How Much Does Wealth Matter in the Racial Health Gap?

Background and Objective The now common public health framework of ‘social determinants of health’ (aka social causation) focuses on how aspects of the social and economic context shape health. Wealth resources such as liquid financial and non-financial assets (home equity, vehicles) have been mostly omitted. Over time an alarming racial wealth gap has emerged in the US due in large part to government policies that have facilitated the differential accumulation and protection of wealth for some racial groups (namely non-Hispanic Whites) at the expense of others (Oliver & Shapiro, 1995; Conley 1999; Krivo & Kaufman, 2004). Despite a growing recognition of wealth as a driver of social and economic stratification (Pfeffer & Schoeni, 2016), little is known about wealth’s role in shaping health inequality. One study showed that within-group net worth position was positively associated with health status for African Americans and Non-Hispanic Whites, but not Latinos (Pollack et al 2013). We are not aware of any work that accounts for changes in patterns over time or specifically quantifies how much wealth matters for determining the racial health gap.

Methods We used eight waves of the nationally representative Survey of Consumer Finances (sampled every three years from 1995 to 2016; n= 243,144 person-years). The dichotomous dependent variable was fair/poor health. Disposable net worth (total assets less total debts) was measured at the household level for working aged household heads. Racial/ethnic groups included non-Hispanic White, African American, and Latino. Within-group wealth quartiles were coded for each race/ethnicity group. Trends over time were estimated with survey-weighted descriptive statistics. A series of linear probability regression models predicted poor health based on within-group wealth quartile, along with demographic variables. Year fixed effects were included to control for macroeconomic context.

Results Large health disparities were observed in the pooled sample: approximately 19% of non-Hispanic Whites reported poor health compared to 29% and 30% of African Americans and Latinos, respectively. Median wealth was significantly lower for respondents with fair/poor health ($13,880/$6,427), compared to those with good/excellent health ($39,267/$63,883). Moreover, median wealth for African Americans and Latinos with excellent health ($10,042/$12,021) was about the same as non-Hispanic Whites with poor health ($11,523).

Regression models (a) revealed a racial health gap that persisted after controls were added, (b) demonstrated a negative relationship between relative wealth position and poor health, (c) illuminated three ethnicity-specific associations for how wealth shapes the racial health gap. Decomposition of the non-Hispanic White/African American and non-Hispanic White/Latino health gaps revealed that if African Americans in the bottom quartile of their wealth distribution had the same returns as similarly positioned non-Hispanic Whites, the health gap would be reduced by about 4.3 percentage points (2.95 percentage points for Latinos).

Conclusions and Implications: Our study provides further evidence that household wealth shapes both health overall, and health inequalities across racial/ethnic groups in the US. Policies that build wealth among African Americans and Latinos may have positive downstream health implications, although our work shows any effects are likely to vary both across and within ethnic groups.