishments employing at least 50, 100, or 250 workers, depending on the industry. Industrial coverage includes mining; construction; manufacturing; transportation, communications, electric, gas, and sanitary services; wholesale trade; retail trade; finance, insurance, real estate; and selected services. The 1982 survey sample comprised 1,516 establishments, designed to represent 21 million employees in 44,288 establishments. Excluded from the survey were executive management employees (those whose decisions have direct and substantial effects on an organization's policymaking) and part-time, temporary, seasonal, and traveling operating employees, such as airline flight crews and long-distance truckdrivers.
Data for the survey were obtained on the number of full-time active employees covered by pension plans, but not the number of retirees or beneficiaries actually receiving annuities. Consequently, it cannot yield direct information on either the proportion of annuitants receiving postretirement pension increases or the average amount of their benefit improvements. However, the magnitude of both can be roughly indicated by weighting the information collected on postretirement increases by the number of active workers participating in plans that granted such increases. This approach was followed to develop the data for this article. Survey parameters
In 1982, 17.6 million full-time workers participated in private pension plans of medium and large firms; 40 percent were covered by plans which gave annuitants at least one postretirement increase during 1978 through 1981. Three percent of the participants were in plans providing these increases automatically. The remainder were under plans with formulas the determined initial pension levels--for example, years of service times a flat dollar amount or a percent of average earnings--but were silent as to postretirement adjustments. In these situations, increases, when made, were on an ad hoc, or discretionary, basis.
This information comes from the Bureau's 1982 survey, the first since the program began in 1979 to include questions on ad hoc postretirement adjustments. Firms were asked to provide information on all such adjustments made during the 1978-81 period. This article examines the extent, value, and methods of determining the adjustments.
Survey respondents supplied information on the effective dates of postretirement pension adjustments, formulas used to determine the amounts of increase, and provisions relating to minimum or maximum increases. Many of the formulas varied the size of the increase according to the pension amount or the date of retirement. Thus, for this analysis, adjustments were determined for four monthly pension values ($250, $500, $750, and $1,000) and three retirement dates (December 31, 1967, 1972, and 1977). The four pension values represent the benefit payable as of December 31, 1977, the day before the period studied for postretirement increases. The three retirement dates are for persons who had retired 10 years, 5 years, and immediately prior to the period studied. The pension adjustments for individual plans were then averaged using the number of active worker participants as weights to provide surveywide estimates for each example.
The analysis showed that automatic adjustments between 1978 and 1981 averaged 13.4 to 15.2 percent among the four pension values studied. Ad hoc adjustments averaged 8.8 to 24.3 percent, depending on the pension value in 1977 and the employee's retirement date. The largest percentage increases under both methods went to retirees with the smallest pensions. This resulted from a variety of factors, such as the use of flat dollar increases, specified minimum and maximum increases, and restriction of percentage increases to a portion of the original pension. Ad hoc adjustments also commonly provided greater increases to those retired the longest.
Despite these increases, the purchasing power of the annuities rarely was maintained. For all but one of the groupings studied, average postretirement increases were less than half the rise in the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) during the 4 years under examination. Ad hoc adjustments
Postretirement annuity adjustments typically are made on a nonautomatic (discretionary or ad hoc) basis (table 1). Such increases may be granted at irregular intervals and their size is at the discretion of the employer (and union), based on such factors as pension fund investment performance, the firm's financial position, and general economic conditions.
Ad hoc adjustments were granted by some employers during the 1950's and 1960's, but became more prevalent during the 1970's. The 1978-81 period studied by the Bureau probably experienced a higher-than-normal incidence of such adjustments, as accelerating inflation focused attention on the adequacy of fixed retirement income. Nearly two-fifths of the pension plans participants covered by the Bureau's 1982 survey were in plans that gave at least one ad hoc increase to retirees during the 1978-81 period. Of the participants in pension plans granting discretionary increases, nearly half were in plans paying more than one increase during the 4 years (table 2). Although there was no sharp difference in the incidence of ad hoc increases among the three occupational groups presented, production workers were the most likely to be in plans with multiple increases.
Ad hoc increases were more widespread during 1980 and 1981 than in the previous two years. This is illustrated in the following tabulation, which provides a percentage distribution of participants in plans granting discretionary increases by year of adjustment. Half or more were in plans paying increases in 1980 and 1981, compared with about one-third in 1978 and 1979:
As described later in this article, nearly all pension plans with automatic adjustment provisions gave annual increases; however, relatively few plans provided discretionary increases in each of the 4 years studied.
Funds needed to finance these ad hoc annuity improvements come either from assets of a pension (or related welfare) fund or wholly or partly from general assets of a business firm. A 1981 study of pension increases indicates that some large corporations elect the latter approach.
Methods of calculating increases. Table 2, which summarizes information on benefit formulas for the most recent ad hoc increase, shows the variety of techniques used to determine the adjustments. Forty-one percent of the participants in pension plans granting discretionary increases were under plans which provided specified increases per year of retirement. Such plans, more common among white- than blue-collar workers, provided percentage increases per year of service that ranged from less than 2 percent to more than 5 percent; frequently, however, there was a ceiling placed on the total amount provided.
A second appraoch, in plans covering 36 percent of the participants, called for flat increases, either in dollars or, more commonly, percentage. The incidence of this type of formula was fairly consistent across the three occupational groups shown in table 2. White-collar employees, however, were more likely to receive percentage increases, while flat-dollar increases were more likely for blue-collar workers. These plans often recognized length of retirement by varying the flat amount or percentage depending on the employee's date of retirement. For example, ad hoc increases might be based on the following formula: Year of retirement Percent increase in benefits Prior to 1970 30 1970 or 1971 25 1972 or 1973 20 1974 oe 1975 15 1976 or 1977 10 1978 or 1979 5
The third basic approach, covering 20 percent of the participants, tied the pension increase to the retiree's length of service. For example, monthly pension checks might be increased by 75 cents per year of service. Such adjustments were common among production workers, who, more often than white-collar employees, have their initial pension determined as a specified dollar amount multiplied by years of service; many of these adjustments were collectively bargained, and increases for retirees were related to improvements in pension accruals of active employees.
Ad hoc adjustment formulas occasionally specify minimum or maximum annuity improvements. Eighteen percent of the participants were under plans whose adjustment formulas provided minimum benefit increases (table 3). Minimums typically were small, usually between $5 and $15 a month among the plans analyzed. Maximums on benefit improvements were twice as common as minimums. These caps typically were specified in percentage terms, and most often were less than 10 percent of current annuities.
Average increases. The amount of ad hoc adjustments generally depends on one or both of the following factors: the size of the annuity prior to the adjustment and the data of retirements. Table 4 presents data on the average 1978--81 increases developed from the survey, given varying assumptions regarding these two factors. (When formulas took length of service into account, 25 years' service was assumed.)
For the particular combinators of annuity size and length of retirement studied, average increases over the 4-year period ranged from $35 to $161 a month, or from 8.8 to 24.3 percent. On average, the greatest dollar increases went to retirees with the highest pensions. In percentage terms, however, the reverse was true, reflecting the influence of flat dollar adjustment formulas and limits on the size of increases.
Three-fifths of the participants in pension plans with ad hoc increases were under plans varying adjustments by length of retirement. The effect of these plans is evident in table table 4. During the 1978--81 period, employees who retired in 1967 received increases averaging nearly twice as much as those who retired in 1977. By granting larger increases to those who retired earlier, employers recognized that their initial pensions had lost the most purchasing power to inflation. In addition, the earlier the retirement date, the greater the likelihood of a relatively small, initial pension.
Frequency of individual adjustments also influenced the size of increases. Thus, plans providing two or more increases yielded smaller gains, on average, in each of the individual adjustments, but a higher total improvement over the 1978--81 period (table 5). Automatic adjustments
As noted, 3 percent of the pension plan participants in 1982 could expect automatic benefit adjustments after retirement. Of those, 2 percent were under plans that granted only automatic adjustments to retirees in the 1978--81 period, while plans covering the other 1 percent gave ad hoc increases in addition to the automatic adjustments (table 1).
Automatic adjustment formulas usually tie into changes in designated statistical series, most often the CPI. Of the 880 defined benefit pension plans.sup.14 examined in the 1982 survey, 23 included provisions for automatic adjustments, and all but one of these plans tied into the CPI. The exception gave automatic 2-percent pension increases each year, independent of price changes. Automatic pension adjustment provisions linked to wage rather than price indexes do exist, but none appeared in the survey sample.
eleven of the 23 plans called for an annual increase equal to the percentage rise in the CPI, up to a maximum of 3 percent. Other plans in the sample gave pension increases that were less than the full CPI increase, but often had caps higher than 3 percent. In some instances, adjustment provisions were not triggered until the CPI had risen a specified percent, such as 3 percent. Some of the plans restricted the percentage adjustments to the first $500 or other initial level of the pension.
three-fifths of the plans with automatic adjustment provisions allowed for decreases in the CPI. Most of these, however, prevented annuities from being reduced below the initial pensions.
Among the four pension values analyzed, automatic adjustments averaged between 13.4 and 15.2 percent over the 1978--81 period (table 6). The most common increase was 12.6 percent--the result of compounding annual increases of 3 percent over the 4 years. The average increases, however, were raised largely by one plan, which granted uncapped increases equal to 100 percent of the rise in the price index.
The infrequency of automatic adjustment provisions can be attributed mainly to cost considerations. When a firm ties pension benefits to a price index without limitations, it assumes a potentially large and indeterminate future obligation. Thus, most plans with automatic adjustments curb costs by restricting the size of the annual increases. In their book, Employee Benefit Planning, Jerry S. Rosenbloom and G. Victor Hallman provide a rule of thumb that says, ". . . pension costs can increase by about 10 percent for each 1 percent increase in benefits provided to pensioners.".sup.17 Employers generally avoid signing a "blank check" for pension increases by providing ad hoc improvements, rather than adopting a formula for automatic benefit adjustments. Benefit adjustments and inflation
The CPI-W rose 51 percent between December 1977 and 1981. Although this index may not be an accurate gauge of changes in retirees' purchasing power (retirees may not have the same spending patterns as active workers) it provides statistical evidence that the real value of retirees' private annuities generally declined over the 4 years. Even for the retiree group associated with the largest percentage increase in annuities during this period, professional and administrative workers retiring in 1967 and having a $250 monthly pension in 1977, the rate of increase was not much more than half the rise in the price index (table 4). Total purchasing power of retirees, however, did not necessarily fall to the degree that the comparison would suggest; most of those with private pension benefits also received social security payments, which are adjusted to take account of changes in the CPI-W.
Which method of adjustment--automatic or ad hoc--provided retirees with the greater degree of protection against inflation? Comparison of table 4 and 6 shows that employees who retired in December 1977, just prior to the 4-year period studied, fared better with automatic increases. The average annuity increase under plans providing automatic adjustments varied between 13.4 and 15.2 percent, while the average ad hoc increase ranged from 8.8 to 14.0 percent. Although individual ad hoc increases were generally larger than individual automatic increases, the latter were granted annually, whereas most plans prividing discretionary adjustments gave only one such increase during the entire 4-year period.
A contrary picture emerges when we focus on 1967 and 1972 retirees. For these individuals, comparisons of tables 4 and 6 favor retirees receiving discretionary increases. Do those contrasting results come about because ad hoc increases tend to favor those who retired earlier or because possibly fewer discretionary increases were granted before 1978, encouraging larger catch-up payments? An answer to this question is not possible from the data collected in this study.
IN ANY EVENT, a minority of pension plan participants were under plans providing for either automatic or ad hoc adjustment. Three-fifths of all participants within scope of the Bureau's employee benefit study were under plans that did not make postretirement pension adjustments in the 1978--81 period.
|Printer friendly Cite/link Email Feedback|
|Author:||Schmitt, Donald G.|
|Publication:||Monthly Labor Review|
|Date:||Sep 1, 1984|
|Next Article:||State and regional employment and unemployment in 1983.|