Prevalence and Predictors of Asset Poverty in Canadian Families with Children

Background: According to UNICEF (2012) four in thirty (13%) children in Canada live in low-income households. While income provides an important proxy measure for short-term well-being, it may not fully capture aspects of long-term well-being. Assets have been theorized to influence child well-being by: decreasing the amount of family stress and by increasing the children’s school focused goals. Research has also shown that the level of assets in a family is related to a child’s educational, cognitive, and behavioral outcomes. In the US, research has shown that families with children compared to those without children have higher rates of asset poverty (Caner & Wolff, 2004; Haveman & Wolff, 2004). This study aimed to close part of a large international research gap by beginning to explore the prevalence and predictors of asset poverty in Canadian families with children under 18.

Methods: We examined individual and household level microdata from two cross-sections of the Survey of Financial Security from 1999 (N =  12,125) and 2005 (N  = 5,282), which were gathered from a nationally representative multi-stage stratified clustered survey design. Four measures were developed to indicate asset poverty. Asset poverty is typically defined as an absolute measure of sufficient wealth-type resources for a limited period of time. In this case, we defined sufficiency as having assets totaling more than the low-income cutoff amount allotted for period of three months with a family equivalence scale. We measured assets in terms of both financial assets (all liquid assets except pensions) and net worth (total financial assets plus durable assets minus debts). Survey weights and bootstrap replicate weights were used in all analyses to adjust for complex survey design and population.

Findings: While 17.4% of families with children were estimated to be income poor in 1999, 34.0% were asset poor based on net worth and 63% were asset poor based on financial assets. In 2005, these rates dropped slightly to 16.4%, 30.4% and 57.9%, respectively. The predictors of asset poverty and income poverty were notably different for families with children. In 1999, families with children had nearly double the odds of those without children of being asset poor, after controlling for age, sex, marital status, education level, language spoken, immigration status, homeownership, income, and full time work (OR = 2.00, p< .001). The same pattern was evident in 2005. Family household heads that were homeowners, with higher education levels, who were older full-time workers were significantly less likely to be asset poor.

Conclusions/Implications: This study elaborates on a new method of measuring well-being for families with children, and finds that the predictors of asset poverty are systematically different from those of income poverty. A majority of Canadian families with children are financial asset poor, indicating that many families may face straightened circumstances should they suffer an income shock. The implications for interventions to improve the long-term well-being of children and families are potentially profound. Policies (e.g. financial access initiatives) are recommended to increase the availability of asset-building mechanisms for vulnerable and middle-income families in Canada.