FRBSF Economic Letter
1996-17 | May 24, 1996
Last week’s Economic Letter (96-16) explored the controversy surrounding the recent estimates of the U.S.’s potential growth rate. It focused particularly on the fact that a slowing in the labor force participation rate is probably the main reason for the slowing in the potential growth rate in the 1990s. This Letter examines the sources of the slowdown in the labor participation rate in detail, and it addresses the question of whether the slowdown is likely to be reversed.
The aggregate labor force participation rate (Figure 1) measures the proportion of the civilian adult population that is in the labor force–either employed or actively looking for work. It is estimated from the Current Population Survey, which is taken monthly and also yields data on employment and unemployment. After rising steadily until the late 1980s, participation has remained roughly constant in the present decade. To judge whether participation is likely to resume its earlier growth, we need to identify the sources of its recent slowing. We begin by breaking the overall change into its components and examining each in turn.
The first breakdown is by gender. Figure 1 shows that the changes in overall participation have been associated largely with changes in the behavior of women. The male participation rate has drifted gradually and steadily lower in recent decades. The female participation rate, however, increased rapidly from about 1970 to the late 1980s, but since then has flattened.
Examination of participation rates for different age groups suggests that the decline in male participation was due mainly to earlier retirement, which caused a large decrease in participation by older men. Between 1960 and 1994, for example, participation by men aged 25 to 55 decreased by five percentage points, whereas that by men over 55 declined by 23 percentage points. By contrast, participation was unchanged among older women and increased strongly in younger age groups.
Since prime-age adults have a higher participation rate than younger or older people do, it seems natural to suspect that the maturing of the baby boom generation might explain the rise in participation in the late 1960s and 1970s–it would increase the proportion of prime-age adults, which would cause the overall participation rate to go up, even if there were no change in participation by any single age-group. However, analysis shows that changes in the age structure of the population had only modest effects on overall participation and played essentially no role in the deceleration since 1990. It appears that any boost to participation associated with the maturing of the baby boom was more than offset by other demographic changes, including the rising share of older persons in the population, which tended to have the opposite effect.
Another way to identify the source of the slowing in the participation rate is to estimate the changes in the proportion of the year that people spend in the labor force. This estimate can be obtained by using the results from the Work Experience Survey, which is conducted annually in March by the Census Bureau and which shows the number of people who worked or looked for work in the preceding year. Unlike the participation rate–which gives the proportion of the population in the labor force in a particular month–the “experience rate” gives the proportion of the population in the labor force at any time during a particular calendar year. Since many people spend only part of the year in the labor force, the experience rate is higher than the annual participation rate, which is simply the average of the monthly data. Figure 2 shows the experience rates and the annual average participation rates for males and females since 1970.
Figure 3 shows an estimate of the proportion of the year (measured in months) that a typical person with some experience spent in the labor force. This estimate is obtained by dividing the average monthly participation rate each year by the annual experience rate. For example, if the average participation rate for a given year was 60 percent, but 80 percent of the population was in the labor force at some time during the year, this means that the average person was actually working or looking for work during only three-fourths of the year (60 divided by 80) or nine months. I call this proportion the “attachment rate,” since people who spend a larger proportion of the year in the labor force are more closely attached to the workforce.
The average participation rate may fall either if a smaller proportion of the population has some labor force experience during the year or if the number of months that a typical person is attached to the labor force during the year falls. The variations in the participation rate in recent decades reflected both types of change, and both have contributed to the slowing in the 1990s.
Figures 2 and 3 show that the long-term decline in male participation has mainly reflected a corresponding decline in the male experience rate. The proportion of the year that men spend in the workforce has remained in a narrow range between 11 and 11-1/2 months with only a slight upward trend. Thus, on average, males do not move in and out of the work force to any great extent during the year, and their behavior in this regard has not changed much in recent decades. Most of the decline in male participation has been due to the decrease in the proportion of the male population that has had any work experience during the year.
In the case of women, on the other hand, participation has risen since the 1960s both because more women now choose to spend some time in the workforce and because these women spend more months of the year in the workforce. Between 1960 and 1994, the proportion of the female population with some labor force experience increased from 48 percent to 64 percent. Over the same period, the average time spent in the labor force rose from nine months in 1960 to eleven months in 1994.
The slower growth in female participation in recent years reflects a slowing in both of these factors. Figure 3 indicates that the rise in female attachment slowed in the 1980s and may have slowed again in the 1990s. The fact that the proportion of the year that women spend in the workforce now is approaching its practical maximum means that there is less room for further growth in female participation from this source. For example, if female attachment were to rise to the present male level of 11-1/2 months with no change in the rate of female work experience, the average female participation rate would rise only modestly from its present 59 percent to 61-1/2 percent. Thus, any significant further rise in female participation must come from increases in the number of women with labor force experience.
Figure 2 indicates that the slowing in female participation since the late 1980s also reflects slower growth in the female experience rate. Between 1988 and 1994, the proportion of women with some labor force experience during the year increased at an annual rate of only 0.2 percent, compared to 0.6 percent in the preceding ten years. However, the proportion of women with some labor force experience during the year still remains well below that of men, so that there is room for significant further increases in female participation from this source.
It is too soon to tell whether the slowing in the growth of the female experience rate will turn out to be permanent or only temporary. The fact that women s participation is lower than that of men across all age groups means that there is plenty of room for further increases. However, we are unlikely to see the same kind of boost to women s participation that occurred in the 1970s and 1980s as fewer women exited the labor force in their mid-twenties and more stayed in the labor force into their forties and fifties. To get such a boost, we would need to see women’s participation rate increase as they got older. However, there is evidence that women (more than men) withdraw from (or delay entry into) the workforce while raising children. This will tend to hold down overall female participation unless there are significant changes in child-rearing customs.
The labor force data examined in this Letter suggest that the potential for economic growth has slowed in the current decade compared to earlier periods. To a significant extent, this slowing is the result of decisions on the part of individuals (especially women) not to increase further the amount of time they devote to paid employment. Prior to the 1990s, female participation increased both because more women chose to enter the labor force and because these women participated for a larger part of the year. A slowing in both factors underlies the slowing in participation that has been observed recently.
There remains plenty of room for further increases in the proportion of women who choose to enter the work force for at least part of the year. On the other hand, the degree of women s attachment to the workforce is now approaching its practical maximum, meaning that there is not much room for further growth in female participation from this source. This suggests that the growth of participation in the near-term future is likely to remain slower than was seen before the late 1980s. Unless productivity growth improves, potential GDP growth also is likely to be less.
Parry, Robert T. 1996. “Economic Growth and Monetary Policy.” FRBSF Economic Letter 95-16 (May 17).
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120