No Decline in Long-Term Employment

In 1969 the average tenure for men in the job they held for the longest period during their careers was 21.9 years. In 2002, the comparable figure was 21.4 years.

For some years it has been that reported that employees in the United States experienced widespread, substantial declines in job security or stability over the past several decades. Various newspaper articles have suggested that big structural changes in labor markets mean that job security is a "myth," that lifetime employment with a single employer is far less likely than it was, say, thirty years ago. Workers themselves worry that their prospects for keeping a job for a long period have shrunk, that they may need several jobs during their careers. "There is, however, a striking lack of solid empirical evidence to support these claims," writes economist Ann Huff Stevens.

In The More Things Change, the More They Stay the Same: Trends in Long-Term Employment in the United States, 1969-2002 (NBER Working Paper No. 11878), Stevens sees stability in the prevalence of long-term employment for men in the United States, contrary to popular views. "Long-term relationships with a single employer are an important feature of the U.S. labor market in 2002, much as they were in 1969," she writes. So, the likelihood is that most workers will have some job during their working lives that lasts for more than 20 years.

Stevens uses data from surveys of men aged 58-62 who were quizzed at the end of their working careers. She finds that in 1969 the average tenure for men in the job they held for the longest period during their careers was 21.9 years. In 2002, the comparable figure was 21.4 years, not much different. Just more than half of men ending their careers in 1969 had been with a single employer for at least 20 years; the same was true in 2002. Around a quarter of those men retiring, anytime in the 1969-2002 period, had stayed with a single employer for 30 or more years.

The focus on complete tenure provides a natural and direct measure of the frequency of "lifetime" or long-term employment, Stevens writes. Moreover, her data -- drawn from the Retirement History Survey, started in 1969, the National Longitudinal Study of Older Men, started in 1966, and the Health and Retirement Study, started in 1992 -- cover a longer time span than those used in previous research on job stability. She also finds that tenure in the longest job held by these older men actually rose to an average of 24 years for those groups retiring in the late 1970s. Then, over the next decade, job tenure declined to levels comparable to those prevailing in 1969.

Looking at the data in more detail, Stevens finds that educated men tend to have longer tenure than less-educated men, that is, men with less than a high school education. The average tenure in the job held longest for those with less than 12 years of completed education was about 21 years in 1969, and 18.6 years in 2002. Tenure for men with 12 or more years of education stood at 22.4 years in 1969 and 22.05 in 2002. Further, non-whites have an average tenure below the comparable measures for white men.

Stevens' findings for the most recent years reflect the career outcomes for the generation of men approaching retirement age in 2002. Whether this level of stability will apply to subsequent generations of men depends on the continued evolution of job retention rates. Job retention rates declined in the 1990s, but it is not yet clear whether these declines will persist. Only with relatively long-lasting reductions in job retention rates will individuals experience corresponding reductions in completed tenure on their longest jobs.

In her paper, Stevens looks at whether job stability during the 1969-2002 period was affected by increased early retirement of men, a rise in average education levels, and the differing numbers of those taking years off from civilian work to serve in the military. But, she concludes that these factors do not bias her major finding, that job stability has remained relatively steady in that period

-- David R. Francis