Staffing agency: A bridge to working during retirement.

BACKGROUND: The ageing population has initiated a debate about a prolonged working life. There is an interest in finding the pre-retirement predictors of bridge employment and retirement decisions, but the understanding of the experiences of bridge employees is still limited.
OBJECTIVES: The aim was to describe the characteristics of the pensioners working for a staffing agency, their motivational aspects, work patterns and types of services they provided.
METHODS: This article analyses the results from a Swedish staffing agency's yearly co-worker questionnaire from December 2017. The response rate was 44% (N = 1741). The design is descriptive, with correlation analyses and construction of typical cases.
RESULTS: Most study participants were aged 65-74 years. Sixty-five percent were men, 66% were cohabitating/married, dominating education level was secondary school or higher education (79%). Important incentives for working were the social context and to gain extra income. A majority of the respondents also stated that their work increased their overall well-being. Seventy-eight percent worked 25 hours per month or less, 37% wanted to work more, 3% wanted to work less. Private services dominated with 61%; most common were gardening (43%), trades (33%) and cleaning (31%).
CONCLUSIONS: More men than women chose this form of work. While single women need to work out of economic necessity, men, to a larger extent, work for the social context and well-being. The highest work frequency in 2017 (14%) in the population was found for those who retired in 2015, i.e. two years after their retirement year. A majority indicated that the work they were doing was different from earlier in their working life.

Introduction

Due to longer life expectancy and better health, a growing group of older employees want to stay active and prolong their time in the labour market after reaching pensionable age. Working for pay after retirement can be a way to smoothen the transition from a career job to withdrawal from the labour market. This has been denoted bridge employment [1]. There are three common ways to conceptualize bridge employment: 1) continued salaried work, full or part-time, for the same employer as before (career jobs); 2) working permanently or temporarily for a new employer; or 3) becoming self-employed [2].

There is an emerging interest in bridge employment in research, particularly in finding the pre-retirement predictors of bridge employment and retirement decisions, well-being and job-seeking patterns, but our understanding of the experiences of people working in bridge employment is still limited [1, 3]. In a review of the literature in the area, Mazumdar et al. [3] found that unmet expectations regarding personal fulfilment, opportunity for mentoring the younger generation and the need for meaningful social relationships, could be detrimental to the experiences during bridge employment, and based on this, jeopardize both employer and employee outcomes. Mazumdar and colleagues performed their study among white collar workers in Canada, and they suggest that future studies include different economic and socio-cultural environments. They also conclude that they did not find other studies in this area, so there is a need for research on experiences of bridge employees, including those hired by staffing agencies. Therefore, this paper focuses on bridge employment in the form of working for a new employer in one of the largest staffing agencies for retired people in Sweden.

What is regarded as “pensionable age” varies over time and place. In 2018 the average effective age of retirement in Sweden was amongst the highest in Europe, 66.4 and 65.0 years of age for men and women respectively [4]. In Sweden, there is no official pensionable age, but 65 was the age when the guaranteed pension was available at the time of this study in 2017 [5]. Accrued pension could be received from the age of 61. Most people seem to view 65 as the normal retirement age even if the way they think about and understand retirement differs according to the time perspective in question. At the age of 67 the rules on priority in employment and the regulations on employment security expired [5]. At this age one could be forced to leave employment, but there were no formal regulations prohibiting employees from continuing to work if they wanted to and the employer allowed it. In October 2019, the Swedish Government decided to raise the age limit for income-based retirement pensions from the age of 61 to 62, and to raise the age limit for the right to remain in employment to 68 years from 2020.

Formal education, work experience and health have had a significant impact on the likelihood of individuals working at higher ages [6-8]. Both internal (meaning of work) and external (rewards) motivational factors can explain a prolongation of working life [9] and there are studies suggesting a transition from external to internal driving forces with increasing age [10].

Several reasons for a continued working life at an individual level have been found: achieving new skills, new learning, a new career, a better pension and expansion of consumption possibilities, and being part of an active lifestyle with stimulating relationships with colleagues [11, 12]. Burnay [13] showed that professional activity at the end of working life remains a factor for social inclusion, since it provides individuals with a means of subsistence, but it is also a source of social identity. The determinants of participation in bridge employment were investigated among male and female retirees in 16 European countries by Dingemans, Henkens and van Soline [14]. They showed that depending on their standing in the social hierarchy, various subgroups of older adults may differ in their access to valuable resources that may enable or constrain participation in bridge jobs. Additionally, they put forward that bridge employment is more likely in countries where the expenditure on pensions is lower than in countries with relatively high pension expenditure, implying that some of the international differences may be driven by people continuing to work out of economic necessity. They also stressed that stronger support for the norm that older adults should be able to work past official retirement age might be associated with a higher likelihood of working in bridge jobs. Consequently, in countries such as Denmark, Sweden and the Netherlands, with normative support for working after retirement, bridge employment is common, but only as one activity among others. This may explain that in these countries, a substantially lower number of hours is worked in bridge jobs [14].

In the Swedish context, employment among older adults has increased since the beginning of the 2000s. Among men in the age group 60–64 years, employment has increased from 55 to 75% and among women in the same age group there is an increase from 45 to 65% [15]. It is noteworthy that in the age group 65–69, almost nothing has changed during the same time period.

Anderson and Guo [16] found that retirees discussed their retirement/bridge employment experience using both negative (uncertainty over financial needs) and positive (excitement) frames. One of the tensions the retirees discussed revolved around the concept of identity during this transition period. Retirees held on to past occupational identities and simultaneously explored new occupational identities. Hence, retirees are finding ways to keep aspects of their past social identity while also exploring new facets of their post-retirement identity through bridge employment.

Two distinct patterns emerge when literature on older worker typologies is brought together: 1) workplace autonomy, and 2) privilege [17]. Workplace autonomy concerns factors that impact on work longevity, such as the nexus of income, qualifications and job status. These factors combined indicate the extent to which workers may design their own post-retirement work situation. Privilege refers to the level of job autonomy and security of pension benefits which make it possible to plan for retirement. On one end of the spectrum, we find workers with high levels of job autonomy and secure pensions who have the possibility to plan for and choose retirement. On the other end, we find workers who are pushed into or out of work in later life due to factors beyond their control [17]. Hence, social class is a determinant for independence during the transfer from working life to retirement. While privileged older adults have the ability to make autonomous decisions, workers with low job autonomy and low wages might be dependent on others for the decision to retire, and they might be pushed to stay longer in the labour market out of necessity. Flynn’s study also highlights the importance of workplace characteristics for the decision to work longer, by pointing for example at factors such as autonomy and secure pensions.

The work-related factors on the intentions of older workers to continue in employment are further emphasized by Shacklock and Brunetto [18]. They state that while health and financial factors impact the decision to continue working later in life, the following four factors were significant predictors: 1) the importance of work to the individual, 2) the flexibility of work arrangements, 3) the individual’s interests outside of work, and 4) management and organisational factors. In addition to this, there must be opportunities for bridge employment for people to be able to prolong working life and transition from full-time jobs to full retirement [19]. A rising phenomenon in Sweden is staffing agencies that offer services to private persons and companies by employed pensioners. The expansion of this phenomenon started in 2007 due to changes in the taxation system and tax exemptions concerning services to consumers and their housing, ROT (Repairs, Conversion, Extension) and RUT (Cleaning, Maintenance and Laundry), but also entailed lower costs for employing older workers [19]. The study in this paper concerns one of the five largest staffing agencies in Sweden, in terms of numbers of employees. The company was established in 2007 and has been operating in its current form by new owners since 2013. They hire solely pensioners for parttime work.

Aim

The aim was to describe the characteristics of the pensioners working for a staffing agency, their motivational aspects, work patterns and types of services they provided.

Material and method

The design of this study is descriptive, and the material emanates from the 2017 yearly co-worker questionnaire from the staffing agency. During the year 2017 the agency had 41 local offices all over Sweden with a total of 4,157 persons employed for shorter or longer assignments. Of the 3,950 persons who had a registered an e-mail address, 1,741 answered the web-based questionnaire. The estimated response rate was 44%.

Ethical considerations

Since this study does not refer to any sensitive or health care data, neither include a vulnerable group of people, an ethical review according to the Swedish Law (2003 : 460) on ethical review of research concerning people [21] was not necessary. However, the authors carefully followed The Research Ethical Principles within humanistic-socialscientific research from the The Swedish Research Council [22] throughout the research process. Those principles include the participants’ right to information, consent, confidentiality and use of data.

Questionnaire

The questionnaire comprised questions on demographics, the staffing agency as an employer, the current work situation, and motivational aspects. Only questions relating to the aim of the study were relevant for data analysis. Questions on demography and work situation had fixed answer alternatives. Responses to statements on motivational aspects were given a value on a scale of 1–5, where 1 = totally disagree and 5 = totally agree.

Data analysis

A description of the sample was conducted in relation to demographic data. Six motivational aspects of working post-retirement constitute the core of this study. Firstly, correlation analysis was performed between these aspects and individual characteristics (in the form of age, gender, educational level, years from pension and marital status). Secondly, we compiled typical cases based on each of these motivational factors by dichotomizing the motivational responses and analysing which of the other motivational factors and personal characteristics typically represented the case (see below). Threshold levels for the dichotomization of the motivational aspects were originally set at “3” in order to capture individuals who more or less “agreed” on the statements but were adjusted upwards in two cases in order to yield reasonably sized groups. The persons that scored at or above the chosen threshold represented the case, since they agreed to a higher degree to the statement of the motivational factor, e.g. “I work because my partner work”. The motivational factors (with dichotomized cut points for the typical case definition, and fraction of the sample representing the case in parenthesis) were:

Fellowship: “I missed the social affiliation in working life”; response limit≥4, (49%).

Extra money: “I work to have an extra income that enhances everyday life”; response limit≥3, (79%).

Confirmation: “I receive appreciation for my proficiency”; response limit = 5, (36%).

Kicked out: “I did not want to be a pensioner, but I was not allowed to/could not continue working at my former workplace”; response limit≥3, (31%).

Partner works: “I work because my partner works”; response limit≥3, (21%).

Poor economy: “The pension is insufficient, and I must work to manage everyday life”; response limit = 5 (15%).

The threshold response values for the typical case definition were set at≥3 for half of the cases, exceptions being Fellowship≥4 and Confirmation = 5, since these groups otherwise became too large. Further, in order to distinguish the group of persons who really have to work for their living, the limit for Poor economy was set = 5 (15 % of the sample, or 243 individuals, belonged to this group).

It should be noted that the pairwise correlation between the six dichotomized motivational factors is limited in size ranging from –0.06 to 0.18 (Poor economy and Extra money) not indicating that problems of collinearity plague the analysis at hand. For the raw data (before dichotomizing), the correlations range from –0.04 to 0.39.

For each case, all possible answer combinations of the other motivational aspects and the independent variables (gender, educational level, marital status, similar work tasks as before) together with answers to the questions of whether the respondents “appreciated new challenges” and perceived their current work as “increasing their well-being”, were sorted in descending frequency. The most common answer combinations (defined as consisting of the 50% highest frequencies cumulatively) were regarded as belonging to the case. The respondents having an answer combination among these were defined as belonging to the case. For each case, a frequency analysis was conducted on how the respondents belonging to the case had answered the questionnaire about the other motivational aspects and the other variables described above. The frequency of responses was also computed for a reference group containing the full sample of respondents (N = 1,741). Finally, the quotient between these frequencies (case/reference) was used as an indicator of how much more common the motivational aspects and independent variables were for the respondents belonging to a particular case compared to the reference group.

For instance, 56.6% of the persons defined as belonging to the typical case Poor economy were women whereas the reference group consisted of 35.0% women. Hence, women were overrepresented in the Poor economy case by a factor 1.62 (56.6/35). This type of comparison of proportions was made for each motivational factor separately with regards to the other motivational factors and variables above.

Results

The results are presented in relation to the study aim: to describe the characteristics of the pensioners working for a staffing agency, their motivational aspects, work patterns and types of services they provided.

Demographics

Among all employees (3950 individuals) invited to participate in the questionnaire, 62% were men and 38% women, and their mean average birth year was 1947.7. Of the 1,741 respondents, 65 % were men and 35 % were women, with a mean birth year of 1947.8. Most people in both samples were born during a ten-year period, 1943–1952, with a peak 1947-50. Hence, the sample of respondents seems to be representative for the full population of employees with respect to age structure and gender. Sixty-six percent were married/cohabiting, 27% were single and 7% lived apart. Forty-nine percent had secondary school, 30% university and 21% elementary school education. Most of them became pensioners between 2008 and 2017. The highest work frequency in 2017 (14%) in the sample was found for those who retired in 2015, i.e. two years after their retirement year, Fig. 1.

Fig. 1

The figure shows the highest work frequency (14%) for those who retired in 2015, i.e. two years after their retirement year (2017).

Motivational aspects (benefits) for working

We begin by conducting bivariate correlations between the motivational aspects and key personal characteristics (Age, Gender, Educational level, Years from pension, Marital status). The only motivational aspect that correlates (negatively) with age is “Partner works”, which is logical because the probability of having a partner that still works decreases with higher age in this sample. The importance of pecuniary motives in terms of “Extra money” and having a “Poor economy” is more pronounced among women. Women are also less likely to state that “Partner works” as a motive for continuing to work. This reflects that women overall have a less favourable economic situation with lower earned pension income rights, and also on average are married to an older man, implying that he will reach pensionable age first. Educational level correlates negatively with “Confirmation” and weakly with “Extra money”, whereas Years from pension correlates with “Partner works” and “Fellowship” –the nearer one was to pension year, the more one agreed (see Table 1). This implies that the importance of missing social interaction, as a motivational factor for continuing to work, fades away over time. Finally, married people are less likely to state pecuniary motives for their continued work, both in terms of “Extra money” and Poor economy”, reflecting gains from pooling resources and specialisation within the household. They are also weakly more likely to state “Confirmation” as a motive for working. Taken together, this correlational analysis confirms that there are associations between the motivational aspects and first and foremost gender and marital status that run in the expected directions. Now we turn to the typical cases in order to deepen the understanding of how people are motivated to continue to bridge their employment.

Table 1

Correlations between demographic and motivational factors (n = 1,622)

Motivational	Age	Gender	Education	Years from	Marital status
aspects		Ref: women	level	pension	Ref: singles
Fellowship	–0.024	–0.007	–0.034	–0.069**	–0.035
Extra money	0.009	–0.188***	–0.051*	–0.010	–0.208***
Confirmation	0.016	–0.024	–0.099***	0.014	0.051*
Kicked out	0.046	0.007	–0.035	–0.004	–0.036
Partner works	–0.157***	0.230***	–0.032	–0.112***	0.284***
Poor economy	0.017	–0.217***	–0.019	0.041	–0.213***

*p < 0.05, **p < 0.01, ***p < 0.001.

Compared to the reference group the respondents in the Fellowship case were characterized by an overrepresentation along the following dimensions: “Kicked out” (1.4 times more likely than the reference group), “Appreciating new challenges” (1.4), “Well-being” (1.4), “Poor economy” (1.3), and “Confirmation” (1.2).

People in the Extra money case were “Single” (1.9), worked because of “Poor economy” (1.6), were “Women” (1.3), had “Secondary school” (1.2), stated “Fellowship” (1.2) and that they had been “Kicked out” (1.1) from their previous work as motivational factors, more than the reference group.

The people in the Confirmation case were characterized by “Partner works” (1.7), being “Married/cohabitant” (1.2), having “Secondary school” (1.2), and were “Women” (1.2) times more than the reference group. They perceived that their work increased their “Well-being” (≥2.0), worked for “Extra money” (1.6) and “Fellowship” (1.6), and “Appreciated new challenges” (1.6) more than the reference group.

The Kicked out subjects were characterized by “Poor economy” (1.4), “Extra money” (1.4), “Fellowship” (1.4), “Partner works” (1.2), “Secondary school” (1.3) and were “Men” (1.2) more than the reference group.

The individuals in the Partner works case were predominately “Men” (1.6), had “Secondary school” (1.5), worked because they had been “Kicked out” (1.4) and “Fellowship” (1.2), and experienced that the work increased their “Well-being” (1.2) more than the reference group.

Finally, the individuals in the Poor economy case were “Single” (2.2), “Women” (1.6), worked because they had been “Kicked out” (1.9) needed the “Extra money” (1.6), for “Fellowship” (1.4) and “Confirmation” (1.2). Their work increased their “Well-being” (1.4) and they “Appreciated new challenges” (1.2) more than the reference group.

Working patterns

Most persons (52%) worked between 5 and 25 hours per month, and about as many worked less (26%) as more (22%). One third wanted to work more than they do, and three percent wanted to work less Table 2.

Table 2

Working patterns as pensioners (N = 1,471)

Working patterns	%
Hours per month
< 5	26
5–10	24
11–25	28
26–50	14
51–75	3
> 75	5
Work characteristics
Other work tasks than before	73
Same work tasks as before	27
Want to work more	37
Want to work less	3

Service provision

The employees provided services to private customers (61%), companies (11%) and 28% provided services to both groups. The most frequent service area was gardening (43%), trades (33%) and cleaning (31%). But there were also assignments such as care (8%), IT services (4%), doing housework (4%), helping with removals and snow clearance.

As seen in Table 3 there were some correlations worth mentioning. Men seem to work more with gardening, trades, IT services, removals and snow clearance, while women to a higher extent work with cleaning, housework/childcare/cooking and care. Persons with higher education seem to work more with IT services, housework/childcare/cooking and care, while persons with lower education work more with removals.

Table 3

Correlations between age, gender, education level and the provided service areas for private customers (n = 1,471)

	Age	Gender	Education
		Ref: women	level
Cleaning	–0.000	–0.619***	0.025
Gardening	–0.051	0.399***	–0.044
Trades	–0.057*	0.539***	–0.041
Housework/Childcare/Cooking	0.023	–0.246***	0.116***
IT services	–0.014	0.102***	0.117***
Care	0.070**	–0.299***	0.080**
Removals	–0.045	0.144***	–0.053*
Snow clearance	–0.007	0.255***	–0.035
Other services	0.031	–0.015	0.035

*p < 0.05, **p < 0.01, ***p < 0.001.

In total, a majority (73%) worked with tasks they had not done before. From a purely economic perspective, this may seem odd because it is likely that the productive capacity, expertise and comparative advantage in the labour market would yield higher returns (wage) as well as appreciation and confirmation in their old line of work. On the other hand, shifting working career may allow people to explore new (or old) interests and cultivate new skills that may well be rewarding. The segment of the labour market sector targeted by the staffing agencies for retirees centres around services covered by the RUT and ROT (see above). Hence, there may be limited possibilities for people to continue in their old line of work within these agencies. Of the motivational aspects addressed above, “Poor economy” and “Kicked out” are probably the strongest determinants of continuing to work for economic reasons. In order to assess whether people shift their line of work out of necessity, we computed correlations between these aspects and working with “Other tasks than before”. The results showed that there are strong positive correlations (p-values < 0.01), but only for women. For men the estimated correlations are close to zero. Hence, women who change their work tasks are more likely to justify their choice to continue to work with “Poor economy” and having been “Kicked out” than women who continue in their old line of work. This indicates that women bridging their employment may be pushed into new lines of work out of necessity.

Discussion

The results will be discussed in relation to the study aim: to describe the characteristics of the pensioners working for a staffing agency (sample), why (motivational aspects), to what extent (working patterns) and with what they worked (service provision).

Sample

In the sample there were more men than women who had chosen this form of work, which is in line with results from other studies showing that men are more likely to work in bridge jobs than women [23, 24]. Most of the employees were aged 65–74 years, which is within the age span in Dingemans, Henkens and van Soline’s [14] study of 16 European countries, where 11% of older adults between the ages of 60 and 75 participated in bridge employment. They found a considerable variation across the investigated countries. Working after retirement was rather common in countries like Estonia (22%), Sweden (21%), Switzerland (20%) and Denmark (12%) compared to countries such as Spain (3%), Slovenia (3%), Poland (5%) and France (5%). In our study 73% had secondary school or higher education, a circumstance that has been found to have a significant impact on the likelihood of individuals working at higher ages [6]. In our sample 61% were married/cohabiting, contrary to Dingemans, Henkens and van Soline’s study [14] where retirees who had experienced divorce or widowhood were more likely to work in bridge jobs compared to married retirees.

Motivational aspects

The three most common motivational factors given by the respondents for working at a staffing agency were Fellowship, Extra money, and Confirmation. The Fellowship case consisted of employees who typically had not been allowed to continue their previous work. They appreciated new challenges and thought that their work increased their well-being. In addition to this, work provided financial support. The Extra money case was typically represented by single women who had a secondary school education. They had been forced to retire and worked out of economic necessity, but they also appreciated the social context. The Confirmation case typically involved married/cohabitant women with secondary school. They worked because they had a partner who worked, to earn extra money and to be part of a social context. They also appreciated new challenges and perceived that their work increased their wellbeing.

Important incentives stated by most of the employees were getting an extra income and belonging to a social context. Most employees also appreciated new challenges and achieved well-being through their work. Previous studies have identified factors that could indicate the propensity to continue working later in life and choose bridge employment as an option. These factors are, for example, health, work experience, level of education and financial factors [6–8, 17, 18]. It was not possible to fully explore health in the present study since accepting a job through a staffing agency demands a certain level of health to be able to perform the job in question. On the other hand, if looking at health as perceived well-being, a majority of the respondents stated that the working life after pensionable age increased their wellbeing. Finance was a strong factor, and although this factor kept recurring in the different cases, the arguments ranged from “enhancing” life, to needing that extra money to get by. There was a positive correlation between needing an extra income and being a single woman.

Among all six cases, there was a concordant agreement that belonging to a social context was important. The staffing agency thus provided the employees with an important social arena in life. However, the cases revealed a possible gender divide in terms of arguments for working in bridge employment via a staffing agency. The material suggests that while single women need to work, men, to a larger extent, work for the social context and well-being. This should be further explored in future studies.

This study has focused on the employees at the staffing agency who are involved in bridge employment, since there was a need for studies on the employees’ perceptions and perspectives. Flynn [17] and Shacklock and Brunetto [18] emphasize the importance of working conditions in order to understand the desire and opportunities for a prolonged working life. Therefore, the workplace characteristic of staffing agencies should be explored in future studies. We propose that this is done in combination with employees’ perception of bridge employment. The reason is that unmet expectations could jeopardize both employer and employee outcomes of bridge employment [3]. This is of particular interest when working for a staffing agency, where the customer is an important part of the work environment for the bridge worker, since work is performed in the customer’s context, not at the staffing agency’s offices.

Most employees worked 25 hours or less per month and with other work tasks than before. They worked the most two years after their pension year. Nearly 40% wanted to work more, although 3% wanted to work less. Nearly half of the sample stated that they wanted to work more, and it could be noted that the total contribution of working hours per month is fairly low. Further research on how much more time they are willing to work per month is wanted. Dingemans, Henkens and van Soline [14] found that though there is high support for working after retirement in countries like Denmark, Sweden and the Netherlands, a substantially lower number of hours is worked in bridge jobs than in countries with lower support.

Most services were done for private customers. Men mostly worked with services like gardening and trades while women to a higher extent worked with cleaning and other housework tasks. Additionally, persons with higher education also worked with IT services and housework to a higher extent than those with elementary school education. This is in accordance with earlier research [6-8] showing that formal education, work experience and health have a significant impact on the likelihood of individuals working at higher ages. It is however interesting to note that the study indicates a willingness to do different tasks, such as performing housework, although people have a high degree of education. In other countries, individuals working in housework may be less educated and foreign-born. This could be an effect of the economic and socio-cultural environment highlighted by Mazumdar et al. [3]. In Sweden, people are often interested in manual labour and although there are increasing social divides in society, working life still is characterized by an openness between hierarchies.

Conclusions

Working for a staffing agency as a form of bridge employment attracted more men than women. The material suggests that while single women need to work out of economic necessity to gain some extra income, men, to a larger extent, work for the social context and well-being. The highest work frequency in 2017 (14%) in the sample was found for those who retired in 2015, i.e. two years after their retirement year. A majority indicated that the work they were doing was different from earlier in their working life.

Strengths and limitations

A strength in this study is the response rate of 44%, which is relatively high for a questionnaire study of this type. Additionally, the number of respondents, 1,741 out of 3,950, was also regarded as sufficient for the analysis performed. Another strength is that the majority of the employees (about 95%) had an opportunity to answer the questionnaire, since only 207 employees were excluded from answering due to a lack of a registered e-mail address. Though the studied sample seems to be highly representative for the full sample of employees in terms of gender and age, a limitation of the study concerns external validity. The lack of an external reference group that does not engage in bridge employment makes it hard to draw any strong conclusions on how the results of this study would translate into other groups of retirees.