Perceived Financial Capability, Financial Knowledge and Financial Behaviors of Canadian Low-Income Households
In recent years, poverty research has focused more on financial decision making and behaviors. A number of social work studies have sought to understand financial capability and savings behaviors among the poor. Most of these efforts report univariate frequencies or summarize observed financial behaviors across domains. Few have taken a “person-oriented” approach to understand unobserved (latent) associations. In this study we address the following questions: (1) How can adults be classified based on patterns of financial behaviors? (2) How do the financial behavior classifications compare across the income distribution? (3) To what extent are perceived financial capability and financial knowledge related to financial behaviors?
Methods
The study is cross-sectional and used data from the 2009 Canadian Financial Capability Survey (CFCS) conducted by Statistics Canada. Data is nationally-representative and includes 15,519 adults selected through a two-phase stratified random sample. Financial behavior was measured through 17 dichotomous items across five domains: cash flow, credit, savings, investment, and other. Age was measured via eight ordinal categories. Household income was measured in deciles. Perceived financial capability was a summary of 5 items scored along a four-point scale. Financial knowledge was measured as the number of correct answers in response to a 14 item assessment of financial knowledge. Latent Class Analysis (LCA) mixture modelling was used to identify latent patterns of financial behavior across the 17 items. Covariates included age, income, financial capability, and financial knowledge.
Results
Seven LCA models were evaluated. A four class model was chosen based on the log likelihood (-116,088), AIC (232,318), BIC (232,854), entropy (0.679), and a parametric bootstrapped likelihood ratio test. Classes were termed (1) "high-functioning higher income" (19.8%); (2) “consumer” (30.2%); (3) "short-term saver” (23.8%); and, (4) "low-functioning lower income" (26.2%). The full paper presents graph of predicted probabilities of class membership by item. Bivariate differences were observed across class membership and age χ2(27, N = 13,870) = 4,463, p < .001 and income χ2(21, N = 13,870) = 2,871, p< .001. The high-functioning higher income class had highest mean scores on financial knowledge (10.1) and financial capability (14.7), while the low-functioning lower income class had the lowest mean scores, 7.42 and 12.7, respectively. Members of the consumer class compared to the short-term saver class had higher financial knowledge but lower financial capability scores.
Conclusion
This paper builds on existing knowledge of financial practices by taking a “person-oriented” approach and showing four distinct latent patterns in financial behaviors across a nationally-representative sample of Canada. We find that about one-fourth of the Canadian population is low-functioning across the observed behaviors. The short-term saver class can be seen as economically vulnerable based on the lack of long-term savings and planning for the future. Findings are relevant to social policy makers as well as financial institutions who aim to enhance financial inclusion. Future research is needed to (a) understand the causal links between financial capability and knowledge and financial behaviors, and (b) better understand the within group variation in financial practices among low-income households.