April 2002     Vol. 3, no. 4

Duration of multiple jobholding

Katherine Marshall

Multiple jobholders are a relatively small but important element of the labour market. Knowing the number of secondary jobs held by workers is necessary to calculate the total number of jobs in the economy. Using cross-sectional data, various aspects of multiple jobholding or 'moonlighting' have been previously studied in Perspectives (Sussman, 1998; Pold, 1995; Cohen, 1994; Webber, 1989).

With the relatively recent availability of longitudinal data, another facet of multiple jobholding can now be examined—the length of time people work at more than one job. This information can give insight into the stability and dynamics of multiple jobholding. Do people work at more than one job for years or only briefly? If multiple jobholding is generally a stable long-term activity, then the incidence rate, as well as the work schedule and income of multiple jobholders, would remain relatively constant. Conversely, if multiple jobholding is mainly a short-term activity, then the turnover rate of secondary jobs would be high, and the yearly incidence rate of multiple jobholding in the population would also be higher.

This article looks at the duration of multiple jobholding and the characteristics of short-, medium-, and long-term multiple jobholders, beginning with a look at monthly and annual incidence rates of multiple jobholding over time (see Data sources and definitions).

One in 10 workers hold multiple jobs each year

During the Labour Force Survey (LFS) reference period,  1  roughly 5% of workers held more than one job in 2001, a rate that did not change much over 15 years (Chart A).

However, since the LFS sample changes each month, it is not possible to determine if multiple jobholders are the same people from month to month.

On the other hand, the Survey of Labour and Income Dynamics (SLID) collects monthly labour market information from the same respondents for six years (longitudinal data), thereby allowing labour market activity changes for individuals to be examined. Therefore, SLID permits the calculation of not only a monthly, but also an annual multiple jobholding rate—that is, the incidence of having at least one multiple jobholding experience over the course of a year. With SLID, persons are considered to be multiple jobholders in any month that they held more than one job concurrently; the spell must have lasted at least 15 consecutive days, and only part of the total period need fall in that month. (Slight differences in multiple jobholding definitions between SLID and LFS may account for some of the monthly rate variation—6% versus 5%, respectively.  2 ) When the SLID reference period is extended to one year, approximately 10% of those employed experienced multiple jobholding for at least 15 days or more, at least once during a year between 1993 and 1999 (see Data sources and definitions).

Both multiple jobholding rates show that women are more likely to be multiple jobholders than men. For example, annual average 1999 LFS data show that 4.4% of employed men held down more than one job, compared with 5.6% of women. In the same year, SLID estimated that 7.3% of men experienced at least one episode of multiple jobholding, compared with 10.1% of women. However, women also have much higher involuntary and voluntary part-time employment rates  3  than men (Marshall, 2001); this difference is likely linked to their higher multiple jobholding rates. For involuntary part-time workers, a second job helps increase total work hours, whereas for voluntary part-time workers, two part-time jobs may offer a more flexible work arrangement than one full-time job.

On average, multiple jobholding lasts eight months

Since in any given month approximately 6% of the employed are multiple jobholders compared with 10% over the course of a year, workers must be moving into and out of this work arrangement. SLID data for 1996 show that multiple jobholding peaked in July and August (6.4%) and dropped to 6.1% in September (Table 1). The 40,000 net drop between August and September resulted from 91,000 workers beginning this work arrangement in September and 131,000 stopping after August. The multiple jobholding exit and entry population in the two months accounted for 1.5% of employment. The surge in multiple jobholders in the summer months is likely a reflection of the annual employment increase during this time (increasing the opportunity to multiple jobhold), as well as an increased student population looking for short-term employment.

The considerable month-to-month turnover indicates that multiple jobholding is not a long-term activity for many. In fact, the median duration of the first spell for all multiple jobholders in 1996 was eight months (Table 2). Furthermore, almost 6 in 10 held multiple jobs for a year or less— 40% for 1 to 6 months, and 17% for 7 to 12 months. Another 15% ended somewhere between 13 months and two years. The rate fell off considerably after two years (Chart B). Nevertheless, 1 in 10 multiple jobholders in 1996 had worked at more than one job for more than four years—making this work arrangement a way of life.

Longitudinal data can also provide information about subsequent multiple jobholding—an activity which turns out to be quite prevalent. Of the 1.1 million (88%) multiple jobholders in 1996 who stopped within the four-year period, almost one-third (31%) resumed at least once before 1999 (Table 2). The median length of time before starting a second spell was nine months.

Multiple jobholders tend to be younger and better educated

A number of demographic and other differences are evident between those with one job and those with two or more. However, some striking differences exist within the multiple jobholding community as well—particularly between those who juggle multiple jobs for six months or less (short-term), and those who do so for three years or more (long-term).

In 1996, the median age among single job workers was 38, compared with 33 among multiple jobholders (Table 3). However, when all multiple jobholders were followed over the next four years, those who turned out to be short-term multiple jobholders had the youngest median age (27), and long-term the oldest (40).

The large age difference between the two groups suggests different reasons for pursuing the activity. Unpublished data from the 1995 Survey of Work Arrangements and similar 1997 U.S. data confirm this supposition. Younger workers (under 35) were the most likely to state either household expenses or debts as the main reason for holding more than one job, whereas older workers (45 and over) were most likely to answer that they enjoyed the work on the second job (Martel, 2000). Money problems, such as debt, suggest a temporary situation that might be resolved with a second job. Enjoyment of the work arrangement, on the other hand, is likely a key influence in extending the duration of the activity.

Being older, long-term multiple jobholders also tended to be more settled—75% were married and 54% had dependent children at home. Only 46% of short-term multiple jobholders and 67% of single jobholders were married.

On average, a higher proportion of multiple jobholders (54%) had a post-secondary certificate, diploma or university degree compared with non-multiple jobholders (47%). However, considerable differences were again seen by type of multiple jobholder. Some 45% of short-term multiple jobholders had completed some form of post-secondary schooling compared with 64% of long-term ones. The higher education levels among long-term multiple jobholders may in part reflect their higher average age—being older they have had more chance to complete all their schooling. Indeed, some 40% of short-term multiple jobholders were attending school in 1996, either part or full time, compared with a very low percentage of long-term ones. However, long-term multiple jobholders were also more likely to work in jobs normally requiring postsecondary schooling (Table 4).

Finally, multiple jobholding rates and durations also varied by province. High rates occurred in Manitoba, Saskatchewan, and Alberta, where 15% of the employed held two or more jobs at least once in 1996. The high rate is not unexpected given that these three provinces have proportionally high levels of employment in agriculture, the occupational group with the highest multiple jobholding rate.

About half of multiple jobholders in the Atlantic region and Quebec were short-term, compared with 39% or less in the other areas. The exceptionally high short-term rate in the Atlantic region (50%) may be linked to the higher-than-average employment levels in the top three seasonal industries: fishing and trapping, logging and forestry, and construction.

One-third of short-term multiple jobholders work in sales and service

Overall, the type of main job held by single and multiple jobholders was relatively similar.

However, a slightly different story emerges when the occupations of main and second jobs held by short-, medium- and long-term multiple jobholders are examined.

Those who started and stopped multiple jobholding after a short time (6 months or less) were much more likely to have a sales and service main or second job (34% and 40% respectively) than were other multiple jobholders, especially long-term ones (Table 4). In fact, only 18% of long-term multiple jobholders had a main job and 29% had a second job in sales and service. Since over one-third of all sales and service jobs are part-time, this field of work is often a source of second jobs. On the other hand, because these jobs offer relatively low average earnings, many in this line of work (full- or part-time) may be driven to take on another job to meet short-term financial needs. Other studies have also found high multiple jobholding rates among occupations with flexible schedules and high rates of part-time (Sussman, 1998; Amirault, 1997).

Conversely, long-term multiple jobholders were more likely to have a main or second job in management, business, or finance and administration (29% and 22%, respectively) than were short-term (20% and 16%). Another noteworthy difference was in farming. Of long-term multiple jobholders, 7% had a main job and 14% had a second job in agriculture, compared with only 4% and 6% of all multiple jobholders.

Indeed, long-term multiple jobholders were more likely to be engaged in either higher-level, better-paying jobs (often found in business or management) or in self-employed jobs such as farming. A full 62% of long-term multiple jobholders were business owners (in either their main or second job), compared with only 23% of short-term and 41% of medium-term multiple jobholders. Obviously, if the goal is to take on a second job for a limited period of time, self-employment is not the answer.

Long-term multiple jobholders earn more, work more

Hours of work and earnings also differ within the multiple jobholding population. Short-term multiple jobholders put in fewer hours at their main job (20) than did medium- (30) or long-term (38) (Table 4). Long-term multiple jobholders had the same median hours in their main job as did single jobholders, mainly because both groups were likely to be working full time (77% and 81%, respectively). Only two-thirds of short- and medium-term multiple jobholders worked full time at their main job. The second job for all multiple jobholders added roughly 10 hours to their workweek, giving long-term multiple jobholders the longest work schedule at 48 hours per week.  4  This extra long workweek reinforces the notion that many long-term multiple jobholders participate in the arrangement because they enjoy the work.

Similar to hours, annual 1996 earnings increased with the length of multiple jobholding-$9,770 for short-term, $15,840 for medium-term, and $29,040 for long-term multiple jobholders. Median earnings for single jobholders were between those of medium and long-term multiple jobholders at $21,720. Although long-term multiple jobholders earned over $7,000 more per year than any other group, it is important to keep in mind their long workweeks. When earnings are divided by hours, long-term multiple jobholders had only the second highest hourly earnings rate ($11.45) (Chart C). Although single jobholders earned less annually than did long-term multiple jobholders, they also worked considerably fewer hours per week. They therefore had the highest average hourly earnings ($13.10). Short-term multiple jobholding is by far the least lucrative work arrangement, with an average hourly rate of only $7.20. It is not surprising to find that such a poorly paid work arrangement ends early, as most in this situation are probably looking for a better-paying single (or second) job.

Summary

The SLID longitudinal data reveal a dynamic alternative work arrangement, embarked on for varying lengths of time and for various reasons.

Over the 1996-to-1999 study period, 10% of the employed had at least one episode of multiple jobholding during the course of a year. Multiple jobholding tended to be relatively short-term. Of the 1.3 million multiple jobholders in 1996, 88% had stopped by the end of 1999, and the median duration of these ended episodes was eight months. But, almost one-third of those who stopped multiple jobholding resumed within the study period.

For a small segment of the multiple jobholding population, this arrangement had become part of their lifestyle. One in 10 multiple jobholders in 1996 were carrying more than one job for at least four years.

Long-term (three years or more) multiple jobholders differed considerably from those who held multiple jobs for a shorter duration. For example, 62% of long-term multiple jobholders were self-employed in their primary or secondary job, compared with only 23% of those who held a second job for six months or less. Furthermore, on average, long-term multiple jobholders were more likely to be older, married, working longer hours, and earning higher hourly wages than other multiple jobholders.

Notes

1. Data collection for the LFS is carried out each month during the week following the LFS reference week (normally the week containing the 15th day of the month).
2. Some variation in multiple jobholding rates between SLID and the LFS may also be due to the difference in sample sizes—15,000 versus 60,000 households, respectively.
3. Involuntary or voluntary part-time workers expressed as a proportion of total employment.
4. Further analysis of the long-term multiple jobholding population showed that almost one in five had a main or second job in agriculture. It is well known that farmers average long workdays. However, even when farmers are excluded from the median hours calculation, long-term multiple jobholders still averaged 48 hours per week.

References

• Amirault, T. "Characteristics of multiple jobholders." Monthly Labor Review 120, no. 3 (March 1997): 9-15.
• Cohen, G.L. "Ever more moonlighters." Perspectives on Labour and Income (Statistics Canada, Catalogue no. 75-001-XPE) 6, no. 3 (Autumn 1994): 31-38.
• Marshall, K. "Part-time by choice." Perspectives on Labour and Income (Statistics Canada, Catalogue no. 75-001-XPE) 13, no. 1 (Spring 2001): 20-27.
• Martel, J.L. "Reasons for working multiple jobs." Monthly Labor Review 123, no. 10 (October 2000): 42-43.
• Pold, H. "Families and moonlighting." Perspectives on Labour and Income (Statistics Canada, Catalogue no. 75-001-XPE) 7, no. 2 (Summer 1995): 7-8.
• Sussman, D. "Moonlighting: A growing way of life." Perspectives on Labour and Income (Statistics Canada, Catalogue no. 75-001-XPE) 10, no. 2 (Summer 1998): 24-31.
• Webber, M. "Moonlighters." Perspectives on Labour and Income (Statistics Canada, Catalogue no. 75-001-XPE) 1, no. 3 (Winter 1989): 21-30.

Katherine Marshall is with the Labour and Household Surveys Analysis Division. She can be reached at (613) 951-6890 or katherine.marshall@statcan.gc.ca.

[ Highlights ] [ Main menu | Editor's corner | More news | Contact us | Survey information | Back issues ] [ Statistics Canada home page | In depth | Français ]

© Statistics Canada - Conditions of  use Published: 2002 04 17