Strategic Recruitment Across Borders: An Investigation of Multinational Enterprises

Strategic Recruiting in MNEs

According to the resource-based view of the firm (Barney, 1991, 2001), firms can develop and identify strategic resources that result in a competitive advantage, which exists if a firm earns a profit that surpasses the industry average (Campbell, Coff, & Kryscynski, 2012). Consequently, the resource-based view of the firm serves as an overarching framework to explain the circumstances under which HRM activities may form positive and negative synergies that can become the strategic resources necessary to develop a sustained competitive advantage (Barney & Wright, 1998). HRM activities that meet the criteria of a strategic resource are critical for not only survival but a competitive advantage (Banks & Kepes, 2015).

Recruiting in MNEs presents unique opportunities for sustaining a competitive advantage. Recruiting practices may lead to critical knowledge acquisition within MNEs (Caligiuri, 2014), as well as to securing the global talent needed to create value (Björkman et al., 2013; Phillips & Gully, 2017). Effective recruiting for MNEs can also influence turnover rates, which affect operating costs (Reiche, Kraimer, & Harzing, 2011). Yet, firms that recruit talent for MNEs must identify effective recruiting signals that attract individuals capable of adding value at both the local and the global level of the organization (Farndale et al., 2010).

Indeed, recruiting talent internationally is recognized as a clear challenge for MNEs (Björkman et al., 2013). The successful balancing of standardized and localized recruiting HRM activities (a) creates value, (b) is not easily executed (i.e., is rare), (c) is not easily substituted, and (d) is a socially complex phenomenon that may not be easily understood by outsiders and thus not easily imitated by competitors (Phillips & Gully, 2015). Hence, the recruiting HRM activities of MNEs meet the requirements for a strategic resource, as specified in the resource-based view of the firm (Banks & Kepes, 2015). The combination of distinct recruiting efforts can lead to rare positive synergies or the avoidance of negative synergies and a subsequent competitive advantage (Phillips & Gully, 2015). A first step toward assessing how recruiting leads to a competitive advantage is to examine how MNEs carry out recruiting efforts across borders.

Recruitment Signals of MNEs

Signaling theory, first introduced into research on labor markets (Spence, 1973), describes how information asymmetries between two distinct parties (a signaler and a receiver) are reduced (Connelly et al., 2011). The signaler has access to private information and uses signals to send select messages to the receiver who lacks information about the signaler. In general, the purpose of a signal is to convey positive messages about the quality of the signaler (e.g., an employer) when quality is otherwise unobservable to the receiver (e.g., a potential job applicant).

In the context of employee recruiting, employers have private information that they selectively send to potential job candidates by way of signals. In the recruiting literature, scholars have applied signaling theory to understand how firms communicate with prospective job applicants and reduce applicant uncertainty in the recruiting process (e.g., Banks, Kepes, Joshi, & Seers, 2016; Lievens & Slaughter, 2016). When recruiting, firms send instrumental and symbolic signals to prospective job applicants (Highhouse, Lievens, & Sinar, 2003). While the former represent objective physical attributes of a firm and position (e.g., working conditions), the latter reflect more intangible qualities (sincerity, prestige; Highhouse et al., 2007). Such signals provide applicants with information that allows them to determine whether they would want to join the organization (Connelly et al., 2011). Thus, firms must understand which signals are most effective in attracting talent (Breaugh, 2013). For instance, firms often desire to obtain “best places to work” certifications to present to prospective employees (Dineen & Allen, 2016).

Firms have at least a basic incentive to send truthful signals if they want to not only attract but also retain the best talent. Theoretical arguments (Breaugh, 1983) and meta-analytic evidence (Phillips, 1998) suggest that misleading signals result in unrealistic job previews and, consequently, dissatisfied employees and higher turnover rates (Phillips & Gully, 2015). Moreover, recruitment signals are related to actual changes in applicant attraction (e.g., Rynes et al., 1991) and hard economic outcomes (e.g., Lievens & Slaughter, 2016; Phillips & Gully, 2015). As such, signals are an important part of the recruitment process.

Recruitment webpages, often labeled “Careers” webpages, are a vital source of information for job candidates and, hence, a significant opportunity for employers to signal their attractiveness to applicants. From the applicant perspective, career pages on firm websites are one of the first places to look to learn what it is like to be a member of the organization and affiliate with those who currently work there (Lievens & Slaughter, 2016). Indeed, the 2015 U.S. National Association of Colleges and Employers survey of college seniors on the job market (domestically and internationally) found that 98.2% of respondents used employer websites in their job search; in fact, websites were the most frequently used source of information. Similarly, from the employer perspective, all employers included in a study on South Korean MNEs relied heavily on websites to recruit prospective job applicants (Kang & Shen, 2013). Thus, employer websites are powerful recruiting tools.

Because of this importance placed on career pages, firms must be selective about the content and strength of the signals that they send to potential candidates via their websites (Rynes et al., 1991). First, the content of the pages, which represents the actual signals, conveys specific messages to applicants. Some firms, for example, signal their support of international careers by describing policies about international assignments on their recruiting websites (Point & Dickmann, 2012); such signals could be useful when trying to attract a global talent pool. Second, firms can vary the strength of particular signals, conveying the relative importance of the messages displayed (Connelly et al., 2011). On websites, signals that are immediately displayed and those that are discussed more frequently are those that a firm wishes to emphasize. Together, the content and strength of signals can be used to convince potential applicants that the firm is a high-quality employer (Dineen & Allen, 2016).

If firms operate solely in a domestic context, the decision regarding what signals to send is limited to a single country. If a firm operates in a domestic context and one or more international contexts, however, there is added complexity. Firms that operate across borders, for instance, face the challenge of recruiting employees who may work in distinct HRM systems (Lepak & Snell, 1999). There may also be pressures to adapt recruiting efforts to the local environment—perhaps to be compliant with local laws regarding diversity (Allen & Vardaman, 2017), such as equal employment opportunity, or to take advantage of tax breaks, such as those for hiring military veterans. Moreover, job candidates in different countries may have different needs and demands, which could result in a firm deciding to signal different qualities when recruiting applicants in different countries. Thus, to understand the recruiting efforts of MNEs, it is necessary to observe recruiting signals in domestic and international contexts.

When deciding how to recruit across national borders, MNEs have two options. First, firms can engage in local customization, adapting their recruiting efforts to each foreign context in which they operate. Second, they can maintain consistent recruiting efforts in domestic and foreign contexts. In terms of signaling, firms that use the same (i.e., consistent) recruiting signals regardless of location engage in global standardization, whereas firms that use different recruiting signals across national contexts use local customization.

On one hand, local customization provides managers the opportunity to better respond to local pressures as well as geographic and national culture concerns (Morris & Snell, 2011). Customization also allows for more culturally sensitive HRM activities, such as supporting team environments in collectivist cultures and increased opportunities for individual advancement or recognition in individualistic cultures (Aycan, 2005). In terms of recruiting, a locally customized approach allows organizations to choose the types of signals that appeal at a local level based on the local social, political, and economic climate (Caligiuri & Stroh, 1995).

On the other, global standardization has its merits too. For example, a firm must attract individuals who can ultimately be promoted to leadership positions in which they need to successfully manage global operations, which calls for more standardized signals (Caligiuri & Stroh, 1995; Phillips & Gully, 2017; Phillips, Gully, McCarthy, Castellano, & Kim, 2014). Such globally oriented managers can help create a competitive advantage for organizations because they increase flexibility and adaptability and acquire international skills along the way (Caligiuri & Stroh, 1995; Phillips et al., 2014).

Moreover, Lepak and Snell (1999) described how systems of distinct HRM activities must be aligned in how they manage employees. Between-HRM system fit represents the “alignment and consistency between distinct HRM systems within an organization,” such as HRM activities that occur across locations in MNEs (Banks & Kepes, 2015: 361). Poorly aligned HRM systems (i.e., systems that vary in terms of the signals sent) send employees mixed signals that could be prone to idiosyncratic interpretation (Bowen & Ostroff, 2004). Global standardization of recruiting signals represents a form of HRM alignment (i.e., between countries), and a low level of standardization could lead to a failure to achieve a “shared mind-set” or an effective organizational climate (Ulrich & Lake, 1990). This can be particularly challenging as MNEs attempt to manage multiple global HRM systems simultaneously (Edwards & Kuruvilla, 2005) and develop the talent pipeline needed to ensure continuity at the upper echelons of organizational leadership (Farndale et al., 2010).

Employing international recruitment signals that are standardized (i.e., consistent across foreign and domestic contexts) with home country recruitment signals may be used as a control function (Brock et al., 2008; Caligiuri & Stroh, 1995; Kang & Shen, 2013). That is, firms may attempt to maintain control over their international operations by using recruiting activities that are aligned with domestic operations. Such control mechanisms serve to mitigate potential misalignment among HRM systems and facilitate the transfer of knowledge across operations (Caligiuri & Stroh, 1995; Kang & Shen, 2013). As such, firms should be motivated to standardize the types of signals presented to job applicants regarding their HRM systems and emphasize these signals to the same degree (Ostroff & Bowen, 2016). Yet, as mentioned previously, firms also have incentives to adapt recruitment signals to the local environment in an attempt to balance global versus local HRM needs (Hinds et al., 2011). Firms must decide whether they should present a set of strong, consistent signals to employees across international boundaries in their organization (Bowen & Ostroff, 2004; Ostroff & Bowen, 2016) or customize recruiting signals to the local environment. We thus ask,

Influence of Contextual Factors on MNE Recruiting Strategies

We previously discussed potential recruiting strategies of firms that operate across borders. We now transition into examining how cultural factors may influence the use of HRM activities in MNE operations. Culture can be characterized as “the values, beliefs, and assumptions learned in early childhood that differentiate one group of people from another” (Schuler & Rogovsky, 1998: 161). Authors have suggested that the role of culture is critical to understand the application of HRM activities in MNEs (Ma & Allen, 2009; Schuler & Rogovsky, 1998). Furthermore, cultural distance between domestic and international operations could play an important role (Dragoni et al., 2014; Rosenzweig & Nohria, 1994). We consider how distance in national culture may influence the global standardization of the recruiting signals sent by firms operating across borders (Phillips & Gully, 2017).

Hofstede’s (1983, 1993) cultural dimensions are perhaps some of the most widely cited and applied measures of national culture (Schuler & Rogovsky, 1998). Hofstede’s seminal studies have been expanded to focus on six dimensions in total: power distance, individualism, masculinity, uncertainty avoidance, long-term orientation, and indulgence. Power distance is a measure of the degrees to which (a) power is distributed unequally in society and (b) a society accepts this distribution (Hofstede, 1983). The individualism-collectivism dimension describes the extent to which a national culture places emphasis on individual interests relative to the interests of a larger, collective group. The masculinity-femininity dimension describes the extent to which a national culture shows a preference for what are considered to be historically “masculine” traits, such as assertiveness and achievement, relative to other, more “feminine” traits, such as cooperation and modesty. Uncertainty avoidance is the extent to which a society is tolerant of ambiguity (Hofstede, 1983, 1993). Long-term orientation is a dimension that describes the extent to which people within a culture are focused on the future or the present and past (Hofstede, 2011). Finally, the indulgence dimension pertains to the extent to which a society pursues gratification relative to control of basic human desires (Hofstede, 2011).

Research has shown that some HRM policies and practices developed in the United States may be less effective in other international contexts due to cultural differences (Brewster, 1995). For example, if a firm adopts a general standardization strategy, the degree to which firms signal that workplace diversity is important in a domestic context (as is the case in the United States) will be relatively equal to the extent that diversity is signaled as important in an international context. However, if firms engage in MNEs in a high–power distance context, the notion of inclusion and equality that is typically associated with diversity signals may conflict with the local cultural context. The reason is that power distance illustrates the degree to which power is distributed unequally in society and a society accepts this distribution.

From a local customization approach, matching HRM recruiting activities to national culture illustrates cultural awareness and sensitivity and attracts individuals whose behavior is likely to be consistent with socially approved norms (Schuler & Rogovsky, 1998). Yet, such customization does not guarantee that the individuals’ values and beliefs will be consistent with the needs of cross-border operations. HRM recruitment activities send signals indicating what is valued, expected, and rewarded within a firm (Bowen & Ostroff, 2004). Such signals are meant to attract applicants as well as build consensus among employees, allowing for shared perceptions to develop (Ostroff & Bowen, 2016). A lack of global standardization may lead to vulnerabilities by a firm, as managerial discretion of local leaders may be suboptimal at times (Kostova, Nell, & Hoenen, 2018). Inconsistent recruitment signals may lead to idiosyncratic interpretations, which could result in negative synergistic effects (Phillips & Gully, 2015). The argument for internal consistency is particularly strong if there is a need for employees to move between geographic locations (Rosenzweig & Nohria, 1994). Firms must make sure that they recruit talent that supports their global business strategy (Edwards & Kuruvilla, 2005; Phillips & Gully, 2017). Given this need for internal consistency, it is not clear the extent to which firms can balance global standardization and local customization in their recruiting practices. Consequently, we ask the following research question:

Data Collection

Fortune 1000 firms represent the 1,000 U.S. firms with the highest revenue from the prior year. The list is assembled annually by Fortune magazine. The firms in the current study were obtained from the 2015 Fortune 1000 list. We focused on recruiting messages displayed on the websites of Fortune 1000 firms, as these are among the first signals that a firm sends to its prospective employees. Moreover, prospective employees use these early signals to form impressions about the organization and the position, which influences their decision to apply (Breaugh & Starke, 2000).

Data were originally collected between May and September 2016. All data were publicly available and derived from websites and online databases. Given that firm reputation is not the same as employer reputation (Lievens & Slaughter, 2016), it was important to focus on the signals sent by the firm directly to applicants via their career webpages. Data collection for domestic recruiting signals began by visiting the main “Careers” or “Jobs” page of the target companies, which we hereafter refer to as the “splash page” because it represents the introduction to a firm’s online recruiting efforts and connects to subsequent pages of recruiting information. As our focus was on firm-level signals (e.g., recruitment policies), we excluded text that fell out of that focus, such as links to external websites, recruitment fraud alerts, webpages explicitly listed as blogs, event information, and information about specific jobs. Similarly, we also excluded recruiting information that was industry specific and related to frequently asked questions (e.g., discussion of interviews, applications, resumes).

We were able to locate a splash page for 947 of the Fortune 1000 firms. All text on the initial splash pages was collected. Next, every recruiting-related hyperlink on the recruiting splash page was opened and the text on the subsequent pages recorded. The text from these combined pages represents the primary and most prominent signals being sent to prospective applicants, given their proximity to the start of the job search process (Bowen & Ostroff, 2004; Breaugh, 2013; Ostroff & Bowen, 2016). On average, we collected text from 4.2 recruiting webpages per firm website.

Measures

Inclusion Bias Check

An inclusion bias check was conducted to compare those firms that were included in the domestic data set versus those for which we were not able to identify a comparison of the English international data set. We used independent samples t tests to examine the mean differences among these firms in terms of rank on the Fortune 1000 list, number of employees, and net income in millions. The findings showed statistically significant differences on all three variables—for example, rank: t(997) = −3.027, p < .01. Yet, the standardized mean differences in all three cases were small in magnitude—for example, rank: Cohen’s d = −0.20. Consequently, there is at least some evidence of an inclusion bias in which the higher-ranked firms, the larger firms, and the firms with more income were more likely to be included in the international data set. However, the bias may be of only minimal practical relevance. To verify the results of these analyses, a second robustness check was completed with the LDA topic model analyses. In this check, we compared the topics that emerged in the domestic data set. In comparing the topics, the results suggested no differences between the firms that had an international recruiting website and those that did not (i.e., domestic recruiting only). In sum, these results suggest that inclusion bias is of little or no concern.

Research Question 1

What signals do firms send when recruiting domestically and internationally via recruiting websites?

We first consider the signals that firms send when recruiting domestically and internationally. Furthermore, we compare the international recruitment signals sent in English and other languages. In total, nine recruiting signals were identified from the domestic websites, seven from the international websites in English, and six from the international websites in a foreign language. The recruiting signals, the operational definitions created, and the lists of the content analysis words with expert validation are presented in Tables 1–3 for the domestic, English international, and foreign international samples, respectively.

OPEN IN VIEWER

Table 1

Domestic Recruiting Signals Sent by Fortune 1000 Firms (n = 947)

Recruiting signal	Definition	Content analysis words with expert validation
Inclusiveness	Signals that a firm engages in inclusive activities to support and meet the needs of all stakeholder groups	accommodation, age, ancestry, Asian, Black, citizenship, disability, disabilities, disabled, discrimination, diverse, diversity, equality, gay, gender, genetic, harassment, Hispanic, identity, inclusion, inclusive, lesbian, LGBT, marital, minority, nationality, origin, protected, race, religion, sex, sexual, transgender, women
Inclusiveness (military specific)	Signals that a firm engages in activities to support and provide opportunities to current and former (i.e., veterans) members of the military and their families	armed, army, duty, GI, hero, heroes, honor, military, navy, proud, reservists, spouse, spouses, troops, veteran, veterans
Achievement-oriented culture	Signals that a firm engages in activities to promote a culture based on attaining results and success through determination and drive	achieve, challenge, challenges, change, deliver, desire, drive, driven, drives, expertise, fast-paced, focused, future, goals, high, ideas, passion, passionate, performance, purpose, reach, results, rewards, strategic, succeed, success, thrive, vision, win
Values-based culture	Signals that a firm engages in activities to promote a culture guided by the values of honesty, ethics, and respect	accountability, accountable, believe, dedication, embrace, ethical, ethics, guide, honesty, integrity, philosophy, principles, recognition, relationships, respect, safety, strive, trust, values
Team-based culture	Signals that a firm promotes a culture based on support and teamwork	ask, fun, give, happy, hear, help, leaders, like, play, positive, say, shared, show, support, team, teammates, teams, together, whole
Development	Signals that a firm engages in activities to advance the KSAOs and opportunities of its employees	ability, advance, advancement, capabilities, courses, develop, developing, development, educational, enhance, exposure, feedback, grow, growing, growth, knowledge, learn, learned, learning, mentoring, objectives, opportunities, opportunity, planning, promote, promoted, rewarding, rotation, rotational, skill, skilled, supporting, talented, training
Development (early career)	Signals that a firm engages in activities to provide opportunities for development specifically for current college students or recent graduates early in their careers	academic, campus, classroom, co-op, co-ops, college, degree, education, enrolled, GPA, grads graduate, graduates, graduation, intern, interns, internship, internships, MBA, school, student, students, study, undergraduate, universities, university
Well-being (individual)	Signals that a firm engages in activities to promote the physical and mental health of employees and their families	assistance, balance, benefit, benefits, care, child, children, counseling, covered, dental, dependent, dependents, families, family, fitness, flexibility, flexible, health, healthcare, healthier, healthy, holidays, illness, incentives, income, insurance, leave, life, lifestyle, medical, pay, pre-tax, prescription reimbursement, retirement, savings, vacation, vision, well-being, wellness, work-life
Corporate social responsibility	Signals that a firm engages in activities to improve the well-being of the local communities in which it resides	charitable, charities, communities, community, donate, donations, foundation, gift, giving, grants, heart, local, non-profit, nonprofit, outreach, partner, partners, partnerships, philanthropic, relief, scholarship, scholarships, volunteer, volunteering, volunteerism, volunteers

Note: KSAOs = knowledge, skills, abilities, and other characteristics.

OPEN IN VIEWER

Table 2

International Recruiting Signals Sent by Fortune 1000 Firms (n = 339)

Constructs	Definition	Content analysis words with expert validation
Inclusiveness	Signals that a firm engages in inclusive activities to support and meet the needs of all stakeholder groups	accommodation, age, Asian, Asians, Black, Blacks, disabilities, disability, disabled, diverse, diversity, equality, gay, gender, genders, Hispanic, Hispanics, identity, inclusion, inclusive, lesbian, lesbians, LGBT, marital, minorities, minority, national, nationality, origin, protected, race, religion, sex, sexual, woman, women, women’s
Creative culture	Signals that a firm engages in activities to promote a culture focused on success and idea generation	connect, creating, creative, creativity, cutting-edge, design, encourage, encourages, entertainment, feedback, fun, ideas, improve, improvement, innovation, lead, leverage, passion, play, playing, result, share, shares, social, strive, success, successful, team, teams, teamwork, think, together, voice
Values-based culture	Signals that a firm engages in activities to promote a culture guided by the values of honesty, ethics, and respect	accountability, belief, beliefs, commit, commitment, committed, empower, empowered, ethical, ethics, exceptional, honesty, integrity, mission, ownership, principles, promise, reputation, respect, responsibility, spirit, trust, value, values, vision
Development	Signals that a firm engages in activities to advance the KSAOs and opportunities of its employees	advice, assignment, assignments, classes, coaching, competencies, competency, developed, developing, development, discover, discoveries, discovery, efforts, explore, explores, functional, goal, goals, grow, growth, hands-on experience, knowledge, leader, leadership, mentor, mentoring, ongoing, opportunities, opportunity, rotation, rotational, skill, skilled, skills, talent, talented, train, training
Development (early career)	Signals that a firm engages in activities to provide opportunities for development specifically for current college students or recent graduates early in their careers	campus, college, degree, graduate, graduates, interns, internship, internships, school, schools, student, students, universities, university
Well-being	Signals that a firm engages in activities to promote the physical and mental health of employees and their families	balance, benefit, care, compensation, coverage, death, dental, dependent, dependents, expenses, families, family, fitness, flexible, health, healthier, healthy, holidays, incentive, insurance, maternity, medical, needs, paid, patient, patients, pay, pension, plan, prescriptions, reimbursement, reward, rewards, savings, treat, treatment, tuition, vacation, vision, well-being, wellness
Corporate social responsibility	Signals that a firm promotes initiatives that benefit society at large	accountable, charitable, charities, clean, communities, community, contributions, efficiency, energy, environment, environmental, food, foot-print, foundation, fuel, natural, partner, partners, philanthropy, planet, plant, plants, power, relief, safety, suppliers, supply, sustain, sustainable, volunteer, volunteering, world

Note: KSAOs = knowledge, skills, abilities, and other characteristics.

OPEN IN VIEWER

Table 3

Foreign Language International Recruiting Signals Sent by Fortune 1000 Firms (n = 162)

Recruiting signal	Definition	Content analysis words with expert validation
Creative culture	Signals that a firm engages in activities to promote a culture based on leading-edge, creative, and industry-leading problem solving that encourages overall success	beyond, change, changing, create, creativity, developing, digital, explore, goals, initiatives, innovation, lead, leader, leaders, leadership, leading, progress, quickly, results, solution, solutions, succeed, success, technologies, technology, world-class
Development	Signals that a firm engages in activities to advance the KSAOs and opportunities of its employees	able, advanced, annual, become, careers, comprehensive, develop, development, encourage, enhance, expand, experienced, expertise, grow, growing, growth, help, highest, improving, knowledge, opportunities, personnel, planning, professionals, programs, promote, provide, quality, recognize, role, senior, share, skills, staff, strategic, support, talent, training
Inclusiveness	Signals that a firm engages in inclusive activities to support and meet the needs of all stakeholder groups	age, basis, color, common, communities, countries, culture, difference, disability, diverse, diversity, equal, gender, identity, inclusion, inclusive, orientation, perspectives, race, religion, respect, sexual, society, status, together, value, values
Well-being	Signals that a firm engages in activities to promote the physical and mental health of employees and their families	access, assistance, benefits, care, economic, family, flexible, free, happy, health, healthy, heart, home, human, improvement, insurance, interests, leave, life, live, lives, medical, needs, paid, partner, partners, physical, protection, public, reward, salary, service, social, welfare
Achievement-oriented culture	Signals that a firm engages in activities to promote a culture based on attaining results and success through determination and drive	achieve, awards, better, efforts, excellence, excellent, great, outstanding, overall, promoting, promotion, proud, recognition, recognized, strategy, talented, top, world’s
Development (early career)	Signals that a firm engages in activities to provide opportunities for development specifically for current college students or recent graduates early in their careers	campus, courses, education, foundation, future, graduates, internship, internships, practical, school, students, study, universities, university

Note: KSAOs = knowledge, skills, abilities, and other characteristics.

Overall, across domestic and international (English and foreign language) webpages, there was overlap in the recruiting signals, thereby providing evidence of signal consistency or global standardization. For instance, domestic, English international, and foreign language international all had inclusiveness as a recruiting signal, defined as text that signals that a firm engages in inclusive activities to support and meet the needs of all stakeholder groups. Here is one example from the data of an inclusive statement:

We foster an environment that attracts a high-performing, diverse, and inclusive workforce. All individuals are respected and valued for their contributions and have the opportunity to achieve their maximum potential.

Other common recruiting signals were development (focused on general activities to advance the knowledge, skills, abilities, and opportunities of employees) as well as development (early career), which characterizes the signal sent by firms indicating that they desire college students or recent graduates early in their careers. An example of this latter recruiting signal is the following statement by one firm:

Our internship and co-op programs are designed to identify bright and passionate student talent. We take a long-term view by offering students a journey—from internship to leadership—through hands-on work experience in a collaborative teaching environment.

An additional mutual recruiting signal was well-being (individual), defined as text that signals that a firm engages in activities to promote the physical and mental health of employees and their families. One firm wrote,

We have the same benefits that you would expect from other high-tech companies like health insurance, retirement benefits, etc. But we also offer more than the basics. Our benefits are part of who we are, and they’re designed to take care of our employees as a whole. We offer benefits to keep you healthy, whether physically, emotionally, financially, or socially.

There were recruiting signals unique to the domestic and international operations. In the U.S. context, a unique subdimension of inclusiveness emerged that was military specific, defined as text signaling that a firm engages in activities to support and provide opportunities to current and former (i.e., veterans) members of the U.S. military and their families. This recruiting signal did not emerge as a stand-alone in the international data (although military and veteran terms did emerge in the inclusiveness topic). Another construct that emerged only within a domestic context included team-based culture.

There were also a few instances where recruiting signals in a domestic context were reflected in one international context but not another, providing some evidence of local customization. For example, values-based culture was signaled in the domestic recruiting websites and the English language international websites but not the foreign language international websites. This recruiting signal was characterized as text signaling that a firm engages in activities to promote a culture guided by the values of honesty, ethics, and respect. For instance, one firm stated, “We conduct ourselves with integrity in everything we do. Our reputation as individuals and as a firm is paramount. Our word is our bond—we say what we mean and we do what we say.” Conversely, achievement-based culture emerged in the domestic recruiting and the foreign language international recruiting but not in international recruiting in English. For example, for achievement-oriented culture, one firm wrote, “We’re innovative, fast-paced, results-oriented and, most importantly, we like to win.”

Interestingly, a corporate social responsibility (CSR) signal appeared in the domestic and English language international data but not in the foreign language international data. When the signal was presented, however, its nature was slightly different. In the domestic context, CSR was defined as text signaling that a firm engages in activities to improve the well-being of the local communities in which it resides (e.g., firms encourage volunteering, donations, community outreach). For instance, one domestic firm wrote, “Our companies and our employees generously invest their time, money, and expertise in building and enriching the communities where our employees live and work.” Conversely, on English language international websites, CSR is characterized by signals that a firm promotes initiatives that benefit society at large (e.g., promotes sustainability). Thus, while CSR did emerge in two data sets, the nature of the CSR signal varied slightly. In summary, firms seem to standardize and send a lot of the same overall signals in domestic and international recruiting. This conclusion does not change dramatically when accounting for the language in which the recruiting signal is being sent.

Research Question 2

To what extent does the strength of signals on domestic recruiting websites relate to the strength of signals on international recruiting websites?

To understand the extent to which firms standardized recruiting signals in MNE operations, one can consider the recruiting signals sent, as described in the previous section. However, the signals alone do not provide enough information about the standardization of recruiting practices. To examine global standardization versus local customization, we considered the correlations of signal strength among common domestic and English language international recruiting signals. We also considered the correlations among the strength of foreign language international recruiting signals with the domestic and English language international signals. The correlations provide evidence of the amount of consistency among the recruiting contexts. Stronger correlations suggest that there is consistency among recruiting contexts—that is, the strength of the signal is similar in foreign and domestic contexts (i.e., the signals are emphasized similarly in both contexts)—while weaker correlations suggest that the strength of the signal may differ to some extent (i.e., be inconsistent) from one context to the next. We present the means, standard deviations, and correlations among the recruiting signals in Table 4 for the domestic and English language international signals. The means in Table 4 represent the average number of words for a given signal divided by the total number of words used on the website.

OPEN IN VIEWER

Table 4

Means, Standard Deviations, and Correlations of Domestic and English Language International Signal Strength

Variable	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Domestic
1. Achievement-based culture	.01	.03
2. Team-based culture	.01	.02	.76(.000)
3. Values-based culture	.00	.01	.71(.000)	.65(.000)
4. Development	.01	.03	.60(.000)	.83(.000)	.41(.000)
5. Development (early)	.00	.01	.00(.962)	−.01(.860)	−.02(.755)	.01(.921)
6. Inclusiveness	.00	.01	.16(.000)	.24(.000)	.25(.000)	.26(.000)	.37(.000)
7. Inclusiveness (military)	.00	.00	.03(.567)	.19(.000)	.02(.777)	.37(.000)	.16(.003)	.37(.000)
8. Well-being	.01	.02	.35(.000)	.34(.000)	.27(.000)	.31(.000)	−.01(.924)	.02(.716)	−.02(.757)
9. CSR	.00	.01	.33(.000)	.29(.000)	.45(.000)	.16(.004)	.11(.054)	.31(.000)	.04(.431)	.14(.012)
International
10. Creative culture	.03	.23	.57(.000)	.17(.002	.53(.000)	.01(.904)	−.02(.685)	−.01(.832)	−.01(.815)	−.02(.664)	.36(.000)
11. Values-based culture	.01	.04	.40(.000)	.62(.000)	.53(.000)	.51(.000)	−.05(.386)	.11(.039)	.03(.542)	.27(.000)	.12(.023)	.05(.359)
12. Development	.03	.11	.15(.005)	.24(.000)	.21(.000)	.31(.000)	−.02(.000)	.07(.201)	.12(.030)	.09(.097)	.06(.243)	.02(.749)	.28(.000)
13. Development (early)	.00	.01	.00(.965)	−.03(.562)	−.01(.846)	−.01(.897)	.21(.000)	.09(.092)	.04(.425)	−.04(.472)	−.02(.688)	−.01(.822)	−.02(.678)	.05(.384)
14. Inclusiveness	.01	.01	.14(.008	.19(.001)	.21(.000)	.24(.000)	.20(.000)	.64(.000)	.26(.000)	.01(.933)	.20(.000)	.00(.959)	.13(.021)	.08(.145)	.16(.004)
15. Well-being	.02	.08	.55(.000)	.18(.001)	.54(.000)	.03(.588)	−.03(.633)	−.04(.479)	−.02(.745)	.02(.665)	.35(.000)	.94(.000)	.05(.317)	.02(.742)	−.01(.828)	.02(.689)
16. CSR	.01	.02	.34(.000)	.48(.000)	.42(.000)	.38(.000)	.06(.265)	.01(.897)	.00(.979)	.23(.000)	.12(.030)	.02(.683)	.60(.000)	.24(.000)	.10(.063)	.11(.034)	.08(.129)

Note: n = 339. Means represent the average number of words for a given signal divided by the total number of words used on the website. Exact p values are provided in parentheses (two-tailed tests). CSR = corporate social responsibility.

We first consider the correlations among English signals common to domestic and international operations. The findings show that the correlations between the strength of the domestic and international signals, such as values-based culture (r = .53, p =.000), development (r = .31, p = .000), and development–early career (r = .21, p =.000), all ranged from large to medium (Bosco, Aguinis, Singh, Field, & Pierce, 2015). This suggests, for example, that firms are likely to emphasize values-based culture and development opportunities in their domestic and international operations (when communicating in English). This result provides evidence of consistency or global standardization. However, there appeared to be almost no correlation between domestic and international well-being signals (r = .02, p = .665). Perhaps because of their slightly different foci, domestic and English language international CSR signals were still correlated, albeit to a minor extent (r = .12, p = .030). Finally, inclusiveness showed a large magnitude correlation among the common recruiting signals (r = .64, p = .000). In general, these findings indicate that domestic and English language international signals certainly appear to be related. However, the degree to which firms standardized the strength of these recruiting signals across national contexts varies; yet, the correlations are not close to 1.0, even in the strongest cases. These changes in points of consistency between the domestic and English language international websites suggest a great deal of local customization.

To the extent to which there was consistency between the strength of the domestic and English language international signals, the foreign language international signals tells a very different story (see Table 5). The magnitude of the correlations was much smaller in this context. For example, the correlation for the development recruiting signal was –.01 (p = .923) across the foreign language and domestic websites and –.01 (p = .954) across the foreign and English language international websites. The same was true for development–early career (r = –.01, p = .881; r = .01, p = .871, respectively) and well-being (r = –.02, p = .798 r = –.03, p = .745, respectively) signals. The only exception was that domestic inclusiveness correlated to a moderate degree with foreign language international inclusiveness (r = .25, p =.001). Consequently, important nuance is added with the inclusion of foreign language international recruiting signals. That is, when international recruiting signals are sent in a foreign language, there appears to be much more local customization as compared with international recruiting signals sent in English, as evidenced by the smaller magnitude correlations.

OPEN IN VIEWER

Table 5

Correlations Between Foreign Language Recruiting Signal Strength and English Language Domestic and International Signal Strength

Variable	Achievement-oriented culture	Development	Development (early)	Inclusiveness	Creative culture	Well-being
Domestic (English) a
1. Achievement-based culture	−.02 (.817)	.01 (.860)	.02 (.845)	.01 (.877)	.01 (.865)	−.05 (.565)
2. Team-based culture	.00 (.958)	.01 (.907)	.04 (.662)	.07 (.368)	−.01 (.937)	−.03 (.706)
3. Values-based culture	−.06 (.418)	−.02 (.771)	−.06 (.471)	.15 (.065)	.04 (.644)	−.04 (.607)
4. Development	.03 (.744)	−.01 (.923)	.06 (.458)	.03 (.703)	.04 (.660)	−.03 (.710)
5. Development (early)	−.02 (.851)	.01 (.865)	−.01 (.881)	.04 (.654)	−.04 (.625)	−.01 (.904)
6. Inclusiveness	.07 (.412)	.00 (.994)	.00 (.963)	.25 (.001)	.05 (.569)	−.02 (.765)
7. Inclusiveness (military)	.02 (.786)	.02 (.813)	.04 (.652)	.07 (.396)	−.04 (.576)	.00 (.967)
8. Well-being	.00 (.960)	−.04 (.593)	−.02 (.822)	−.01 (.880)	.00 (.990)	−.02 (.798)
9. CSR	.01 (.894)	−.01 (.860)	−.04 (.582)	.29 (.000)	.00 (.927)	.01 (.860)
International (English) a
10. Creative culture	.14 (.079)	.01 (.927)	−.05 (.532)	−.04 (.619)	.19 (.017)	−.03 (.680)
11. Values-based culture	.18 (.024)	.01 (.932)	−.05 (.505)	−.01 (.893)	.30 (.000)	−.03 (.696)
12. Development	.23 (.003)	−.01 (.954)	−.02 (.834)	.02 (.835)	.32 (.000)	−.03 (.700)
13. Development (early)	−.06 (.479)	−.01 (.938)	.01 (.871)	.05 (.502)	−.03 (.690)	−.02 (.824)
14. Inclusiveness	−.04 (.575)	−.01 (.888)	−.02 (.806)	.02 (.825)	−.05 (.544)	−.03 (.710)
15. Well-being	−.06 (.467)	−.05 (.568)	−.06 (.424)	−.08 (.308)	−.04 (.620)	−.03 (.745)
16. CSR	.03 (.712)	−.01 (.872)	−.07 (.379)	−.07 (.354)	.25 (.002)	−.05 (.564)

Note. n = 162. The following information is for the variables in the first row: achievement-oriented culture (M = .01, SD = .03), development (M = .08, SD = .40), development (early) (M = .01, SD = .02), inclusiveness (M = .01, SD = .04), creative culture (M = .02, SD = .03), and well-being (M = .05, SD = .33). CSR = corporate social responsibility.

a

Exact p values are provided in parentheses (two-tailed tests).

Research Question 3

How does cultural distance between domestic and international locations relate to global standardization of recruiting signals used by MNEs?

In total, six common recruiting signals emerged from our data between the domestic and English language international recruiting signals. There were five common recruiting signals between the domestic and foreign language international recruiting signals. To test the role of cultural distance on the relationship between domestic and international signal strength, we regressed international signal strength (English language) on domestic signal strength, on cultural distance, and on the interaction of domestic signal strength and cultural distance. For cultural distance, we selected the Hofstede dimensions that we believed most closely related to these recruiting signals: inclusiveness (power distance), development (long-term orientation), development–early career (long-term orientation), well-being (masculinity; indulgence), and CSR (individualism; in the English language international data set).

The results from the hierarchical regression analyses are presented in Table 6 for the English language international sample. In each case, Model 1 shows that the dependent variable (i.e., the international signal strength) regressed onto the control variable (the number of employees). Number of employees was controlled for, as there is evidence that larger firms tend to use more standardized, objective criteria when recruiting job applicants as compared with smaller firms, perhaps owing in part to the larger number of applicants (Barber, Wesson, Roberson, & Taylor, 1999). Model 2 shows that the dependent variable regressed onto the control variable and the mean-centered domestic signal and cultural distance variables. An interaction term was computed with mean-centered variables for the domestic signal and cultural distance variables. The interaction term is introduced in Model 3. The same analyses were conducted again with foreign language international signals as the dependent variable (see Table 7).

OPEN IN VIEWER

Table 6

Effects of Cultural Distance on the Relations between Domestic and English Language International Signal Strength

	Model 1a			Model 2a			Model 3a
International inclusiveness	B	SE	β (p; t)	B	SE	β (p; t)	B	SE	β (p; t)
No. of employees	−3.49E–9	.00	−.05 (.444; −0.77)	−1.39E–9	.00	−.02 (.741; −0.33)	−1.45E–9	.00	−.02 (.729; −0.35)
Domestic inclusiveness				.58	.09	.41 (.000; 6.54)	.45	.14	.32 (.001; 3.27)
Cultural distance (PD)				1.17E–5	.00	.02 (.721; 0.36)	4.46E–5	.00	.08 (.293; 1.05)
Domestic Inclusiveness × Cultural Distance (PD)							.01	.01	.13 (.221; 1.23)
ΔR2			.00			.17			.01
R 2			.00			.17			.17
International development	Model 1b			Model 2b			Model 3b
No. of employees	−2.126E–8	.00	−.03 (.656; −0.45)	−1.03E–8	.00	−.02 (.827; −0.22)	−1.213E–8	.00	−.02 (.799; −0.26)
Domestic development				.83	.36	.16 (.021; 2.32)	.62	.59	.12 (.297; 1.05)
Cultural distance (LTO)				.00	.00	.08 (.228; 1.21)	.00	.00	.14 (.351; 0.93)
Domestic Development × Cultural Distance (LTO)							.02	.04	.07 (.664; 0.44)
ΔR2			.00			.03			.00
R 2			.00			.03			.03
International development (early)	Model 1c			Model 2c			Model 3c
No. of employees	−2.84E–9	.00	−.06 (.394; −0.85)	−2.03E–9	.00	−.04 (.534; −0.62)	−1.74E–9	.00	−.04 (.594; −0.53)
Domestic development (early)				.21	.08	.17 (.009; 2.64)	.27	.09	.22 (.003; 3.06)
Cultural distance (LTO)				6.94E–5	.00	.14 (.038; 2.09)	8.35E–5	.00	.17 (.016; 2.43)
Domestic Development (Early) × Cultural Distance (LTO)							.01	.01	.12 (.126; 1.53)
ΔR2			.00			.05			.01
R 2			.00			.05			.06
International well-being	Model 1d			Model 2d			Model 3d
No. of employees	−1.79E–8	.00	−.03 (.636; −0.47)	−1.94E–8	.00	−.03 (.611 −0.51)	−1.94E–8	.00	−.03 (.612; −0.51)
Domestic well-being				−.14	.36	−.03 (.695; −0.39)	−.02	.47	.00 (.963; −0.05)
Cultural distance (masculinity)				.00	.00	−.04 (.579; −0.56)	.00	.00	−.03 (.716; −0.36)
Domestic Well-Being × Cultural Distance (Masculinity)							.02	.05	.04 (.693; −0.40)
ΔR2			.00			.00			.00
R 2			.00			.00			.00
International well-being	Model 1e			Model 2e			Model 3e
No. of employees	−1.75E–8	.00	−.03 (.649; −0.46)	−1.67E–8	.00	−.03 (.668; −0.43)	−1.66E–8	.00	−.03 (.669; −0.43)
Domestic well-being				−.16	.36	−.03 (.664; −0.44)	−.14	.68	−.03 (.838; −0.21)
Cultural distance (indulgence)				.00	.00	−.05 (.465; −0.73)	.00	.00	−.05 (.648; −0.46)
Domestic Well-Being × Cultural Distance (Indulgence)							.00	.04	.00 (.974; 0.03)
ΔR2			.00			.00			.00
R 2			.00			.00			.00
International CSR	Model 1f			Model 2f			Model 3f
No. of employees	−4.71E–9	.00	−.06 (.352; −0.93)	−4.91E–9	.00	−.07 (.336; −0.96)	−4.69E–9	.00	−.06 (.357; −0.92)
Domestic CSR				.03	.14	.01 (.859; −0.18)	.14	.17	.07 (.393; −0.86)
Cultural distance (individual)				2.26E–5	.00	.05 (.480; 0.71)	3.57E–5	.00	.08 (.288; 1.07)
Domestic CSR × Cultural Distance (Individual)							.01	.01	.11 (.198; 1.29)
ΔR2			.00			.01			.01
R 2			.00			.01			.01

Note. n = 223~225. B = unstandardized regression coefficient; SE = standard error; β = standardized regression coefficient. Cultural distance and domestic recruiting signals are mean centered. Model 1 = base model. ΔR² = change in R² between models. Exact p values are provided (two-tailed tests). PD = power distance; LDO = long-term orientation; CSR = corporate social responsibility.

OPEN IN VIEWER

Table 7

Effects of Cultural Distance on the Relations Between Domestic and Foreign Language International Signal Strength

	Model 1a			Model 2a			Model 3a
Foreign inclusiveness	B	SE	β (p; t)	B	SE	β (p; t)	B	SE	β (p; t)
No. of employees	−1.05E–8	.00	−.05 (.509; −0.66)	−3.70E–9	.00	−.02 (.811; −0.23)	−5.07E–9	.00	−.03 (.667; −0.43)
Domestic inclusiveness				1.28	.40	.25 (.002; 3.19)	2.96	.35	.59 (.000; 8.54)
Cultural distance (PD)				.00	.00	.15 (.059; 1.90)	.00	.00	.53 (.000; 7.49)
Domestic Inclusiveness × Cultural Distance (PD)							.24	.02	.79 (.000, 10.3)
ΔR2			.00			.08			.39
R 2			.00			.08			.47
Foreign development	Model 1b			Model 2b			Model 3b
No. of employees	−3.18E–8	.00	−.06 (.482; −0.71)	−2.85E–8	.00	−.05 (.527; −0.63)	−2.91E–8	.00	−.05 (.521; −0.64)
Domestic development				.08	.26	.03 (.745; 0.33)	−.07	.66	−.02 (.921; −0.10)
Cultural distance (LTO)				.00	.00	.14 (.088; 1.72)	.00	.00	.09 (.717; −0.36)
Domestic Development × Cultural Distance (LTO)							−.01	.04	−.07 (.809; −0.24)
ΔR2			.00			.02			.00
R 2			.00			.02			.02
Foreign development (early)	Model 1c			Model 2c			Model 3c
No. of employees	7.97E–11	.00	.00 (.991; 0.01)	−6.35E–10	.00	−.01 (.931; 0.09)	−6.23E–10	.00	−.01 (.932 −0.09)
Domestic development (early)				−.08	.21	−.03 (.687; −0.40)	−.07	.26	−.03 (.796; −0.26)
Cultural distance (LTO)				.00	.00	−.19 (.025; 2.27)	.00	.00	−.18 (.031; −2.18)
Domestic Development (Early) × Cultural Distance (LTO)							.00	.01	.01 (.906; 0.12)
Foreign development (early)	Model 1c			Model 2c			Model 3c
ΔR2			.00			.03			.00
R 2			.00			.03			.03
Foreign well-being	Model 1d			Model 2d			Model 3d
No. of employees	−4.91E–8	.00	−.03 (.728; −0.35)	−5.28E–8	.00	−.03 (.712; −0.37)	−5.28E–8	.00	−.03 (.713; −0.37)
Domestic well-being				−.72	2.70	−.02 (.792; −0.26)	−.72	3.26	−.02 (.825; −0.22)
Cultural distance (masculinity)				−3.98E–5	.00	.00 (.982; −0.02)	−3.24E–5	.00	.00 (.990; −0.01)
Domestic Well-Being × Cultural Distance (Masculinity)							.00	.20	.00 (.997; −0.00)
ΔR2			.00			.00			.00
R 2			.00			.00			.00
Foreign well-being	Model 1e			Model 2e			Model 3e
No. of employees	−4.97E–8	.00	−.03 (.726; −0.35)	−5.93E–8	.00	−.03 (.679; −0.42)	−6.21E–8	.00	−.04 (.666; −0.43)
Domestic well-being				−.47	2.68	−.02 (.861; −0.18)	−.64	2.71	−.02 (.813; −0.24)
Cultural distance (indulgence)				.00	.00	.07 (.382; −0.88)	.00	.00	.03 (.836; −0.21)
Domestic Well-Being × Cultural Distance (Indulgence)							−.10	.19	−.06 (.616; −0.50)
ΔR2			.00			.01			.00
R 2			.00			.01			.01

Note: n = 150~151. B = unstandardized regression coefficient; SE = standard error; β = standardized regression coefficient. Cultural distance and domestic recruiting signals are mean centered. Model 1 = base model. ΔR² = change in R² between models. Exact p values are provided (two-tailed tests). PD = power distance; LTO = long-term orientation.

In all six cases, cultural distance did not moderate the relation between domestic signals and English language international signals. Furthermore, in four of the five cases, cultural distance did not moderate the relation between domestic and foreign language international signals. These findings suggest that culture distance largely does not relate to the global standardization of recruiting signals. Specifically, it appears that changes in the emphasis of the signals seems to not depend on cultural differences. There was one exception to this when considering the Domestic Inclusiveness × Power Distance interaction in the foreign language international data set.

Key Findings and Insights

Standardization versus local customization

There were several insights of note from this work that facilitate a future research agenda (see Table 8 for a summary). As such, we offer a set of propositions to provide explicit direction for future research based on our findings. First, while our findings indicate some consistency across all three contexts in the signals that are sent (evidence of global standardization), we found differences in the strength of those signals in each context (evidence of local customization). That is, when operating across borders, firms appear to send many of the same types of signals to potential job applicants, which supports the notion of standardization. Yet, the extent to which firms emphasize the signals appears to vary across contexts. Particularly, the international foreign language context provided the least amount of consistency with the other two contexts. The practice of translating one’s recruiting signals from one language to another may in fact serve as a form of customizing one’s recruiting signals for the local context. Thus, it appears that firms are less likely to standardize recruiting signals from one international context to another if a different language is used.

OPEN IN VIEWER

Table 8

A Future Research Agenda

Key findings and insights	Future research implications	Managerial implications
Standardization versus local customization	● Distinction between the types of signals sent and emphasis when investigating HRM activities in one or more HRM system● Some recruitment signals may be universally relevant (e.g., inclusiveness) for MNE operations● Evaluation of reasoning for MNE standardization or localization● Consistency in standardization and localization across multiple MNE activities	● The role of language in the implementation of HRM policies● Consider not only what recruiting messages are provided to applicants but how much those messages are emphasized
Role of cultural distance	● Drivers of the decision of whether to consider culture (e.g., local competitors or laws)	● Evaluate intentional or unintentional role of culture in recruiting activities
Benefits of standardization	● Effects of misaligned recruiting signals in MNE contexts● Conditions under which positive and negative synergies among HRM activities are realized● Effects of localization versus standardization on the quality and types of applicants	● Decision to localize or standardize should align with business model and strategic firm activities
Taxonomy of recruiting signals	● Evaluation of actual recruiting signals sent domestically and internationally by Fortune 1000 firms	● Best practice recommendations for recruitment signalsIdentify means for differentiation in recruiting

Note: HRM = human resource management; MNE = multinational enterprise.

As an example, we found that firms signaled an inclusive work environment to job applicants in all three contexts, which is not surprising given that we sampled MNEs, where it is important to highlight inclusiveness across employees from diverse geographic regions. However, the strength of the inclusiveness signal in domestic recruiting correlated more closely to international recruiting in English (r = .64) than in a foreign language (r = .25). Furthermore, the strength of domestic recruiting signals was consistent across contexts for values-based culture (r = .53) and development (r = .31) when international recruiting was conducted in English, but this was not the case when recruiting was done in a foreign language (see Table 5). Consequently, the findings contribute to the literature by offering insights into how MNEs balance standardizing versus customizing recruiting activities across international locations, particularly when recruiting in different languages.

Based on the findings in the current study, it may be expected that MNE HRM policy signals are more likely to be standardized when they are sent in the same language as the domestic language of the firm. Yet, a question that remains is why this might be the case. One argument is that language is essentially a proxy for standardization (i.e., changing the language of the recruiting message is itself a form of customization): the MNE operational needs in a particular country perhaps require or expect job applicants in these locations to speak English in their day-to-day activities, so it only follows that recruiting signals would be similar as well. Extending signaling theory, we propose that policy signals from other types of HRM activities (e.g., incentive systems, progressive disciplinary practices) are more likely to be locally customized in MNE operations when they are sent in a language different from the one used in domestic signals. An HRM policy is an employment-related statement that facilitates the enactment of one or more HRM practices (Banks & Kepes, 2015; Lengnick-Hall et al., 2009). Firms that operate MNEs must create and disseminate policies that provide guidance for how practices related to recruiting, hiring, and performance evaluations are to be implemented. Based on the findings in the current study, it may be expected that MNE HRM policy signals are more likely to be standardized when they are sent in the same language as the domestic language of the firm.

Proposition 1: MNE HRM policy signals that are sent in the same language are more likely to be standardized than signals sent in different languages.

We expect that the findings regarding MNEs’ standardization of recruiting signals are also likely to generalize to the standardization of other types of HRM activity signals. Firms determine how much or how little control they desire over their operations (Brock et al., 2008), and this applies to their HRM activities. When firms standardize their MNE HRM activities, they are attempting to maintain control over such activities (Caligiuri & Stroh, 1995; Kang & Shen, 2013) and sustain a strong, coherent HRM system (Bowen & Ostroff, 2004). Conversely, when firms make the strategic decision to release tight control and allow for local customization, they may do so for all their HRM activities.

Past research on signaling theory was limited in that it was developed with a focus on “one-to-one or transaction specific communication” (Connelly et al., 2011: 44). Thus, research on signaling theory has largely neglected the fact that multiple and potentially competing signals can be observed (Dineen & Allen, 2016). This was particularly true when signaling theory was applied in a recruiting context (Banks, Kepes, et al., 2016; Highhouse et al., 2007; Rynes et al., 1991). It would be interesting to establish whether HRM activities are standardized across multiple and potentially competing recruiting signals. In sum, we present the following proposition:

Proposition 2: The degree of standardization of HRM recruiting signals in MNEs is similar to the standardization of other HRM activity signals.

Limitations

As is the case with any research study, the current work had several limitations. These restrict the inferences that can be drawn from the current data but provide opportunities for future research. First, the current study focused on HRM activities specific to recruiting at the HRM policy level of HRM systems. The extent to which our findings generalize to other HRM activities (e.g., selection, training and development) is not known. Furthermore, the degree to which our results generalize to the HRM practice level (i.e., practices are the activities taken to implement policies) is not clear. In the current study, we collected data at the policy level of MNE HRM systems (Banks & Kepes, 2015; Lengnick-Hall et al., 2009). The findings here leave open the question of whether cultural distance plays a stronger role in influencing the local customization of HRM practices than HRM policies.

Second, there may be moderating variables that we were not able to include in the current work due to the constraints of the data. For example, the degree to which firms select to send globally standardized versus locally customized recruiting signals could depend on the dynamic capabilities of the firm (see Teece, Pisano, & Shuen, 1997). Firms with greater capabilities may differ in terms of the extent to which they standardize recruiting activities. Third, the present study focused on the use of website text versus other media (e.g., videos, pictures on websites). While the amount of website text was staggering (a total of 2,312,504 words across all three contexts), future research should expand the current work by considering other mechanisms of delivering recruiting signals. We argue that the current work represents, at the very least, the first step in a line of important research. The firms under investigation offered so much text that one could argue that it seems to be the most prominent and consistent form of communication on these websites. Furthermore, empirical data suggest that other media in the form of pictures may not add more value than text and may even detract from signals (Lievens & Slaughter, 2016). Future research should continue to consider these issues.

As a fourth limitation, we focused exclusively on Fortune 1000 firms. It is not clear the extent to which the findings of the current work generalize to the recruiting activities of smaller firms as well as government and nonprofit organizations. Nonprofits, in particular, are known to operate MNEs. Future research should consider the standardization strategies of other organizations that conduct MNEs. Another potential limitation is that the translations of the international foreign language websites via the GNMTS resulted in some discrepancies. These may provide an alternative explanation for the low correlations between the foreign language text and the English language text. However, a recent report illustrated an almost identical translation between humans and the GNMTS (Wu et al., 2016). While it is true that a word could be mistranslated by the GNMTS, it is important to keep in mind that our analysis focused on broader topics rather than on individual words. Thus, if a word is mistranslated and, as a consequence, erroneously assigned to a different topic, that misclassification would not be expected to have a large effect on the conceptual interpretation of the topic or on the topic scores generated in our analyses.