Income Research Paper Series – Research Paper
Low-income Dynamics and Determinants under Different Thresholds: New Findings for Canada in 2000 and Beyond
The existing analysis on low-income dynamics and persistence in Canada builds on the data of different sources available until the end of the 1990s and on single low-income thresholds. What are the patterns of low-income dynamics and persistence in Canada in 2000 and beyond? Would different low-income thresholds matter to the analysis of low-income dynamics and persistence? This paper attempts to address these and other questions to fill the void in the literature. Our findings are as follows:
First, the year-to-year low-income-transition probabilities are broadly consistent under all three thresholds and across the two panels. Following these people year to year by their low-income state, we found that the low-income-transition pattern is quite robust. While two-thirds of the low-income population and 97% of the non-low-income population remained in the same income state as the previous year, about one-third of the low-income population emerged from low income and about 3% of the non-low-income population fell into low income. The low-income transition for women improves under all three low-income thresholds over the two panels. A similar pattern can be found in the male or the total population under some low-income thresholds, but this pattern is not robust across all thresholds.
Second, we find that many Canadians experienced transitory low income during the study period. We saw remarkable patterns of life cycle transitions: young people, students, unattached individuals and lone parents are likely in low income for a short period, reflecting their life cycle transitions. Those who experience persistent low income typically account for a very small percentage of the total population. We found that certain groups are at high risk of falling into low income—people with less than high school education, people with activity limitations, members of visible minorities and recent immigrants. These findings are prevalent under all three low-income thresholds and are robust across Panels 3 and 4 of the SLID data.
Third, we found that the life cycle factors such as family composition dynamics, number of children, age and student status affect the probability of falling into (or staying in) low income.
Fourth, we disentangle the determinants for transitory and persistent low income. Student status and recent immigration are more likely key determinants for transitory low income; family composition (unattached and lone parents), activity limitation and less education are more likely key determinants for both transitory and persistent low income. Although we identified gender differences in low income for various durations in both panels, gender does not appear to play a significant role in determining persistent low income after we control for confounding factors. The observations are quite robust—no matter which low-income thresholds and no matter which panels of the SLID data are used.
Finally, we find that low-income persistence becomes less severe over time for several vulnerable groups—for example, women and lone parents. But the overall low-income duration worsened slightly from Panel 3 to Panel 4, under all three thresholds.
We notice that this research has some limitations. On the one hand, we note that some senior citizens have low income but may have accumulated some wealth. In this research we were not able to link the flow (income) to the stock (wealth): we could only focus on income. On the other hand, LICO, LIM and MBM are taken as given in this research. Therefore, our estimates on low-income dynamics and persistence are conditional on the appropriateness of the thresholds. Nevertheless, this research serves as a timely update on the trend in low-income dynamics and persistence in Canada.
- Date modified: