CPS estimates of cohort-specific immigrant populations peak two to five years after immigration. Because cohort populations should decline due to emigration and mortality, this suggests undercounting in prior periods, particularly of younger immigrants and female immigrants. Assuming that this peak represents the true cohort population, this brief proposes using systematic deviations prior to the peak to upweight recent immigrant observations in the CPS. Summing these adjusted weights provides an alternative estimate of the number of recent immigrants. This approach predicts that the CPS typically undercounts recent immigrants by about 1 million and that this undercount has recently increased to about 2 million. These adjustments lead to relatively small changes in headline employment statistics; the most substantial is an estimated 0.05 percentage point increase in the unemployment rate. However, these corrections also increase recent CPS estimates of job growth by about 30,000 per month.
View the Full Brief
Brief
Delayed Sampling of Recent Immigrants in the Current Population Survey
January 2025
Research Brief — We show that the Current Population Survey (CPS) undercounts immigrants for up to five years after arrival. We provide methods to interpolate year of immigration and correct for undercounting by upweighting recent immigrants observed in the CPS.
Share
Download