The Impact of Market Factors on Racial Identity: Evidence from Multiracial Survey Respondents 
In 2003, the US Census Bureau and the US Bureau of Labor Statistics introduced multiple-race response options to the US Current Population Survey (CPS). Before then, respondents could identify themselves only as a single race. This paper uses the expansion of race-reporting options in the CPS to study whether certain market factors—specifically, racial composition and unemployment and wages by race—help causally determine racial identity.
Key Findings
- In the CPS, the pre-2003 self-reported race of multiracial individuals responds to the racial composition of their state’s population. Responses are due primarily to the racial composition of an individual’s household as opposed to the racial composition of people residing in the same state but outside the individual’s household.
- State-level unemployment rates also seem to influence racial identity but only at a one-month lag, suggesting that respondents may need time to learn about labor market conditions.
- Lagged state-level wages affect a respondent’s reported race when the wages pertain to the respondent’s gender- or education-defined subgroup. This finding might indicate that individuals need narrower reference frames for wages to affect their racial identity.
- Together, these results suggest that race may be an endogenous variable. However, estimation of how race affects individual-level labor market outcomes such as unemployment or wages indicates that any bias due to the omission of market factors from such analysis is likely minimal.
Implications
This paper’s findings offer encouragement that including race measures as regressors in individual-level estimation of racial disparities in labor market outcomes is econometrically permissible. However, given the relatively small population and period on which the paper focuses—multiracial persons in 2002—the findings may not be applicable to broader populations and periods. The paper’s results also suggest survey participants may react strategically and in unanticipated ways when their options for race-related responses are limited, and such behavior could affect the accuracy with which demographic trends are measured.
Abstract
This paper examines the reported race of multiracial persons in the US Current Population Survey (CPS) before 2003, when limited response options exogenously constrained respondents to identify as a single race. Using this survey attribute and the 16-month longitudinal design of the basic monthly CPS, I explore whether market factors help causally determine racial identity. I find that pre-2003 race responds to state-level (1) racial composition, due largely to household composition, and (2) unemployment rates and wages by race. Although these findings suggest potential endogeneity of race, estimation of how race affects individual-level labor market outcomes indicates minimal bias.