Which Combinations of Human Resource Management and National Culture Optimize Government Effectiveness?

Introduction

As management reform has become globalized in recent decades, human resource management (HRM) has displaced traditional models of personnel administration in public organizations in most industrialized countries (Brown, 2004; Im & Yoo, 2016), introducing the era of new public management (NPM) (Dunleavy & Hood, 1994; Hood, 1991). HRM has so-called performance-bounded characteristics that differ from traditional modes of rule-bounded bureaucracy in that they are intended to create a more flexible and responsive system for the management of public employees. Due to the large and persistent disparities in good government practices across countries (Holmberg & Rothstein, 2012), employing culturally congruent HRM practices would likely enhance government effectiveness.

However, as NPM principles have expanded to different contexts, the transferability of uniform performance-based HRM practices to different cultures has come into question (Cho & Yoon, 2009; Flynn, 2005; Mendonca & Kanungo, 1996; Pillay, 2008; Rhodes et al., 2012). Critics have pointed out that the inherent cultural influence of Anglo-Saxon ideologies on NPM doctrines renders these practices incompatible with the frameworks of developing countries, and even with those of developed countries such as France, Germany, and the Mediterranean states (Earley, 1994; Fletcher, 2001).

Existing studies on the compatibility between management practices and cultures are concentrated in the private sector (Dahlström et al., 2011), as large amounts of data on the subsidiaries of multinational corporations are out there (Howells, 1981; Lindholm, 1999). Also, few countries share common human resource strategies (Teorell et al., 2011), inevitably limiting scholarship to case studies (Adkisson & McFerrin, 2014; Pillay, 2008; Pimpa, 2012; Taylor & Beh, 2013). This paper will fill the gap in the literature by using Hofstede’s four original cultural dimensions (1980): power distance (PD), individualism (IV), masculinity (MAS), and uncertainty avoidance (UA), along with particular management practices like recruitment, appraisal, and compensation.

This study explores the best universal combinations of national culture and human resource management practices to promote government effectiveness, and which HRM practices are most suitable for different cultures. We use “government effectiveness”—which refers to the quality of civil service, policy formulation and implementation, and government credibility and commitment—as an outcome variable and Fuzzy-set qualitative comparative analysis (FsQCA), a method that enables researchers to investigate causal complexity even within small sample sizes, is employed. The expected outcome is identifying possible ideal synergistic combinations of HRM practices and national cultures (Guest, 1997).

The remainder of the paper is organized as follows. First, we discuss the relationships among HRM practices, government effectiveness, and national cultures, drawing largely from international HRM literature. The following sections describe the data and methods used, and justify the use of fuzzy-set analysis. Finally, the results section describes the empirical analyses, focusing on necessary (but not sufficient) and sufficient (but not necessary) conditions, followed by a conclusion providing implications for policy.

Managing People to Develop Government Effectiveness

Government effectiveness has been long on discussion, which have covered a wide swath of topics, such as the characteristics of bureaucracy, quality of government, government competitiveness, and good governance (M. Andrews, 2010; Im & Hartley, 2019; P. Kim & Im, 2019; Lee & Whitford, 2009; Rothstein & Teorell, 2012). With government effectiveness, countries can achieve good performance in infant mortality and literacy (Kaufmann et al., 2009; Tennant & Gilmore, 2020), health and environment (Brooks et al., 2005), citizens’ well-being (Guisan, 2009), and functioning of democracy (Magalhães, 2014). The long discussion seems to have reached a consensus that reform measures for government effectiveness are likely to individually implemented across differing cultural contexts (Hintea, 2020), for good governance can mean different things in different contexts (M. Andrews, 2010).

However, a set definition of performance-based government effectiveness would be helpful to link those varying perceptions into more productive concept that countries can pursue (Rothstein & Teorell, 2012). Lee and Whitford (2009) pointed out that government effectiveness is a slippery concept compared to organizational effectiveness, which is well-understood in the study of management (T. Kim, 2016; Whetten & Cameron, 1991), and tried to define it by interpreting organization-level findings through the lens of effectiveness. According to Rainey and Steinbauer (1999), effectiveness is an indicator that shows whether agencies and their employees perform well and achieve their mission as conceived by stakeholders (p.13). Centering the public interest as a priority complicates the definition of government effectiveness (Ferlie et al., 2002). While effectiveness in organizational studies have focused on goal achievement (Barnard, 1938), we define government effectiveness as the quality of the civil service, policy formulation, and implementation, and the credibility of the government’s commitment to such policies. This definition follows the long discussion on government’s role from Easton, Lipset, Dahl and Linz that formulated the World Bank’s ‘government effectiveness’ indicator (Pedro, 2014). From such a definition, we assume that government effectiveness is a key determinant by which national governments can achieve competitiveness in a globalized era (Im & Hartley, 2019; Pollitt & Bouckaert, 2004).

Given the labor-intensive nature of public organizations, and their responsibility for determining organizational interest, the way manages people would matter for government effectiveness (Im & Yoo, 2016; Pynes, 2009). The role of public staffing policy and corresponding professional competence of governments goes back at least as far as Weber (1922/2013). His well-organized ideological bureaucracy was recognized as a prerequisite for implementing economic and social government policies, linking a well-managed bureaucracy with government effectiveness and performance. Since Wilson’s (1887) worldwide call for comparative studies, numerous investigations have used a range of theoretical insights to determine the key ingredients for effective national governments (Evans & Rauch, 1999; Government Competitiveness Center, 2019), but only a few studies have investigated on HRM practices that differs among countries. One distinctive study is Evans and Rauch (1999)'s work that credited the relative ease of measurement of qualified public employees and effective personnel management for their ubiquity in comparative administration studies. However, including Evans and Rauch (1999), comparative works have focused on the structural dimensions of bureaucracy rather than HRM strategies establishing more flexible and responsive ways to recruit, appraise, and compensate public employees. Meanwhile, micro-level studies have devoted significant interest to HRM practices in order to investigate outcome variables at individual and organizational levels (French & Goodman, 2012).

It partially comes from new public management (NPM) administrative paradigm in recent decades that have been applying one-best-way model reforms (Brunetto & Beattie, 2020). NPM, followed by centre-right government in 1980s, were considered to be a cure to “big governments”, but now it is recognized to be not suitable at times, especially in HRM practices. A focus on measurement may cause goal displacement due to goal ambiguity (Huizinga & de Bree, 2021), and decentralization of authorities may cause a roll-back to bureaucracy (Sakdiyakorn & Voravivatana, 2015).

Government effectiveness depends on management practices that are chosen according to institutional context (Ingraham et al., 2003). NPM has failed to keep pace with the unique administrative traditions of different countries. For example, in NPM paradigm, recruitment process can become informal and diverse. This can result in two ways; in countries where public sector jobs are popular for their job security, diverse recruitment might face corruption issue. For instance, South Korea, public-sector jobs have recently become highly desirable among job seekers, and they consider current standardized tests relatively fair to the recruitment in private sector (E-today News, 2021). In contrast, in countries where public sector jobs are disregarded for their less competitive compensation, applicants may be overwhelmed by the intensity of the application process, which is usually characterized by formality. In the United States, for instance, the lack of attractiveness in federal employment is no doubt in part due to the perceived administrative burden of applying for a federal job (Riccucci et al., 2019). For the former, informal recruitment might lead to fairness issue, on the other hand, for the latter, formal recruitment may hinder the system to sort out the most talented candidates.

Andrews et al. (2019) used survey data measuring perceptions on the efficiency, effectiveness, and equity of healthcare program reforms to discuss how the conditions for what constitutes NPM might vary among 14 European countries. Im and Hartley (2019) asserted the need for a new paradigm for government competitiveness; previous literature is rooted in neo-liberal ideology that limits governments’ roles in assisting market economies. Despite those critics, thanks to the worldwide spread of NPM, HRM practices geared towards government effectiveness resist cultural differences, preferring to conform to market logic and prioritize competition, cost reduction, and outsourcing. In response, this paper investigates possible “better sets” of institutional and cultural context and HRM practices (Flynn, 2005; Mendonca & Kanungo, 1996).

The Fit as Gestalt: Human Resource Management and National Culture

The benefit of good matches between HRM and its applying context may be coined by the term “the fit as gestalt”, which expresses the idea that synergy between workplace cultures and practices can facilitate outcome effectiveness (Guest, 1997). Here, we argue that national culture, a non-replicable element of this dynamic, can provide organizations with a competitive advantage when dealing with human resources. This model implies that aspects of human resource management, such as pay or appraisal, should be considered alongside the cultural context in which they function.

Oddly enough, while “business model” of public HRM being emphasized, a large amount of researches on multinational companies that point out the need for culturally suitable HRM has not been discussed in depth. Multinational companies in the private sector have long struggled to find the right HRM approach for their local subsidiaries (Hofstede, 1980, 1993; Jackson & Schuler, 1995; Schuler & Rogovsky, 1998; Taras et al., 2011). Investigations of the relationship between public management practice and national culture also remain insufficient, as many focus on only one national context in single case studies (Milikić, 2009; Pillay, 2008; Pimpa, 2012). An approach to investigating government effectiveness might lean on the literature that discusses multinational companies’ management practices and national cultures while recognizing the unique nature of the public sector. Rather than provide a description of the interaction between a specific management practice and a specific workplace culture, this study focuses on the possible causal relationship between human resource management practices and effectiveness.

Four Dimensions of National Culture (and Their Critics)

National culture has been called the software of the mind (Hofstede, 1980). Common theories of behavior (Jaeger, 1986, p. 179) have noted differences among citizens of different nations, and related culture to good governance (Adkisson & McFerrin, 2014; Porcher, 2019). Adkisson and McFerrin (2014) explored possible relationships between culture and governance using World Bank’s Governance Indicators of 68 nations, finding that real per capita GDP and cultural dimensions like tradition preference and general tolerance influence good governance. Porcher (2019) also found that, after controlling for institutional and economic differences, elements of national cultures such as embeddedness, autonomy, individualism, and collectivism had a strong impact on the quality of government. Culture also plays a unique role in understanding the effectiveness of HRM practices; scholars have long questioned the generalizability of North American behavior theories (Chiang & Birtch, 2010; Đorđević, 2016; Fletcher, 2001).

Culture can function simultaneously at group, organizational, and even international levels (e.g., ‘European culture’). Scholars have focused on national culture in comparative administration for two reasons. First, national culture is seldom subject to change, and thus functions as a constant rather than a variable.^[1] Birnberg and Snodgrass (1988, p. 448) argued that culture at a national/societal level should be distinguished from culture at an organizational level in that “it is the set of norms and values which the managers and workers bring to the job, rather than the norms and values which management and/or the workers develop in their work environment”. Therefore, to establish national culture as the main predictor of the outcome variable, one should understand existing cultural environments, rather than simply choose the best culture to pursue.

Second, in the aggregate, the precision of national culture makes it a useful key predictor of outcome variables. Taras and his colleagues (2011) argued that culture resembles the weather: predictions of its effects become more precise as they accumulate. One can’t know whether it will rain on a certain day next month, but it is possible to know which months, generally, will be rainy (p.192). Some argue that macro-level administrative tradition can interact with meso-level organizational characteristics (Bach et al., 2020), or that individuals can moderate the relationship between HRM practices and individual performance (Cho & Yoon, 2009), but this weather metaphor demonstrates the significant impact that national culture has on organizational and individual experiences.

Hofstede (1980) identified four dimensions along which national cultures vary: power distance (the extent to which people believe that power and status are distributed unequally), individualism or collectivism (the extent to which identity is derived from the self or the collective), masculinity or femininity (the extent of preference for doing and acquiring rather than thinking and observing), and uncertainty avoidance (the extent to which people are threatened by uncertain, unknown, or unstructured situations).^[2] As the most widely-used classification of its kind (Newman & Nollen, 1996; Schuler & Rogovsky, 1998; Snape et al., 1998; Taras et al., 2011), criticisms of Hofstede’s model should be discussed. McSweeney (2002) criticized its lack of clarity, enriched conception of culture, and overall misguided intention to measure the unmeasurable. However, this critique was partly rooted in functionalist theories, which are not easily used to examine social phenomena like culture (Williamson, 2002). House et al. (2004), while proposing the GLOBE model, claimed that Hofstede failed to measure important cultural dimensions like wealth. Nevertheless, Hofstede (2006) argued that GLOBE’s dimensions do not capture the same phenomena as his concepts despite their homonymous names. Finally, Hofstede’s dimensions might favor Western culture and thinking (Chow et al., 1994). But to our knowledge, no alternative theories exist to rectify this bias. As these critiques limit the studies in this area, it is necessary to draw on multiple methods from several paradigms, including Hofstede’s model, to determine the effect of national culture. Despite criticism, many studies have adopted Hofstede’s framework for use in the public sector (Goldbach et al., 2014; S. Kim, 2017; Pimpa, 2012).

Three Culturally Controversial HRM Practices

The national culture embedded in the theories of managing and organization can neither be divorced from society nor guided by universals. It shapes the behavior and structures of the perceptions of managers, subordinates, and prospective hires. There has been a discussion on whether the practices can be compatible across countries. We explore the three most debatable HRM practices: recruitment, appraisal, and compensation.

Recruitment

Recruitment practices, crucial for hiring the most qualified candidates for government positions, allow potential employees to gauge the cultural environment of a workplace (Oh & Kang, 2021; Prince & Kabst, 2019; Riccucci et al., 2019). Traits of a formalized recruitment strategy could include a formal intermediary, a regular announcement, and a standardized examination. In the traditional Weberian approach, a bureaucratic employment policy should be consistent, formalized, and systematic, with rules and processes (Brown, 2004; Schroeder, 1992). However, according to some scholars, formalized recruitment might not be the best path towards increasing government effectiveness (Im, 2018; Lavigna & Hays, 2004; Riccucci et al., 2019), though its emphasis on official procedures might be more effective in high power-distance cultures (Chang, 2020; Ma & Allen, 2009).

Formality has a different effect on pre-hire outcomes in collectivist and individualistic societies. In collectivistic cultures, job applicants prefer personal or relationship-oriented recruitment sources (Stone et al., 2007). Applicants are more enthusiastic in masculine cultures; in feminine cultures, potential applicants react passively to a routinized annual recruiting process (Ma & Allen, 2009). Accordingly, formal intermediaries and regular announcements of public jobs tend to attract qualified applicants more effectively in feminine cultures. Due to their preference for clear instructions and expectations, job applicants in high uncertainty avoidance cultures feel comfortable in more formalized hiring situations. Using formalized practices, like interviews that include fixed lists of questions, can make organizations more effective in the context of high UA cultures (Ryan et al., 1999; Singh, 2009). In sum, formalized recruitment is expected to lead to government effectiveness in cultures that contain high rates of femininity, individualism, power distance, and uncertainty avoidance.

Appraisal

Performance appraisal, defined by Fletcher (2001) as official mechanisms for evaluating individual employees’ performance, is widely acknowledged to be important for supporting the achievement of organizational priorities (Chiang & Birtch, 2010). Employee perceptions of cultural norms, values, connections, and beliefs strongly influence appraisal systems’ effectiveness, particularly with respect to notions of fairness and satisfaction (Chiang & Birtch, 2010; Jawahar, 2010; Raboca & Dodu, 2019; Thurston & McNall, 2010).

Therefore, a performance appraisal that does not reflect a workplace’s cultural tone can lead to a loss of legitimacy within the organization (Schuler & Rogovsky, 1998). Lindholm (1999) discussed managerial actions to establish accurate performance goals and encourage efforts to meet them in high power distance (PD) cultures, in which upper classes are expected to exercise command on lower classes (Aguinis et al., 2012; Fletcher, 2001). In such cultures, managers who encourage their subordinates to participate in the performance evaluation process may be viewed as incompetent. Huo and Von Glilnow (1995) found that managers in China, a high PD culture, were unlikely to be involved in a mutual appraisal system. Despite its capacity to deliver multi-source feedback, the bottom-up style of performance appraisal might not work in hierarchical culture.

Individualistic and collectivist cultures also require different appraisal methods. An employee’s trust in the rating system conditions their acceptance of an appraisal. In an individualistic culture that emphasizes the self, this level of trust rises when the appraisal centers the concept of individualism. Alternatively, a collectivist culture would prefer a rating system that spreads accountability across the whole team (Chow et al., 1994; Lindholm, 1999).

Highly masculine cultures, with their prevalent desire for career advancement, have reported positive acceptance of performance appraisals, while highly feminine cultures have reported resistance, as their assessments tend to involve interpersonal factors (Mendonca & Kanungo, 1996). Finally, high uncertainty avoidance cultures, such Germany, tend to view formalized performance evaluation and feedback in a positive light (Lindholm, 1999), reflecting their desire to work without constant supervision. When executed correctly, performance appraisal will likely result in government effectiveness, although the degree may depend on the methodology.

Compensation

The widely accepted practice of pay for performance, in which workers receive financial or non-financial rewards from an organization in exchange for their labor, is believed to motivate employees and increase productivity. Increased input from workers about pay will eventually lead to better organizational outcomes (Ljungholm, 2015). Thus, pay for performance establishes practical equity, ensures the continued effectiveness of high-performing employees, and allows other members to learn from them (Gerhart & Milkovich, 1990), although its results may be tempered by intrinsic elements, such as the complexity of public service, multiple principals, multi-task problems and collaborative undertakings, misallocation of effort, and the risks of gaming or cheating.

Aside from sector sensitivity, common doubts about pay for performance are closely related to culture. Cross-cultural studies have shown that people express a preference for the means of reward distribution that match their culture. In high PD cultures, where the position or period of service is most important, people are less receptive to performance-based equity (Martocchio, 2017). Taylor and Beh (2013) revealed that in low power distance cultures, the application of performance-based pay has a positive impact on members’ in-role behavior. Similarly, pay for performance schemes that employ direct feedback would likely be unacceptable to members of a collectivist culture (Elenkov, 1998; Fletcher, 2001), as would bonuses and pay systems based on performance (Muduli, 2011). Meanwhile, highly individualistic cultures would likely favor these methods (Schuler & Rogovsky, 1998).

Newman and Nollen (1996) discovered that countries with highly masculine cultures, such as the United States and Germanic nations, found performance-based pay to be effective, while more feminine cultures responded better to non-performance-based methods. Finally, uncertainty avoidance cultures rarely welcome performance-based pay, due to its uncontrollable environmental effects on performance (Schuler & Rogovsky, 1998). Chow et al. (1994) linked environmental uncertainty to performance-based pay, arguing that Japan, a high UA culture, also prefers predetermined pay. Aoki and Rawat (2020) investigated adoption of performance-based pay across countries, focusing on the education sector, and found out that the technique was more popular among less-liberal economies and in cultures with a lower degree of UA and higher degree of IV. In sum, performance-based pay may produce sufficient conditions for government effectiveness when combined with low power distance, individualism, masculinity, and low uncertainty avoidance cultures.

Figure 1.Framework of External and Internal Organizational Contexts for Government Effectiveness

The study’s framework, shown in Figure 1, sums up our view that combining the external context of national culture with the internal context of public HRM practices results in government effectiveness. Specifically, we view four dimensions in national culture (power distance, individualism, masculinity, and uncertainty avoidance), and three HRM practices (recruitment, appraisal, and compensation) as possible routes towards government effectiveness.

Data and Method

Data

Considering national culture and human resource management as ingredients for government effectiveness presents an important methodological issue: sampling. We took an exploratory approach to the relationship between cultural and management dimensions by using qualitative comparative analysis, which leaves no room for control variables. Our comparative approach to suitable public management was to group countries by their economic scales (M. Andrews, 2010). We restricted our sample Organization for Economic Co-operation and Development (OECD) member countries,^[3] because they are assumed to have less variance in terms of resources (M. Andrews, 2010).

This article uses data from four independent sources: Worldwide Governance Indicators (WGI), the Quality of Government Expert Survey (QoG Expert Survey), Strategic Human Resource Management (SHRM), and Hofstede Insights. The World Bank’s 2016 WGI, which covers 215 countries, pinpoints government effectiveness as an outcome variable. In 1996, the World Bank published survey reports on national perceptions of governance quality in six dimensions, one of which was government effectiveness, representing the quality of public services, the quality of civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies (Kaufmann et al., 2009). Researchers used a 7-point Likert scale to assess government effectiveness; the scale featured six items based on perception (e.g., “Government decisions are effectively implemented”) (see Appendix). Most respondents expertly assessed organizations’ quality of operation and quality (Kaufmann et al., 2009).

While some content measured by the WGI survey does not meet the standard of construct validity (Thomas, 2010), we argue that subjective measures are well suited to study outcomes, since public performance is based on the consensus and perception of multiple stakeholders (Brewer, 2006). Likewise, subjective measures are often observed to be consistent with objective measures, although neither can claim supremacy over the other (Lee & Whitford, 2009). The World Bank has accomplished methodological advancement over time by including hundreds of individual variables from 15 sources (Kaufmann et al., 2009, p. 7). In doing so, it has generated margins of error for each governance estimate, enabling cross-country comparisons. The indices are normalized to fit in a range of approximately -2.5 to 2.5 (with higher numbers representing better outcomes), and confidence intervals are reported along with the indices (M. Andrews, 2010). While some say that evidently coherent theories seldom support countries’ international rankings (Pollitt & Bouchaert, 2004), researchers have used WGI scores to investigate good governance due to their straightforwardness, reliability, and transparency (M. Andrews, 2010; Lee & Whitford, 2009). Researchers also found a relationship between good governance, national culture (Porcher, 2019), per capita income, infant mortality, literacy (Kaufmann et al., 2009), and several health- and environment-related outcomes (Holmberg & Rothstein, 2012).

Second, we utilized the Quality of Government (QoG) Expert Survey and Strategic Human Resource Management (SHRM), to investigate HRM practices and their potential causal configuration sets. The QoG Expert Survey is a unique dataset that enables scrutiny of the structures and behaviors of public administrations across 159 countries. In 2012 and 2015, researchers conducted surveys of government officials, asking them to score public bureaucracies’ degree of politicization, impartiality, and openness. More than 80% of respondents were born in the surveyed country, 76% were 2012 residents, and 91% were 2015 residents. We measured each item of formal examination and performance-based pay on a 7-point Likert scale. Given that fuzzy-set analysis does not allow panel or pooled analysis, we estimated the mean values of the two years’ datasets.

SHRM holistically examines the capacity of government workforces. The OECD’s Directorate for Public Governance conducted a survey of high-level government officials responsible for recruiting and managing public employees. While this data is not often used in quantitative analysis due to its small size (N=30), the fuzzy-set approach enables small-N diversity-oriented studies to develop quantitative analyses (Ragin, 2000, p. 25). Thus, this analysis used performance appraisal variables from SHRM to measure the total normalized value for existence of a formalized performance assessment, use of performance assessment tools, performance assessment criteria, etc.

Third, we derived national culture-related antecedents from data drawn from the Hofstede Insight website. We estimated that the index was within the range of 0-100 and recorded each country’s scores on indexes for the four cultural dimensions: power distance, individualism, masculinity, and uncertainty avoidance. As discussed above, scholars have been using old/dated culture variables because national culture does not change rapidly (Hofstede, 1993). Moreover, partial updates have been provided on the current status of dimensions from 2001, so it seems reasonable for researchers to use those scores at present (Jwijati & Bititci, 2014; Milikić, 2009; Peretz & Fried, 2012; Porcher, 2019). Finally, we combined the above variables together into one dataset.^[4] Table 1 shows descriptive statistics for the variables, and the appendix contains details of the measures in relation to antecedents and outcomes.

Method

To test our core assumption that HRM practices and cultural dimensions have a secure fit to each other to result in strong government effectiveness, we utilized fuzzy-set qualitative comparative analysis, a set-theoretic strategy increasingly popular with social science researchers (Ragin, 1987, 2000). By managing causal conditions and possible outcomes into sets and set-theoretic relationships, a fuzzy-set qualitative analysis overcomes the limits of the dichotomized tradition of variable- and case-oriented studies. The most distinctive feature of this analysis is the visualization of a case combined in different configurations. To discuss a particular outcome, according to the fuzzy-set approach, it is crucial to see the sets of variables rather than each variable’s effect on the outcome. This configurational perspective is based on Boolean algebra, while a variable-oriented approach has its roots in linear algebra. It identifies equifinality, in which more than one combination is linked to the same outcome.

A fuzzy-set qualitative comparative analysis is appropriate for our study for three reasons. First, it empowers researchers to cope with concepts which are typically expressed qualitatively in theoretical discourse but allow for varying degrees, (Ragin, 2000, p. 12) such as variation in the degree of a culture’s individualism. Second, it has strength in presenting complex causality,(Longest & Vaisey, 2008) using configurations and the cases conforming to each configuration as analytic units. Cultural dimensions and management practices may operate simultaneously so that one can act as a substitutes for or complements of the another. Lastly, fuzzy-set qualitative comparative analysis can be performed on a mid-size sample (N=10 to 30). Regardless of its importance, studies on the interaction effect between HRM practice and national culture have been scarce due to methodological limitations. In this study, we used the data from 30 OECD countries, which provide ample evidence of modern HRM practices compared to non-member countries.

In fuzzy-set analysis, the researcher first select cases and define the initial configuration (Ragin, 2000, p. 144). Theoretically, there are possible configurations wherein k would be the number of possible conditions for the outcome. In this stage, fuzzy-set calibration for the outcome and antecedents is processed so that each case can attain a fuzzy-set score ranging from full membership (1) to non-membership (0). This score represents the extent of each country’s membership in the set; a country with a threshold score does not belong to any set. Ragin (2000) introduced various ways to set a threshold value: crisp set style sets (using only 0 and 1), three-value fuzzy-set (using 0, .5, 1) as a rudimentary form, five-value fuzzy-set (using 0, .25, .5, .75, 1), seven-value fuzzy set (using 0, .17, .33, .5, .67, .83, 1), and continuous fuzzy-set. Ragin (2000, p. 166) recommended that researchers should understand theoretical concepts in depth so that they can identify relevant evidence for assessing fuzzy membership scores, and specify appropriate anchors related to full membership, full non-membership, and the crossover point. We employed a three-value fuzzy-set because it is the most intuitive for understanding threshold points. As an explanatory study, we categorized countries into sets without specific qualitative anchors in theory but with standardized means. Therefore, after normalizing the data, we defined fully-in cases as those with a value exceeding 0.95, and set fully-out cases as those with a value of less than 0.05. We calibrated the variables, which resulted in two fuzzy-sets per-variable, and set three threshold values. The calibrated fuzzy-set membership values for each variable are shown in Table 2.

The second step is to use truth tables to test the necessity and the sufficiency of causal conditions. Necessary conditions, which can be tested individually, must be present to produce accurate outcomes. Accordingly, we tested the outcome to develop configurations for every possible combination. First, we made an eight (23)-row truth table with three HRM practices before adding each cultural dimension, thus creating four 1 (24)-row truth tables. As cultural dimensions overlap, we operated the dimensions one-by-one and provided sufficient condition for the outcome. It should be noted that our study proposes a configurational understanding of government effectiveness, while other fuzzy-set studies have focused on identifying ideal case types. Finally, we evaluated the results and looked for simplifying assumptions, the findings of which are located in the discussion section.

Results

There are two primary sets of measures for the assessment of fuzzy-set solutions: set-theoretic consistency and set-theoretic coverage. Consistency refers to the extent to which observations of the same causal combinations lead to the same outcome; the consistency scores for this study are shown in Table 3. The other important measure is coverage, which “expresses how much of the outcome is covered by the sufficient condition” (C. Q. Schneider & Wagemann, 2010, p. 325). According to this definition, a necessity test does not provide a coverage value, as the existence of an outcome is always guaranteed. As no single condition is usually sufficient to generate an outcome, solutions are developed from a combination of conditions at varying degrees of consistency and coverage. The usual cutoff consistency value for a necessity test is 0.9 (M. R. Schneider et al., 2010), and 0.75 or 0.8 for sufficiency(Ragin, 2008).

The consistency scores in Table 3 indicate that the two highest values for consistency, neither of which exceed the standard point of 0.9, are individualistic workplace cultures and performance pay in human resource management practice. Nevertheless, performance pay reaches a generally significant value of 0.73. Notably, power distance and uncertainty avoidance have relatively low scores.

We explored which set of sufficient conditions for management practices allows governments to be most effective. An analysis of sufficiency is presented in Table 4. Practices like formal recruitment, performance appraisal, and performance-based pay cannot produce outcome variables by themselves. As is widely known, a configuration combined with a cultural dimension induces synergistic effects for government effectiveness along with the three systems. Adding the power distance dimension, government effectiveness creates a configuration composed of informal recruitment, substantial performance-related pay, and low PD culture. Nine countries fit into this category: Australia, Denmark, Estonia, Finland, the Netherlands, Norway, Sweden, Switzerland, and the United Kingdom. This configuration has the highest coverage value (0.536). Another configuration for government effectiveness is made up of individualistic cultures with informal recruitment, robust performance appraisal, and substantial performance-related pay: The Netherlands, Sweden, and the United Kingdom fit here. The effects of masculine culture are more complex. In highly masculine cultures, government effectiveness can often result from weak performance appraisals, but in less masculine cultures the opposite is true. As shown in Figure 2, this leads to a climate in which two groups of countries have sufficient conditions for strong government effectiveness: Australia and Switzerland (with masculine cultures), and the Netherlands and Sweden (with feminine cultures). Lastly, in low uncertainty avoidance cultures, three conditions form the proper measure of government effectiveness: informal recruitment, strong performance appraisal, and strong performance-related pay. This configuration shows the highest consistency (0.939). Again, three countries, the Netherlands, Sweden, and the United Kingdom, had sufficient conditions for strong government effectiveness. Of these countries, the Netherlands and Sweden showed strong government effectiveness regardless of the cultural dimension used in their analysis.

Figure 2.Reversed Impacts of Performance Appraisal and Masculinity

FR=Formal Recruitment, PA=Performance Appraisal, PFP=Pay for Performance

Sufficiency plot of the specified configuration; the X-axis indicates government effectiveness and the Y-axis indicates sufficient condition of the configuration

Discussion

Recruitment, appraisal, and compensation systems alone cannot achieve government effectiveness, even in the most industrialized countries. This supports institutionalists’ argument that management practices cannot solve the decoupling phenomena facing the public sector without an understanding of cultural context (Mendonca & Kanungo, 1996; Pillay, 2008). Researchers are encouraged to identify external factors that may influence the success of management practices in different contexts. Aside from power distance, only three practices held causal combinations with added dimensions: flexible recruitment, performance appraisal, and performance-based pay. By our interpretation, the PD dimension did not embrace performance appraisal because of the limited, ambiguous variable used. Further, configuration outputs were associated only with low power-distance, high individualism, and low uncertainty avoidance, all of which were subject to the idea of NPM embedded in management practices (Fletcher, 2001; Ryan et al., 1999; Snape et al., 1998).

When combined with the proper cultural dimensions, the usage of informal recruitment practices consistently results in more potent government effectiveness. This is in line with Sundell (2014), who argued that a formalized recruitment process could be too slow, unresponsive, and inflexible in countries with a prior background in professional merit system domains. Dahlström and colleagues (2011) found that formal recruiting processes in the developed world represent an isolated bureaucracy more than a merit system. Chen et al. (2002) recently reported that overly competitive, overly standardized public service exams might discourage high-PSM people from entering the public sector. As expected, flexible recruitment contributed to government effectiveness in cultures with low power distance, high individualism, and low uncertainty avoidance.

Performance appraisal yielded the least clear results. For example, while the presence of highly masculine cultures shapes the effectiveness of performance appraisals in the private sector, our study displayed the inverse relationship, as shown in the right-side table in Figure 2. We connect these findings to public sector characteristics. Government performance is beholden to public interest and is thus socially constructed by multiple stakeholders (Brewer, 2006). Further, most government performance is not measurable; the production of public services such as national defense and public safety, the qualities of which lie in equity, justice, due process, accountability, and other intangibles, are not easily quantifiable (Naff et al., 2013). Public sector jobs, especially those with high levels of responsibility, tend to be more complex than private sector jobs. Therefore, appraisal reports in the public sector are often subjective and hard to compare. In this light, essential organizational consensus is more easily reached when individuals are less aggressive or confrontational in their relationships. As expected, cultures with high individualism and low uncertainty avoidance adapted well to performance appraisal.

The results of this study’s examination of performance-based pay were consistent with previous literature: scholars have predicted that work units with more merit-based reward practices can perform better in cultures with low power distance, high individualism, and low uncertainty avoidance. As shown in Figure 3, countries that implement performance-based pay show a distinct distribution in terms of national culture. Specifically, countries with high power distance and uncertainty avoidance are distributed in the second and fourth quadrants, whereas individualistic cultures displayed opposite results. Although masculine culture plays a vital role in the effectiveness of performance-based pay, it did not have a significant effect when combined with performance appraisal systems (Newman & Nollen, 1996).

Figure 3.Pay for Performance and Four Dimensions of National Culture

Conclusion

Although ample evidence in the field suggests that national culture can either facilitate or hinder the effectiveness of management practices (Hofstede, 2001; Schuler & Rogovsky, 1998), our trial was the first to seek out sets of causal configurations related to government effectiveness. Our theoretically and practically relevant findings show that flexible, performance-based HRM practices improve government effectiveness if cultural context is taken into account. As in the private sector, more flexible recruitment, performance appraisal, and performance-based pay function better in cultures with strong ties to individuality and high tolerance for uncertainty. This finding supports the results of previous studies, conducted in few countries with a limited scope, but similar links were not found between performance appraisal and masculine cultures. We interpreted this unexpected finding to be a result of the nature of the public sector, in which masculinity constitutes a complex dimension (Im & Hartley, 2019). Future studies need to conduct in-depth examinations of masculinity and its associated characteristics, such as trait competitiveness.

The study was subject to a few limitations. First, the research sample was not necessarily representative of all OECD countries, because the study’s methodology required the exclusion of observations with missing values. Second, some may criticize the omission of major variables, like resources, when identifying which configurational sets were sufficient to promote government effectiveness. Third, we explored the fit between HRM practices and national culture, placing two in parallel, but some might doubt whether suitable practices inherently come from culture that were previously out there. Thus, interaction between cultural dimensions and each HRM practice needs to be explored in a more descriptive way. Fourth, we used datasets come from different years, which is sometimes acceptable in country-level studies but obviously not the best way, so the causal relationship might be questioned. We hope that more country-level HRM datasets be available in the future. Fifth, we draw our logic mostly from private sector literature but with clarifying the distinction between sectors, yet some explanation of results might be apart from the actual conditions of public administration. These limitations could be best addressed by future studies.

Acknowledgements

This study was supported by a National Research Foundation of Korea Grant from the Korean Government (NRF-2017S1A3A2065838)