Beyond the Averages: A Decomposition Analysis of Racial Poverty Disparities Among White and Asian American Ethnic Subgroups

Background/Purpose: Racial and ethnic disparities in poverty are a persistent feature of inequality in the United States. While the empirical literature has extensively examined Black—white and Hispanic—white poverty gaps, the Asian American population remains understudied, particularly when disaggregated by ethnic group. As an aggregate, Asian Americans share some similarities with the white population, but this masks considerable variation. Many Asian ethnic groups experience poverty rates well above those of white Americans, with some nearing the levels observed among Black, Hispanic, and American Indian populations. This includes Chinese Americans—the largest Asian ethnic group in the U.S.—who alone comprise a quarter of the Asian American population.

To better understand poverty disparities, this study examines the poverty gaps between white Americans and the six largest Asian ethnic groups (Chinese, Indian, Filipino, Korean, Vietnamese, Japanese) using a decomposition approach. Specifically, I estimate the extent to which household- and individual-level demographic characteristics account for observed disparities in Supplemental Poverty Measure (SPM) rates.

Methods: The study draws on the 2015-2019 American Community Survey (ACS), a nationally representative dataset that provides rich demographic, labor market, and economic indicators, and uniquely identifies 21 Asian ethnicities. I use Fairlie decompositions for nonlinear models to quantify how much of the poverty gap between each Asian ethnic group and the white population can be explained by demographic characteristics.

Given prior findings that immigration and geography are major drivers of poverty differences between Asian and white populations, I focus on these indicators in the decomposition analysis. Immigration measures include nativity, citizenship status, and years since immigration; geographic indicators include urbanicity and state. Additional controls include education, employment, family structure, English proficiency, disability status, and welfare receipt.

Results: Measurable demographic characteristics explain 76%, 63%, and 67% of the poverty gaps between white Americans and Chinese, Vietnamese, and Korean Americans, respectively. Immigration factors, especially nativity, are key drivers across these groups. Educational attainment helps narrow the gap for Chinese and Korean Americans but is not relevant in explaining the Vietnamese–white gap.

Conversely, poverty gaps between white Americans and Indian, Filipino, and Japanese Americans are small and less interpretable. Still, their distinct profiles suggest meaningful variation and caution against assuming uniform advantage or disadvantage across all Asian ethnicities.

Conclusion and Implications: Studying the Asian American population requires a disaggregated approach to fully uncover racial disparities in poverty, which remains a key gap in the literature. The sizable unexplained portions of the decompositions highlight the potential influence of unmeasured factors, including structural racism and discrimination. These findings shed light on the importance of Asian ethnic-specific approaches in research, policy, and social services aimed at addressing economic insecurity among Asian American communities.