The wealth and finances of
employed low-income families

View the most recent version.

By May Luong

Full article in PDF

• Profile of employed low-income families
• Among low-income families, the employed have higher median wealth
• Assets
• Debts and liabilities
• Employed low-income families less likely to be behind on payments than others in low income
• Employed low-income families more likely than other low-income families to have enough savings to cover unexpected expenses
• Employed low-income families more likely to prepare for retirement than other low-income group
• Conclusion
• Data source and definitions

In 2008, over 4.1 million individuals were living in low-income families.¹ While many people in low income relied on government transfers, 37% of these people were also part of a family in which someone was employed for at least one-half of the year.²

Employed low-income families have been the subject of several recent studies.³ One study found the average income of individuals living in employed low-income families to be less than one-third of the income of individuals in other employed families (Fleury and Fortin 2004). Although fewer individuals in employed low-income families work full year, full time, their average hours worked are on par with other workers at around 2,000 hours (Fleury and Fortin 2006).

Low-paying jobs are often associated with employed low-income families. However, while low pay was found to be a significant risk factor, it was not the most important determinant of low-income status. Instead, the presence of one earner (compared to multiple earners) and other family characteristics were found to be more important than pay (Fleury and Fortin 2006). Fortin calculated that 3.4 million of the employed in 2002 would drop under the low-income line if they experienced a separation or divorce in the family, or if other earners in the family experienced unemployment (Fortin 2007). In addition, certain groups like immigrants were found to be more likely to be part of an employed low-income family (Fortin 2007).

Other studies compared spending in employed low-income families to other groups to assess their living standards. Results indicate that despite their stronger labour force attachment and slightly higher income than those in other low-income families, employed low-income families were more likely to borrow or liquidate assets to make ends meet, and they had more work-related expenses and less access to subsidized housing (Fleury et al. 2005). Nevertheless, individuals in employed low-income families experience health outcomes comparable to the employed non-low-income and score higher than the not-employed low-income on a number of health measures, both in a given year and over the longer term (Fortin 2008).

Although these studies shed light on the current income and consumption of employed low-income families, there remains a research gap regarding their wealth and financial situation. Wealth studies typically treat low-income families as a single group, rather than separating the employed from other low-income families. For example, one study reported that while not all low-income families have low wealth, the vast majority of low-income families have very little financial wealth (Morissette 2002).

Wealth is a key aspect of long-term well-being since some assets can be converted into cash for consumption during periods of economic hardship (Wolff 1998). Other assets may be more difficult to liquidate in a short period of time but often can be used as security for loans. Studying the wealth and financial security of individuals in employed low-income families can provide a more complete picture of their long-term financial well-being and ability to weather short-term difficulties.

Using the 2009 Canadian Financial Capability Survey (CFCS), this study examines the financial situation of individuals living in employed low-income families compared to not-employed low-income families and employed families not in low income. The CFCS provides the unique opportunity to look at respondents' perceptions of their financial situation and their estimates of household assets and debts during a labour market downturn. Since the response rates for the asset and debt questions were relatively low, the tabulations were replicated using the 2005 Survey of Financial Security (SFS). The comparison validated the main conclusions of this study but also indicated that the CFCS measures of asset and debt levels should be interpreted with caution (see Data sources and definitions).⁴

Although CFCS data were collected and processed at the individual level, the primary groups of interest were defined based on family income and wealth. Thus 'families' is used synonymously with 'individuals in families' for brevity in this article.

Profile of employed low-income families

In 2009, there were over 1.6 million people age 25 to 64 living in employed low-income families,⁵ representing 9% of the population in this age group (Table 1). There were just fewer than 1.2 million people living in not-employed low-income families, which represented 7% of the target population. Individuals in employed non-low-income families were in the majority, accounting for 77% of the target population of 14 million. The remaining 8% is made up of those in not-employed non-low-income families and is excluded from subsequent analyses (see Data sources and definitions).

The average age of those in employed low-income families is younger than all other groups at 42.3 years. Almost 60% of those in this group are from age 25 to 44. This group also had the largest proportion of women (58%), the largest household size (3.6) and more children (1.4), on average, than all other groups. Almost one-half of employed low-income families are couples with children and 15% are lone parents. Compared to those in low-income families with no one employed, those in employed low-income families are somewhat better-educated with 24% holding a college diploma and 18% holding a university degree (undergraduate and graduate).

The median household income for employed low-income families was $25,000 in 2009 compared to $15,000 for low-income families with no one employed. Since the not-employed families were smaller, the difference between the two groups shrank after adjustments were made for family size: from $10,000 to $4,000. Similar to previous findings, the adjusted household income of employed low-income families was less than one-third of that of employed non-low-income families.

Among low-income families, the employed have higher median wealth

Wealth or net worth is defined as the difference between a family's total assets and total debts.⁶ Since wealth varies by age, the results are standardized to the age distribution of individuals in employed low-income families to counteract the effect of age differences among the groups.

On average, employed low-income families have greater wealth than other low-income families, but lower wealth than employed non-low-income families. The median net worth of employed low-income families was $19,000 compared to $1,000 for other low-income families, and $257,700 for employed non-low-income families (Table 2).

Assets

The assets contributing to net worth highlight further differences among groups. The median value of total assets for employed low-income families ($60,000) fell between the other two groups: significantly higher than the other low-income group ($3,000) but less than one-sixth of the median of the other employed group ($389,200).

Home equity is the most valuable asset for many Canadians. More than one-half of employed low-income families owned their homes compared to just over one-third of the other low-income group.

Employed low-income families were also more likely to hold other financial assets [excluding Registered Retirement Savings Plans (RRSPs)] than other low-income families. Furthermore, 42% of the employed low-income had RRSPs compared to 22% among other low-income families. The liquidity of such assets can help families weather temporary shocks like the loss of a job or an unexpected expense.

Although Registered Education Savings Plans (RESPs) are used by relatively few low-income families, employed low-income families are twice as likely to hold an RESP compared to other low-income families (20% versus 10%).

About 1 in 6 (16%) employed low-income families held business assets—slightly less than the 18% of employed non-low-income families who held such assets.

Debts and liabilities

In addition to having a higher level of assets, employed low-income families also carried more debt, on average, than other low-income families. Just over one-half (56%) of not-employed low-income families carried no debt compared to 31% of employed low-income families.

The higher incidence of debt among employed low-income families was primarily due to mortgages. Notably, 42% of employed low-income families had mortgages on their homes compared to 16% of other low-income families. Employed low-income families also surpassed other low-income families in the incidence of all other types of debt except student loans. Counter to most financial advice, 4 in 10 employed low-income families carry outstanding credit card balances.

In summary, the average wealth of employed low-income families exceeded that of other low-income families, but was significantly lower than that of employed non-low-income families. Did these differences in wealth translate into other indicators of financial security? The next section looks at the financial security of employed low-income families compared to the other two groups.

Employed low-income families less likely to be behind on payments than others in low income

Over one-half of all families reported having a household budget (Table 3). A slightly smaller proportion of employed low-income families reported having a household budget (54%) than other groups. However, they were also less likely to report rarely or never staying within their budget (11%) than the other low-income group (14%). In other words, they were a bit better at staying on budget than the other low-income families.

Just over one-half of employed low-income families have monthly expenses under $2,000 compared to 3 out of 4 in the other low-income group. Despite higher spending, a smaller proportion of employed low-income families reported falling behind on payments.⁷

Employed low-income families more likely than other low-income families to have enough savings to cover unexpected expenses

Having a 'rainy day' fund helps during periods of financial hardship or given an unexpected expense. When asked what individuals would do given a $500 unexpected expense, 46% of those in employed low-income families said they would use savings to cover such an expense (Chart A). This is higher than the other low-income group (with no employed family members), among whom less than one-third would use savings to cover such an expense.

However, if the unexpected expense were $5,000, the proportion of employed low-income families who would use savings to cover the expense would only be slightly higher (17%) than the other low-income group (14%) (Chart B).

Employed low-income families would be less likely to borrow from friends or relatives than other low-income families (11% and 16%, respectively) but a higher proportion would draw on lines of credit or credit cards (39% versus 28%). For a larger unexpected expense, both low-income groups would be less likely to go to their friends or families for a loan. Instead, over one-half of employed low-income families would borrow the $5,000 from a line of credit or credit card, compared to just over one-third of the other low-income group. The not-employed low-income were more likely to report that they would be unable to pay (52%) than employed low-income families (30%).

Employed low-income families more likely to prepare for retirement than other low-income group

Over one-half of employed low-income families reported preparing financially for retirement compared to less than one-quarter of the other low-income families (Table 4). A slightly higher proportion of not-employed low-income families plan to rely on government pensions than employed low-income families (86% and 81%, respectively). However, a higher proportion of the employed low-income included employer pensions as a planned source of retirement revenue than others in low income (44% and 33%, respectively).

RRSPs also figured into many families' retirement plans. Employed low-income families were more likely to include RRSPs (65%) in their retirement plans than others in low income (50%). Although both low-income groups had lower intentions of using RRSPs in retirement than employed non-low-income families, RRSPs may not be the best retirement savings vehicle for many in low income. Since the main transfer program for low-income seniors, the Guaranteed Income Supplement (GIS), is reduced by 50 cents for each dollar of additional income above an income threshold, the advantages of investing in RRSPs are diminished for low-income families. Previous research indicated that low-income non-savers may be better off than those with modest savings given the GIS eligibility requirements in place at the time (Shillington 2003).

Working during the retirement years is another option. Despite the differences in their current situations, similar proportions of each group reported that they would at least partially rely on employment earnings when they retire—ranging from 50% to 53%. While working during the retirement years is likely to be financially driven, it is becoming more common and previous research has concluded that it may often be a choice rather than a necessity (Hébert and Luong 2008).

For those who reported they were not financially preparing for retirement, the most frequent reason was "can't afford to, don't earn enough, income too low." Employed low-income families were most likely to cite this reason (50%) followed by the employed non-low-income group (42%). On the other hand, 40% of the not-employed low-income group reported "don't have a job, haven't worked long enough" as their reason for not financially preparing for retirement compared to 14% of the employed low-income group—a reminder that the concept of retirement is, after all, linked to long-term labour force attachment.

When asked how confident respondents were that their household income in retirement would provide the standard of living they anticipated, few low-income families were very confident (12% for employed families and 14% for not-employed families). However, employed low-income families were more likely to report being fairly confident than the other low-income group (38% versus 28%). Although one-half of employed low-income families were very or fairly confident in their income adequacy in retirement, just one-quarter knew how much money would be needed to maintain their desired standard of living. This likely reflects the range of factors that can affect income retirement adequacy, as well as the range of opinions on the topic.

Conclusion

In addition to income, wealth is an important indicator of well-being since some assets could presumably be converted into cash for immediate consumption needs, especially during periods of economic hardship. This study examined the wealth, financial security and retirement plans of individuals living in employed low-income families compared to those in not-employed low-income families and those in employed non-low-income families.

On the whole, the wealth of employed low-income families was higher than that of not-employed low-income families, but was significantly lower than that of the employed non-low-income group. An examination of assets and debts adds nuances to this finding. While 69% of employed low-income families carried debt compared to 44% of the other low-income group, a large proportion of their debt took the form of residential mortgages. Much of their debt thus supported the long-term advantages of home ownership: greater wealth and lower housing expenses when the mortgage is paid off. However, employed low-income families were also more likely to carry consumer debt than the other low-income group. Notably, 4 in 10 employed low-income families carry outstanding credit card debt.

Indicators of financial security again highlight some differences between employed and not-employed low-income families, as well as their position relative to families not in low income. Employed low-income families were less likely to report not keeping up with payments than other low-income families, despite higher expenses. Nevertheless, when compared to the other employed group, employed low-income families were twice as likely to be behind in their payments.

Another indicator of financial security is how families would deal with an unexpected expense. Compared to others in low income, a smaller proportion of employed low-income families reported that they would not be able to cover the expense, whether the amount were $500 or $5,000. Moreover, the employed low-income group would be more likely to use savings to cover such an expense than the other low-income group. Altogether, these results indicate that employed low-income families were likely to feel more financially secure than the other low-income families but likely to feel less secure than families who weren't in low income.

Retirement planning also differed for the two low-income groups. Employed low-income families were more likely to have a plan that included more diverse sources of income than other low-income families. Families with a weaker connection to the labour market would be less likely to include workplace pensions or group RRSPs in their plans. Moreover, retirement planning may be a moot point for some since government pensions and other transfers to seniors replace a higher level of pre-retirement income for those near the bottom of the income distribution (LaRochelle-Côté et al. 2010).

Data sources and definitions

The Canadian Financial Capability Survey (CFCS) is a voluntary survey conducted in 2009, targeting persons 18 years of age and over. Full-time residents of Yukon, the Northwest Territories and Nunavut were excluded. Since the survey was conducted using a sample of telephone numbers, the 8% of households without telephones or with cell phones only were excluded. One respondent was selected from each household. The CFCS sheds light on respondents' personal knowledge, abilities and behaviours concerning financial decision-making. Information on their families' assets, debts, and net worth is also available. Although family-level responses may not apply to all family units in the sampled household, the data are weighted to represent all individuals in the target population.

One limitation of the CFCS is that only about 50% of the respondents completely reported their assets and net worth. Given the high item non-response rate, biased estimates of wealth differences among groups were a possibility. Prior to 2009, asset and debt information was most recently collected in the 2005 Survey of Financial Security (SFS). Although the number of responses to the 2005 SFS was smaller than the number of responses to the CFCS (6,000 compared to 15,500), the item response rates for wealth items were higher and imputed if missing. The SFS thus provides a ready source for the validation of CFCS estimates even though its sample size limits the precision of estimates for smaller population groups. As such, the asset and debt tabulations were replicated using the 2005 SFS. The results presented in this article would be substantially the same for each survey even though many estimates differed in level. Thus the CFCS can identify statistically significant differences in assets and debts among groups, although the levels may be biased and should not be used to infer trends in relation to the 2005 (or 1999) SFS.

The target population for this study includes individuals from age 25 to 64. Students are excluded. The target population is divided into four groups:

• individuals in employed low-income families
• individuals in not-employed low-income families
• individuals in employed non-low-income families
• individuals in not-employed non-low-income families.

The target population included 10,875 respondents and represented over 18 million individuals in 2009. The sample of individuals living in employed low-income families was 1,010. Only the first three groups are examined in the main analysis.

The employed low-income group must be framed within the household and family contexts, as household income is used to determine the group's low-income status and the family is used to determine employed or not-employed status. An employed family is defined as a family with at least one employed individual. Therefore an individual living in an employed low-income family may not actually be employed himself or herself. Assets and debts are also reported at the family level in the CFCS. Thus, the major units of analysis in this report are defined along family concepts. On the other hand, the household reference person rather than the family is the unit of analysis in the CFCS, and questions relating to financial security are directed to that individual. Furthermore, Low Income Measure (LIM) also uses adjusted household income observed at the person level.⁸ Therefore, this study more accurately examines 'individuals living in employed low-income families' rather than 'employed low-income families' or 'employed low-income individuals.' However, for simplicity, this paper will refer to 'individuals living in employed-low-income families' as 'employed low-income families,' and similarly so for comparison groups.⁹ See the appendix for a comparison of low income calculated using the CFCS and Survey of Labour and Income Dynamics (SLID).

Wealth (net worth) is defined as the difference between a family's assets and its total debts. Future entitlements to social security provided by the government such as Old Age Security, and Canada Pension Plan and Quebec Pension Plan benefits are not included as they were not available in the CFCS.

Appendix
Employment and low-income definition comparisons between SLID and the CFCS

Using the CFCS, individual employment status is identified using the variable LF_Q01, which asks about the respondent's employment status. Respondents are flagged as employed if they reported currently being employed or self-employed (regardless of the number of hours worked per week). Additionally, the variable LF_Q05 is similarly used to determine the employment status of the spouse.

In this study, low-income status is defined by adjusting the self-reported total household income before taxes by the square root of the household size.¹⁰ The low-income threshold for 2008¹¹ is $21,189¹² and is used to determine whether families are living in low income. Those who had adjusted total household income¹³ before taxes below the LIM threshold are flagged as living in low income. Finally, individuals are categorized as employed low-income, not-employed low-income or employed non-low-income based on their employment and low-income status.

Two employment definitions using the 2008 Survey of Labour and Income Dynamics (SLID) are used for comparison with the CFCS. Previously, Fleury and Fortin (2006) identified 910 hours as the threshold for being employed. They reasoned that an individual (or his or her spouse) should work for at least half the year in order to be considered employed. However, hours of work information was not available in the CFCS, thus an alternate definition was used: whether an individual (or their spouse) was employed at the time of the survey.

Results indicate the proportion employed estimated by the CFCS falls between the two SLID estimates (Table 5). The estimates using both surveys for the low-income estimates are very close, with the CFCS higher by 1 percentage point.

The proportion of individuals in employed low-income families in the CFCS matches that of the positive hours estimate in SLID (9%). The CFCS estimates for all the other groups fall somewhere between the two SLID definitions.

Overall, the proportion by employment and low-income status estimated by the CFCS is comparable to that for both measures using SLID. A closer examination of the sample profile by family type shows similar distributions between the SLID self-reported definition and the CFCS definition. Therefore, the samples are sufficiently consistent between the two surveys to conclude that the CFCS provides an accurate representation of the employed low-income group.

---

Notes

1. Estimated using the Low Income Measure (LIM) from the 2008 Survey of Labour and Income Dynamics. LIM is defined as 50% of the median of the adjusted household income over the population of individuals.
2. Using the 2008 Survey of Labour and Income Dynamics, the proportion of the low-income who were part of an employed family is based on the definition of an employed family where either the reference person or the spouse was employed a minimum of 910 hours during the reference year (Fleury and Fortin 2006). This proportion increases to 51% of all families when those with any work hours are included.
3. Previous studies have used the term 'working poor.' Statistics Canada does not measure poverty—it measures low income.
4. The inter-group differences in assets and debts were in the same direction and were statistically significant in both surveys, but varied in level. There was no clear pattern in the SFS–CFCS-level differences—they were negative in some cases, positive in others.
5. An individual was defined as living in an employed family if the respondent and/or his or her spouse was employed at the time of the survey.
6. Morissette et al. (2002) used the same definition of wealth as this study. However, it was not possible to examine 'financial wealth' using the CFCS since net housing equity and net business equity cannot be separated from total asset value.
7. The CFCS asked respondents whether they had been behind on various payments for two consecutive months or more.
8. LIM previously estimated the median over the population of families. However, this has been revised and it now estimates the median over the population of individuals. LIM is now defined as 50% of the median of the adjusted household income observed at the person level (Murphy et al. 2010).
9. Individuals living in not-employed low-income families may be referred to as 'the not-employed low-income' or 'the other low-income group.' Individuals living in employed non-low-income families may be referred to as 'the employed non-low-income' or 'the other employed group.'
10. In 2010, the equivalence scale was changed from a given weight depending on the age and number of family members to simply taking the square root of the household size (Murphy et al. 2010).
11. Although the CFCS was conducted in 2009, income is reported for 2008.
12. LIM is defined as 50% of the median of the adjusted household income over the population of individuals. In this analysis, the LIM threshold for 2008 ($21,189) was used as the threshold for determining whether an individual was in low income. This threshold was calculated using income data from the Survey of Labour and Income Dynamics and can be found in CANSIM Table 202-0808. Although the CFCS was conducted in 2009, the reference year for the income information is 2008. Therefore, the 2008 LIM threshold was used.
13. Another change made to LIM is the use of household income rather than economic family income (Murphy et al. 2010).

References

Fleury, Dominique and Myriam Fortin. 2006. When Working is not Enough to Escape Poverty: An Analysis of Canada's Working Poor. Working Paper. Ottawa.Human Resources and Social Development Canada. 174 p. (accessed June 28, 2011).

Fleury, Dominique, Myriam Fortin and May Luong. 2005. What Does it Mean to Be Poor and Working? An Analysis of the Spending Patterns and Living Conditions of Working Poor Families in Canada. Working Paper Series 007. Ottawa. Policy Research Initiative. 50 p. (accessed June 28, 2011).

Fleury, Dominique and Myriam Fortin. 2004. "Canada's working poor." Horizons.Vol. 7, no. 2. December. Policy Research Initiative. p. 51-57. (accessed June 28, 2011).

Fortin, Myriam. 2008. "How (un)healthy are poor working-age Canadians?Policy Options.Vol. 29, no. 08. September. p. 71-74. (accessed June 28, 2011).

Fortin, Myriam. 2007. The Role of Family and Government Financial Supports in Helping Canadian Workers Avoid Poverty. Catalogue no. HS28-117/2007E-PDF. Ottawa. Human Resources and Social Development Canada. 87 p. (accessed June 28, 2011).

Hébert, Benoît-Paul and May Luong. 2008. "Bridge employment." Perspectives on Labour and Income. Vol. 9, no. 11. November. Statistics Canada Catalogue no. 75-001-X. (accessed June 28, 2011).

LaRochelle-Côté, Sébastien, Garnett Picot and John Myles. 2010. "Income replacement during the retirement years." Perspectives on Labour and Income. Vol. 11, no. 8. August. Statistics Canada Catalogue no. 75-001-X. (accessed June 28, 2011).

Morissette, René. 2002. "Families on the financial edge." Perspectives on Labour and Income.Vol. 3, no. 7. July. Statistics Canada Catalogue no. 75-001-X. (accessed June 28, 2011).

Morissette, René, Xuelin Zhang and Marie Drolet. 2002. "Are families getting richer?" Canadian Social Trends.No. 66. Autumn. Statistics Canada Catalogue no. 11-008. p. 15-19. (accessed June 28, 2011).

Murphy, Brian, Xuelin Zhang and Claude Dionne. 2010. Revising Statistics Canada's Low Income Measure (LIM). Statistics Canada Catalogue no. 75F0002M – No. 004. Income Research Paper Series. Ottawa. 31 p. (accessed June 28, 2011).

Shillington, Richard. 2003. New Poverty Traps: Means-Testing and Modest-Income Seniors. Backgrounder. No. 65. April. Toronto. C.D. Howe Institute. 13 p. (accessed June 28, 2011).

Wolff, Edward N. 1998. "Recent trends in the size distribution of household wealth." The Journal of Economic Perspectives.Vol. 12, no. 3. Summer. p. 131-150. (accessed June 28, 2011).

Author

May Luong is with the Labour Statistics Division. She can be reached at 613-951-6014 or perspectives@statcan.gc.ca.