The Role of Health and Disability in the Trend in Prime-Age Labor Force Participation

Recent work has found declines in health and increases in mortality over the past few decades among certain US subpopulations (Case and Deaton 2015). Krueger (2017) briefly examined the potential role of pain, disability, and concomitant opioid abuse in declines in the labor force participation of prime-age workers. In this paper, I examine more formally and comprehensively the trends in health and disability and their role in the decline in labor force participation across the prime-age (25-54) working population in the US using data from the National Health Interview Survey for 1997-2016. First, I find health is declining significantly across all racial and education groups, as measured by self-reported health, even after controlling for shifts in the age composition of the population. The percent of prime-age population reporting they are unable to work because of a health problem rose by about 1.5 percentage points over this period. While some risk factors are improving such as smoking and educational attainment, the obesity rate in this population rose by around 50 percent (from about 20 to about 30 percent) over the two decades. The diagnosed prevalence of associated conditions such as diabetes, hypertension, stroke, low back pain, and asthma also rose significantly. I find explanatory factors for the trend in labor force participation by including a trend in a logit regression of labor force participation on covariates and testing the trend coefficient after adding and subtracting covariates. Education and family status explain only a small fraction of the decline in prime-age men’s labor force participation but adding measures of health and disability removes the trend entirely. For prime-age women, raw labor force participation has no trend in the NHIS but, after controlling for the considerable increase in women’s educational attainment in this period, labor force participation exhibits a significant decline. This decline disappears again after controlling for health and disability status. Health and disability are measured with self-reported health, diagnosed conditions, and activity limitations. The last are the most important factors in the decline in labor force participation holding everything else equal. The particular limitations whose increases appear to be most important for the trend in labor force participation are: difficulty standing for two hours, difficulty socializing, not being able to walk without equipment, activities limited by memory problems, and difficulty with any instrumental activity of daily living (shopping, taking medicines, moving around the community etc.). The rates of all these limitations in the prime-age population in the NHIS each rose by over a percentage point between 1997 and 2016, which is about equal to the overall decline in labor force participation. The trends in functional limitations seem to be driven primarily by reported increases in arthritis, back problems, musculoskeletal disorders, and nervous disorders. The role of obesity in the development of these conditions and therefore in the drop in labor force participation will be discussed.