| Prior research has documented the trend since 1974 away from defined-benefit plans toward defined-contribution plans as the primary vehicle for employees' retirement income security. No published research has examined the specific impact of the four major legislative acts passed during 1986 and 1987 on this trend. The purpose of this research was to examine the United States population of plans with over 100 participants to determine the extent of the reaction away from defined benefit plans resulting from the 1986 and 1987 legislation.;This research organized the Internal Revenue Service form 5500 records into a time-series panel-data format covering the years 1984 through 1989 for each unique Employer Identification Number. The LIMDEP statistical computer package was used to formulate a pooled time-series, intervention-type, random-effects model. A separate multinomial logit regression on the population of defined-benefit plans existing in 1984 and 1985 predicted the probability of plan termination by 1990.;Prior research on the population of plans was achieved by performing cross-sectional regressions on selected years with explanatory variables including size of firm, one-digit SIC industry code, and union status. The present study is the first research of which the author is aware that examined the issue using a time-series approach tracking a specific firm through time. For the logit regression, additional variables unique to a plan (top heavy, integrated, maximum over/under funding, existence of a funding waiver request, change in retirement age) were examined.;Results indicated a decrease in defined-benefit (DB) coverage for 1986 and 1987 greater than expected, given the pre-existing downward trend. Size was positively correlated with the existence of a defined-benefit plan when addressing the entire population of firms reporting for any qualified plan. Surprisingly, size had minimal DB-plan-continuation prediction ability for firms with a pre-existing defined-benefit plan. Union existence and plan integration with Social Security appeared to exert a strong influence against DB plan termination. |
