Using the Joint Distribution of Income and Asset Poverty to Reveal Economic Vulnerability
Relatively recently scholars have begun to contrast income poverty with measures of asset poverty. As such, poverty scholarship tends to focus on either income poverty or asset poverty. In this study I exploit the intersection of income and assets to identify three poverty groups: (1) the income poor non-asset poor, (2) the joint income and asset poor, and (3) the income non-poor but asset poor. This analysis provides an enhanced picture of economic vulnerability in a time of increasing inequality. The results have major implications for understanding economic insecurity and the role of the welfare state in addressing the condition.
Methods
Data from the Statistics Canada 2005 Survey of Financial Security (SFS) were used for analysis. The SFS is a cross-sectional multi-stage stratified clustered survey design providing nationally representative income, expenses, assets, and debt data from more than 5,200 households. Survey weights adjusted for population totals for province, age, sex, household size. A household was "income poor" if it fell below the low-income cutoff (LICO) threshold established by Statistics Canada. A household was asset poor when financial assets were insufficient to meet basic needs for three months (based on LICO 2005). Headcount poverty rates were calculated across and within demographic groups. A multinomial logistic regression model was used to predict membership in the three categories of poverty (non-poor as the reference category). Multivariate analyses accounted for the complex survey design by using bootstrap replicate weights for parameter and sampling variance estimation.
Results
Overall, 53% of Canadian society was financial asset poor in 2005, 95% CI [51, 55]. Female lone parent families, renters, and younger households were disproportionately represented among the asset poor. The joint income asset poverty analysis revealed the following groups: 4% income poor only, 95% CI [3, 4]; 14% joint income and asset poor, 95% CI = [12, 15]; 39% asset poor only, 95% CI = [37, 41]. In the multivariate results, female lone parents were more likely to be income poor only and joint income asset poor but not asset poor only. Married households were less likely to be in any of the three forms of poverty (average marginal effect of being non-poor = 13%). Speaking a non-official language (i.e., not English or French) was associated with an increased likelihood of income only poverty (OR = 2.59), but not the other two poverty categories. The full paper presents odds ratios and marginal effects across all age and family characteristics. Results are presented with numerous graphs and tables.
Implications
In a time of rising economic inequality, this study adds to the literature on economic vulnerability. For the first time, this paper shows that 39% of Canadian households are above the low income thresholds but do not possess sufficient financial assets to maintain living standards at the poverty line for three months. To decrease the prevalence of asset poverty, policies should target economic stability among female lone parent families, renters, and younger households. Future research is needed to better understand policy and behavioral factors that affect asset poverty.