Content uploaded by Hendrik P. Van Dalen
Author content
All content in this area was uploaded by Hendrik P. Van Dalen on Nov 19, 2020
Content may be subject to copyright.
Content uploaded by Hendrik P. Van Dalen
Author content
All content in this area was uploaded by Hendrik P. Van Dalen on May 14, 2016
Content may be subject to copyright.
© The Author 2014. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com.
814
The Gerontologist
cite as: Gerontologist, 2015, Vol. 55, No. 5, 814–824
doi:10.1093/geront/gnu048
Advance Access publication June 4, 2014
Research Article
Recharging or Retiring Older Workers?
Uncovering the Age-Based Strategies of
European Employers
Hendrik P.van Dalen, PhD,*,1,2 KèneHenkens, PhD,1,3,4 and MoWang, PhD5
1Netherlands Interdisciplinary Demographic Institute (NIDI), The Hague. 2Tilburg University, Tilburg
School of Economics and Management (TISEM) and CentER, The Netherlands. 3University Medical
Center Groningen, University of Groningen, The Netherlands. 4Department of Sociology and Anthropology,
University of Amsterdam, The Netherlands. 5Department of Management, Warrington College of Business
Administration, University of Florida, Gainesville.
*Address correspondence to Hendrik P.van Dalen, PhD, Netherlands Interdisciplinary Demographic Institute (NIDI), PO Box
11650, NL-2502 AR, The Hague, The Netherlands. E-mail: dalen@nidi.nl
Received January 17, 2014; Accepted April 8, 2014
Decision Editor: Nicholas G.Castle, PhD
Purpose of the Study: We offer an empirically based taxonomy of the human resource
policies of European employers in relation to older workers. In particular, 3 age-based
strategies are discussed and analyzed in a simultaneous fashion: a focus on exit through
retirement, workplace accommodation measures, and employee development measures.
Design and Methods: A sample of 3,638 organizations in 6 European countries (Denmark,
Germany, Italy, The Netherlands, Poland, and Sweden) is analyzed to discover which of the
7 antecedents investigated herein are associated with the implementation of these human
resource strategies. The key predictors are the proportion of older workers (aged 50 or
older), organization size, seniority-based compensation, labor union involvement, training
requirement, recruitment problems, and knowledge intensity. Structural equation mod-
eling is used to assess whether these predictors are associated with the 3 latent factors.
Results: The 7 key predictors of the 3 strategies show that these strategies are used
simultaneously, but that the employers clearly use exit policies more intensively than
they use development measures. Organizations thus use a dual approach to managing
the employment of older workers. They may sort older workers either upwards (e.g., by
encouraging career development and training) or outwards (by promoting early retire-
ment). The same division can be detected when examining the effects of labor union
involvement and seniority-based wages. When recruitment problems are encountered,
more effort is directed toward accommodation and investment.
Implications: Despite the warnings of policymakers about the possible consequences of
an aging population, European employers are not yet formulating strategies that promote
active aging, often still opting for the easy way out, via exit strategies.
Key Words: Early retirement, Human resource policy, Employers, Older workers
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
In response to recent pension and social security reforms,
the age at which people retire from the labor force is gradu-
ally increasing (D’Addio, Keese, & Whitehouse, 2010). The
Great Recession may add more pressure on workers to
extend their working career (Szinovacz, Martin, & Davey,
2014), although at the same time unemployment may ham-
per realizing those intentions (Cahill, Giandrea, & Quinn,
2013). Effective management of an aging workforce, with
a particular focus on older workers, is therefore a key chal-
lenge (Walker, 2005). Understanding the position of older
workers in an aging labor force demands an exploration
of how employers behave. In particular, researchers must
identify which human resource policies are being designed
and applied to older workers.
Over the past three decades, rms have become aware of
the need to adapt to an aging workforce (Clark & Ogawa,
1996; Taylor, 2002), and certain practices are now speci-
cally targeted at their older workers. However, the literature
on employers’ policies and behavior toward older workers
is often fragmented, viewing each policy separately and not
interrelated. Some studies of older workers focus on employ-
ers’ policies with regard to training (Zwick, 2011), phased
retirement (Hutchens & Grace-Martin, 2006; Johnson,
2011; Oude Mulders, Henkens, & Schippers, 2013), early
retirement (Munnell & Sass, 2008; Vickerstaff, Cox, &
Keen, 2003), and workplace accommodation (McMullin &
Shuey, 2006), but few authors have thus far examined such
policies in an interrelated and comprehensive way. As such,
it is unclear from previous ndings whether emerging poli-
cies are isolated initiatives or tied together with personnel
policies.
Furthermore, previous studies of age-based policies are
mostly descriptive in nature, enumerating the wide vari-
ety of policies that deal with older workers but not going
onto uncover the drivers of these policies. Moreover, stud-
ies that have comprehensively focused on age-based poli-
cies often rely on qualitative case studies (Frerichs, Lindley,
Aleksandrowicz, Baldauf, & Galloway, 2012; Fuertes,
Egdell, & McQuaid, 2013; Timmons, Hall, Fesko, &
Migliore, 2011) or on focus groups (Loretto & White,
2006). While these authors describe a variety of approaches
to age-based management, their ndings cannot easily be
generalized to other organizations, particularly when the
cases have been selected as so-called “best practices.” Finally,
studies that use employers’ surveys that cover a number of
policy measures are often country specic (Midtsundstad,
2011; Taylor & Walker, 1994), which limits insights into
the degree to which employers’ policies are tied to domestic
institutions and the prevailing socioeconomic contexts.
The present paper contributes to the body of knowledge
on this topic in three ways. First, this study reviews prac-
tices targeted at older workers and provides a three-factor
conceptual taxonomy—measures aimed at accommoda-
tion, development, and exit routes—to organize and inter-
pret the ndings presented in the literature. In theory, each
of these specic strategies has its own internal logic, but
just as production can be characterized as the combina-
tion of various inputs, so a human resource practice can be
seen as the clustering of individual practices (Ichniowski,
Shaw, & Crandall, 1995). To offer an encompassing and
theory-driven view, we use structural equation modeling
(SEM) to analyze organizational policies targeted at older
workers. The use of SEM enables us to examine rms’ age-
based strategies as latent variables that account for various
age-related human resource practices. As such, we clearly
capture the interrelatedness of age-related human resource
practices following our conceptual taxonomy, rather than
arbitrarily treating them as empirically derived composites.
The second contribution of this study is an examination
of the various antecedents or predictors of practices related
to older workers. Previous studies have generally identied
a number of practices with referral to older workers, but
few have classied these practices. As such, we know little
about which factors inuence organizational decision mak-
ing in this regard.
Third, this study broadens empirical research on older
workers by presenting the rst comprehensive study of
European employers’ age-based policies for older workers.
By studying older worker-related policies in a representative
sample of organizations in six European countries (France,
Germany, Sweden, Poland, The Netherlands, and Italy), our
database offers a variety of work and welfare state cultures.
To our knowledge, no employer-related studies have pro-
vided a similar examination on such a largescale.
In summary, to address the foregoing gaps in our under-
standing, we offer an empirically based taxonomy of the
policies of European employers in conjunction with an anal-
ysis of the various predictors of practices related to older
workers. The analysis is based on archival data derived
from an EU project (Conen, van Dalen, & Henkens, 2012)
the aim of which was to collect employer-based data on
practices with respect to older workers. The pooling of these
diverse experiences provides more robust and clearer per-
spectives on how aging and labor market institutions affect
employers’ behavior.
Older Worker-Related Policies
The lens through which to view rm policies targeted at
older workers is quite diverse. Different disciplines focus on
different aspects of the challenges that older employees face
in the workplace. Gerontological, demographic, and occu-
pational health studies primarily examine the occupational
needs and capacities of older workers. With this view, human
The Gerontologist, 2015, Vol. 55, No. 5 815
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
resource strategies aim to accommodate these needs as far
as possible (for an overview, see Crawford, LePine, and Rich
(2010) and to alleviate the stress that may arise as a result
of the skills gap between job demands and the capabilities
of the employee. Job strain has been conceptualized as a
health risk by some authors (Bakker & Demerouti, 2007;
Karasek, 1979; Siegrist, 1996). Economic studies stress the
importance of long-run investment in human capital in order
to develop and maintain individual capabilities and maxi-
mize productivity (Skirbekk, 2008) and resolving incentive
problems and keeping workers motivated over the lifetime
of an employee (Lazear, 1979). In contrast to the previous
perspective, this stream of the literature explicitly focuses on
the divergence of productivity and wages over the life of the
worker as means of aligning the interests of the employee
and the rm; a central element which was neglected by the
noneconomic disciplines. Upward sloping age earning pro-
les implicitly increase the cost of shirking and act thereby as
a motivator. Although there are clear signs that the implicit
contract is being eroded with respect to employment pro-
tection (Hallock, 2009; Kahn, 2010), it is still is relevant in
understanding career incentive structures and pension issues
(Lazear, 2000). Economic studies also take account of the fact
that mismatches between rms and workers occur over time
as labor market institutions change (e.g., when the retirement
age increases) or when worker productivity declines unex-
pectedly. Based on the foregoing, it is clear that organizations
implement a number of policies specically to recharge or
dismiss older workers. In the remainder of this section, we
review the following three main strategies in this regard:
accommodation practices, development practices, and offer-
ing an exit route to retirement.
Accommodation Practices
Accommodation practices comprise workplace measures
that compensate for the possible fall in the physical and
cognitive capacities that accompany the process of aging.
Meta-studies or reviews of aging show that the relationship
between age and productivity in the workplace is diverse (cf.
Posthuma and Campion, 2009). For instance, OECD (2006)
concluded that any decline in the physical and mental abili-
ties of workers aged 50 or older is gradual and that there is
substantial variation among individuals; indeed, other func-
tions may remain unchanged or even improve. For example,
Crawford and colleagues (2010) found that physical and
psychological deterioration can be moderated by increases
in physical activity, intellectual activity, and other lifestyle
factors. In short, physical and cognitive changes associated
with aging are modiable (Koopman-Boyden & Macdonald,
2003; Prenda & Stahl, 2001; Schaie, 1996; Schalk et al.,
2010; Skirbekk, 2004), and the impact of these changes on
productivity and performance is primarily dependent on the
work environment and on how job-related tasks are organ-
ized (Bloom & van Reenen, 2011). Measures used to com-
pensate for aging generally include reductions to working
hours, decreases in workload, and preventing older workers
from working overtime or irregular shifts.
Development Practices
Development practices are those measures that aim to
increase the productive capacity of older workers. The pro-
cess of investing in human capital is becoming more formal-
ized within human resource practices by means of annual
reviews and career development planning. In particular,
strategies for older workers merit special research atten-
tion thanks to a number of related factors, including prob-
lems with recruitment, unexpected technological advances
that destroy accumulated knowledge (Daveri & Maliranta,
2007), and low investment in human capital.
In the context of career development planning, the pro-
vision of specic training measures is often mentioned as
a rm policy that can stimulate the productivity of older
employees. In knowledge-intensive sectors in particular,
constantly updating workers’ knowledge of the latest tech-
nologies is vital (Göbel & Zwick, 2012). The promotion of
internal job mobility is another career development option
that not only allows the rm to “groom” its own talent but
also facilitates the allocation of workers to positions that
make the most of their individual capabilities. Encouraging
job mobility within the rm is more likely in large organiza-
tions, which can mimic an internal labor market (Doeringer
& Piore, 1971).
Offering Exit Options
Finally, early retirement measures enable older workers to
retire from the labor force either fully or partially, by taking
up some form of bridge employment. Dismissing or retir-
ing older workers suggests that the benet of keeping them
engaged in the organization no longer offsets the costs of
employing them for the remainder of their careers, which
ofcially end at the mandatory retirement age (65 in most
countries), or on the date at which an occupational pension
starts to payout.
The use of a wage structure based on seniority, in con-
junction with a mandatory retirement age, is one reason
for adopting a proactive approach toward dismissal and
early retirement, because these types of seniority-based
contracts come under increasing pressure where the
workforce in general is aging (Lazear, 1990). Contracts
which use wage structures based on seniority imply a
redistribution of income from younger to older workers.
The Gerontologist, 2015, Vol. 55, No. 5816
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
An aging workforce tends to disrupt the balance in this
regard, making it difcult for organizations to uphold
their promise or implicit contract. Another reason why
managers prefer using early retirement as a policy instru-
ment is that they suppose that more recent technologies,
to which younger workers more easily adapt, are only
learned and used by older workers at a prohibitive cost.
Offering an exit route through early retirement has for
many years therefore been the easy and well-accepted
route. Only recently have institutional changes made dis-
missing older workers a more difcult option for employ-
ers, because exit routes are now becoming increasingly
restrictive (OECD, 2011).
Predictors of Older Worker-Related Policies
In this section, we focus on three categories of antecedents
that are often viewed in the literature as determining the
makeup of older worker-related policies: organization char-
acteristics, job requirements within the organization, and
management restrictions in dealing with older workers.
Organization Characteristics
In terms of organizational characteristics, we focus on the
effects of the number of older employees in the establish-
ment and rm size, in addition to the industry sector in which
organizations operate. First, having a higher percentage of
older workers increases the likelihood that the employer has
rst-hand experience of the challenges faced by older work-
ers. Second, employers accustomed to a relatively old work-
force are likely to be more active in accommodating their
needs or investing in improving their capabilities, but at the
same time employers may also consider for specic employ-
ees that investing in the human capital of older workers or
accommodating their needs may not be worth the costs and
early retirement may the preferable action totake
The second organizational characteristic that deserves
attention is rm size, as measured by the number of
employees. Like most organizational domains, personnel
policies are expected to be subject to economies of scale.
For example, while older worker-related policies in small
organizations are bound by government regulations on
implementing certain measures that facilitate the inter-
ests of workers, they are often still of an informal nature.
However, the exibility—in terms of time and nancial
funds—to accommodate workers or offer opportunities for
training and educational programs is expected to be lower
among small businesses than among larger rms. The same
may apply to offering exit options within small organiza-
tions because exit of older workers may imply an abrupt
loss of valuable rm-specic knowledge.
Job Requirements
Technological change is one of the most important drivers
of productivity growth (Bosworth & Collins, 2003; Easterly
& Levine, 2001). Organizations that operate in dynamic
environments, typically characterized by those in which
the state of knowledge changes at a continuous and rapid
rate, face the challenge of keeping their workers up-to-date
and innovative. Knowledge can be divided into two types:
rm-specic and general knowledge. Firm- or sector-specic
knowledge refers to the type of knowledge necessary to
function in the rm as well as in the industry sector. General
knowledge (e.g., ICT know-how), however, transcends the
rm or sector in question and can in principle easily be
transferred to another organization.
Both rm-specic and general knowledge can be acquired
by employees through extensive training. When training
requirements are high, organizations may be more likely to
invest actively in, and accommodate, older workers, who
have accumulated signicant amounts of knowledge and
are hard to replace within short timeframes or when train-
ing resources are limited. Similarly, when knowledge inten-
sity in an organization is high, they may be more likely than
not to invest in their older employees and provide facili-
ties to help maintain the organization-specic knowledge
embedded in the most experienced staff members. Forcing
the retirement of older employees serves as a measure of
last resort.
Employer’s Restrictions
The ability of managers to deal with older workers is often
restricted by institutional characteristics such as the strength
and inuence of labor unions and the prevailing structure
of incentives, but also the restrictions which become bind-
ing when managers encounter the problems of a tight labor
market. In particular, employers in western European coun-
tries typically encounter the seniority wage system when
tackling the consequences of an aging labor force (D’Addio
etal., 2010). The responses of employers to workforce aging
can run in two directions. On the one hand, they may per-
ceive age-related productivity and the rm’s wage structures
as entrenched and believe that the only way to sustain the
organization is by offering exit routes to older workers. On
the other hand, the alternative response is to invest in build-
ing the knowledge and thereby the productivity of older
workers in order to ensure that the wage–productivity gap
does not widen as the workforce ages. We hypothesize that
because accommodation measures, like extra paid leave or
ergonomic measures, are likely to increase the wage costs of
older workers, employers facing steep age–wage proles are
less tempted to offer those measures than those facing at
age–wage proles.
The Gerontologist, 2015, Vol. 55, No. 5 817
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
In addition, labor union involvement is expected to
inuence personnel policies by stimulating management to
develop formal systems and procedures that aim to improve
efciency and organizational effectiveness. For instance, the
review by Verma (2005) of the various channels through
which unions affect the workplace concluded that this effect
can be seen most clearly in the higher incidence of formal
training programs in unionized organizations. Previous
studies, however, also suggest that unions are tempted to
opt for exit routes in the face of downsizing operations or
mass layoffs (Kalwij, Kapteyn, & De Vos, 2010).
Finally, the state of the business cycle within a sector is
known to affect personnel decisions as pool of potential
workers is larger when it is a buyer’s market than when it is
a seller’s market (Karpinska, Henkens, & Schippers, 2013).
In case employers encounter problems in recruiting new
personnel, they may be more likely to back away from retir-
ing older employees and shift their attention toward accom-
modation and investment measures.
Methods
Data
Data on employers’ policies and behaviors were col-
lected between March and November 2009. The coun-
tries included in this study were geographically dispersed
throughout Europe and represented different types of
European welfare states. Sweden and Denmark represent
Esping-Andersen’s (1990) social-democratic welfare state,
The Netherlands, Germany, and France stand for the con-
tinental welfare state. Italy represents the Mediterranean
type of welfare state. Finally, Poland is a representative of
the “new” EU-member States from Eastern Europe. In each
country, a stratied sample was drawn according to size and
sector covering establishments with 10 employees or more
in all sectors of the economy, excluding the agricultural sec-
tor. Within countries, samples were randomly selected with
respect to geographic area. The questionnaire used in the
different countries was identical. However, the interview
techniques differed across countries, depending on what
was perceived to be the best mode to address respondents
in a particular country. Denmark used computer-assisted
web interviewing (CAWI); Germany, The Netherlands, and
Sweden used paper and pencil interviewing (PAPI); and
Italy and Poland used computer-assisted telephonic inter-
viewing (CATI). The surveys were addressed to directors,
owners, and heads of HR departments. The response rates
of the survey for the respective sample countries were 11%
(Germany), 17% (Italy), 23% (the Netherlands), 23%
(Poland), 28% (Denmark), and 53% (Sweden). These rates
were lower than the average response rates for individual
surveys but in line with those generally found in corporate
surveys. In Europe and the United States, for instance,
response rates are generally found to be 20%–30% at most
(Baruch & Holtom, 2008; Brewster & Hegewisch, 1994;
Kalleberg, Knoke, Marsden, & Speath, 1996; Van Dalen,
Henkens, & Schippers, 2009). The total sample covered
3,780 observations of employers. The survey was com-
pleted by directors (32%), general managers (28%), human
resource managers (34%), and other employees(6%).
Since we created a stratied sample of the characteristics
of the sectors and sizes of the investigated business units, we
carried out multivariate analyses using weights accounting
for the sampling design. Weights are constructed according to
the population of establishments from the national statistical
bureaus. We pooled the data for the six countries and con-
structed a weighting factor that takes the net sample size of
the different countries into account. This was done to prevent
large samples from inuencing the results more than countries
with smaller sample sizes. All countries were given an equal
weight. We did not weight the data according to population
size of the country, because in that case results for Germany—
the largest country—would dominate the ndings.
Measures
Dependent Variables
As discussed earlier, organizations’ age-based policies con-
sist of three measures aimed at early retirement, develop-
ment, and accommodation. These latent variables are based
in turn on a number of rm strategies. Exit route measures
are based on two items, namely “Part-time retirement” and
“Early retirement schemes.” Development measures con-
sist of the following age-based policy measures: “Training
plans for older workers,” “Promoting internal job mobil-
ity,” and “Continuous career development.” Finally,
accommodation measures comprise “Reduction of work-
ing time,” “Decreasing workload,” “Ergonomic measures,”
and “Age limit for irregular work or shifts.”
Independent Variables
The inuence of the following six antecedents are central to
this study (the exact wording of the questions is presented
in Table1):
•Proportion of older workers (aged 50years and older)
•Organization size
•Importance of seniority-based compensation
•The perceived inuence of unions over human resource
policies
•The knowledge intensity of the organization as approxi-
mated by (a) the need for training and (b) the perceived
level of knowledge intensity
•Recruitment problems.
The Gerontologist, 2015, Vol. 55, No. 5818
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
Further, in order to control for the inuences of industry sec-
tor and the country in which the rm is located, we added a
set of dummy variables.
Results
Descriptive Statistics
Table1 shows the usage of various employer strategies. On
average, approximately one-third of the sampled employers
have applied some form of policy within each of the three
studied strategies: 31% offer early retirement schemes, 32%
continuous career development, and 33% ergonomic meas-
ures. Nonetheless, barely 1% of European employers use
all three strategies to the full, whereas 20% use at least one
of the measures of each of the three age-based strategies. In
short, this nding signies that age-based policies directed
toward “active aging” are the exception in the organizations
concerned.
Table1. Descriptive Statistics (N=3,638)
Mean SD Wording
Exit measures “Are the following measures regarding older workers
currently implemented in your establishment?” (0=no;
1=yes)
Part-time retirement 0.26 0.44
Early retirement schemes 0.31 0.46
Development measures
Training programs for older workers 0.20 0.40
Promoting internal job mobility 0.28 0.45
Continuous career development 0.32 0.47
Accommodation measures
Reducing working time before retirement 0.24 0.43
Decreasing workload for older workers 0.20 0.40
Ergonomic measures 0.34 0.47
Age limit for irregular work/shift work 0.11 0.31
Predictor variables
Proportion of older workers 24.65 17.26 “What percentage of employees is 50years or older?”
Organization size 544.17 8102.17 “Approximately how many people are currently employed
in this establishment?”
Importance of seniority-based compensation 2.55 0.81 “To what degree do wages rise with tenure? (i.e., number
of years employees have worked in your establishment)”
(1=not at all; 2=fairly low extent; 3=some extent;
4=high extent
Labor union involvement 2.71 1.28 “The inuence of labor unions on personnel policies is
clearly visible in this establishment” 1=completely
disagree to 5=completely agree
Training requirement 3.50 1.19 “Working in our establishment requires regular additional
training” 1=completely disagree to 5=completely agree
Knowledge intensity 3.55 1.10 “The knowledge intensity in our establishment is high”
1=completely disagree to 5=completely agree
Recruitment problems 1.52 0.61 “Has your establishment experienced recruitment problems
in the last two years?” 1=no, generally not; 2=Yes, with
some vacancies; 3=Yes, with relatively many vacancies
Control variables
Industry sector “Within which of the following industry sectors does your
establishment operate?” Public sector 0.14 0.34
Industry 0.36 0.48
Service 0.30 0.46
Health and education 0.20 0.40
Sector information missing 0.01 0.11
Country
Netherlands 0.22 0.41
Italy 0.18 0.39
Denmark 0.13 0.34
Sweden 0.09 0.28
Poland 0.20 0.40
Germany 0.18 0.39
The Gerontologist, 2015, Vol. 55, No. 5 819
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
Age-Based Policies
As a rst step in the analysis, we performed a conrma-
tory factor analysis with categorical indicators in order to
examine the construct validity of the three age-based poli-
cies. Athree-factor model was tested by loading items on
their respective latent variables. The results showed that
the items were all signicantly loaded on their respective
latent factors (standardized factor loadings ranged from
0.66 to 0.85 and were all statistically signicant). The
information criteria of the three-factor model were also
obtained (Akaike Information Criterion [AIC]=23958.91,
Bayesian Information Criterion [BIC]=24089.10, sample-
sized adjusted BIC=24022.37). An alternative one-factor
model was also specied by loading all items on the same
latent factor (AIC=24311.20, BIC=24422.79, sample-size
adjusted BIC=24365.59). Since all three information crite-
ria of the three-factor model are smaller than those of the
one-factor model, the former model shows the better t and
therefore this was used for further analysis.
Predictors of Age-Based HR Strategies
Next, we specied a structural equation model by regress-
ing the set of control variables (i.e., country and industry
sector) in Step 1 and the key predictors (i.e., proportion of
older workers, organization size, seniority-based compensa-
tion, labor union involvement, training requirement, knowl-
edge intensity, and recruitment problems) in Step 2 against
the three latent factors obtained in the conrmatory factor
analysis model above. Table2 presents the coefcients of the
estimated models of Steps 1 and 2.The rst step reveals that
sector and country-specic factors are important, for exam-
ple, public sector organizations were more age conscious
than private sector rms. And between countries, it becomes
clear that Italian employers are across the board less set on
dealing with an aging work force than other countries. But
the main focus of this paper is on the key predictors which
apply to organizations in general. We show that the key pre-
dictors included in the model explain 6%, 5%, and 7% of
the additional variances beyond the control variables in the
organization’s use of early retirement schemes, accommoda-
tion strategies, and development strategies, respectively.
In particular, the proportion of older workers was posi-
tively correlated with the use of exit schemes (γE=2.73),
development strategies (γD = 1.86), and accommodation
strategies (γA= 0.79), indicating that organizations with a
high percentage of older workers offered all three of the
strategies discussed more often than those with a low per-
centage of older workers. Organization size, however, was
not signicantly correlated with any of these three strate-
gies. Moreover, the use of seniority-based compensation
was only positively related to accommodation strategies
(γA=0.20). Labor unions involvement was positively related
to both early retirement schemes (γE = 0.22) and accom-
modation strategies (γE=0.19). Training requirement was
related to all three types of strategies (γE=0.23, γA=0.17,
γD=0.35). Knowledge intensity was only positively related
to development strategies (γD =0.34). As a nal observa-
tion, recruitment problems were positively related to both
accommodation strategies (γA = 0.33) and development
strategies (γD=0.23).
Table 3 compares the predictive effects of each key
antecedent across the three dependent variables. Predictive
effects were examined by comparing the likelihood ratio-
based χ2s of the free model (i.e., the model estimated with
both covariates and predictors as in Step 2, Table 2) with
those of a constrained model in which the predictive effects
of a particular antecedent on two dependent variables were
constrained to be equal. Signicant Δχ2 means that the
constrained model tted the data signicantly worse than
the free model, indicating that the predictive effects of the
particular antecedent on the two dependent variables were
signicantly different.
The results shown in Table3 reveal that the proportion
of older workers was a better predictor of the organiza-
tion’s use of exit schemes compared with accommodation
strategies and development strategies as well as of its use
of accommodation strategies compared with development
strategies. In addition to the degree of workforce aging,
the other predictive variables also shed light on the foci
of employers’ strategies. Labor union’s involvement was a
better predictor of organization’s use of early retirement
schemes and accommodation strategies than of develop-
ment strategies. Finally, there was no difference in the pre-
dictive effect of recruitment problems for all three age-based
strategies.
Discussion and Conclusion
We have herein offered an empirically based taxonomy of
the age-based strategies of European employers in relation
to older workers. Three strategies were discerned: the use of
accommodation measures, development measures and exit
through early retirement. We found that the higher the pro-
portion of older workers in an organization, the more likely
it was to use the three investigated strategies for older work-
ers. Further, the presented results showed that organizations
in which the compensation is strongly related to seniority,
where the inuence of labor unions is highly visible in per-
sonnel measures, and where training requirements are high
are more likely to adopt these three strategies than those that
do not share these particular characteristics.
Another important nding is that the key predictors
examined in the present study inuence the three age-based
The Gerontologist, 2015, Vol. 55, No. 5820
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
strategies to differing degrees. For instance, having a higher
percentage of older workers is associated with a higher
likelihood of formulating development policies for this
age group but is much more strongly related to the use of
exit measures. Meanwhile, the impact of workforce aging
on accommodation measures takes an intermediate posi-
tion. These results suggest that workforce aging is a cause
of highly divergent HR policy responses. Specically, this
nding suggests that organizations use a dual approach
toward managing the employment of older workers. In
simple terms, enterprises may sort older workers either
upwards (e.g., encourage career development and training)
or outwards (promote early retirement packages); however,
given a straightforward choice between these two alterna-
tives, employers have a clear preference for offering older
workers exit routes.
The same division can be detected when we examine
the effects of labor union involvement and seniority-based
wages. Although we nd that labor union involvement
increases the likelihood of the provision of development
programs for older workers, it has an even stronger asso-
ciation with the use of exit and accommodation policies.
The same nding holds for seniority-based wage systems,
namely that exit strategies and accommodation measures
Table2. Coefficients of the Estimated Structural Equation Model
Predictor Exit strategies Accommodation strategies Development strategies
Coefcient βSE Coefcient βSE Coefcient βSE
Step 1: only control variables
Sector (public sector=ref.)
Industry −0.69** −.13** 0.24 −0.91** −.16** 0.26 −0.97** −.23** 0.23
Service −1.38** −.29** 0.29 −1.28** −.26** 0.30 −1.11** −.29** 0.24
Health and education −0.94** −.15** 0.25 −0.58* −.09* 0.27 −0.45 −.09 0.23
Sector information missing 0.22 .01 0.63 −0.85 −.05 0.66 −1.42** −.10** 0.55
Country (Germany=ref.)
Italy −1.51** −.26** 0.29 −2.02** −.33** 0.39 −1.33** −.28** 0.31
Denmark −0.89** −.16** 0.28 1.46** .24** 0.33 −0.07 −.02 0.24
Sweden 0.27 .04 0.33 1.33** .19** 0.32 0.41 .08 0.27
Poland 1.74** .22** 0.31 0.65** .08** 0.25 0.89** .14** 0.25
Netherlands 1.75** .29** 0.32 1.45** .23** 0.27 0.05 .01 0.22
R2.29 .29 .14
Step 2
Key predictors
Proportion of older workers 2.73** .22** 0.46 1.86** .14** 0.54 0.79* .07* 0.40
Organization size 0.01 .02 0.01 0.01 .02 0.01 0.02 .04 0.01
Seniority-based compensation −0.02 −.01 0.09 0.20* .06* 0.10 0.02 .01 0.08
Labor union’ s involvement 0.22** .12** 0.06 0.19** .10** 0.07 0.00 .00 0.06
Training requirement 0.23** .12** 0.07 0.17* .08* 0.08 0.35** .21** 0.08
Knowledge intensity −0.08 −.04 0.09 0.19 .09 0.10 0.34** .18** 0.09
Recruitment problems 0.19 .05 0.11 0.33** .08** 0.12 0.23* .07* 0.11
Control variables
Sector (Public sector=ref.)
Industry −0.32 −.06 0.24 −0.52* −.09* 0.25 −0.67** −.15** 0.23
Service −0.97** −.21** 0.28 −0.89** −.15** 0.27 −0.95** −.24** 0.25
Health and education −0.88** −.14** 0.25 −0.55* −.08* 0.27 −0.48 −.09 0.25
Sector information missing 0.50 .03 0.62 −0.60 −.03 0.61 −1.20* −.08* 0.55
Country (Germany=ref.)
Italy −1.17** −.20** 0.31 −1.34** −.22** 0.38 −0.07 −.01 0.27
Denmark −0.82** −.14** 0.27 1.41** .24** 0.33 0.17 .04 0.26
Sweden 0.09 .01 0.32 1.20** .17** 0.35 0.55 .10 0.29
Poland 1.75** .23** 0.30 0.64* .08* 0.25 1.00** .15** 0.27
Netherlands 1.92** .32** 0.32 1.48** .23** 0.28 0.31 .06 0.23
R2.35 .34 .21
ΔR2.06 .05 .07
Note: β denotes the standardized coefcients. N = 3,638.
*p < .05, **p < .01.
The Gerontologist, 2015, Vol. 55, No. 5 821
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
are much more likely to be deployed for older workers
than measures aimed at the development of human capital.
Finally, we note that when employers encounter recruit-
ment problems, their effort is directed more toward accom-
modation and investment and less so in the direction of
early exit arrangements. In other words, when the labor
market changes from a state of excess supply to one of
excess demand, this may offer a positive stimulus to the
position of the older worker.
This paper has a number of noteworthy strengths. First,
considering the policy responses of employers in such an all-
encompassing way increases our insights into the divergent
organizational consequences of changing management restric-
tions. Making the seniority-based wage systems less promi-
nent, as suggested by the OECD (2006), for improving the
relative position of older workers in the employee hierarchy,
is indeed likely to result in lower exit rates. However, employ-
ers are also likely to reduce their provision of training and
development programs for older workers when wage proles
atten. Second, the use of a large European sample of employ-
ers covering all industry sectors improves the empirical base of
age-related personnel policy research and shows that substan-
tial policy-related differences between organizations remain
in place. We believe that the results offer insights as to the
strategy choices of employers in continental European wel-
fare states. Alimitation of the current study is that it did not
cover Anglo-Saxon welfare states. It remains an open question
whether the results are generalizable to these types of welfare
states where employment protection is relatively low and sen-
iority wages are less prominent.
Another limitation is the cross-sectional nature of
the data used in this study. It restricts us from analyzing
whether employers are gradually changing their attitudes
toward older workers and thus evolving their policy direc-
tions over time. Current human resources strategies should
be seen as the result of current conditions and causality
issues cannot be uncovered. Furthermore, a number of the
key predictor variables are based on one-item indicators.
Future research could aim to develop independent variables
in order to reduce the bias in the estimation results.
The retirement landscape is undergoing rapid change.
The strong tendency of employers to choose exit strategies
for older workers is a legacy of past organizational strate-
gies which were until recently rmly embedded in the institu-
tional arrangements of social security and pensions (cf. Wise,
2010) and the employer’s mindset (Van Dalen, Henkens, &
Schippers, 2010). In most European countries, a transition is
currently taking place in which the incentive to leave the labor
force at a relatively young age is being curtailed and alterna-
tive exit routes (e.g., disability insurance and early retirement
programs) are gradually disappearing. As a result, employ-
ers are being challenged to nd new avenues to keep older
workers motivated and productive (Damman, Henkens, &
Kalmijn, 2013). The present ndings conrm that employers
have thus far been relatively inactive in this area up to now.
The fact that organizations that have a relatively large per-
centage of aging workers and are more inclined to recharge
them—or at least to consider policies that aim to lengthen
their working lives—is a positive sign, as well as a possible
indication of a future in which older workers and their ben-
ets for organizations can no longer be ignored.
Funding
This article was funded by a grant from NEUJOBS, a research project
nanced by the European Commission, under the 7th Framework
Programme, grant agreement #266833.
References
Bakker, A. B., & Demerouti, E. (2007). The job demands-resources
model: State of the art. Journal of Managerial Psychology, 22,
309–328. doi:10.1108/02683940710733115
Baruch, Y., & Holtom, B. C. (2008). Survey response rate levels and
trends in organizational research. Human Relations, 61, 1139–1160.
doi:10.1177/0018726708094863
Bloom, N., & van Reenen, J. (2011). Human resource management
and productivity. Handbook of Labor Economics, 4, 1697–1767.
doi:10.1016/S0169-7218(11)02417-8
Table3. Model Comparisons for Free Model and
Constrained Models
Model −2 Log-likelihood Δχ2 (1) Conclusion
Free model 22509.87
Predictive effect comparisons
Proportion of older workers
γE vs. γD22537.16 27.30** γE > γD
γE vs. γA22514.89 5.02* γE > γA
γD vs. γA22520.79 10.93** γA > γD
Labor union’s involvement
γE vs. γD22529.41 19.54** γE > γD
γE vs. γA22510.16 0.29 γE = γA
γD vs. γA22529.67 19.80** γA > γD
Training requirement
γE vs. γD22512.85 2.98 γE = γD
γE vs. γA22510.88 1.01 γE = γA
γD vs. γA22517.98 8.11** γD > γA
Recruitment problems
γE vs. γD22510.07 0.20 γE = γD
γE vs. γA22512.01 2.14 γE = γA
γD vs. γA22511.04 1.18 γD = γA
Note: γE=the structural coefcient in predicting exit strategies; γD=the struc-
tural coefcient in predicting development strategies; γA=the structural coef-
cient in predicting accommodation strategies.
*p < .05, **p < .01.
The Gerontologist, 2015, Vol. 55, No. 5822
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
Bosworth, B., & Collins, S. M. (2003). The empirics of growth: An
update. Brookings Papers on Economic Activity, 2003, 113–206.
doi:10.1353/eca.2004.0002
Brewster, C., & Hegewisch, A. (1994). Policy and practice in
European human resource management: The Price Waterhouse
Craneld Survey. London: Routledge.
Cahill, K. E, Giandrea, M. D., & Quinn, J. F. (2013). Retirement
patterns and the macroeconomy, 1992–2010: The preva-
lence and determinants of bridge jobs, phased retirement, and
reentry among three recent cohorts of older Americans. The
Gerontologist. Advance online publication. doi:10.1093/geront/
gnt156
Clark, R. L., & Ogawa, N. (1996). Human resource policies and older
workers in Japan. The Gerontologist, 36, 627–636. doi:10.1353/
eca.2004.0002
Conen, W. S., van Dalen, H. P., & Henkens, K. (2012). Ageing and
employers’ perceptions of labour costs and productivity: Asurvey
among European employers. International Journal of Manpower,
33, 629–647. doi:10.1108/01437721211261796
Crawford, E. R., LePine, J. A., & Rich, B. L. (2010). Linking job
demands and resources to employee engagement and burnout:
Atheoretical extension and meta-analytic test. Journal of Applied
Psychology, 95, 834. doi:10.1037/a0019364
D’Addio, A. C., Keese, M., & Whitehouse, E. (2010). Population age-
ing and labour markets. Oxford Review of Economic Policy, 26,
613–635. doi:10.1093/oxrep/grq035
Damman, M., Henkens, K., & Kalmijn, M. (2013). Late-career work
disengagement: The role of proximity to retirement and career
experiences. Journals of Gerontology, Series B: Psychological
Sciences and Social Sciences. doi:10.1093/geronb/gbt001
Daveri, F., & Maliranta, M. (2007). Age, seniority and labour costs:
Lessons from the Finnish IT revolution. Economic Policy, 22,
117–175. doi:10.1111/j.1468-0327.2007.00175.x
Doeringer, P. B., & Piore, M. J. (1971). Internal labor markets and
manpower analysis. London: ME Sharpe.
Easterly, W., & Levine, R. (2001). What have we learned from a
decade of empirical research on growth? It’s not factor accu-
mulation: Stylized facts and growth models. The World Bank
Economic Review, 15, 177–219. doi:10.1093/wber/15.2.177
Frerichs, F., Lindley, R., Aleksandrowicz, P., Baldauf, B., &
Galloway, S. (2012). Active ageing in organisations: Acase study
approach. International Journal of Manpower, 33, 666–684.
doi:10.1108/01437721211261813
Fuertes, V., Egdell, V., & McQuaid, R. (2013). Extending working
lives: Age management in SMEs. Employee Relations, 35, 272–
293. doi:10.1108/01425451311320477
Göbel, C., & Zwick, T. (2013). Are personnel measures effective in
increasing productivity of old workers? Labour Economics, 22,
80–93. doi:10.1016/j.labeco.2012.11.005
Hallock, K. F. (2009). Job loss and the fraying of the implicit employ-
ment contract. The Journal of Economic Perspectives, 23, 69–93.
Hutchens, R., & Grace-Martin, K. (2006). Employer willingness to
permit phased retirement: Why are some more willing than oth-
ers? Industrial and Labor Relations Review, 59, 525–546.
Ichniowski, C., Shaw, K., & Crandall, R. W. (1995). Old dogs and
new tricks: Determinants of the adoption of productivity-
enhancing work practices. Brookings papers on economic activ-
ity. Microeconomics, 1995, 1–65.
Johnson, R. W. (2011). Phased retirement and workplace exibility
for older adults opportunities and challenges. The ANNALS of
the American Academy of Political and Social Science, 638, 68–
85. doi:10.1177/0002716211413542
Kahn, L. M. (2010). Employment protection reforms, employ-
ment and the incidence of temporary jobs in Europe:
1996–2001. Labour Economics, 17, 1–15. doi:10.1016/j.
labeco.2009.05.001
Kalleberg, A. L., Knoke, D., Marsden, P. V., & Speath, J. L. (1996).
Organizations in America. Analyzing their structures and human
resource practices. London: Sage.
Kalwij, A., Kapteyn, A., & De Vos, K. (2010). Retirement of older
workers and employment of the young. De Economist, 158,
341–359. doi:10.1007/s10645-010-9148-z
Karasek, R. A. Jr. (1979). Job demands, job decision latitude, and
mental strain: Implications for job redesign. Administrative
Science Quarterly, 24, 285–308.
Karpinska, K., Henkens, K., & Schippers, J. (2013). Retention of
older workers: Impact of managers’ age norms and stereotypes.
European Sociological Review, 29, 1323–1335. doi:10.1093/esr/
jct017
Koopman-Boyden, P. G., & Macdonald, L. (2003). Ageing,
work performance and managing ageing academics. Journal
of Higher Education Policy and Management, 25, 29–40.
doi:10.1080/13600800305744
Lazear, E. P. (1979). Why is there mandatory retirement? The Journal
of Political Economy, 87, 1261–1284.
Lazear, E. P. (1990). Adjusting to an aging labor force. In Wise, D. A.
(Ed.) Issues in the economics of aging (pp. 287–316). Chicago,
IL: University of Chicago Press.
Lazear, E. P. (2000). The future of personnel economics. The
Economic Journal, 110, 611–639.
Loretto, W., & White, P. (2006). Employers’ attitudes, practices and
policies towards older workers. Human Resource Management
Journal, 16, 313–330. doi:10.1111/j.1748-8583.2006.00013.x
McMullin, J. A., & Shuey, K. M. (2006). Ageing, disability and
workplace accommodations. Ageing and Society, 26, 831–847.
doi:10.1017/S0144686X06004958
Midtsundstad, T. I. (2011). Inclusive workplaces and older employees:
An analysis of companies’ investment in retaining senior work-
ers. The International Journal of Human Resource Management,
22, 1277–1293. doi:10.1080/09585192.2011.559099
Munnell, A. H., & Sass, S. A. (2008). Working longer: The solution
to the retirement income challenge. Washington, DC: Brookings
Institution Press.
OECD. (2006). Live longer, work longer. Asynthesis report. Paris,
France: Organization for Economic Development.
OECD. (2011). Pensions at a glance 2011: Retirement-income sys-
tems in OECD and G20 countries. Retrieved April 5, 2011, from
www.oecd.org/els/social/pensions/PAG
Oude Mulders, J., Henkens, K., & Schippers, J. (2013). Organizations’
ways of employing early retirees: The role of age-based HR
policies. The Gerontologist. Advance online publication.
doi:10.1093/geront/gnt114
Posthuma, R. A., & Campion, M. A. (2009). Age stereotypes in
the workplace: Common stereotypes, moderators, and future
research directions. Journal of Management, 35, 158–188.
doi:10.1177/0149206308318617
The Gerontologist, 2015, Vol. 55, No. 5 823
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
Prenda, K. M., & Stahl, S. M. (2001). The truth about older workers.
Business and Health, 19, 30.
Schaie, K. W. (1996). Intellectual development in adulthood: The Seattle
longitudinal studies. Cambridge, UK: Cambridge University Press.
Schalk, R., van Veldhoven, M., de Lange, A. H., De Witte, H., Kraus,
K., Stamov-Rossnagel, C., … Bal, M. (2010). Moving European
research on work and ageing forward: Overview and agenda.
European Journal of Work and Organizational Psychology, 19,
76–101. doi:10.1080/13594320802674629
Siegrist, J. (1996). Adverse health effects of high-effort/low-reward
conditions. Journal of Occupational Health Psychology, 1, 27.
doi:10.1037/1076-8998.1.1.27
Skirbekk, V. (2004). Age and individual productivity: Aliterature
survey. In Feichtinger, G. (Ed.) Vienna Yearbook of Population
Research (Vol. 2, pp. 133–153). Vienna: Austrian Academy of
Sciences Press.
Skirbekk, V. (2008). Age and productivity potential: Anew approach
based on ability levels and industry-wide task demand. Population
and Development Review, 34, 191–207.
Szinovacz, M. E., Martin, L., & Davey, A. (2014). Recession and
expected retirement age: Another look at the evidence. The
Gerontologist, 54, 245–257. doi:10.1093/geront/gnt010
Taylor, P. E. (2002). New policies for older workers. Bristol, UK: The
Policy Press.
Taylor, P. E., & Walker, A. (1994). The ageing workforce: Employers’
attitudes towards older people. Work, Employment and Society,
8, 569–591. doi:10.1177/095001709484005
Timmons, J. C., Hall, A. C., Fesko, S. L., & Migliore, A. (2011).
Retaining the older workforce: Social policy considerations for
the universally designed workplace. Journal of Aging and Social
Policy, 23, 119–140. doi:10.1080/08959420.2011.551623
Van Dalen, H. P., Henkens, K., & Schippers, J. (2009). Dealing with
older workers in Europe: Acomparative survey of employers’
attitudes and actions. Journal of European Social Policy, 19, 47–
60. doi:10.1177/0958928708098523
Van Dalen, H. P., Henkens, K., & Schippers, J. (2010). Productivity
of older workers: Perceptions of employers and employees.
Population and Development Review, 36, 309–330. doi:http://
dx.doi.org/10.1111/j.1728-4457.2010.00331.x
Verma, A. (2005). What do unions do to the workplace? Union effects
on management and HRM policies. Journal of Labor Research,
26, 415–449. doi:10.1007/s12122-005-1013-5
Vickerstaff, S., Cox, J., & Keen, L. (2003). Employers and the man-
agement of retirement. Social Policy and Administration, 37,
271–287.
Walker, A. (2005). The emergence of age management in Europe.
International Journal of Organisational Behaviour, 10,
685–697.
Wise, D. A. (2010). Facilitating longer working lives: International
evidence on why and how. Demography, 47(Suppl.), S131–
S149. doi:10.1353/dem.2010.0000
Zwick, T. (2011). Why training older employees is less effective.
ZEW-Centre for European Economic Research Discussion Paper
(11–046).
The Gerontologist, 2015, Vol. 55, No. 5824
Downloaded from https://academic.oup.com/gerontologist/article/55/5/814/2605195 by University of Groningen user on 19 November 2020
