Section 5: Descriptive analysis

5.1 University-educated graduates with low earnings — demographics
5.2 Low earnings university-educated graduates – what are they doing?
5.3 Low earnings university-educated graduates – what did they study?
5.4 Low earnings university-educated workers – family situation
5.5 College-educated low earners - demographics
5.6 College-educated low earners – what are they doing?
5.7 College-educated low earners – what did they study?
5.8 Low-earnings college-graduates – family situation

This section provides a descriptive overview of college- and university-educated workers in Ontario in 2006. Data are provided at the Canada level for comparative purposes. For ease of comparison, tables for Ontario and for Canada are provided in two separate appendices (Ontario – Appendix A; Canada – Appendix B). Descriptive statistics for university and college graduates in 2006 are shown in Tables A.1 – Ontario and B.1 – Canada. The discussion begins with university graduates and follows with college graduates.

5.1 University-educated graduates with low earnings — demographics

As was noted in Table 2.1, 16% of university-educated graduates aged 25 to 64 in Ontario earned less than half of the national median income in 2006. The corresponding figure for Canada is 18%.

As might be expected given previous literature on gender differences in fields of study, industry and occupation and the amount of labour offered, university-educated females in Ontario were slightly over-represented in the three lowest employment earnings categories compared to males (ratios of 1.3, 1.2 and 1.3) and under-represented in the highest earnings category, with a ratio of 0.6 (Table A.2). Similar results are found at the Canada level (Table B.2).

The distribution across earnings categories by age was very polarized. Older university-educated adults in Ontario, particularly those aged 60 to 64, were over-represented in the lowest earnings category (ratio of 2.4). This was also true of workers aged 55 to 59 (ratio of 1.5). The higher ratios for these age groups suggest that some of these workers may have entered into retirement and that employment earnings were no longer their main source of income in 2006.

The other group that was over-represented in the lowest earnings categories in Ontario consisted of university-educated adults aged 25 to 29 (ratio of 1.4). One of the reasons why this might be is that 34% of university graduates in this age group who had low earnings reported that being a student was their main activity for the year. In addition, given their age, many would be relatively recent entrants to the labour market and it is expected that their earnings would be lower than in the case of older workers with more experience. Job-matching theory would suggest that some of them may be undergoing a period of employment instability while searching out the right job match.

University-educated workers most likely to be in the highest earnings category in Ontario in 2006 were those aged 45 to 49 and 50 to 54 (ratios of 1.4 and 1.6, respectively); they were also least likely to be in the lowest earnings category.

Who the worker lived with also had a relationship with whether or not they would be over-represented in the various earnings categories. For example, unattached individuals who were living in multi-person households and people living in 'other' family arrangements were over-represented in the lowest earnings category and under-represented in the highest earnings category. Conversely, married or common-law couples living with children under the age of 18 were slightly under-represented in the lowest earnings category and slightly over-represented in the highest.1

As might be expected given the literature on the challenges immigrants face in the labour market, immigrants, particularly those who had immigrated to Canada within the last ten years, were over-represented in the lowest earnings categories. In this respect, however, immigrants living in Ontario tended to be better off than in Canada as a whole. In Ontario, 20% of recent, university-educated immigrants (in Canada for 10 years or less) were in the lowest earnings category (at or below the national median employment income) in 2006, compared to 15% of their Canadian-born counterparts. Comparable figures at the Canada level were 23% of recent immigrants compared to 17% of the Canadian-born.

The earnings gap was much larger at the other end of the scale. In Ontario, only 14% of recent immigrants were in the highest earnings category, compared to 40% of the Canadian-born. A similar proportion of recent immigrants was in the highest earnings category at the Canada level in 2006 (14%); however, a smaller share of Canadians overall were in the highest earnings category, at 34%.

The Canada-level analysis also included a component that explored the extent to which differences were apparent by region. With the exception of British Columbia, university graduates' province of residence had little effect on whether or not the graduate would fall into a given earnings category. In British Columbia, however, fully 27% of university-educated adults aged 25 to 64 fell into the lowest earnings category in 2006 (Table B.2).

5.2 Low earnings university-educated graduates – what are they doing?

In Ontario, slightly more than half (52%) of low-earnings university graduates reported that working at a job or business or being self employed was their main activity in 2006 (data not shown.). The remaining 48% of this group reported taking care of children (12%), being retired (14%), being a student (13%) or other activities (9%). These shares were very similar at the Canada level.

As might be expected, those who did not list working at a job as their main activity for the year were over-represented in the lowest earnings category. Thus, 72% of retirees, 61% of students, and 46% of people who were caring for a child or a family member were in the lowest earnings category in Ontario in 2006 (Table A.3). These figures were higher at the Canada level in the case of students (71%) and people who were caring for a child or family member (53%) (Table B.3). In Ontario, just 10% of those who reported that working was their main activity for the year fell into the lowest earnings category; this figure was 11% at the Canada level.

The strength of an individual's attachment to the labour force also played a role. Workers who worked full-time for the whole year were, by far, the least likely to be in the lowest earnings category in Ontario in 2006 (ratio of 0.4), compared to part-time workers (ratio of 2.8), those with mixed schedules (ratio of 2.4) and those who did not report 'working' as their main activity for the year, though they did report non-zero earnings (ratio of 4.7).2 These ratios are very similar at the Canada level. Of those low-earnings university graduates who were working part-time, 16% reported wanting to work full-time; in Ontario, this proportion was only 6%.

Strong differences are observed when earnings distributions are examined by major source of income. As one might expect, workers whose main source of income was wages and salaries were under-represented in the lowest earnings category in Ontario in 2006 (ratio of 0.6). Self-employed workers, on the other hand, were over-represented in the lowest earnings category (ratio of 1.7). However, workers whose main source of income was government transfers, investment income and retirement pensions had employment earnings such that they were strongly over-represented in the lowest employment-earnings category (ratios of 6.1, 4.9 and 5.2, respectively).3

Box 5.1
A note on self-employed workers

The earnings distribution of self-employed workers differs from that of workers who work for wages and salaries. As Survey of Labour and Income Dynamics (SLID) income data is derived from tax files, this is likely related to the way in which self-employed workers report their income. Since self-employed workers are able to claim expenses for their businesses, they frequently report negative self-employment earnings. Tables A.4 and B.4 show the earnings distribution for self-employed workers versus workers who are not self-employed. As can be seen from this table, self-employed workers were over-represented in the lowest earnings category.

In this study, workers were considered self-employed if they reported that their main job4 was as a self-employed worker. These workers have, over the past 13 years, accounted for about 16% of the total population aged 25 to 64 who earned non-zero employment income. As illustrated in Charts A.1 and B.1, this percentage has been fairly constant, although it did dip below 15% in the years 1996 to 1999. Workers who had less than a high school education were more likely to be self-employed, while college-educated workers were less likely (although in recent years, this percentage has been rising). Workers with a university education or with high school or a trade reported similar rates of self-employment to that of the total population.

Certain occupations were also associated with being in the lowest earnings category. Workers in occupations in art, culture, recreation and sport (ratio of 1.9) and in sales and service (ratio of 1.5) were over-represented in the lowest earnings category in Ontario and at the Canada level. Conversely, those who were working in occupations in management, natural and applied sciences and health were over-represented in the highest earnings category.5

5.3 Low earnings university-educated graduates – what did they study?

The earnings distributions of university-educated workers by field of study are shown in Table A.5 (Ontario) and Table B.5 (Canada). Overall, for most fields of study, similar proportions of the university-educated fell into the lowest earnings category in Ontario and at the Canada level. Those most likely to report low earnings were individuals who had studied psychology (ratio of 1.6 in Ontario and at the Canada level) and, at the Canada level, the humanities (Canada 1.4; Ontario: 1.1).

Having completed one's postsecondary studies outside Canada was also related to earnings. As was seen in the analysis by province, graduates who had received their degree in British Columbia were over-represented in the lowest earnings category and under-represented in the highest one. In addition, graduates who received their degree outside Canada were also over-represented in the two lowest earnings categories.

5.4 Low earnings university-educated workers – family situation

Individuals' choices regarding their participation in the labour market often are determined with respect to earnings of other family members. Data placing individuals within their family context are shown in Table A.6 (Ontario) and Table B.6 (Canada).

Among the population aged 25 to 64 in Ontario in 2006, the group that was most over-represented in the lowest earnings category was children with some employment income living at home (ratio of 2.7) followed by spouses of the major income earner (ratio of 1.7). In contrast, the major income earners were under-represented in the lowest earnings category (ratio of 0.5) and over-represented in the highest earnings category (ratio of 1.1). Findings are similar at the Canada level.

That being said, individual earnings and family income were highly correlated. Individuals in lower income families in Ontario were themselves over-represented in the low earnings category (ratio of 4.3). High-earning individuals also tended to be in families for which earnings were also high. In other words, higher earners tended to contribute to higher income families while lower earnings workers belonged to lower income families. As well, workers who came from a family where there was only one earner were over-represented in the lowest earnings category. Again, findings are similar at the Canada level.

5.5 College-educated low earners - demographics

In 2006, 24% of college-educated adults aged 25 to 64 in Ontario earned less than half the national median employment earnings (Table 2.1). Females were more likely than males to fall into the lowest earnings category (ratio of 1.3, compared to 0.7) and less likely to fall into the highest earnings category (ratio of 0.5, compared to 1.5) (Table A.7). These gaps between males and females were similar at the Canada level (Table B.7).

As was the case for those with a university education, the earnings distributions of college workers were highly polarized by age, with individuals aged 20 to 29 and those aged 55 to 59 and 60 to 64 being over-represented in the lowest earnings category in Ontario in 2006. The situation for college-educated individuals aged 55 to 59 is worthy of note. Compared to the Canada level, larger proportions of individuals in this group were found in the lowest earnings category (ratios of 1.3, Canada, and 1.5, Ontario) and smaller proportions were found in the highest earnings category (ratios of 1.2, Canada, and 1.0, Ontario). Workers aged 60 to 64 was the group that was most likely to fall into the lowest earnings category in Ontario and at the Canada level (both with ratios of 1.8).

With respect to family situation, college-educated lone-parent families and unattached individuals living in multi-person households were over-represented in the lowest earnings category in Ontario (ratios of 1.4 and 1.3, respectively). This finding differs from that found at the Canada level, where it was married or common-law couple with children who were over-represented in the lowest-income group (ratio of 1.3). At both the Ontario and Canada levels, unattached individuals in one-person households were under-represented in the lowest-income group (ratios of 0.6 and 0.7, respectively).

As in the case of those with a university education, some differences are apparent when college-educated immigrants are compared to their Canadian-born counterparts. In Ontario, college-educated immigrants who were born in Asia or in Europe were over-represented in the lowest income group (ratios of 1.3 and 1.4, respectively). Similarly, immigrants who had been in Canada for any period of time (that is, both recent and established immigrants) were slightly over-represented in the lowest income group compared to the Canadian-born.

Some small differences are apparent by province of residence, with higher proportions of the college-educated population living in Newfoundland and Labrador and in Nova Scotia being at or below half of the national median employment earning compared to other provinces (ratios of 1.4 and 1.2, respectively) and smaller proportions being in the highest earnings category (Table B.7); the ratio for Ontario was 1.1. At the other end of the scale, those with a college education in Alberta and Ontario were slightly over-represented in the highest earnings category (ratios of 1.3 and 1.1, respectively).

5.6 College-educated low earners – what are they doing?

An individual's main activities for the year play a key role in determining earnings levels. College-educated individuals who were employed, whether full-time or part-time, were under-represented in the lowest earnings category (ratio of 0.7, Ontario and Canada), compared to those whose main activity for the year was keeping house / caring for children, being retired or being a student (each with ratios close to 3.0 at both the Ontario and Canada levels) (Tables A.8 and B.8). Working hours also play a key role – college-educated individuals working full-time, full-year were least likely to be in the lowest earnings category (ratios of 0.5, Ontario and Canada), compared to those working part-time and those with mixed schedules (ratios close to 2.0, Ontario and Canada). For the small proportions of individuals who reported having some employment earnings in 2006 even though working was not their main activity for the year, those ratios were closer to 3.0.

On a related note, earnings distributions are strongly influenced by an individual's major source of income. At both the Ontario and Canada levels, individuals whose major source of income for the year consisted of wages and salaries were under-represented in the lowest earnings category (ratios of 0.5). Self-employed workers, on the other hand, were over-represented in the lowest earnings category (ratios of 2.4 for both Ontario and Canada). In both cases, these ratios were much higher than was the case for university-educated self-employed workers. Finally, individuals who reported having some employment income in 2006, but whose major source of income consisted of government transfers, investment income or income from retirement pensions were highly over-represented in the lowest earnings category.

Occupations which tended to be over-represented in the lowest earnings category among college-educated individuals were those in sales and services (Ontario and Canada) and in occupations unique to primary industry, with Ontario ranking above the Canada level (ratio of 2.2, Ontario, and 1.8, Canada).

5.7 College-educated low earners – what did they study?

Among college-level fields of study, college-educated individuals whose fields of study were in agriculture, personal and culinary services, family and consumer services, and visual and performing arts were over-represented in the lowest earnings category at both the Ontario and Canada levels (Tables A.9 and B.9). Conversely, college graduates who had studied in security and protective services, engineering, mechanics, construction trades, and precision production were over-represented in the highest earnings categories at both the Ontario and Canada levels. Notably, the 2006 Census showed that the fields of study which were over-represented in the lowest earnings category were among the top 10 fields of study for women (personal and culinary services, family and consumer services and visual and performing arts) while the fields of study that were over-represented in the highest earnings category were among the top 10 fields of study for men (mechanic and repair technologies, construction trades and precision production) (Statistics Canada 2008b). As noted earlier, gender differences in fields of study and occupation may explain in part why women are over-represented in the lowest earnings category compared to men.

5.8 Low-earnings college-graduates – family situation

College-educated workers who were the major income earner in their family were under-represented in the lowest earnings category and over-represented in the highest, whereas their spouses and children living at home were over-represented in the lowest earnings category (Tables A.10 and B.10).

Family income and worker earnings were highly correlated, with workers in lower-income families being over-represented in the lowest earnings category themselves, while individuals in high-income families were over-represented in the highest earnings category. Thus, it was not the case that low earners were living in high income families – the vast majority of them were living in families who were in the two lowest income categories.

Having identified key worker characteristics that are associated with falling into a low employment-earnings situation, the analysis now turns to considering these variables in a multivariate statistical analysis in order to determine their independent effects when controlling for the effects of the other variables in the model.


Note

  1. Picot et al. (2005) note that single-earner families have a higher risk of low income in all of the countries they studied.
  2. Although 15% of all respondents in our population of interest reported something other than working as their main activity for the year, they also reported employment earnings and so fell into the population of interest.
  3. The earnings data refer only to earnings from employment and do not include income from other sources, such as government transfers, investment income and retirement pensions. As a result, the information provided here does not indicate whether individuals with other sources of income in addition to earnings from employment are in a situation of low total income or not.
  4. The main job for the year is defined as the one with the most paid hours in the year. If hours are identical between two jobs, the main job is the one with the greatest earnings or the longest tenure (if earnings are identical).
  5. Small sample sizes preclude a detailed discussion of earnings distributions by industry at the Ontario level, though the data do point to a higher proportion of earners in retail trade falling into the lowest earnings category.
Date modified: