1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023].

See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE
REACTIONS TO TALENT
IDENTIFICATION
I N G M A R B J Ö R K M A N , M AT S E H R N R O O T H , K R I S T I I N A
MÄKELÄ, ADAM SMALE, AND JENNIE SUMELIUS

In this article, we examine the effect of talent identification on employee at-

titudes. Building on social exchange theory, we analyze the association be-
tween employees’ perceptions about whether or not they have been formally
identified as “talent” and the following attitudinal outcomes: commitment to
increasing performance demands, building skills, and supporting strategic
priorities; identification with the unit and the multinational enterprise; and
turnover intentions. Our analyses of 769 managers and professionals in nine
Nordic multinational corporations reveal a number of differences between
employees who perceive that they have been identified as “talent” and those
who either perceive that they have not been identified or do not know whether
they have been identified. We found only limited differences between the two
latter categories.

Keywords: talent, talent management, attitudes, career, selection, multina-

tional enterprises

Introduction most strategic roles (those roles necessary to

nterest in talent management has prolifer- achieve organizational strategic priorities) on a

I ated over the last decade, with the global

shortage of leadership talent being touted
as one of the highest HR concerns for mul-
tinational enterprises (MNEs) today (Cap-
pelli, 2008; Guthridge, Komm, & Lawson,
2008). Consequently, MNEs have directed in-
creasing attention to global talent manage-
global scale” (Scullion, Collings, & Caligiuri,
2010, p. 106). Although approaches vary, tal-
ent management usually focuses on a pool of
employees who rank at the top in terms of per-
formance and competencies, and are therefore
considered leaders or key professionals either at
present or at some point in the future (Collings
ment (Farndale, Scullion, & Sparrow, 2010; & Mellahi, 2009; Lewis & Heckman, 2006). In
McDonnell, Lamare, Gunnigle, & Lavelle, MNEs, talent management decisions are in-
2010; Stahl et al., 2012; Tarique & Schuler, creasingly global in that employees may be
2010), defined as “all organizational activities identified as “talent” or “high potentials” re-
for the purpose of attracting, selecting, devel- gardless of whether they are parent-country
oping, and retaining the best employees in the nationals, expatriates, or local employees

Correspondence to: Ingmar Björkman, School of Business, Aalto University, P.B. 21210, FI-00076 Aalto, Finland,
Phone: +358407040291, E-mail: [email protected].

Human Resource Management, March–April 2013, Vol. 52, No. 2. Pp. 195–214
© 2013 Wiley Periodicals, Inc.
Published online in Wiley Online Library (wileyonlinelibrary.com).
DOI:10.1002/hrm.21525
 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
196 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

working in foreign subsidiaries (Collings, In this article, we seek to address this

Scullion, & Morley, 2007). research gap by analyzing the association
At the core of talent management is the between employees’ perceptions about whether
assumption that the “talent” must be found, or not they have been formally identified as
segmented, nurtured, and placed in pivotal “talent,” and a number of attitudes that have
positions that are crucial for the competitive been associated with positive organizational
advantage of the firm (Boudreau & Ramstad, outcomes in previous research, and which
2005). Some attention has been paid to the thus are central to effective talent manage-
question of how to identify talent, and schol- ment systems. Building on social exchange
ars have also begun to examine factors that theory, we develop a range of hypotheses with
influence whether or not someone is classified regard to how individuals’ perception of their
as talent (Mäkelä, Björkman, & Ehrnrooth, talent status is related to the following attitu-
2010; Mellahi & Collings, 2010). This is dinal outcomes: acceptance of increasing per-
in line with calls for literature on HRM to formance demands, commitment to building
shift emphasis from HR practices to include competencies, support of company strategic
an increased focus on the human capital priorities, identification with the focal unit
that constitutes the underlying and the MNE, and turnover intent. Our analy-
resource of these practices (Wright ses of 769 managers and professionals in nine
At the core of talent & McMahan, 2011). Nordic multinational corporations reveal a
management is the One key issue that many MNEs number of differences between employees
wrestle with is whether or not who perceive that they have been identified
assumption that the to inform high potentials about as talent and those who either perceive that
their status after talent reviews they have not been selected or do not know
“talent” must be have been conducted and corpo- whether they have been identified.
found, segmented,
rate talent pools decided upon
(Evans, Pucik, & Björkman, 2011).
Talent Identification and Employee
nurtured, and placed The differential treatment of such
employees in terms of develop-
Attitudes: A Social Exchange
in pivotal positions mental support or compensation Perspective
can be a sensitive matter. If the sta- Social exchange theory suggests that when
that are crucial for
tus of high potentials is not made corporations invest in their employees, they
the competitive public, this may, for instance, are likely to reciprocate these corporate in-
lead to frustration among high vestments in positive ways (Cropanzano &
advantage of the performers who do not feel ade- Mitchell, 2005), providing a useful lens
quately recognized. On the other through which to understand the mecha-
firm.
hand, if talent pool membership nisms involved in how employees interpret
is publicized, the motivation of and react to organizational talent manage-
those not on the list of talent may drop. While ment practices. Within this strand of research,
the question has been posed of whether or there are a number of different ways of
not to inform individuals about their possible conceptualizing the employee–organization
status as talent, there is little, if any, empirical relationship. An employer perspective, for ex-
research on the extent to which firms do this, ample, features in research on employment
how, and why. More important, there is no modes (March & Simon, 1958), in which it is
research that analyzes this issue in terms of analyzed in terms of the inducements an or-
the effects this may have on the individuals ganization offers and the contributions it ex-
themselves. This is a serious omission since pects from its employees (e.g., Guest &
employee perceptions of talent management Conway, 2002; Tsui, Pearce, Porter, & Tripoli,
practices and decisions are likely to influence 1997).
attitudes that are important for the perfor- The employee view of this organization–
mance of the organization (Boxall & Macky, employee exchange relationship has, in
2009; Wright & Nishii, forthcoming). turn, been extensively studied in relation to

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE REACTIONS TO TALENT IDENTIFICATION 197

psychological contracts (Rousseau, 1995), to be more committed toward issues that are
which focuses on individuals’ perceptions important for their employer than those who
of what the organization offers and what either perceive that they have not been identi-
employees are obliged to contribute in return fied or those who do not know whether they
(Conway & Briner, 2002, 2005; Guest, 2004). have been selected or not. Similarly, based
Previous empirical studies have established on research into the effects of perceived psy-
a link between perceived organizational chological contract breach (or violation) on
inducements and employee obligations (e.g., employee attitudes and behaviors (Morrison &
Rousseau, 1990; Shaw, Dineen, Fang, & Vellella, Robinson, 1997; Zhao, Wayne, Glibkowski, &
2009; Shore & Barksdale, 1998), with individ- Bravo, 2007), we further expect the atti-
uals perceiving that when the organization tudes of those who perceive that they are not
has invested in the employment relationship, included in talent pools to differ
they have an obligation to reciprocate the from those of individuals who
investment (Cole, Schaninger, & Harris, 2002; do not know whether they have Inclusion in a talent
Kuvaas & Dysvik, 2010). A related research been selected. The attitudes of pool is perceived
stream—organizational support theory—has the former will be based on per-
examined employee reactions to their beliefs ceptions of unmet expectations as a signal that the
regarding how the organization values their and negative signals sent by the
contributions and cares about their well- organization concerning its sup- focal individual’s
being (Eisenberger, Fasolo, & Davis-LaMastro, port and future investments in
contribution to the
1990). Employees interpret corporate actions, them. Thus, employees who do
especially HRM decisions (Guzzo & Noonan, not know whether they are iden- organization has
1994; Rousseau & Greller, 1994), as commit- tified as talent are likely to be
ments or support on the part of the organi- more positive toward the organi- been valued and that
zation, which they then reciprocate through zation than those who perceive
the employer has
positive attitudes and behaviors that support they have not been singled out as
the attainment of organizational goals (Coyle- talent. fulfilled its contract
Shapiro & Conway, 2005).
Building on the social-exchange perspec- by deciding to invest
Acceptance of Increasing
tive, we argue that talent identification, which
Performance Demands in his or her future
explicitly assumes differential treatment of
selected employees, is likely to be viewed From the organization’s point of career.
by talent pool members as an indication of view, how employees perform in
their employer’s commitment toward them their job is of obvious importance
(Meyer & Allen, 1991) and as discretionary, and the performance of individuals labeled
future organizational support (Eisenberger as talent is likely to be particularly important
et al., 1990; Eisenberger, Cummings, Armeli, & (Collings & Mellahi, 2009). Individual per-
Lynch, 1997). In other words, inclusion in a formance is to a significant extent an out-
talent pool is perceived as a signal that the come of their knowledge and skills and their
focal individual’s contribution to the organi- personal efforts. Related to the latter, in ac-
zation has been valued and that the employer cordance with social exchange theory, we ex-
has fulfilled its contract by deciding to invest pect individuals who perceive that they have
in his or her future career. This, in turn, is been identified as talent to be more likely to
likely to lead to an internalized, normative accept increasing demands to do well in their
obligation to act in a way that meets organi- jobs. This expectation is in line with the pos-
zational goals and interests (Allen & Meyer, itive relationship that was found in a meta-
1990; Lee, Liu, Rousseau, Hui, & Chen, 2011; analysis of the relationship between perceived
Wiener, 1982). organizational support and employee perfor-
We therefore expect individuals who mance (Riggle, Edmondson, & Hansen, 2009).
believe that they know or, in other words, per- Therefore, and although there is little existing
ceive that they have been identified as talent research that specifically examines employee

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
198 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

acceptance of increasing performance de- mitted to building competencies than are those
mands, the following hypotheses will be who perceive that they are not identified as talent.
tested in this study:
Hypothesis 2b: Individuals who perceive that they
Hypothesis 1a: Individuals who perceive that they are identified as talent are more likely to be com-
are identified as talent are more likely to accept mitted to building competencies than are those who
increasing performance demands than those who do not know whether they are identified as talent.
perceive that they are not identified as talent.
Hypothesis 2c: Individuals who perceive that they
Hypothesis 1b: Individuals who perceive that they are not identified as talent are less likely to be com-
are identified as talent are more likely to accept mitted to building competencies than those who
increasing performance demands than those who do not know whether they are identified as talent.
do not know whether they are identified as talent.
Support of Strategic Priorities
Hypothesis 1c: Individuals who perceive that they
are not identified as talent are less likely to accept Again following social exchange logic, we ex-
increasing performance demands than those who pect employees to differ in the extent to
do not know whether or not they are identified as which they support the strategic objectives of
talent. the corporation based on whether they per-
ceive themselves to be identified as talent.
Commitment to Building Competencies Van Riel, Berens, and Dijkstra (2009) found
that organizations can support strategically
A central tenet of talent management is that aligned behaviors in workgroups by stimulat-
corporations need to meet their future need ing employee motivation, informing employ-
for human capital in particular competencies ees, and encouraging capability development.
needed in positions that are important for the Similarly, talent management activities, such
competitiveness of the organization (Collings as the inclusion of a person in a corporate tal-
& Mellahi, 2009). One central strategy for ent pool, that shape employee interpretations
meeting these needs is employee development. of the terms of their psychological contract
Applying social exchange theory, individuals (Guzzo & Noonan, 1994; Rousseau & Greller,
who perceive that they are part of talent pools 1994) and signal the kinds of behaviors that
can be expected to reciprocate by building are desired and rewarded within the
company-specific competencies by seeking organization (Bowen & Ostroff, 2004) create
out developmental job experiences (De Pater, stimuli that increase individual motivation to
Van Vianen, Bechtoldt, & Klehe, 2009; Mc- actively support the strategic priorities of the
Call, Lombardo, & Morrison, 1988), feedback employer. We expect employees who perceive
from their colleagues, mentors, and others they have received talent status to draw more
whose knowledge and insights they may favorable interpretations about their psycho-
learn from, as well as striving to develop their logical contract and be more likely to support
knowledge and skills in other ways. This such priorities. Thus, we propose the follow-
tendency may be further strengthened by ef- ing hypotheses:
forts on the part of talented individuals to
maintain current perceptions about their Hypothesis 3a: Individuals who perceive that they
knowledge and skills (Firfiray, 2009), with are identified as talent are more likely to actively
management and leadership competence support the strategic priorities of the firm than are
development being one of the ways in which those who perceive that they are not identified as
they can achieve this. Therefore, we hypoth- talent.
esize the following:
Hypothesis 3b: Individuals who perceive that they
Hypothesis 2a: Individuals who perceive that they are identified as talent are more likely to actively
are identified as talent are more likely to be com- support the strategic priorities of the firm than are

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE REACTIONS TO TALENT IDENTIFICATION 199

those who do not know whether they are identified lead to views of an imbalance in the social
as talent. exchange process, which in turn may induce
individuals to level the playing field by reduc-
Hypothesis 3c: Individuals who perceive that they ing their attachment to the organization (e.g.,
are not identified as talent are less likely to active- Turnley & Feldman, 1998).
ly support the strategic priorities of the firm than In MNEs and other large organizations,
those who do not know whether they are identified however, employees may identify with more
as talent. than one organizational entity. For instance,
a manager may identify with both the cor-
poration as a whole and/or the focal unit in
Organizational Identification
where he or she is working. This has been
Further, we expect talent identification to shown to be true for managerial employees in
have an impact on how employees identify subsidiaries (e.g., Reade, 2001a) as well as for
with their employer. Organizational identifi- expatriates (e.g., Stroh, Black, Mendenhall, &
cation refers to the strength of an employee’s Gregersen, 2005). Moreover, there is evidence
identification with the organization in which that subsidiary managers (Vora, Kostova, &
the person works, indicating a “perception of Roth, 2007) and even expatri-
oneness with or belongingness to an organi- ates from the MNE home country We expect
zation, where the individual defines him or (Gregersen & Black, 1992) may
herself in terms of the organization(s) in identify more with the local unit employees who
which he or she is a member” (Mael & than with the MNE.
perceive they have
Ashforth, 1992, p. 104). Organizational iden- Identification with the
tification has been found to have several pos- Unit. A study by Reade (2001b) received talent
itive effects for individuals and organizations reveals that employee identifi-
(Ashforth, Harrison, & Corley, 2008). For ex- cation with the local unit and status to draw
ample, if employees identify with the organi- identification with the whole cor-
zation and have positive feelings about its poration are driven by different more favorable
leadership, their self-esteem and self- sets of antecedents. Three factors interpretations about
motivation will be enhanced. Organizational shown to lead to greater identifi-
identification also facilitates cooperation cation with the unit were the per- their psychological
across individuals and units since employees ceived support of the individual’s
share values and loyalties. immediate supervisor, perceived contract and
Identification is a perception of belong- opportunities for career advance-
be more likely
ing to an organization that is influenced by ment and fulfillment of poten-
situational cues highlighting common inter- tial within the local unit, and to support such
ests or shared outcomes between an individ- perceptions that the individual’s
ual and an organization (Ashforth & Mael, nationality is not a barrier to the priorities.
1989). However, situational cues (e.g., per- organizational hierarchy within
ceived similarity to others) do not necessar- both the local company and the global cor-
ily resolve the ambiguity that can exist about poration. Building on our earlier arguments
organizational membership. We argue that regarding positive distinctiveness and organi-
being formally identified as talent will reduce zational support, we argue that being identi-
individuals’ ambiguity about organizational fied as talent sends a strong message that will
membership, strengthen their perceptions of serve to increase or support these perceptions
positive distinctiveness, and lead to enhanced in the minds of employees.
organizational identification (Firfiray, 2009). Furthermore, employee identification
This is consistent with research showing per- with the local unit may be particularly strong
ceived organizational support to be positively among those perceived as talent since the
associated with organizational identification talent review process in MNEs is typically
(Edwards & Peccei, 2010). Similarly, percep- carried out at different levels in the organi-
tions of not being identified as talent may zation, with local and regional units being

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
200 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

responsible for assessing local talent (Evans Hypothesis 5a: Individuals who perceive that they
et al., 2011). Similar to the positive effect of are identified as talent are more likely to identify
perceived support of the immediate super- with the multinational corporation than those
visor cited earlier, employees are therefore who perceive that they are not identified as talent.
likely to attribute their talent status (or lack
of it) to decisions heavily influenced by deci- Hypothesis 5b: Individuals who perceive that they
sion makers in the local unit. Hence: are identified as talent are more likely to identify
with the multinational corporation than those
Hypothesis 4a: Individuals who perceive that they who do not know whether they are identified as
are identified as talent are more likely to identify talent.
with their unit than are those who perceive that
they are not identified as talent. Hypothesis 5c: Individuals who perceive they are
not identified as talent are less likely to identify
Hypothesis 4b: Individuals who perceive that they with the multinational corporation than those
are identified as talent are more likely to identi- who do not know whether they are identified as
fy with their unit than those who do not know talent.
whether they are identified as talent.
Turnover Intentions
Hypothesis 4c: Individuals who perceive that they
are not identified as talent are less likely to iden- Finally, from a talent management perspec-
tify with their unit than those who do not know tive, it is crucial for the corporation to retain
whether they are identified as talent. high-performing individuals with valuable
and rare competencies. Perceptions of psy-
Identification with the MNE. Reade’s chological contract breach are commonly
(2001b) study identified two antecedents of shown to have a strong correlation with turn-
employee identification with the MNE that over intentions (Zhao et al., 2007). There is
are relevant for the identification of talent— also extensive evidence that employees who
support and appreciation of superiors at MNE perceive that they receive support from the
corporate headquarters, and the opportu- organization are less likely to consider leav-
nity for career advancement and fulfillment ing it: a meta-analysis revealed a mean cor-
within the global corporation. Although rected correlation of –.51 between perceived
decisions concerning talent identification are organizational support and turnover inten-
likely to be influenced by people at the local tions (Rhoades & Eisenberger, 2002). Simi-
or regional level, being identified as talent and larly, in the context of international assign-
placed in a corporate talent pool as a result of ments, Lazarova and Caligiuri (2001) and
a formal talent review process is still likely to Stahl, Chua, Caligiuri, Cerdin, and Taniguchi
be perceived by an individual as evidence of (2009) found support for a negative relation-
his or her positive distinctiveness and a clear ship between international assignees’
sign of organizational support from the MNE, satisfaction with company support and their
and provide individuals with a stronger sense turnover intentions. Both studies emphasized
of membership within the MNE. Moreover, the centrality of perceptions, arguing that
being identified as talent may lead to more employees’ intent to stay or leave the organi-
and/or better opportunities, perceived or zation is influenced by their subjective per-
actual, for training and development oppor- ceptions of the value and usefulness of the
tunities and career advancement on a global company support available to them, rather
scale. Indeed, an explicit objective of the tal- than by an objective evaluation of the exis-
ent management systems of many MNEs is tence or effectiveness of a certain practice.
to improve the possibilities for talented indi- This continuance commitment (Meyer &
viduals from units worldwide to develop an Allen, 1991) can make it psychologically
international career within the corporation. more difficult for an employee to leave the
We thus hypothesize: organization.

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE REACTIONS TO TALENT IDENTIFICATION 201

In line with the arguments forwarded Norwegian MNEs joining the project. The re-
earlier, perceptions of having been selected sulting 11 Nordic MNEs represent a variety of
as talent can be viewed by employees as an industries, ranging in size from 2,500 to
indication that the employer values their 60,000 employees, and have units in an aver-
contributions and has fulfilled its obligations age of 30 different countries. According to the
by deciding to invest in their future develop- corporate HR representatives, all MNEs sought
ment (Lee at al., 2011). Based on this reason- to achieve a corporate-wide talent manage-
ing, we expect that talent identification will ment system. It should be noted that none of
influence employees in terms of their turn- the firms had an explicit policy of always in-
over intentions as follows: forming individuals about their talent status.
The data for this article were collected
Hypothesis 6a: Individuals who perceive that they through a web-based survey of 930 managers
are identified as talent are likely to have lower and professionals in 106 subsidiaries of the
turnover intentions than those who perceive that 11 corporations. This part of the study began
they are not identified as talent. with an e-mail sent to the HR manager of each
of the subsidiaries (whom we had interviewed
Hypothesis 6b: Individuals who perceive that they over the phone approximately six
are identified as talent are likely to have lower months earlier). In this e-mail, we
Perceptions of
turnover intentions than those who do not know described the survey and requested
whether they are identified as talent. names and e-mail addresses of ten having been
respondents in each participating
Hypothesis 6c: Individuals who perceive that they subsidiary. We proposed the fol- selected as talent
are not identified as talent are likely to have higher lowing criteria for the respondents
turnover intentions than those who do not know can be viewed by
in order to achieve a consistent
whether they are identified as talent. frame but with enough variation: employees as an
(1) there should be a fairly even
Method balance between managers (with indication that the
direct subordinates) and profes-
employer values
sionals/specialists (with no direct
Data Collection subordinates) from each unit, (2) their contributions
the managers should be one or
This study is based on data collected in the two hierarchical steps from the and has fulfilled
context of a large-scale research project on general manager (i.e., they report
global HRM. The first step of the entire data- to the general manager or to a its obligations by
collection process was to identify the largest manager that reports to the gen- deciding to invest
Finnish MNEs in terms of number of employ- eral manager), and (3) managers
ees. We also checked that the scope of their and professionals/specialists were in their future
international operations was suitable for the accepted from a range of different
purpose of our project. Our aim was to gain departments/functions, but not development.
access to at least ten subsidiaries in ten MNEs, from the HR function.
one home-country and nine foreign units Upon receipt of the list of names, we con-
(excluding representative offices), in each tacted the prospective respondents directly via
MNE by asking the corporate HR representa- e-mail, briefly describing the project and its
tive to select those units that fit these criteria. authorization from corporate and subsidiary
The result was that eight MNEs chose to par- HR, and inviting the respondents to complete
ticipate; however, this was reduced to seven a web-based survey. We created a unique ques-
since one MNE was forced to postpone data tionnaire for each company, which enabled
collection due to major restructuring. We the inclusion of some company-specific ter-
then targeted additional Swedish and Norwe- minology, such as the specific name of their
gian MNEs of similar size to increase compa- talent review process where applicable. After
rability, resulting in one Swedish and three one to two weeks, all respondents received

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
202 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

TABLE I Characteristics of Participating Managers and Professionals

N = 769 Category %
Gender Male 74.2
Female 25.8
Tenure in MNE Years (mean) 5.22
Tenure in unit Years (mean) 4.61
Reports to unit GM Yes 40.2
No 59.8
Nationality Host-country national 92.4
Foreign national 7.6

an e-mail reminder. In some units where the Operationalizations

response rate remained low, we contacted the Independent Variable
unit HR managers for a second time, asking
them to remind the respondents or provide Talent identification. As discussed, all
additional names. The total number of ques- nine corporations in our sample had formal
tionnaires sent out was 1,230, and the final yearly corporate talent review systems in
number of individuals responding to the place, which they used for identifying high-
survey was 930, producing a response rate of performing and high-potential individuals.
76 percent. The survey was answered anony- In order to assess whether the respondent was
mously, with individual respondents being identified as talent, we asked the respondents
unidentifiable. For the purpose of this study, the question “Are you formally identified by
we excluded the individuals from two MNEs [the MNE] as belonging to a talent pool?”1
that did not have an explicit formal talent Based on this, we created a three-category
review process. This resulted in a sample of grouping variable in which group 1 = those
787, and after having removed missing val- who perceive that they are identified as talent
ues, the final sample was 769 managers and (n = 185),2 group 2 = those who do not know
professionals from 90 MNE units in nine cor- whether they are identified as talent (n =
porations. Characteristics of the sample are 494), and group 3 = those who perceive that
provided in Table I. they are not identified as talent (n = 90).
The survey questionnaire was developed
through multiple rounds of iteration, and
Dependent Variables
pretested in pilot interviews with four exter-
nal managers in equivalent positions to the In order to test the discriminant validity of
respondents. Based on these, some questions our dependent variables and ensure that each
were reworded in order to make them easier of our constructs comprised only one under-
to understand. The questionnaire language lying dimension (Cascio, 2012), we did a
was English and took between 15 and 20 Varimax rotated factor analysis, extracting six
minutes to complete. We chose to use the factors. The analysis revealed five factors with
English-language version in all subsidiaries, Eigenvalues over 1, and a sixth factor with an
as this was the official language used in inter- Eigenvalue slightly below 1. Despite this, we
unit communications within the MNEs. In decided to maintain the distinction between
addition, for practical reasons as well as for the six constructs based on theoretical argu-
comparability, we did not consider it realistic ments and because there were no significant
to translate the questionnaire into the more cross-loadings exceeding the .50 level (Hair,
than ten national languages across the MNE Anderson, Tatham, & Black, 1998). The factor
units. loadings ranged between .730 and .886, and

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE REACTIONS TO TALENT IDENTIFICATION 203

the individual factors explained the follow- Reade (2001a), we measured identification
ing amount of variance: factor 1: 39.2 per- with the local unit by asking respondents to
cent, factor 2: 16.9 percent, factor 3: 9.1 per- indicate the extent to which they agreed with
cent, factor 4: 6.1 percent, factor 5: 5.5 the following statements: (1) The practices of
percent, and factor 6: 5.1 percent. this local unit/subsidiary are in line with my
Acceptance of increasing perfor- personal values, (2) What this local unit/sub-
mance demands. The operationalization sidiary stands for is important to me, and (3)
of this construct was adapted from previous My values and the values of the local unit/
research (Rousseau, 2000). We asked respon- subsidiary that I work for are the same. The
dents to indicate the extent to which they had questions were rated on a seven-point scale
made the following commitments to their ranging from 1 = “do not agree” to 7 = “agree
employer: (1) to accept increasingly challeng- entirely.” The alpha value for this construct
ing performance requirements, (2) to adjust was 0.88.
to changing performance demands, and (3) to Identification with the MNE. We
accept new and different performance require- measured identification with the MNE by
ments. The questions were rated on a seven- asking respondents to indicate the extent to
point Likert scale ranging from 1 = “not at which they agreed with the following three
all” to 7 = “to a great extent.” The Cronbach’s statements: (1) [The MNE’s] global practices
alpha value for this construct was 0.92. express my own values, (2) [The MNE] rep-
Commitment to building compe- resents values that are important to me, and
tencies. Following Rousseau (2000), we oper- (3) I see no difference between my values
ationalized the respondents’ commitment and the corporate values of [the MNE]. The
to building competencies by asking them to operationalization was also adapted from the
rate the extent to which they had made the values-based construct validated by Reade
following commitments to their employer: (2001a). The questions were rated on a seven-
(1) to seek out developmental opportunities point scale where 1 = “do not agree” and 7 =
that enhance my value to my employer, (2) “agree entirely.” This construct had an alpha
to build skills to increase my value to my value of 0.89.
employer, and (3) to make myself increas- Turnover intentions. We operation-
ingly valuable to my employer. The questions alized the respondents’ turnover intentions
were rated on a scale from 1 = “not at all” to by asking them to indicate the extent to
7 = “to a great extent.” This construct had an which they agreed with the three following
alpha value of 0.92. statements: (1) I intend to look for a job out-
Support of strategic priorities. We side of [the MNE] within the next year, (2)
measured the support of strategic priorities I often think about quitting my job at [the
of the firm by asking respondents to rate the MNE], and (3) I intend to remain with [the
extent to which they had made the follow- MNE] for the near future (reverse-scored).
ing commitments to their employer: (1) to The questions were adapted from Konovsky
actively support the strategic priorities of my and Cropanzano (1991) and rated on a seven-
employer in my daily work, (2) to actively point scale ranging from 1 = “do not agree”
help colleagues and subordinates focus on the to 7 = “agree entirely.” The alpha value for
strategic priorities of my employer, and (3) to this construct was 0.81.
actively discuss the strategic priorities of my
employer with my peers. This operationaliza-
Control Variables
tion is in line with previous research (van Riel
et al., 2009). The questions were rated on a To control for individual-level heterogeneity
scale where the scale anchors were 1 = “not at in terms of demographic and organizational
all” and 7 = “to a great extent.” The construct characteristics (Felin & Hesterly, 2007), we
had an alpha value of 0.88. controlled for gender, tenure in the MNE,
Identification with the unit. In line number of subordinates, and nationality
with the values-based construct validated by in terms of whether the respondent was a

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
204 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

host-country national or not. Gender and na- they are identified as talent are more likely
tionality were operationalized as dummy to accept increasing performance demands
variables, whereas tenure in the MNE and than those who perceive that they are not
number of subordinates were linear and mea- identified as talent, is supported (β = 0.377,
sured in number of years and number of sub- p < 0.01). Similarly, Hypothesis 1b, suggest-
ordinates, respectively. ing that individuals who perceive they are
identified as talent are more likely to accept
increasing performance demands than those
Results
who do not know whether they are identi-
In order to assess the hypothesized differ- fied as talent, is also supported (β = 0.306, p <
ences between the three different groups (tal- 0.01). Conversely, Hypothesis 1c, which sug-
ent, not talent, and don’t know if talent), we gests a difference between the groups “no”
conducted our analysis using mul- and “don’t know” regarding the likelihood of
tivariate analysis of covariance accepting increasing performance demands,
Hypotheses 2a and
(MANCOVA), which enabled us to is not supported (β = 0.070, p > 0.05).
2b, which posit examine group differences while Our second set of hypotheses concerns atti-
also taking interdependencies be- tudinal differences in commitment to building
that individuals tween the different dependent skills. Hypotheses 2a and 2b, which posit that
variables into account (Hair et al., individuals who perceive that they are identi-
who perceive that
1998). Since we have a nested da- fied as talent are more likely to have a commit-
they are identified taset in which the respondents be- ment to building competencies than are those
long to one of 90 different MNE who perceive that they are not identified as tal-
as talent are more units, we also control for this by ent, and those who do not know whether they
including the MNE unit as a cate- are identified as talent, are both supported (β
likely to have gorical blocking variable. (We also = 0.649, p < 0.001) and (β = 0.400, p < 0.001).
a commitment ran a separate analysis in which Hypothesis 2c (“no” vs. “don’t know”) is not
we used “MNE” rather than the supported (β = 0.249, p > 0.05).
to building MNE unit as the categorical group- In Hypothesis 3a, we receive support (β =
ing variable; the overall results re- 0.456, p < 0.01) for our argument that indi-
competencies than mained similar in terms of both viduals who perceive that they are identified
sign and significance to those re- as talent are more likely to actively support
are those who
ported here.) the strategic priorities of the firm than those
perceive that they The correlation matrix of the who perceive that they are not identified as
variables in the study indicates talent. Our results also support Hypothesis
are not identified that all correlations in the model 3b (“yes” vs. “don’t know”) (β = 0.373, p <
are below 0.70. This suggests that 0.001), but not Hypothesis 3c (β = 0.084,
as talent, and those
our model does not suffer from a p > 0.05), which suggests a difference between
who do not know serious collinearity problem since the groups “no” and “don’t know” regarding
Kline (2005) argues that the first the support of strategic priorities.
whether they are indication of substantial multi- In Hypotheses 4a and 4b, we argue for
collinearity is correlations above an increased likelihood of identifying with
identified as talent,
0.85. In Table II, we present the the unit if individuals perceive that they are
are both supported. correlation matrix and descriptive identified as talent, as opposed to if they per-
statistics of the variables in this ceive that they are not identified as talent, or
study. In Table III, we present the do not know whether they are talent. These
results of our models.3 hypotheses are both supported (β = 0.369,
Our first set of hypotheses argues for dif- p < 0.05) and (β = 0.230, p < 0.05). However,
ferences between the three groups (yes, don’t our results show no support for Hypothesis 4c
know, and no) in terms of accepting increasing (“no” vs. “don’t know”) (β = 0.139, p > 0.05).
performance demands. Hypothesis 1a, which Further, we find no support for Hypothesis
posits that individuals who perceive that 5a (β = 0.173, p > 0.05), which posits that

Human Resource Management DOI: 10.1002/hrm

 TABLE II Means, Standard Deviations, and Pearson Correlations
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13
Mean 0.66 0.84 0.27 5.79 5.65 5.82 5.55 5.45 2.42 0.74 5.22 0.08 1.15
SD 0.47 0.37 0.45 1.01 1.10 1.01 1.09 1.14 1.39 0.44 1.71 0.26 0.88
1. Talent (1 = yes,
0 = no)
2. Talent (1 = yes,
0 = don’t know)
3. Talent (1 =
don’t know,
0 = no)
4. Acceptance 0.24*** 0.14*** 0.01

Human Resource Management DOI: 10.1002/hrm

of increasing
performance
demands
5. Commitment 0.32*** 0.18*** 0.10* 0.63***
to building com-
petencies
6. Support of 0.29*** 0.20*** 0.06 0.58*** 0.59***
TALENT

strategic
priorities
OR

7. Identification 0.26*** 0.12** 0.12** 0.33*** 0.26*** 0.31***

with unit
8. Identification 0.17** 0.13** 0.02 0.33*** 0.28*** 0.34*** 0.58***
with MNE
9. Turnover −0.22*** −0.09* −0.10* −0.15*** −0.17*** −0.14*** −0.40*** −0.30***
intentions
10. Gender (male 0.08 0.02 0.04 −0.06 − 0.06 0.00 0.00 −0.01 0.00
= 1, female = 0)
NOT? EMPLOYEE REACTIONS

11. Tenure in MNE −0.15* 0.02 −0.12 −0.03 −0.16*** 0.00 0.04 0.04 0.03 0.13***
12. National- 0.03 −0.02 0.04 0.05 0.03 0.08* −0.02 0.06 0.04 0.02 0.02
ity (foreign =
1, host-country
national = 0)
13. No. of subor- 0.26*** 0.10** 0.11** 0.15*** 0.10** 0.23*** −0.18*** 0.19*** −0.05 0.24*** 0.17*** −0.01
dinates
TO TALENT IDENTIFICATION

All two-tailed tests. *p < .05, **p < .01, ***p < .001.
N = 782–787 for variables 4, 5, 6, 7, 8, 9, 10, 11, 12, and 13; N = 281–283 for variable 1;
N = 685–688 for variable 2; and N = 591–594 for variable 3.
205

1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 206

TABLE III Multivariate General Linear Regressions

Yes vs. No Yes vs. Don’t Know Don’t Know vs. No

β Std. Error t-statistic β Std. Error t-statistic β Std. Error t-statistic

Unita
Controlsb
1. Acceptance of increas- 0.377 0.14 2.76** 0.306 0.09 3.45** 0.070 0.12 0.59
ing performance de-
mands
2. Commitment to build- 0.649 0.15 4.35*** 0.400 0.10 4.13*** 0.249 0.13 1.91
ing competencies
HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

3. Support of strategic 0.456 0.14 3.28** 0.373 0.09 4.12*** 0.084 0.12 0.69
priorities
4. Identification with unit 0.369 0.14 2.60* 0.230 0.09 2.49* 0.139 0.12 1.12
5. Identification with 0.173 0.15 1.17 0.296 0.10 3.07** −0.123 0.13 −0.95
MNE
6. Turnover intentions −0.479 0.18 −2.61** −0.182 0.12 −1.53 −0.297 0.16 −1.85

R2 0.196−0.267 0.196−0.267 0.196−0.267

F 3.362*** 3.362*** 3.362***
N 769 769 769
All two-tailed tests. *p < .05, **p < .01, ***p < .001.
a The blocking variable “unit” is not significant for variable 3, significant at a p < .05 level for variable 2, and significant at a p < .001 level for variables 1, 4, 5, and 6.
b Of the control variables, gender (being male) was positively related with variable 1 (β = .359, p < .001) and variable 2 (β = .278, p < .01). Tenure was negatively associated with variable 2 (β = −.105,
p < .001), and number of subordinates was positively related with variable 1 (β = .162, p < .001), variable 2 (β = .128, p < .05), variable 3 (β = .244, p < .001), variable 4 (β = .208, p < .001), and
variable 5 (β = .203, p < .001). Nationality in terms of if the respondent was of foreign nationality (rather than being a host-country national) was negatively associated with variable 3 (β = −.382, p < .05).

Human Resource Management DOI: 10.1002/hrm

1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE REACTIONS TO TALENT IDENTIFICATION 207

individuals who perceive that they are identi- identified as “talent” and a number of indi-
fied as talent are more likely to identify with vidual attitudes. Our contribution to the
the MNE than those who perceive that they are literature lies in using a social exchange per-
not identified as talent. Hypothesis 5b (“yes” vs. spective to develop and empirically test a
“don’t know”) on the other hand is supported number of hypotheses concerning the previ-
(β = 0.296, p < 0.01), while Hypothesis 5c (“no” ously unexplored association between talent
vs. “don’t know”) is not (β = –0.123, p > 0.05). identification and employee attitudes that
Finally, in our last set of hypotheses, we are central to talent management and in pre-
suggest differences between the three differ- vious research have been associated with pos-
ent groups concerning their turnover inten- itive organizational outcomes (Boxall &
tions. Hypothesis 6a is supported (β = –0.479, Macky, 2009; Wright & Nishii, forthcoming).
p < 0.01), suggesting that individuals who Our findings show that there are signifi-
perceive that they are identified as talent are cant differences between those who perceive
less likely to have turnover intentions than they have been identified as “talent” and both
those who perceive that they are not identi- those who perceive they haven’t and those
fied as talent. Hypotheses 6b (“yes” vs. “don’t who don’t know. First, comparing those who
know”) and 6c (“no” vs. “don’t know”) are perceive that they have been identified with
not supported by our results (β = –0.182, p > those who haven’t, we found that those who
0.05) and (β = –0.297, p > 0.05). perceive that they have been identified as tal-
Of the control variables, the number of ent are more likely to be associated with all
subordinates came out as significant in all three attitudes examined (commitment to increas-
models (“yes” vs. “no,” “yes” vs. “don’t know,” ing performance demands, to building com-
and “don’t know” vs. “no”), showing a strong petencies that are valuable for their employers,
positive relationship with all dependent vari- and to actively support its strategic priorities;
ables, except turnover intent, as follows: accep- identification with the focal unit; and lower
tance of increasing performance demands (β turnover intent), with the exception of iden-
= 0.162, p < 0.001), commitment to building tification with the MNE.
competencies (β = 0.128, p < 0.01), support of Second, we found that those who per-
strategic priorities (β = 0.244, p < 0.001), iden- ceive they have been identified as talent are
tification with the unit (β = 0.208, p < 0.001), more likely than those who don’t know their
and identification with the MNE (β = 0.203, talent status to be associated with all the atti-
p < 0.001). The respondent being of foreign tudes examined, except turnover intentions.
nationality (as opposed to being a host-coun- In other words, those who perceive they are
try national) was negatively associated with identified as talent and those not knowing
the support of strategic priorities (β = –0.382, have the same likelihood of leaving the corpo-
p < 0.05) in all three models, and tenure in the ration. Taken together, these findings suggest
MNE was negatively related to commitment to that informing talented individuals of their
building competencies (β = –0.105, p < 0.001). status has a motivational effect in line with
Finally, in all three models, gender (being male) the predictions of social exchange theory and
was positively associated with the acceptance thus support the general logic of talent man-
of increasing performance demands (β = 0.359, agement. At the same time, the nonsignifi-
p < 0.001) and the individual’s commitment to cant findings concerning identification with
building competencies (β = 0.278, p < 0.05). the whole corporation (“yes” vs. “no”) and
turnover intention (“yes” vs. “don’t know”)
point to the intriguing possibility that these
Discussion individuals know their value and that reten-
tion issues may arise should their expecta-
Theoretical Implications tions and needs not be met. Longitudinal
The objective of this article was to examine the studies of this issue are needed to enhance
association between employees’ perceptions our understanding of the dynamics involved
about whether or not they have been formally (cf. Robinson, Kraatz, & Rousseau, 1994).

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
208 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

The lack of significant differences between implementation within the various units of
individuals who perceived that they had not one MNE, and differences within supervisor–
been identified as talent and those who didn’t subordinate relationships. For instance, recent
know was surprising, as it seems to indicate case-study research on Western MNE subsid-
that informing them that they are not talent iaries in China suggested a heightened inter-
has little negative effect. We can only specu- est in identifying internal talent to address the
late why this was the case. For instance, cogni- high turnover of qualified Chinese employees
tive dissonance theory (Festinger, 1957) would (Hartmann, Feisel, & Schober, 2010). It was
lead us to expect that individuals who haven’t beyond the scope of the present study to inves-
been identified as talent might cope with this tigate the influence of such cultural or insti-
knowledge by downplaying the importance of tutional factors further, but this would be an
being a member of a talent pool. interesting avenue of future research.
This would be similar to the “sour Another key exogenous effect is that aris-
The lack of grapes” effect, where individuals ing from differences in supervisor–subordinate
significant lower their opinion of anything relationships and leadership/communication
they don’t or can’t get (Mann, styles of supervisors. Supervisors influence
differences between Janis, & Chaplin, 1969). In social employees’ attitudes toward the HR practices
identification terms, although of the organization (Kuvaas & Dysvik, 2010)
individuals who group membership (i.e., “talent”) as well as their perceptions of the employment
is perceived to be known (cogni- relationship (Rousseau & Tijoriwala, 1998).
perceived that
tive identification), the emotional There is likely to be significant variation—
they had not been attachment (affective identifica- regardless of company policies—in terms of
tion) and value connotation (eval- how individual supervisors communicate
identified as talent uative identification) assigned to about their subordinates’ talent status, ranging
this group are perhaps not strong from explicit formal assertions to informal and
and those who
enough among those not identi- indirect clues. More comparative and qualita-
didn’t know was fied as talent to influence their atti- tive research is called for to shed light on the
tudes or behaviors (Tajfel, 1981). effects of the communication strategies of
surprising, as it In either case, a perception of not supervisors, and how individuals make sense
being identified as a talent would of and react to the signals they are sent, for-
seems to indicate
not have significant attitudinal mally and informally, about their talent status.
that informing them effects. However, more research is
clearly warranted on this issue.
Limitations and Future Research
that they are not In line with Scullion et al.’s
(2010) definition of global talent Like all research, this study is subject to a
talent has little management, which includes the number of limitations. First, given its cross-
negative effect. need for organizations to take into sectional nature, we cannot rule out the pos-
account their “global strategic pri- sibility of common method variance having
orities as well as the differences influenced the results. However, while the
across national contexts for how talent should same individual provided data on both inde-
be managed” (p. 106), it is reasonable to expect pendent and dependent variables, the former
that there may be contextual differences in the was a grouping variable (as opposed to a Lik-
preferred communication strategy and overall ert-type scale variable) and placed in the
impact of talent identification on employee background section of the questionnaire.
attitudes (Farndale et al., 2010). Indeed, This is likely to have created a reasonable de-
although all nine corporations in our sample gree of psychological separation, which helps
had corporate-wide talent management strat- in reducing other potential sources of com-
egy and practices, there are a number of poten- mon method variance (Podsakoff, MacKen-
tial sources of exogenous variation that may zie, Lee, & Podsakoff, 2003).
influence their impact on employee attitudes, Second, we measure individual percep-
including different company policies, differing tions of whether they are formally identified as

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE REACTIONS TO TALENT IDENTIFICATION 209

talent or not. This may or may not correspond firms from three Nordic countries. It may be
with objective talent pool inclusion. However, that Nordic samples differ from, for example,
a key point we are making is that it is the per- Anglo-American ones with regard to how pre-
ception of being identified as a “talent” that pared companies are to differentiate between
matters for the dependent variables, rather employees and employee attitudes with regard
than any “objective” identification, regard- to differentiation (both reflecting Nordic
less of whether this involves membership in equalitarian values). Although any general-
a formal talent pool, training, accelerated pro- izations beyond the Nordic coun-
motion, or a certain type of job. Nevertheless, tries should be done with caution
it would also have been of interest to analyze and we can only speculate, it may The results suggest
the correlates of talent pool membership. be that our results would be more
that it is not only
Furthermore, in line with equity theory and pronounced in cultures in which
social comparison (Carrell & Dittrich, 1978), individual-level differentiation better to inform
knowledge about the talent status of fellow is traditionally more accepted.
colleagues could also be included as a variable While the findings from a recent employees that
in order to help shed further light on employ- study conducted by a consultancy
they have been
ees’ attitudinal reactions. firm for the Chartered Institute of
Third, we acknowledge a possibility of Personnel and Development in identified as talent,
reverse causality. It may be that individuals the United Kingdom are broadly
who are seen to exhibit some of the attitudes similar to those presented here but that it is perhaps
examined in this study are more likely than regarding the positive attitudes
also better to tell
others to be included in talent pools, rather of those in talent pools and non-
than (only) the other way around. However, detrimental attitudes of those those who have not
while this might be feasible concerning, for who are not (Chartered Institute
example, the acceptance of increasing per- of Personnel and Development, made it into talent
formance demands, it is unlikely that corpo- 2010), future research on MNEs
rate executives involved in decision making from other countries is needed to pools—at least, if
about talent-pool inclusion have in-depth investigate whether or not there both talent reviews
knowledge about the organizational identi- are contextual limitations to the
fication and turnover intentions of individ- generalizability of the results. and communication
ual employees. The findings concerning the
groups “yes” vs. “don’t know” also suggest about inclusion
Managerial Implications
that reverse causality is unlikely. Assuming
are conducted in a
that the “don’t know” category includes both The results suggest that it is not
those who actually have and haven’t been only better to inform employees transparent and fair
identified as talent, we would expect the dif- that they have been identified as
ferences between “yes” vs. “don’t know” to be talent, but that it is perhaps also way, and if those
less significant than the differences between better to tell those who have not
who haven’t been
“yes” vs. “no.” Since this isn’t the case (see made it into talent pools—at least,
Table III), the more logical explanation is if both talent reviews and commu- identified as talent
that identification leads to attitudinal reac- nication about inclusion are con-
tions. Nonetheless, feedback loops may also ducted in a transparent and fair have a real chance
exist in that inclusion in a talent pool might way, and if those who haven’t
of making it next
predict inclusion in the future, partly due to been identified as talent have a
the attitudinal differences examined earlier. real chance of making it next time time around.
We therefore call for longitudinal research to around. At best, this transparency
examine the nature of causality within the may create a continuous tourna-
relationships analyzed in the present study. ment concerning who is viewed to have tal-
Finally, considering the external validity ent and thus have a motivational effect on
of the findings from our study, it should be employees. We know that talent reviews are
noted that the sample of MNEs consisted of susceptible to a number of biases stemming

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
210 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

from, for example, cultural differences, gen- broader and more inclusive approaches to
der, and network centrality (Collings & Mel- talent management.
lahi, 2009; Greer & Virick, 2008; Mäkelä et al.,
2010). It is therefore likely that the reactions
Acknowledgments
of employees to their talent status will be me-
diated by their perceptions of whether the The authors are grateful to the Finnish Fund-
talent review process was carried out fairly ing Agency for Technology and Innovation
(Firfiray, 2009; Turnley & Feldman, 1998). (Tekes) (Dec. no.53/31/08) and the Academy
It is worth noting that explicit, differen- of Finland (Dec. no. 122402) for financing
tial treatment of employees has the poten- this research.
tial to reinforce competitive organizational
climates in which the few go forward and Notes
the many are left behind (Cooper, 2008).
Such climates might, on the one hand, lead 1. This question was customized for the various
to self-selection of performance-oriented MNEs by using the company-specific term (i.e., in
individuals to the organization, but may on some cases “Are you identified as talent?” and in
the other hand also have long-term effects others “Are you identified as a high potential?”)
on employee morale and be demeaning for The choice of wording was based on the term used
those who are solid workers but lack the in earlier face-to-face interviews carried out with
ambition to compete for higher positions corporate HR representatives at HQ.
or for some reason are perceived by corpo- 2. We can only speculate that this relatively high fig-
rate decision makers to lack future potential. ure (i.e., 20 percent of the sample) is partly due to
Indeed, it is argued that a company’s long- our focus on managers and professionals no more
term performance depends on the unsung than two levels from the general manager, but may
commitment and contributions of their also be due to the respondents’ inflated expecta-
“B players” (i.e., those who are capable, tions of their standing.
steady performers, and provide an impor- 3. Since one group always constitutes the reference
tant counterbalance to the ambitions of group, we conducted two separate runs in order
the high-performing “A players”; DeLong to get the comparisons between all the groups.
& Vijayaraghavan, 2003). It is therefore In the first run, “no” was the reference group,
important to consider the potential long- thus enabling the comparison between the “yes”
term implications of identifying talent, and vs. “no” and “don’t know” vs. “no” groups,
to counterbalance the focus on top talent while in the second run “don’t know” constituted
with creative “talent solutions” (Beechler & the reference group in order to allow for the
Woodward, 2009) that capitalize on diver- comparison between the “yes” and “don’t know”
sity (Greer & Virick, 2008), and involve groups.

INGMAR BJÖRKMAN is a professor and dean of the Aalto University School of Business
in Finland. His research interests focus on international HRM, knowledge creation and
transfer in multinational corporations, and integration of international mergers and ac-
quisitions. His research has been published in journals like the Journal of International
Business Studies, the Journal of Management Studies, Human Resource Management,
and Organization Studies. His latest books are Global Challenge: International Human
Resource Management (2011, McGraw-Hill), co-authored with Paul Evans and Vladimir
Pucik, and Handbook of Research in International Human Resource Management (2012,
Edward Elgar), co-edited with Günter Stahl and Shad Morris.

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE REACTIONS TO TALENT IDENTIFICATION 211

MATS EHRNROOTH (PhD in business administration) is currently a lecturer in the Hank-

en School of Economics in Helsinki, Finland. His research focuses on HRM, international
HRM, and the management of multinational corporations and has appeared in the Jour-
nal of Management Studies, Journal of World Business, and International Journal of Hu-
man Resource Management, among others.

KRISTIINA MÄKELÄ (PhD in business administration) is a visiting associate professor in

international business at Aalto University in Finland. Her research focuses on HRM and
knowledge issues in multinational corporations and has appeared in Human Resource
Management, the Journal of Management Studies, the Journal of World Business, In-
ternational Business Review, the Journal of Managerial Psychology, the International
Journal of Human Resource Management, International Studies in Management and Or-
ganization, and the Scandinavian Journal of Management, among others. She teaches
international business and human resource management both in Finland and interna-
tionally.

ADAM SMALE is a professor in the Department of Management at the University of

Vaasa, Finland, where he also acts as vice program manager of the Master’s Degree Pro-
gramme in International Business. His research interests focus on HRM, electronic HRM,
and knowledge transfer in multinational firm settings. He is also part of the international
5C network that is conducting multicountry research on global careers.

JENNIE SUMELIUS (PhD in business administration) works as a postdoctoral researcher

at the Department of Management and Organization of the Hanken School of Economics
in Helsinki, Finland. Her research focuses on HRM-related issues in multinational corpo-
rations and has been published in International Business Review, the International Jour-
nal of Human Resource Management, and the Journal of World Business, among others.

References decision science paradigm for a new strategy defi-

nition. Human Resource Management, 44, 129–136.
Allen, N. J., & Meyer, J. P. (1990). The measurement
Bowen, D. E., & Ostroff, C. (2004). Understanding
and antecedents of affective, continuance and nor-
HRM-firm performance linkages: The role of the
mative commitment to the organization. Journal of
“strength” of the HRM system. Academy of Man-
Occupational Psychology, 63, 1–18.
agement Review, 29, 203–221.
Ashforth, B. E., Harrison, S. H., & Corley, K. G. (2008).
Identification in organizations: An examination of Boxall, P., & Macky, K. (2009). Research and theory on
four fundamental questions. Journal of Manage- high-performance work systems: Progressing the
ment, 34, 325–374. high-involvement stream. Human Resource Man-
agement Journal, 19, 3–23.
Ashforth, B. E., & Mael, F. E. (1989). Social identity
theory and the organization. Academy of Manage- Cappelli, P. (2008). Talent on demand: Managing talent
ment Review, 14, 20–39. in an age of uncertainty. Boston, MA: Harvard Busi-
Beechler, S., & Woodward, I. C. (2009). The global “war ness Press.
for talent.” Journal of International Management,
Carrell, M. R., & Dittrich, J. E. (1978). Equity theory: The
15, 273–285.
recent literature, methodological considerations,
Boudreau, J. W., & Ramstad, P. M. (2005). Talentship, and new directions. Academy of Management
talent segmentation and sustainability: A new HR Review, 3, 202–210.

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
212 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

Cascio, W. (2012). Methodological issues in interna- treatment, and job satisfaction. Journal of Applied Psy-
tional HR management research. International chology, 82, 812–820.
Journal of Human Resource Management, 23,
Eisenberger, R., Fasolo, P., & Davis-LaMastro, V. (1990).
2532–2545.
Perceived organizational support and employee
Chartered Institute of Personnel and Development. diligence, commitment, and innovation. Journal of
(2010). The talent perspective: What does it feel Applied Psychology, 75, 51–59.
like to be talent-managed? Survey report, Summer Evans, P., Pucik, V., & Björkman, I. (2011). Global chal-
2010. London, UK: Author. lenge: International human resource management
Cole, M. S., Schaninger, W. S., & Harris, S. G. (2002). (2nd ed.). New York, NY: McGraw-Hill.
The workplace social exchange network: A multi- Farndale, E., Pauuwe, J., Morris, S. S., Stahl, G. K.,
level, conceptual examination. Group & Organiza- Stiles, P., Trevor, J., & Wright, P. M. (2010). Context-
tion Management, 27, 142–167. bound configurations of corporate HR functions
Collings, D. G., & Mellahi, K. (2009). Strategic talent in multinational corporations. Human Resource
management: A review and research agenda. Hu- Management, 49, 45–66.
man Resource Management Review, 19, 304–313. Farndale, E., Scullion, H., & Sparrow, P. (2010). The role
Collings, D. G., Scullion, H., & Morley, M. J. (2007). of the corporate HR function in global talent man-
Changing patterns of global staffing in the multi- agement. Journal of World Business, 45, 161–168.
national enterprise: Challenges to the conventional Felin, T., & Hesterly, W. S. (2007). The knowledge-based
expatriate assignment and emerging alternatives. view, nested heterogeneity, and new value crea-
Journal of World Business, 42, 198–213. tion: Philosophical considerations on the locus of
knowledge. Academy of Management Review, 32,
Conway, N., & Briner, R. B. (2002). A daily diary study
195–218.
of affective responses to psychological contract
breach and exceeded promises. Journal of Organi- Festinger, L. (1957). A theory of cognitive dissonance.
zational Behavior, 23, 287–302. Stanford, CA: Stanford University Press.

Conway, N., & Briner, R. B. (2005). Understanding Firfiray, S. (2009). Social identification among employ-
psychological contracts at work: A critical evalua- ees in talent-intensive organizations. Presented at
tion of theory and research. New York, NY: Oxford the 9th EURAM conference, Liverpool, UK.
University Press. Greer, C. R., & Virick, M. (2008). Diverse succession
Cooper, M. (2008). The inequality of security: Winners planning: Lessons from the industry leaders. Hu-
and losers in the risk society. Human Relations, 61, man Resource Management, 47, 351–367.
1229–1258. Gregersen, H. B., & Black, J. S. (1992). Antecedents to
Coyle-Shapiro, J.A.-M., & Conway, N. (2005). Ex- commitment to a parent company and a foreign
change relationships: Examining psychological operation. Academy of Management Journal, 35,
contracts and perceived organizational support. 65–90.
Journal of Applied Psychology, 90, 774–781. Guest, D. (2004). The psychology of the employment
relationship: An analysis based on the psychologi-
Cropanzano, R., & Mitchell, M. S. (2005). Social ex-
cal contract. Applied Psychology: An International
change theory: An interdisciplinary review. Journal
Review, 53, 541–555.
of Management, 31, 874–900.
Guest, D., & Conway, N. (2002). Communicating
DeLong, T., & Vijayaraghavan, V. (2003). Let’s hear it for
the psychological contract: An employer perspec-
B players. Harvard Business Review, 81, 96–102.
tive. Human Resource Management Journal, 12,
De Pater, I., Van Vianen, A., Bechtoldt, M., & Klehe, U. 22–38.
(2009). Employees’ challenging job experiences Guthridge, M., Komm, A. B., & Lawson, E. (2008).
and supervisors’ evaluations of promotability. Making talent a strategic priority. McKinsey Quar-
Personnel Psychology, 62, 297–325. terly, 1, 49–59.
Edwards, M. R., & Peccei, R. (2010). Perceived or- Guzzo, R. A., & Noonan, K. A. (1994). Human resource
ganizational support, organizational identification, practices as communications and the psychologi-
and employee outcomes: Testing a simultaneous cal contract. Human Resource Management, 33,
multifoci model. Journal of Personnel Psychology, 447–462.
9, 17–26.
Hair, J., Anderson, R. E., Tatham, R. L., & Black, W.
Eisenberger, R., Cummings, J., Armeli, S., & Lynch, P. (1997). (1998). Multivariate data analysis with readings.
Perceived organizational support, discretionary Englewood Cliffs, NJ: Prentice-Hall.

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TALENT OR NOT? EMPLOYEE REACTIONS TO TALENT IDENTIFICATION 213

Hartmann, E., Feisel, E., & Schober, H. (2010). Talent Meyer, J. P., & Allen, M. J. (1991). A three-component
management of Western MNEs in China: Balanc- conceptualization of organizational commitment.
ing global integration and local responsiveness. Human Resource Management Review, 1, 61–89.
Journal of World Business, 45, 169–178.
Morrison, E. W., & Robinson, S. L. (1997). When
Kline, R. (2005). Principles and practice of structural employees feel betrayed: A model of how psycho-
equation modeling. New York, NY: Guilford Press. logical contract violation develops. Academy of
Management Review, 22, 226–256.
Konovsky, M. A., & Cropanzano, R. (1991). Perceived
fairness of employee drug testing as a predictor of Podsakoff, P., MacKenzie, S., Lee, J.-Y., & Podsakoff,
employee attitudes and job performance. Journal N. (2003). Common method biases in behavioral
of Applied Psychology, 76, 698–707. research: A critical review of the literature and rec-
ommended remedies. Journal of Applied Psychol-
Kuvaas, B., & Dysvik, A. (2010). Exploring alterna-
ogy, 88, 879–903.
tive relationships between perceived investment
in employee development, perceived supervisor Reade, C. (2001a). Dual identification in multinational
support and employee outcomes. Human Resource corporations: Local managers and their psychologi-
Management Journal, 20, 138–156. cal attachment to the subsidiary versus the global
organization. International Journal of Human
Lazarova, M., & Caligiuri, P. (2001). Retaining repatri-
Resource Management, 12, 405–424.
ates: The role of organization support practices.
Journal of World Business, 36, 389–401. Reade, C. (2001b). Antecedents of organizational iden-
Lee, C., Liu, J., Rousseau, D. M., Hui, C., & Chen, Z. X. tification in multinational corporations: Fostering
(2011). Inducements, contributions, and fulfillment psychological attachment to the local subsidiary
in new employee psychological contracts. Human and the global organization. International Journal
Resource Management, 50, 201–226. of Human Resource Management, 12, 1269–1291.

Lewis, R. E., & Heckman, R. J. (2006). Talent manage- Rhoades, L., & Eisenberger, R. (2002). Perceived
ment: A critical review. Human Resource Manage- organizational support: A review of the literature.
ment Review, 16, 139–154. Journal of Applied Psychology, 87, 698–714.

Mael, F., & Ashforth, B. E. (1992). Alumni and their Riggle, R. J., Edmondson, D. R., & Hansen, J. D.
alma mater: A partial test of the reformulated (2009). A metaanalysis of the relationship between
model of organizational identification. Journal of perceived organizational support and job out-
Organizational Behaviour, 13, 103–123. comes: 20 years of research. Journal of Business
Research, 62, 1027–1030.
Mäkelä, K., Björkman, I., & Ehrnrooth, M. (2010). How
do MNEs establish their talent pools? Influences on Robinson, S. L., Kraatz, M. S., & Rousseau, D. M.
individuals’ likelihood of being labelled as talent. (1994). Changing obligations and the psychological
Journal of World Business, 45, 134–142. contract: A longitudinal study. Academy of Man-
agement Journal, 37, 137–152.
Mann, L., Janis, I. L., & Chaplin, R. (1969). The effects
of anticipation of forthcoming information on Rousseau, D. M. (1990). New hire perceptions of their
predecisional processes. Journal of Personality and own and their employer’s obligations: A study of
Social Psychology, 11, 10–16. psychological contracts. Journal of Organizational
Behavior, 11, 389–400.
March, J. G., & Simon, H. A. (1958). Organizations.
New York, NY: Wiley. Rousseau, D. M. (1995). Psychological contracts in or-
ganizations: Understanding written and unwritten
McCall, M. W., Lombardo, M. M., & Morrison, A. M. agreements. Newbury Park, CA: Sage.
(1988). The lessons of experience: How successful
executives develop on the job. New York, NY: Free Rousseau, D. M. (2000). Psychological contract inven-
Press. tory: Technical report (3rd ed.). Pittsburgh, PA:
Carnegie Mellon University.
McDonnell, A., Lamare, R., Gunnigle, P., & Lavelle, J.
(2010). Developing tomorrow’s leaders—Evidence Rousseau, D. M., & Greller, M. M. (1994). Human
of global talent management in multinational en- resource practices: Administrative contract makers.
terprises. Journal of World Business, 45, 150–160. Human Resource Management, 33, 385–401.

Mellahi, K., & Collings, D. G. (2010). The barriers to Rousseau, D. M., & Tijoriwala, S. A. (1998). Assessing
effective global talent management: The example psychological contracts: Issues, alternatives and
of corporate élites in MNEs. Journal of World Busi- measures. Journal of Organizational Behavior, 19,
ness, 45, 143–149. 679–695.

Human Resource Management DOI: 10.1002/hrm

 1099050x, 2013, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hrm.21525 by Singapore Management, Wiley Online Library on [15/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
214 HUMAN RESOURCE MANAGEMENT, MARCH–APRIL 2013

Scullion, H., Collings, D. G., & Caligiuri, P. (2010). Glo- and suggestions for further research. Journal of
bal talent management. Journal of World Business, World Business, 45, 122–133 .
45, 105–108.
Tsui, A. S., Pearce, J. L., Porter, L. W., & Tripoli, A. M.
Shaw, J. D., Dineen, B. R., Fang, R., & Vellella, R. F. (1997). Alternative approaches to the employee-
(2009). Employee–organization exchange relation- organization relationship: does investment in em-
ships, HRM practices, and quit rates of good and ployees pay off? Academy of Management Journal,
poor performers. Academy of Management Jour- 40, 1089–1121.
nal, 52, 1016–1033.
Turnley, W. H., & Feldman, D. C. (1998). Psychological
Shore, L. M., & Barksdale, K. (1998). Examining contract violations during corporate restructuring.
degree of balance and level of obligation in the Human Resource Management, 37, 71–83.
employment relationship: A social exchange ap- van Riel, C. B. M., Berens, G., & Dijkstra, M. (2009).
proach. Journal of Organizational Behavior, 19, Stimulating strategically aligned behaviour among
731–744. employees. Journal of Management Studies, 46,
Stahl, G., Björkman, I., Farndale, E., Morris, S., Paau- 1197–1226.
we, J., Stiles, P., . . . Wright, P. (2012). Six principles Vora, D., Kostova, T., & Roth, K. (2007). Roles of sub-
of global talent management. MIT Sloan Manage- sidiary managers in multinational corporations: The
ment Review, 53(2), 25–32. effect of dual organizational identification. Manage-
Stahl, G., Chua, C. H., Caligiuri, P., Cerdin, J.-L., & ment International Review, 47, 595–620.
Taniguchi, M. (2009). Predictors of turnover Wiener, Y. (1982). Commitment in organizations: A
intentions in learning-driven and demand-driven normative view. Academy of Management Review,
international assignments: The role of repatriation 7, 418–428.
concerns, satisfaction with company support, and
perceived career advancement opportunities. Hu- Wright, P. M., & McMahan, G. C. (2011). Exploring
man Resource Management, 48, 89–109. human capital: Putting human back into strategic
human resource management. Human Resource
Stroh, L. K., Black, J. S., Mendenhall, M. E., & Management Journal, 21, 93–104.
Gregersen, H. B. (2005). International assignments:
Wright, P. M., & Nishii, L. H. (forthcoming). Strategic
An integration of strategy, research and practice.
HRM and organizational behavior: Integrating mul-
Mahwah, NJ: Lawrence Erlbaum Associates.
tiple levels of analysis. In D. Guest (Ed.), Innova-
Tajfel, H. (1981). Human groups and social categories: tions in HR. Oxford, UK: Blackwell.
Studies in social psychology. Cambridge, UK: Cam-
Zhao, H., Wayne, S. J., Glibkowski, B. C., & Bravo, J.
bridge University Press.
(2007). The impact of psychological contract breach
Tarique, I., & Schuler, R. S. (2010). Global talent man- on work-related outcomes: A meta-analysis. Per-
agement: Literature review, integrative framework, sonnel Psychology, 60, 647–680.

Human Resource Management DOI: 10.1002/hrm