Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Within a decade, the share of households refusing to participate in the Current Population Survey (CPS) tripled. We show households that refuse 1 month but respond in an adjacent month account for an important part of the rise. Leveraging the labor force status of survey participants in the months surrounding their nonresponse, we find that rising refusals suppressed the measured labor force participation rate and employment–population ratio but had little effect on the unemployment rate. Notably, nonresponse bias accounts for at least 10% of the reported decline in the labor force participation rate from 2000 to 2020.