January 2005
Vol. 6, no. 1

In the early 1990s, non-standard employment—temporary work, part-time work and self employment—seemed to be escalating at a worrisome rate. This led to fears that employment conditions were deteriorating for a substantial percentage of workers (OECD 2002; Krahn 1995; Schellenberg 1996). From 1997 to 2003, temporary work grew much more rapidly than permanent employment, despite a period of economic growth and favourable employment conditions. Indeed, GDP expanded during the entire period and the unemployment rate stayed below 8%.

Temporary jobs are generally less well paid than permanent ones, offer fewer employee benefits, are less likely to offer on-the-job training, and are precarious. Indeed, temporary work does not seem to be the first choice of many who do it. Recent data are sparse, but in 1994, two-thirds of temporary employees would have preferred permanent employment (Schellenberg 1996). Nonetheless, temporary employment has certain advantages for both employee and employer. Greater flexibility may allow employees to achieve a better balance between work and personal life, while employers can adjust operations to meet fluctuations in demand. Temporary work may also provide a gateway into the labour market or serve as a springboard to a permanent position (Kapsalis and Tourigny 2004).¹

On the whole, temporary work can be viewed as an inexpensive buffer that reduces frictional unemployment and improves the overall functioning of the labour market. However, the workers involved pay part of the price for this adjustment mechanism. Using the Labour Force Survey, this study attempts to explain the earnings gap between temporary and permanent employees based on some of their characteristics. For the first time, this comparison is made by type of temporary employment, since the characteristics of such employees vary greatly depending on whether their jobs are term or contract, seasonal, casual, or obtained through an employment agency (see Data source and definitions). Economic vulnerability is also addressed: How persistent is the earnings gap when hours worked, earnings of other members of the household, and number of dependants are taken into account?²

A significant contribution

Temporary employment³ accounted for almost one-fifth of overall growth in paid employment between 1997 and 2003.⁴ This stood against the 12.5% of paid employment that temporary employment represented in 2003, up slightly from 11.4% six years earlier. Contract employment⁵ accounted for the largest percentage of this growth (56.3%), the rest being taken up by seasonal and casual employment. On the other hand, jobs obtained through employment agencies and other temporary jobs decreased (Table 1).

In 2003, 40% of temporary employees were between the ages of 15 and 24. Flexibility and the short-term commitment associated with temporary positions were likely the major drawing cards for these young people, especially since 44% were students. For many, this type of work may have been a way to pay for their education (see Young temporary workers). Workers 55 and over made up 9% of the temporary labour force, comparable to their percentage among employees as a whole. The lower level of commitment and greater flexibility of temporary employment likely facilitates the transition from working life to retirement for some.

Persons aged 25 to 54 represented slightly more than half (809,200) of temporary workers. They have been focused on in this article in order to eliminate those making the transition from school to work or from work to retirement. Between 1997 and 2003, temporary workers represented 8% to 9% of employees in this age group. Slightly more than half (4.7% in 2003) were contract employees, while the other half consisted of seasonal employees (1.9%) and casual employees (including those using employment agencies and other types of temporary employees) (1.6%).

A varied profile

Temporary employees aged 25 to 54 vary greatly. For example, they are overrepresented among employees with low levels of education, but also among workers with university degrees. Temporary jobs are more common in rural areas, but the Ottawa-Gatineau metropolitan area is also distinguished by a high proportion of temporary workers. They are overrepresented in small workplaces and in public administration (Charts A and B).

Temporary employees also work in a wide variety of industries and occupations. They are overrepresented both in occupations requiring few skills and in ones requiring the opposite. This diversity notwithstanding, younger workers (25 to 34 year-olds), women, and part-time workers were overrepresented among temporary employees.

These varied characteristics can be explained by the type of temporary employment (seasonal, contract, through an agency, casual, or other). In fact, substantial differences exist in the characteristics of employees (sex, level of education, seniority, rate of part-time employment, occupation, industry) according to type of temporary employment, and this is reflected in the earnings gap with permanent employees.

The earnings gap

In 2003, temporary workers earned 16% less per hour than their permanent counterparts, or $16.69 versus $19.98. From 1997 to 2003, the gap varied between -16% and -19% (reached in 2000).⁷

Of the four types of temporary employees, contract employees showed the smallest gap, earning an average of 8% less than their permanent counterparts in 2003. Seasonal, casual and others, and those using employment agencies earned 28%, 24% and 40% less respectively (Table 2).

The gap also varied by sex. In 2003, women temporary employees earned 12% less per hour than their permanent counterparts, compared with a difference of -19% for men. Women posted a smaller earnings gap than men throughout the entire period. This could be attributed partly to their higher concentration in contract employment. Furthermore, the earnings gap for women holding contract or casual positions or positions obtained through employment agencies was lower (-4%, -16% and -34% respectively) than for men (-9%, -29% and -44%). Seasonal employment was the only category in which men showed a smaller gap (27% versus 38%).

Is the earnings gap between temporary and permanent employees greater in the private sector than in the public sector because of generally more uniform employment conditions in the public sector? Average hourly earnings are higher for public-sector employees; however, the gap between temporary and permanent employees in each sector is not always consistent with this finding. For example, contract employees in the public sector had higher hourly earnings ($20.65 versus $16.56). However, they earned 14% less than permanent employees in the public sector ($24.02), compared with the -11% gap in the private sector (Table 3). Furthermore, in the private sector, casual and other employees as well as those hired through employment agencies showed larger gaps compared with permanent employees.⁸ Irrespective of the type of temporary employment, earnings gaps in comparison with permanent employees can be explained largely by the characteristics of each type of temporary employment.

Contract workers

In 2003, 57% of contract workers were women. Holding contract employment appears to have less effect on the earnings gap, given the high levels of education among contract workers (38% had a university degree compared with 23% of permanent employees) and unionization (41% versus 36%) (Table 4).

Moreover, their young age (44% are aged 25 to 34), their greater tendency to work part time (21% versus 10% for permanent employees), and their low levels of seniority (only 34% had worked for the same employer for more than two years, compared with 77% of their permanent counterparts) likely had the opposite effect.

Another factor that may tend to reduce the earnings gap is the strong concentration of contract employees in the professional group (Table 5).⁹ More than a third of women working on contract held professional positions, the vast majority as teachers (21%), in contrast with only one-fifth of women in permanent employment.¹⁰ Almost twice as many men working on contract were employed as professionals (28%) compared with their permanent colleagues (15%). Almost half were teachers, while another large group worked in natural and applied sciences, mainly as computer analysts or programmers. On the other hand, fewer men and women worked on contract in management and in less-skilled occupations.

Education, health care and social assistance, and public administration employed close to 6 in 10 women contract workers; construction, education, public administration and manufacturing accounted for the majority of men (55%).

Seasonal workers

More than two-thirds of seasonal workers were men. Seasonal workers were slightly younger than permanent employees, although they were on average the oldest among temporary employees. They tended to work full time (87%) and more hours per week than permanent employees (42.7 hours versus 40.3).

Their low levels of education, their concentration in small workplaces, their low levels of seniority, as well as their low unionization rate compared with permanent workers could serve to widen their earnings gap (Table 4).

Virtually no professionals worked as seasonal employees. However, 1 in 5 men in seasonal employment were specialized workers, mostly in construction. Close to 1 in 4 were in primary-sector occupations, most often related to fishing, agriculture (and nurseries) and forestry. One in 10 seasonal workers drove trucks or buses. In summary, nearly two-thirds of these men were employed in construction, primary industries, or manufacturing.

The positions held by women were largely unskilled. Women were highly concentrated in sales and services (43%), where they worked as sales clerks, cashiers and cooks, for example. A high percentage also worked in processing, manufacturing and utilities occupations (14%) or those related to the primary sector (13%). Manufacturing (16%), accommodation and food services (16%), the primary industries (13%) and trade (11%) employed a substantial percentage of these women. This concentration by occupation and industry should tend to expand the earnings gap, more so for women than for men.

Given that primary industries tend to dominate seasonal employment to some extent, a higher percentage of workers holding this type of employment lived in rural areas compared with permanent workers (43% versus 21%) and in the Atlantic provinces (23% compared with 7%).

Casual employees

Casual workers represented less than 20% of temporary employees. Women accounted for 68%, perhaps explaining the high part-time rate among casual employees (over half) compared with only 10% of permanent employees. Casual employees tend to be younger than their permanent counterparts and generally have fewer years of seniority with the same employer (only 43% had more than two years compared with 76%).

Despite fairly comparable levels of education, close to two-thirds of male casual employees held occupations requiring no more than a high school education, in contrast with slightly over one-third of permanent employees. The gap for women was smaller but nonetheless substantial (59% compared with 44%).

A mere one-tenth of men employed on a casual basis held professional positions, compared with 15% of permanent employees. As for women, one-fifth worked as professionals (mainly registered nurses or teachers), a proportion comparable to that of permanent employees.

The strong concentration of casual workers in occupations requiring few skills should tend to increase the earnings gap between them and permanent workers, particularly for men. On the other hand, the relatively high unionization rate of casual employees (45% compared with 36%) should tend to close it.

Agency placements

Some 19,000 workers were hired through employment agencies in 2003—an estimate comparable to previous years. Men were somewhat more likely to belong to this group. These workers displayed some surprising characteristics that set them apart. For example, 40% of the men held a university degree, a percentage similar to that of male contract employees. For women, the percentage holding a university degree was the same as for permanent workers (24%).

More than half of workers using employment agencies lived in Toronto, and close to 6 in 10 worked in business, building and other support services. Most were full time. In general, these employees had very low levels of seniority, less than 10% having worked for the same employer for more than two years. They were relatively young—close to 44% between the ages of 25 and 34—and almost all (96%) were not unionized.

Despite sometimes high levels of education, 86% of the men were in unskilled occupations, mainly in the large group related to processing, manufacturing and utilities (36%) and trades, transport and equipment operating (33%), often as equipment operators and construction trades helpers. Close to three-quarters of the women were in occupations requiring a high school education or less, often as office clerks (34%), or in processing and manufacturing (31%). This concentration within occupations requiring few skills should once again tend to widen the earnings gap with permanent employees.

Explaining the earnings gap

A multi-factor analysis revealed that a combination of variables (age, education, province of residence, size of workplace, part-time employment, being non-unionized and living in a rural community) accounted for between one-quarter and one-half of the earnings gap between temporary and permanent employees (Table 6, adjustment 1).¹¹

Contract employees were the exception. The relatively small earnings gap between them and permanent employees made the exercise less revealing. Nor did these variables seem to explain the gap in the case of men hired through employment agencies.¹²

Depending on the type of temporary employment, some of the variables had more or less influence on the gap. For example, the lower education level of seasonal workers, the small size of the workplaces they worked in, and their concentration in the Atlantic provinces played a determining role in explaining their earnings gap.¹³

For casual workers, the predominance of part-time work appeared quite crucial in explaining the gap.¹⁴ Their concentration in certain provinces also played a role, but to a lesser extent. Being young and being employed in a small workplace were especially significant for men, who constituted less than one-third of casual employees. The gap for women who obtained their jobs through an employment agency was particularly sensitive to their young age and tendency not to be unionized.

Including industry and occupation for female seasonal workers, male casual workers, and those who obtained their jobs through an employment agency (both sexes) went a long way toward explaining the gap (adjustment 2). Doing so indicated that a substantial portion of the wage gap could be attributed to the heavy concentration of certain temporary workers in unskilled occupations and in industries with lower-paid jobs. For example, adding industry and occupation to the list of variables narrowed the gap for female seasonal employees from -21% to -12%. A similar effect was noted for male casual employees (from -23% to -15%) and for those using employment agencies (men from -58% to -38%; women from -26% to -16%).

The influence of industry and occupation tended to be minimal for contract employees, for men holding seasonal employment, and for women casual employees. The gap for these groups decreased little if at all after the variables were added.

Overall situation

The finding that temporary employees earned less per hour than their permanent counterparts looks at only part of their financial situation. To reflect their overall situation, it is important to also consider hours of work, earnings of other household members, and number of dependants.¹⁵

Only couples were used for this exercise, making it possible to consider elements such as spouse's earnings and number of dependants. Couples represent the majority of employees: Close to three-quarters of women temporary employees were in a spousal relationship as were two-thirds of their male counterparts. For permanent employees as a whole, the proportion was three-quarters.

Temporary employees tend to work fewer hours per week than permanent employees. If weekly earnings are compared, the gap widens for men (from -19% to -22%) and women (from -12% to -26%). The more pronounced change for women is due to the much larger difference between permanent and temporary employees in the number of hours worked (-15%) than is the case for their male counterparts (-3%) (Table 7).

Most temporary workers in a couple relationship had a working spouse (72% of men and 86% of women). The percentages for permanent employees were also higher: 77% and 89% respectively. Spouses made a significant contribution to family earnings, especially for temporary workers. For example, for men, the spouse's contribution was 42% in the case of temporary employees, compared with 38% for permanent employees. For women, the proportions were 63% and 56%.

With respect to weekly family earnings, which include those of all members, the gap is lower for both men and women. However, the effect is much more significant for women: Their gap decreases by 12 percentage points, compared with 7 points for men. This reflects women's smaller role in family earnings, but also shows that spouses partly compensate for temporary workers' lower earnings since they contribute substantially to family earnings. Furthermore, the gap in weekly earnings of spouses of temporary and permanent workers is fairly small (between -6 and -4%), as is the gap in number of hours (between 0% and 2%).

Adjusting weekly family earnings according to family size has little effect for either men or women. For men, the smaller percentage of temporary employees with children under 18 (53% compared with 60% for permanent employees) narrows the gap marginally (from -15% to -14%); for women, the slightly higher percentage of temporary employees with children under 18 (58% compared with 55%) has the opposite effect—the gap widening from -14% to -15%.

When hours of work, earnings of other household members, and number of dependants are taken into account, significant gaps persist between permanent and temporary workers. These factors affect all types of temporary employees fairly similarly. In other words, with a few exceptions, taking the number of hours into account widens the gap as a whole, taking the contribution of spousal earnings into account narrows it, while adjusting for the number of dependants has little effect.

After these factors were taken into account, contract workers still posted the smallest gaps in relation to permanent workers (-8% for men and -6% for women) (Table 8).

Male seasonal workers showed smaller gaps than initially (from -28% to -18%), primarily because of more hours worked (6% more than permanent employees) and, to a lesser extent, the contribution of spousal earnings. The negative effect of their low education level and their concentration in small workplaces and in the Atlantic provinces is therefore partially offset by their long hours of work and their family situation. A substantial average gap of -18% nonetheless persists for the 85,000 male seasonal workers (in couples).

For the 30% of temporary employees in spousal relationships (164,000 workers), an average earnings gap of -24% to -28% persists even when family situation is taken into account.¹⁶ This group consists of female seasonal employees, casual workers, and workers using agencies.

To the extent that temporary jobs provide a gateway into the labour market or a springboard to a permanent position, the earnings gap has fewer consequences. Kapsalis and Tourigny (2004) have shown that certain non-standard jobs, particularly temporary full-time jobs, can sometimes lead to permanent employment.

However, these more economically vulnerable employees earn a substantial portion of the family income (between one-half and two-thirds) and are mainly women (78%). They are only slightly younger (39.5 years) than permanent workers (40.0) and most have some postsecondary education (64%), but they hold largely unskilled jobs (two-thirds held a position requiring a high school education or less). They tend to have few years of seniority with the same employer; they often work part time (45%) and only about half of them do so by choice.

Summary

From 1997 to 2003, temporary workers earned 16% less per hour, on average, than their permanent counterparts. The gap varied greatly depending on the type of temporary employment. Contract employees posted the smallest gap relative to permanent employees (-4% for women and -9% for men in 2003). Seasonal and casual employment and employment obtained through an agency showed larger gaps (-28%, -24% and -40% respectively).

The smaller gap observed for contract employees could be attributed to their high level of education and the numerous skills required by their positions. In general, the characteristics of workers holding other types of temporary employment (age, education, province of residence, workplace size, part-time employment, not being unionized, or living in a rural community) went a long way toward explaining their differences in earnings compared with permanent employees.

Industry and occupation were major factors in the case of female seasonal employees, casual employees, and those using employment agencies. This demonstrates that the gap for these employees stems partly from their concentration in poorly paid industries and in occupations that require few skills.

With respect to temporary employees in spousal relationships, when number of hours of work, earnings of other household members, and number of dependants are taken into account, contract workers continue to show the smallest gap. Male seasonal workers display a smaller gap once these factors are taken into consideration, mainly because of the large number of hours they worked (6% more than permanent employees) and the contribution of spousal earnings.

For female seasonal employees, casual employees (73% of whom are women), and those who used employment agencies (50% women), the average earnings gap remained significant: between -24% and -28%. The gap persists despite the significant contribution of spousal earnings to total family earnings. Almost one-third of temporary employees held this type of employment. Despite their relatively high level of education, their concentration in part-time, often unskilled jobs accounts for much of their large earnings gap.

To the extent that temporary workers eventually gain access to less precarious employment, the earnings gap may prove less consequential. The study did not take into account differences related to employee benefits (such as pension plans, drug insurance, or disability insurance), an element that could tend to increase the disparity in earnings.

Notes

1. According to Kapsalis and Tourigny, this applies particularly to full-time temporary employment.

2. Despite their importance as a means of remuneration, employee benefits could not be included because of data limitations. In 1995, according to the Survey of Work Arrangements, temporary employees received distinctly fewer benefits than permanent employees (Grenon and Chun 1997).

3. Since temporary employment by definition cannot apply to self-employed workers, this article pertains only to employees.

4. The Labour Force Survey has included data on earnings since 1997 only.

5. In this article, 'contract employment' includes employment for a set period of time.

6. The 1.7% gap for those in the 15 to 17 age group was significantly different (at the 5% level) from the -1.9% recorded for those aged 18 to 21.

7. The -19% gap in 2000 is significantly different (at the 5% level) from the -16% recorded in 2003.

8. Analyzing the earnings gap between employees in the public and private sectors is a complex issue and beyond the scope of this article.

9. The professional group includes occupations such as accounting, medicine, law and teaching, which generally require a university degree. These fall under the B0, C0, D0, D1, E0, E1 and F0 categories of the 2001 National Occupational Classification (Statistics Canada 2001).

10. Some figures do not appear in Table 5 because they correspond with the 4-figure level of detail in the National Occupational Classification.

11. Other non-observable characteristics such as differences in skills, productivity or motivation can also help to explain the earnings gaps but cannot be taken into account here.

12. These results were obtained using regressions pertaining to the logarithm of earnings of temporary and permanent employees. Average earnings for each type of temporary employment were compared with that for permanent employment, for men and women separately. The regressions included the following variables: being a temporary employee, age, age squared, province, education, working part time, being non-unionized, living in a rural area, and workplace size. The seniority of temporary and permanent employees was tested but ultimately removed from the final model since temporary employees by definition have low levels of seniority with the same employer. In a second step, for each comparison, all the above variables were included, along with the industry and occupation. Results are available on request.

13. These results were taken from a Oaxacaÿdecomposition model involving 10 regressions (for permanent employment, for the four types of temporary employment and for both sexes). The variables included in the model were age, age squared, province, education, working part time, not being unionized, workplace size, living in a rural area, and industry and occupation.

14. Cranford, Vosko and Zukewich (2003) show that the earnings gap is larger for part-time than for full-time temporary employees.

15. For a truly complete table, it would also be necessary to include other income sources for all household members. However, this information is not part of the LFS.

16. Even larger gaps (of between -31% and -71%) were observed for workers who were lone parents or living alone and in seasonal jobs (12,000 women), jobs obtained through an employment agency, casual jobs or other (38,000 men and women) .

References

• Cranford, Cynthia, Leah F. Vosko and Nancy Zukewich. 2003. "The gender of precarious employment in Canada." Industrial Relations (Quebec) 58, no. 3 (Summer): 454-483.

• Grenon, Lee and Barbara Chun. 1997. "Non-permanent paid work." Perspectives on Labour and Income (Statistics Canada catalogue no. 75-001-XPE) 9, no. 3 (Autumn): 21-31.

• Kapsalis, Costa and Pierre Tourigny. 2004. "Duration of non-standard employment." Perspectives on Labour and Income (Statistics Canada catalogue no. 75-001-XIE) 5 no. 12. December 2004 online edition.

• Krahn, Harvey. 1995. "Non-standard work on the rise." Perspectives on Labour and Income (Statistics Canada catalogue no. 75-001-XPE) 7 no. 4 (Winter): 35-42.

• Organisation for Economic Co-operation and Development (OECD). 2002. "Taking the measure of temporary employment." OECD Employment Outlook 2002, Chapter 3: 135-196.

• Schellenberg, Grant and Christopher Clark. 1996. Temporary employment in Canada: Profiles, patterns and policy considerations. Ottawa: Canadian Council on Social Development.

• Statistics Canada. 2001. National occupational classification for statistics. Catalogue no. 12-583-XPE. Ottawa.

Full article in PDF

Author
Diane Galarneau is with the Labour and Household Surveys Analysis Division. She can be reached at 951-4626 or perspectives@statcan.gc.ca.

You need to use the free Adobe Reader to view PDF documents. To view (open) these files, simply click on the link. To download (save) them, right-click on the link. Note that if you are using Internet Explorer or AOL, PDF documents sometimes do not open properly. See Troubleshooting PDFs. PDF documents may not be accessible by some devices. For more information, visit the Adobe website or contact us for assistance.