The full set of revisions to state employment and unemployment data, along with new information for January 2025, will be released next week. Meanwhile, the Bureau of Labor Statistics has issued revised data for the Local Area Unemployment Statistics (LAUS). This years LAUS revisions were more extensive than normal, with data revised back to the beginning of the series in 1976. In this article, we will focus on the more recent history revealed by the revised data.
The figures below compare the pre-revision time series to the newly revised data for total household employment and unemployment. As is typical, the revisions tend to smooth out some of the fluctuations in the originally published data.
The originally published data for the number of employed showed two brief periods of contraction, in 2022 and 2023. The revised data show positive monthly increases for the entire period from May 2020 through December 2024. The revised series ends up slightly lower than the original estimates, but by only 1,850 individuals (0.14%).
Source: Bureau of Labor Statistics, Local Area Unemployment Statistics (LAUS)
The number of unemployed individuals was revised lower for the second half of 2023, then somewhat higher during the second half of 2024. As a result, the new path of unemployment shows a slow, but relatively steady increase in the number of unemployed over the past two years. As of December 2024, the number of unemployed was revised higher by approximately 2,740 individuals.
Source: Bureau of Labor Statistics, Local Area Unemployment Statistics (LAUS)
The impact on the unemployment rate was a smoother profile path, attenuating some of the extreme changes. The record-low unemployment rate of 2.8% in early 2023 has been revised up to 2.9%; the transitory peak in late 2023 was lowered from 3.7% to 3.5%; and the new end-of year unemployment rate was revised up from 3.4% to 3.6%.
Source: Bureau of Labor Statistics, Local Area Unemployment Statistics (LAUS)
The revisions also affected the calculated estimates for the state’s labor force participation rate. The new data show a more sustained rate of increase over the past two years, culminating in a rate of 58.3% in December 2024—slightly higher than the $58.2% originally reported.
Source: Bureau of Labor Statistics, Local Area Unemployment Statistics (LAUS)
Recent conditions in the nation’s labor markets have been characterized as a labor shortage, prompting attention to trends in labor force participation. The issue is particularly relevant for Arkansas, where the rate of unfilled job openings has been running higher than the national average while labor force participation rates are consistently lower than the U.S.
This article presents an analysis of Arkansas’ labor force participation rate, examining the demographic and behavioral factors that underlie the gap between the overall state and national rates.
Long-term Trends
As shown in Figure 1, the U.S. labor force participation rate rose steadily from the early 1960s to around 2000, and has been declining over the first quarter of this century. The participation rate in Arkansas has generally followed the same pattern over time, but has been persistently lower than the U.S. rate. The gap between the two rates has varied between 3% and 5%, and has recently been near the top of that range.
Figure 1:
The forces driving the long-term increase and decline in labor force participation rates are generally attributed to underling participation rates of men. From around 1960 through 2000, the participation rate of women rose dramatically, levelling off during the first decade of the century and declining more recently. Meanwhile, participation rates for men have gradually but consistently trended lower.
Figure 2:
Data for the breakdown of participation rates by sex for Arkansas are available only for 1997 through 2023. As shown in Figure 3, participation rates for both men and women in Arkansas have trended below the national averages, with the gaps varying in the range of 3% to 5%. Recent gaps for both men and women have been near 5%.
Figure 3:
Participation Rates by Age Cohorts
Another important factor for understanding overall labor force participation rates is age. As shown in Figure 4, there is a clear life-cycle patter: Participation rates are relatively low for both the youngest and oldest cohorts, and they peak during the prime working ages of 25-44. While there is a tendency for participation rates to drop off in the 45-54 and 55-64 age ranges, the decline is more pronounced for Arkansas. Note that the participation rates for younger workers (16-19 and 20-24) are higher for Arkansas than for the U.S. The gap reverses and widens as we move to higher age brackets.
Figure 4:
The population distribution matters as well. Figure 5 shows that Arkansas’ population shares of both younger and older workers are higher than the U.S. average, while shares of those in the ranges of 25 to 53 are lower than average.
Figure 5:
Consequently, even if participation rates for each age cohort was the same for Arkansas and the U.S., Arkansas would have a lower overall participation rate due to the prevalence of younger and older workers (with relatively low participation rates) and the shortfall of middle age workers (who tend to have significantly higher participation rates).
Decomposition by Age
As suggested in the previous section, the overall labor force participation rates for Arkansas and the U.S. can differ for two reasons. Both the differences in age distributions, and the differences in participation rates within age cohorts contribute to the total gap between U.S. and Arkansas labor force participation rates.
To decompose the total gap into two distinct components, we can write the labor force participation rate, R, for either Arkansas or the U.S. as the share-weighted sum of age-bracket components. For n distinct age-group categories,
Where Siis the share of the population in cohort i, and Riis the participation rate for that cohort. The overall labor force participation rate is then the sum of the n components.
With this decomposition in hand, we can investigate two hypothetical questions:
What would the Arkansas labor force participation rate be if the age distribution in Arkansas matched that of the U.S.?
What would the Arkansas labor force participation rate be if participation rates within each age group matched those of the U.S.?
The data used for this exercise are the 2022-23 averages shown in Figures 4 and 5, which show the U.S. participation rate to be 62.4% and the Arkansas participation rate to be 57.4% — a gap of a little over 5.0 percentage points.
The results of this decomposition are shown in Table 1. The first line of Table 1 shows the actual participation rate for Arkansas, as well as the U.S. minus Arkansas gap. The breakdown by age group for the first line of the table displays each of the Si Ri terms from the equation above. The contributions from the youngest two age groups are negative (i.e., narrowing the gap) since participation rates within these cohorts are slightly higher in Arkansas than for the U.S.
Table 1:
The second line refers to Question 1: If we use U.S. population shares instead of Arkansas population shares, the Arkansas participation rate would be 58.3% — nearly a full percentage point (0.95) higher than the actual rate. The components by age group in the second line show the portions of the gap that are explained by the hypothetical. The contributions are negative at both the low end and high-end of the age ranges because U.S. population shares are smaller for those age groups. If the population distribution in Arkansas was the same as the U.S., those high and low age brackets would be smaller and therefore contribute less to the overall labor force participation rate. In the middle-age ranges U.S. population shares are higher, so raising the Arkansas population shares would tend to raise the overall participation rate, closing the gap between the U.S. and Arkansas.
The next line refers to Question 2: If labor force participation rates for all age cohorts were the same in Arkansas as for the U.S.—holding population shares constant—the overall labor force participation rate for Arkansas would be 61.5%, a gap of less than one percentage point. We call this the “behavioral” component of the gap because it refers to the labor force decisions made by individuals within each of the age brackets. In this hypothetical, the lower age brackets show negative contributions because imposing the (lower) U.S. participation rates for these age groups would reduce their contribution to the total. For the older age brackets, the effect of raising participation rates would help close the overall gap between the U.S. and Arkansas.
The final line in the table is labeled as a residual. It is the portion of the gap that is not explained by either of the two hypothetical exercises individually, but represents the small component that is explained by the interaction between differences in population shares and participation rates.
Among the contributions by age group in Table 1, the greatest impact in closing the gap between the U.S. and Arkansas participation rates would be raising the participation rates for those in the 45-54, 55-64 and 65+ age cohorts, each of which would close the gap by more a percentage point or more. The demographic mix matters too. As noted above, the smaller population shares for the middle age cohorts (25-54) are important contributing factors to the gap.
Decomposition by Age and Sex
In the previous section, we considered n=7 distinct age cohorts. The decomposition can be expanded to consider more detailed breakdowns as well. In the available data for Arkansas, the youngest and oldest age brackets are not broken down by sex. However, the age brackets in the range of 20-64 are reported separately for men and women.
Figures 6 and 7 show the labor force participation rates and population shares, respectively, for the U.S. and Arkansas.
In Figure 6, the breakdown of the rates for men and women shows some interesting differences. For the 20-24 age bracket, Arkansas participation rates are higher for both men and women. In the 25-34 cohort, however, the Arkansas participation rate is higher than the national average for men, but notably lower for women. Arkansas participation rates are lower than the U.S. for both men and women in the 35-64 age brackets, particularly in the 45-54 and 55-64 age ranges. Among men aged 45-54, the U.S. minus Arkansas gap is over 11 percentage points.
Figure 6:
Figure 7 shows that the relative sizes of population shares are generally similar for men and women. The exception is for the 25-34 cohorts, where the population share in Arkansas is smaller for men, but larger for women.
Figure 7:
Table 2 shows the decomposition of the U.S. minus Arkansas participation rate gap, with the more detailed breakdown of population cohorts. The two hypothetical calculations show the same overall result (with tiny differences associated with using the sums of averages over two years). The contributions of the youngest cohort and the oldest cohort are identical, since those to categories are the same.
In the second row of Table 2, using U.S. population shares and holding the labor force participation rates of the individual cohorts constant, the effects for the 20-24 and 55-64 cohorts widen the gap for both men and women, reflecting the fact that Arkansas has higher population shares in each of these categories. The contributions of men 25-34 and women 25-34 differ in sign, reflecting the relative magnitudes of the U.S. and Arkansas shares for those cohorts. The positive contributions to closing the gap are primarily for men and women in the 35-54 brackets, with a more substantial impact coming from the low population share for men aged 45-54.
The third row calculates the reduction in the gap that would take place if all participation rates within the age and sex cohorts were the same for Arkansas as the U.S. For older workers, raising the Arkansas participation rate would significantly close the gap, particularly for men. An important difference between the gender cohorts is for the 25-34 age range: Arkansas men in that age group display slightly higher participation rates in Arkansas, while the participation rates of women in that age range are lower than the U.S.
Table 2:
Conclusions
The decompositions described in this article are merely an arithmetic exercise, designed to identify how the relative sizes and labor force participation rates of demographic age and sex cohorts contribute to the gap between the overall participation rates of the U.S. versus Arkansas. The key results are quantified in two numerical decompositions, but they derive from some basic observations about the patterns in the data.
If Arkansas had more prime working-age workers (25-54) and fewer younger workers (16-24) and fewer older workers (55-64 and 65+), the higher participation rates for prime age workers would increase Arkansas’ labor force participation rate, but only by about 1 percentage point.
If the participation rates of older workers in Arkansas was higher, particularly in the 45-54 and 55-64 age brackets, a considerable portion of the gap between the U.S. and Arkansas would be narrowed. This is particularly true for men, but applies to a slightly lesser extent to women as well. The net effect of having participation rates in Arkansas as high as the national averages would be to narrow the gap from 5 percentage points to 1 percentage point.
The most notable difference between the participation rates of men and women can be seen in the 25-34 age range. The participation rate for Arkansas men is higher than the national average, while the rate for women is noticeably lower.
The differences between U.S. and Arkansas participation rates for individual age and sex cohorts have been referred to as behavioral, but a deeper examination would likely yield demographic explanations for those differences as well. For instance, the higher participation rates for Arkansas men and women in the 20-24 age range is probably related to the lower rate of higher education enrollment in Arkansas. People tend to enter the labor force at a younger age.
Similarly, the relatively sharp drop-off in participation rates for older workers in Arkansas could be associated with a higher prevalence of jobs for which disability or other early-retirement motivations are factors.
The distinctive pattern of participation rates for men and women in the 25-34 age range corresponds to prime child-rearing years. We might speculate that Arkansas has more families in which the man is working and the woman is at home taking care of the children. This could, in turn, be related to the proportion of children (<16 years old) in the population.
These speculative explanations for the labor force participation gap raise interesting questions that are worthy of further investigation and research.
The labor force participation rate is defined as the Labor Force (employed plus unemployed) expressed as a proportion of the population (civilian noninstitutional population age 16 and over). We can divide both the numerator and denominator into the sum of distinct components, i = 1, 2, …n. It is straightforward to show that this ratio of sums can be expressed as a share-weighted sum of individual participation rates for each cohort Ri , where the share-weights are defined as population shares, Si .
