Employers' Adjustment to Longer Working Lives.

BACKGROUND AND OBJECTIVES: The aging of the workforce and the trend toward longer working lives has substantial implications, not only for employees, but also for employers. The aim of this study is to quantitatively investigate the extent to which employers have implemented human resource (HR) practices in adjustment to longer working lives. We distinguish between information practices, health practices, and person-job fit practices. RESEARCH DESIGN AND METHODS: Data from 1,296 Dutch employers, collected in 2017, are used to study the ways in which employers have adjusted their HR practices. We estimate a structural equation model to identify predictors of implementing adjustment measures.
RESULTS: Employers have largely started to adjust their HR practices to make longer working lives more attainable. Especially larger organizations are highly active in using HR practices to enhance the long-term employment perspectives for their employees. Employers who are concerned about the mental fitness of their workforce in relation to longer working lives are especially likely to invest in information and health adjustment practices. Organizations with a high proportion of older workers are mostly focused on providing information. DISCUSSION AND IMPLICATIONS: The HR practices that are implemented by employers to facilitate longer working lives are different from those traditionally associated with an older workforce, implying a fundamental shift in employers' focus. Instead of promoting the exit of older workers and accommodating older workers specifically, employers are now moving toward providing information, promoting healthy habits, and improving long-term person-job fit, also targeting early- and mid-career workers.

Translational Significance

Employers have started to implement adjustment HR practices to make longer working lives of their employees more attainable, focusing on information, health, and person–job fit. Policy initiatives that promote the implementation of such HR practices may ease the transition toward longer working lives, especially for workers in smaller organizations and organizations with a comparatively low proportion of older workers.

Background and Objectives

The previous two decades have seen significant changes to the retirement process of older workers in countries all around the world. Population aging has created the need for higher labor market participation rates of older workers to keep pension systems financially sustainable (Organisation for Economic Co-operation and Development [OECD], 2017). Many countries have implemented policy changes to facilitate longer working lives, for example by discouraging early retirement, abolishing mandatory retirement, or increasing the state pension age (also known as normal retirement age). The implications of these changes have been substantial for older workers. Current cohorts of older workers, especially in Western European countries, have worked the majority of their careers in what has been characterized as a “culture of early retirement,” in which retirement several years before the normal retirement age was a completely accepted and normal phenomenon, and working up to and beyond normal retirement age was very uncommon (Hofäcker & Unt, 2013). As a result of the institutional changes in recent decades, the current cohort of older workers is no longer able to retire early without accepting a severe income cut, and has had to adjust their retirement timing (Euwals, Van Vuuren, & Wolthoff, 2010). As Van Solinge and Henkens (2017) have shown, these institutional changes have often led to resentment and anger among older workers, since they have to work much longer than they previously envisioned. These strong negative emotions are especially prevalent among those in more physically demanding jobs, and among those with fewer personal and social resources. Such feelings may hamper productivity of older workers and may lead to stereotyping of older workers as unmotivated.

However, not only employees, but also employers have to adjust to the changing composition of the workforce and the institutional reforms that have been implemented in recent years. Organizations are challenged to find ways to accommodate larger absolute and relative numbers of older workers into their operations and keep them productive until higher ages. Older workers have different skills, capacities, and needs than younger workers, which means organizations may need to adapt, for example, their management styles or job design and operational process to facilitate sustainable careers. The aging workforce and institutional reforms thus have a direct effect on organizations’ workforce composition, staff planning, opportunities for internal promotions, organizational restructuring, and the way in which organizations operate. How organizations, in practice, adjust to longer working lives and its effects have, however, received little attention in the literature. Some studies explore how employers adapt to the general aging of the workforce (e.g., Conen, Henkens, & Schippers, 2011), and there are studies that investigate the types of (age-based) human resource (HR) policies employers use (e.g., Fleischmann, Koster, & Schippers, 2015; Kooij, Jansen, Dikkers, & De Lange, 2014; Lössbroek, Lancee, Van der Lippe, & Schippers, 2017; Van Dalen, Henkens, & Wang, 2015). These studies generally focus on formal written-down age-related personnel policies that aim for specific goals such as increasing older workers’ access to formal training, accommodating their workplace needs, or promoting (part-time) exit. Often, these policies are part and parcel of collective labor agreements (Smits, Beeksma, Feenstra, & Junger-Van Hoorn, 2010).

However, next to these more formal and specific personnel policies, there are also more informal and generalistic (i.e., available and beneficial for all employees of an organization) adjustment practices that can be taken to adjust to an aging workforce and institutional reforms (Ollier-Malaterre, McNamara, Matz-Costa, & Pitt-Catsouphes, 2013). These are likely to have the more general goal of making an organization’s workforce better adapted to the changed aged structure and longer working lives, and more resistant to the potential long-term negative effects of workforce aging. We argue that these informal adjustment measures have been implemented only fairly recently in response to the aging workforce and the institutional reforms that have led to longer working lives. While evidence for this has been presented in qualitative studies (e.g., Conen, Henkens, & Schippers, 2014; Moen, Kojola, & Schaefers, 2017), a quantitative assessment of the implementation of such measures is missing (Henkens et al., 2018). In this article, we aim to fill this gap.

We aim to contribute to the literature in three main ways. First, we present a quantitative measure to capture different ways in which employers have adjusted their HR practices in recent years within the context of an aging workforce and longer working lives. The level of implementation of these practices indicates how macro-level institutional retirement-related reforms start to percolate down and affect organizational practices.

Second, we conceptually group these adjustment measures and present three main categories: (1) measures to provide information to managers and older workers about longer working lives and retirement planning; (2) measures to enhance workability and health of (older) workers; and (3) measures to optimize person–job fit of employees. While measures in each of these three main categories can contribute to making longer working lives more attainable, the specific measures underlying these main categories can differ strongly in their ease of implementation, costs, and effectiveness. We use a structural equation modeling (SEM) approach to identify these main categories as latent variables that represent various adjustment HR practices.

Third, we study how the extent to which employers have implemented adjustment measures toward longer working lives can be accounted for. We examine whether structural characteristics of organizations, such as their size and proportion of older workers, the influence of unions, and experiencing recruitment problems, account for differences in implementing adjustment HR practices. In addition, we examine whether the implementation of HR practices is affected by employers’ worries regarding mental and physical fitness of their employees, limited employability of employees with health problems, and the extent to which older workers block career opportunities for younger workers. Recent research has shown such worries are common among older workers themselves (Van Solinge & Henkens, 2017). In a context where exit before the public pension age is no longer easily attainable, employers’ concerns might offer a strong stimulus to design organizational policies that facilitate and stimulate active and healthy aging.

We examine these relationships with survey data collected among a large sample of Dutch employers in 2017. The Netherlands is a country that is at the forefront of extending working lives and promoting active labor market participation of older individuals. Although the culture of early retirement was prominent in the Netherlands until the early 2000s, several policy changes have since been implemented to promote longer working lives. For example, financially attractive early retirement schemes have been abolished, and access to “alternative routes” into early retirement, through for example unemployment or disability benefits, have been severely restricted. The policy changes culminated in 2012, when the Dutch government decided to gradually raise the eligibility age for the public state pension (also known as statutory retirement age or normal retirement age) from 65 to 67 years, after which it will be tied to changes in life expectancy projections (OECD, 2014; Sonnet, Olsen, & Manfredi, 2014). Under the current law, each year of added life expectancy will lead to an increase in the public pension age of 1 year. As a result, the Netherlands is projected to have one of the highest normal retirement ages of the world in 2060, at around 71 years (OECD, 2017). (There is an ongoing public debate in the Netherlands about this law. Commentators have, for example, called for a lower public pension age for those with a lower level of education or a physically demanding job. At the moment of writing, however, there was not enough political support for such proposals.) Other relevant parts of the institutional framework of the Dutch labor market include mandatory retirement at the public pension age for most workers, obligatory pension savings for wage-and-salary employees, and a comparatively high level of employment protection legislation (OECD, 2014). The Netherlands is also known for its high prevalence of part-time work, especially among women (Wielers & Raven, 2013). Labor unions bargain with employer representatives to establish sectoral collective labor agreements that specify additional conditions of employment for employees in different sectors.

Employment participation of older workers in the Netherlands before the policy changes of the last 15 years was comparatively low, due to generous early retirement schemes. But the policy changes, together with increased demand for labor and a higher educated cohort of older workers, have led to an unprecedented increase in labor force participation rates of older workers, average age at retirement, and employees working until the (increasing) public pension age (Visser, Gesthuizen, Kraaykamp, & Wolbers, 2016). For example, employment participation of 60–65 year olds has increased from 28.5% in 2007 to 55.6% in 2018, and the average age at retirement has increased from 61.7 years to 64.8 years in that time span (Statistics Netherlands, 2018).

Employers’ Adjustment HR Practices

The need for specific HR policies and practices for older workers stems from the notion that older workers have different skills and needs than younger workers, which need to be catered to in order to prevent a decline in productivity (Beier, 2015; Kooij et al., 2014). Recognizing the needs of older workers and managing their employment and retirement in a way that is congruent with the overall business strategy of the organization may furthermore contribute to the organization’s success (Rau & Adams, 2013).

Research on age-based policies of European employers reveals that organizations’ formal HR policies mostly can be categorized as either aimed at exit, training, or accommodation (e.g., scheduling flexibility, ergonomic support) of older workers (Van Dalen et al., 2015). However, research has also shown that many of the age-based HR practices are developed in an ad hoc way, without any formal written statements, rules, or procedures (Conen et al., 2014; Flynn, 2010). There may thus be a distinction between more formal HR policies that are based on, for example, large corporate policies or collective labor agreements, and more informal HR practices. These informal HR practices are likely specifically tailored to the context of one particular organization. In this study, we assert that, in the initial stages of organizational responses to institutional changes, such practices are likely to be more prevalent than formal HR policies.

Furthermore, we argue that gradually the focus is now shifting from ad hoc policies designed to accommodate current cohorts of older workers to a more fundamental approach to longer working lives, which includes policies and practices aimed at workers at all stages of their careers (De Lange, Kooij, & Van der Heijden, 2015; Moen et al., 2017). After all, due to persistent population aging, the relative share of older workers in the workforce will keep increasing, after which it will plateau and stabilize at a high level (i.e., there will be no return to a “young population”). This implies that longer working lives are not a temporary phenomenon, but will be necessary in aging societies across the globe to protect the sustainability of welfare states (Hofäcker & Radl, 2016). Preparing early- and mid-career workers will be necessary to accomplish longer working lives.

The lens through which to view HR practices in an aging workforce is quite diverse. Different disciplines focus on different aspects of the challenges that employees face in the workplace. Gerontological, demographic, and occupational health studies primarily examine the occupational needs and capacities of older workers (for an overview, see Crawford, LePine, & Rich, 2010), and study how to alleviate the stress that may arise as a result of a mismatch between job demands and employees’ capabilities. HR practices may then be used to enhance productivity with health investments to prevent health-related productivity declines due to aging. Also, HR practices may aim at increasing person–job fit to accommodate diversity among staff as they age. Economic theory emphasizes that mismatches between firms and workers may occur over time as labor market institutions change (e.g., when the retirement age increases) or when worker productivity declines unexpectedly (Lazear, 1979). While employers’ opportunities to lay off older workers via early exit routes have been blocked, they may opt for information strategies for managers and employees to enable more flexible transitions into retirement. Based on these perspectives, we argue that there are three main strategies employers may use to facilitate the long-term longer working lives of their employees: information-related strategies, health-related strategies, and strategies to enhance person–job fit. These three strategies may be implemented separately, sequentially or simultaneously. We elaborate on these strategies below.

Information Practices

Relevant information about opportunities for early, gradual, phased, late, or “normal” retirement and its financial consequences plays a major role in decisions made by older workers (Adams & Rau, 2011; Fisher, Chaffee, & Sonnega, 2016). In order to have some influence and control over their employees’ retirement, organizations may therefore choose to play an active role in informing older workers about different retirement opportunities, especially since organizational and managerial involvement and support for either early or late retirement may influence older workers’ decision-making process (Davies, Van der Heijden, & Stephenson, 2018; Münderlein, Ybema, & Koster, 2013; Van Solinge & Henkens, 2014). Discussing older employees’ retirement plans, for example as part of their yearly performance appraisal, may provide clarity about expectations and possibilities for both employer and employee, and can therefore be a good example of organizations’ adjustment to the reality of longer working lives. Another way in which employers may utilize information to manage longer working lives is by providing information and training to (line) managers about aging, longer working lives, and ways to manage older workers (Truxillo, Cadiz, Rineer, Zaniboni, & Fraccaroli, 2012), thereby promoting active and successful aging at work.

Health Practices

Older workers’ health is one of the central factors associated with their productivity (Robertson & Tracy, 1998). Investments in occupational health may translate into lower rates of sickness leave and increased productivity. In the Netherlands, employers have an extra incentive to invest in workers’ health after a series of policy reforms in the 1990s and 2000s made employers liable to pay the first 2 years of disability insurance in case an employee falls ill (Euwals, A. Van Vuren, & D. Van Vuuren, 2012). Indeed, Conen and colleagues (2014) observed an increase in health-related investments by Dutch organizations, primarily those featuring physical labor, in an attempt to prevent long-term disability of employees. Next to focusing on occupational health, employers may also try to promote a more healthy lifestyle of their employees outside of work, for example by stressing the importance or giving workshops in healthy eating, exercise, and a healthy work–life balance (Grawitch, Trares, & Kohler, 2007). Such measures may all contribute to healthy aging and longer and more productive working lives (Goetzel & Ozminkowski, 2008).

Person–Job Fit Practices

The relationship between age and productivity is central to concerns about the aging workforce. Although a general negative association between age and core task performance has not been consistently found (Ng & Feldman, 2008), many employers subscribe to the stereotypical view that productivity declines with age (Van Dalen, Henkens, & Schippers, 2010). In general, older age is associated with a decrease in factors such as cognitive skills and physical abilities, but often compensated by fewer counterproductive activities and more experience (Ng & Feldman, 2008; Skirbekk, 2004). Still, the factors that contribute to employee work satisfaction, engagement, and productivity often change with age (Truxillo et al., 2012), making the consideration of person–job fit and job design over the course of the career an important one. With more older workers in the workplace, employers may be concerned with tailoring job design to accommodate older workers, or retraining older workers for jobs within or outside of the organization in which they may be better suited to retain their productivity until retirement. Since training generally becomes more difficult with increasing age, the issue of current and future person–job fit is also important for early- and mid-career employees (Moen et al., 2017).

Method

Data

Data were collected from Dutch employers between December 2016 and March 2017. First, a sample of 6,000 organizations with at least 10 employees was drawn. Organizations with fewer than 10 employees were excluded because they commonly have little formal HR management and deal with aging in an ad hoc way (Cardon & Stevens, 2004). The sample was stratified according to size and sector, meaning large organizations and those in the public sector were oversampled, and small organizations and those in the services sector were undersampled, in order to ensure sufficient responses from all types of relevant organizations. This approach was successful with an approximately equal distribution of respondents over sizes and sectors. As a result, the data are not fully representative of the population of organizations in the Netherlands, but we control for size and sector to test for variation on those variables.

A hard-copy questionnaire was sent to the organizations, along with an accompanying letter inviting them to participate in the study. The letter also contained a unique code with which the employers could access an online version of the questionnaire. Two reminders were sent, one containing a letter reminding them of the survey and the code for the online questionnaire, and one also containing a new hard copy of the questionnaire. The questionnaire was addressed to the director or CEO of the organization, although the letter stated that also other employees knowledgeable about the organizations’ background and practices could participate. In total, 1,312 organizations participated in the study, for a response rate of 22%. This rate is lower than that commonly found in surveys targeting individuals, but comparable to other large surveys among organizations (Baruch & Holtom, 2008) and earlier organizational surveys in the Netherlands (Conen et al., 2011). The surveys were completed by owners (23%), directors (24%), HR managers (27%), HR employees (12%), general managers (6%), and other employees (8%). Half of the responses came from the hard-copy questionnaire, while the other half came from the online version.

Sixteen observations were deleted because they did not answer the question on their organization’s adjustment practices, leaving 1,296 observations for analysis. Item non-response on the independent variables was generally low, for all but one variable in the 0.5%–3% range, and 8% on the remaining variable (recruitment problems). We imputed missing values using single stochastic imputation (Stata command mi impute chained), which has been found acceptable with such low prevalence of missing values (Enders, 2010).

Measures

Dependent variables

As mentioned above, we distinguish three conceptual categories of HR practices that employers may use to make longer working lives more attainable: information practices, health practices, and person–job fit practices. These are latent constructs that are each based on three specific measures that were derived from the questionnaire. Respondents were asked: “How does your organization deal with the fact that employees need to keep working until higher ages?”, providing a yes or no answer on a number of proposed HR practices. First, information-related measures were based on the following items: “We provide information to managers,” “We talk to all workers over age 60 about their desires and plans,” and “We actively support employees in making retirement choices.” Second, health-related measures were based on the following items: “We pay more attention to the work–life balance of employees,” “There is increased attention for healthy working conditions,” and “We stimulate a healthy lifestyle.” Third, person–job fit measures were based on the following items: “In case of health restrictions, we are more likely to look for another job within this organization,” “In case of limited employability of an older worker, we actively support their search for another job outside this organization,” and “During the entire career of employees, there is increased attention for person–job fit.”

Independent variables

Several structural characteristics of the organizations were included to study whether they affected adjustment measures implemented by employers. In particular, we asked for the sector of industry in which the organization operated (reduced to represent three broad sectors: industry, services and trade, and public) and the size of the organization (the number of employees; reduced to three categories: small [10–49 employees]; medium [50–249 employees], and large [≥250 employees]; categories based on Eurostat’s Structural Business Statistics classification). We also asked for the percentage of their workforce that was female, older than age 50, were low educated, were on a flexible (e.g., temporary) contract, and that worked part-time. In addition, we asked for the perceived influence of unions on their organization’s personnel policies (answers on a 4-point scale from no influence to a large influence), and the extent to which their organization experienced recruitment problems (answer categories “no,”, “sometimes,” or “often”).

Employers’ worries were measured in the context of longer working lives. We asked employers: “People need to keep working much longer than before. In that context, to what extent do you, as an employer, worry about …?” (1) whether employees are physically capable to do so, (2) whether employees are mentally capable to do so, (3) about limited career opportunities for younger employees, and (4) about limited employability of employees with health problems. They answered on a scale from 1 to 5 (1 = Not at all; 2 = A little; 3 = Fairly; 4 = Very; 5 = Extremely). The wording of the questions is presented in Table 1.

Table 1.

Descriptive Statistics

	Mean	SD	Wording
Dependent variables
Information practices			“How does your organization deal with the fact that employees have to work much longer than before? Are the following measures applied?” (1 = yes; 0 = no)
Information for managers	0.34	0.47
Talking with older workers about retirement plans and desires	0.54	0.50
Actively supporting employees in their retirement decisions	0.44	0.50
Health practices
Healthy work–life balance	0.61	0.49
Healthy working conditions	0.82	0.39
Supporting a healthy lifestyle	0.63	0.48
Person–job fit practices
Limited employability, look for other job within organization	0.45	0.50
Limited employability, support job search outside organization	0.30	0.46
Continuous attention for person–job fit	0.65	0.48
Independent variables
Sector			“Within which of the following industrial sectors does your organization operate?”
Industry	0.34	0.48
Services and trade	0.29	0.45
Public sector	0.38	0.48
Size	403.70	1895.12	“How many employees are currently employed in this organization?” (small = 10–49; medium = 50–249; large = ≥250)
Small	0.33	0.47
Medium	0.39	0.49
Large	0.28	0.45
Proportion of female employees	0.42	0.31	“What percentage of employees is female / older than 50 years of age / low-educated / on a flexible contract / working part-time?” Transformed to proportions by dividing by 100
Proportion of older employees	0.32	0.18
Proportion of low- educated employees	0.53	0.32
Proportion of flexible contract employees	0.14	0.17
Proportion of part-time employees	0.40	0.32
Influence of unions	1.89	0.98	“To what extent do unions influence personnel policies in your organization?” (answers range from 0 [no influence] to 4 [large influence])
Recruitment problems			“Has your organization experienced recruitment problems in recent times?”
No	0.35	0.48
Sometimes	0.34	0.47
Often	0.30	0.46
Employers’ worries			“Nowadays, older workers have to work much longer than before. To what extent are you as an employer worried…”
Physical fitness	3.24	1.14	... whether employees are physically capable to do so?”
Mental fitness	3.07	0.96	... whether employees are mentally capable to do so?”
Career opportunities younger workers	2.71	0.93	… about limited career opportunities for younger employees?”
Limited employability due to health problems	3.43	1.05	... about limited employability of employees with health problems?”

Results

Descriptive Statistics

Table 1 presents descriptive statistics for the variables used in this study. On average, health-related adjustment measures are most frequently implemented by employers. More than 80% of employers have focused more on safe and healthy working conditions in recent years, while more than 60% of employers focus on a healthy work–life balance and support a healthy lifestyle. Information-related measures and person–job fit measures are less frequently implemented, although still a majority of employers have implemented at least one of the measures. Approximately 4% of employers have not implemented any of the adjustment HR practices, whereas about 2% has implemented all of the adjustment practices.

With regard to employers’ worries within the context of longer working lives, we observe that employers are on average most worried about the limited employability of workers with health problems. The worries about whether employees will be physically and mentally able to keep up with longer working lives are also noteworthy. Employers are least worried about limited career opportunities for younger workers due to longer working lives.

Multivariate Analyses

Because our dependent variables are conceptualized as latent variables, constructed from dichotomous variables, we first performed confirmatory factor analysis to examine the construct validity of the adjustment measures (which is, in the case of dichotomous variables, equivalent to a two-parameter logistic Item Response Theory model; StataCorp, 2015, pp. 63–64). We used Stata’s gsem command with logit link function to fit a three-factor Generalized Structural Equation Model with items loading on their respective latent constructs, and the variance for latent variables constrained to 1 (StataCorp, 2015, pp. 311–314). Factor loadings, which can be interpreted as slopes or logit coefficients, for the measurement model were as follows: Information: information for managers (1.94), talking about retirement plans and desires (1.39), and supporting employees (1.37); Health: work–life balance (1.56), working conditions, (1.69), and supporting healthy lifestyle (1.53); and Person–job fit: other job in organization (0.90), support external job search (0.62), and continuous attention for person–job fit (1.53). The information criteria for the three-factor model were lower than for the one-factor model (Aikake Information Criterion [AIC]: 13,926.56 vs. 13,978.78; Bayesian Information Criterion [BIC]: 14,035.07 vs. 14,071.99), thus showing better fit of the three-factor model.

Next, we specified a full structural equation model by regressing the latent adjustment measures on the organizations’ structural characteristics and employers’ worries. The results are presented in Table 2. The RMSEA measure for goodness-of-fit is not available for SEM models with dichotomous variables (StataCorp, 2015, p. 142); AIC (13,912.06) indicates model improvement over the empty three-factor measurement model (although BIC increases to 14,253.09, it places a heavier penalty on model complexity, and is thus less informative in this case). Since we used the logit link function, we present odds ratios (ORs), which indicate the likelihood of having implemented the relevant adjustment practice (vs. having not implemented) when the independent variable increases with 1, meaning that ORs below 1 indicate a negative effect, and ORs above 1 indicate a positive effect. The results show no significant differences in applying adjustment HR practices for organizations from different sectors. Organizational size, on the other hand, is one of the main drivers of the implementation of adjustment measures: Large organizations are more likely than medium-sized organizations, which are more likely than small organizations, to implement adjustment HR practices to deal with longer working lives of their employees, especially when it comes to information-related practices.

Table 2.

Structural Equation Modeling Results of the Effect of Organization’s Structural Variables and Employer’s Worries Regarding Longer Working Lives on the Implementation of Information, Health, and Person–Job Fit Measures (N = 1,296)

	Information practices		Health practices		Person–job fit practices
	OR	SE	OR	SE	OR	SE
Sector (ref. = industry)
Services and trade	0.71	0.14	0.87	0.10	0.92	0.14
Public	1.46	0.37	1.16	0.18	0.88	0.17
Size (ref. = small)
Medium	2.40***	0.46	1.48***	0.16	1.87***	0.29
Large	4.26***	1.07	2.76***	0.44	2.72***	0.62
Proportion female employees	0.57	0.26	1.15	0.32	1.60	0.52
Proportion older employees	7.31***	3.31	0.88	0.21	0.64	0.19
Proportion low-educated employees	0.74	0.19	1.08	0.16	0.75	0.13
Proportion flexible contract employees	0.72	0.32	0.55*	0.15	1.10	0.35
Proportion part-time employees	0.51	0.22	0.70	0.18	0.43*	0.14
Influence of unions	1.24**	0.10	1.08	0.05	1.13*	0.06
Recruitment problems (ref. = no)
Sometimes	1.16	0.19	1.00	0.10	0.99	0.11
Often	0.65*	0.12	0.80*	0.09	0.89	0.12
Employers’ worries
Physical fitness	1.14	0.10	1.18**	0.07	1.08	0.07
Mental fitness	1.50***	0.14	1.12*	0.06	1.12	0.07
Career opportunities younger workers	0.87	0.07	0.97	0.05	0.98	0.05
Limited employability health problems	0.91	0.08	0.95	0.05	0.97	0.06

Note. OR = odds ratio.

*p <.05. **p <.01. ***p <.001.

An organization’s workforce composition also affects the implementation of adjustment measures. The proportion of older employees in an organization is especially strongly related to the implementation of information adjustment practices (OR = 7.31), but is not related to the implementation of health-related or person–job fit adjustment practices. Organizations with a larger proportion of flexible workers are less likely to implement health-related adjustment measures (OR = 0.55), and organizations with many employees working part-time are less likely to implement person–job fit measures (OR = 0.43). Organizations where unions have a strong influence on personnel policies are more likely to have implemented information (OR = 1.24) and person–job fit measures (OR = 1.13), but not health measures. Organizations that regularly have recruitment problems are significantly less likely to implement information (OR = 0.65) and health measures (OR = 0.80), but not person–job fit measures.

Additionally, the results show that employers’ worries about workforce aging influence their adjustment HR practices. Employers who are worried about the physical fitness of their employees are more likely to invest in the workability and health of their employees (OR = 1.18). Employers who are worried about the mental fitness of their employees are most likely to invest in information-related HR practices (OR = 1.50), but also more likely than employers who do not worry about the mental fitness of their employees to invest in health-related measures (OR = 1.12), but not person–job fit adjustment practices. While worries about the limited employability of employees with health problems are most common among employers (Table 1), they are not related to the implementation of adjustment practices. Worries about limited career opportunities for younger workers also do not influence the implementation of adjustment HR practices.

Discussion and Implications

In this article, we have examined the ways in which employers have adjusted their HR practices to deal with longer working lives of their employees. We have conceptually distinguished three types of adjustment measures: information practices, health-investment practices, and HR practices to increase person–job fit of employees. The results show that employers have, on a large scale, started to implement mostly informal HR practices that are aimed at facilitating longer working lives of their employees. A noteworthy shift is that employers no longer primarily focus on promoting the exit of older workers (cf. Van Dalen et al., 2015), but rather focus on providing information to both older workers and managers about longer working lives. We found that especially employers’ worries about whether employees will be mentally capable to keep on working until higher ages than before drives the implementation of information and health-related adjustment practices, whereas worries about employees’ physical fitness were only related to the implementation of health-related measures. Other worries that employers may have, such as worries about limited career opportunities for younger workers and limited employability of employees with health problems, were not related to adjustment practices.

Another main finding was that larger organizations are much more likely to have implemented adjustment measures to deal with longer working lives. This is in line with earlier findings that showed that larger organizations are more active in terms of supporting older workers to continue working and in general are more occupied with age management (Oude Mulders, Henkens, & Schippers, 2017). This finding can be attributed to the notion that larger organizations need to implement uniform ways to deal with employees, whereas smaller organizations will deal with issues such as longer working lives in a more ad hoc fashion. Next, we found that organizations with a larger proportion of older workers are much more likely to implement information-related adjustment measures, but not measures related to health or person–job fit of employees. This is somewhat surprising, given that proportion of older workers is usually positively related to any outcome measure related to aging (e.g., Oude Mulders, Henkens, & Schippers, 2015). Presumably, organizations with a larger share of older workers are mostly concerned with the productivity of those workers in the short and middle-long term, whereas adjustment measures focusing on health and person–job fit are more related to longer-term productivity of the entire workforce of an organization. A strong influence of labor unions on personnel policies was related to the implementation information and person–job fit adjustment practices, showing that labor unions may not only affect formal policies (Van Dalen et al., 2015), but also the more informal policies studied here. Finally, market forces were also found to be relevant for the implementation of adjustment practices. Organizations that experience regular problems finding suitable personnel were less likely to implement information and health-related adjustment practices, presumably because solving their recruitment problems in the short term takes precedent over supporting longer working lives in the long term.

Although employers are more engaged in the development of strategies to accommodate higher retirement ages in their organizations, it is unclear whether all workers have equal access to the HR practices of employers. Earlier research suggest that such HR policies may be predominantly available to high-skilled and high-income employees, since employers are mainly interested in their continued employment, but not to employees in lower-skilled and lower-income jobs (Hofäcker, Hess, & König, 2016). Furthermore, it is important to note that other labor market developments might weaken the position of older workers. Continued flexibilization of the labor market, with more temporary and contingent jobs, may seduce employers with an aging staff to select only those older workers with the highest expected productivity.

Some limitations of this study should be noted. First, we have used cross-sectional data whereas adjustment procedures are by definition dynamic in nature. As a result, we cannot make causal inferences about the relation between employers’ worries and the implementation of adjustment measures. Conceptually, when worries would lead to the implementation of adjustment measures, the effect of these measures could lead to reduced worries. However, the policy change in the Netherlands that gradually increases the public pension eligibility age from 65 to 67 was implemented in 2012, and the gradual increase in public pension age was still ongoing at the time of data collection (2017). In addition, the policy change and its effects on employees and employers were still strongly discussed at the time of data collection. This leads us to belief that employers’ worries would not have had the time to subside strongly after implementing adjustment measures to deal with longer working lives in the period between the policy change and data collection, thus reinforcing the proposed mechanism. Still, a longitudinal investigation into the mechanism which leads to implementing adjustment measures would be welcome.

Second, we analyze data from the Netherlands. While the Dutch labor market is comparable to that of other Western European countries, the findings in this study may not be generalizable to countries with a more liberal labor market, such as the United States and the United Kingdom. In particular, the lower level of employment protection, the lower level of pension coverage, and the lack of mandatory retirement regulations in such countries may lead to employers adjusting in different ways than Dutch employers. Still, employers in the United States are also adapting their policies and practices in response to the aging workforce (Moen et al., 2017), and the practices studied here are likely to also be observed there, with differences most likely to relate to the prevalence of certain adjustment practices and the structural characteristics associated with the implementation of adjustment HR practices. Third, the study is carried out among employers with at least 10 employees. The results can therefore not be generalized to very small organizations.

This study has shown that employers have, on a large scale, started to implement HR practices to adjust to the reality of longer working lives. Many of these HR practices that are to make longer working lives more attainable are different from the HR policies that used to be associated with an older workforce, such as trying to promote the early retirement or providing older workers with benefits such as additional leave or flexible working hours (Kooij et al., 2014; Lössbroek et al., 2017; Van Dalen et al., 2015). Instead, these new HR practices are more informal and may also be more generalistic. They are aimed at providing information, enhancing the workability and health of employees, and improving their person–job fit. These practices are not restricted to older workers, but can also be applied to early- and mid-career workers (De Lange et al., 2015; Moen et al., 2017). Policy initiatives that promote the implementation of such HR practices can help make longer working lives more attainable for everyone. Especially smaller organizations and those with a comparatively low proportion of older workers in their workforce should be encouraged to prepare for longer working lives, to prevent future problems.

Funding

This work was supported by Netspar (IRG2016.03 to J. Oude Mulders), the Netherlands Organisation for Scientific Research (VICI 453-14-001 to K. Henkens), and Instituut GAK.

Conflict of Interest

None declared.