The purpose of this paper is to investigate the influence of national culture on the adoption of four different incentive pay bundles (incentive maximizer, contingent rewarder, profit rewarder, and incentive minimizer) using GLOBE national culture dimensions in 14 countries. It uses incentive pay bundles derived by Prince et al. (2016).
The study adopted multilevel random-intercept logistic modeling using firm incentive practice usage from the CRANET database and country culture scores from the GLOBE study.
Evidence suggests that in-group collectivism is associated with increased use of the incentive maximizer approach, in which firms use a combination of high levels of individual, team, and profit sharing incentives, and decreased use of the incentive minimizer approach (where firms minimally employ incentives) and the individual and team bonus focused contingent rewarder configuration. Higher uncertainty avoidance is linked to increased use of the profit rewarder approach (where only profit sharing is emphasized) and decreased use of the contingent rewarder approach. Performance-orientation cultures appear to support using the incentive maximizer and avoiding the incentive minimizer bundles.
This study investigates incentive practice bundles that firms use verses separate analysis of practices and use the GLOBE culture metrics. It utilizes multilevel modeling, which has been lacking in past studies of culture and incentives.
Prince, N., Prince, J. and Kabst, R. (2018), "Incentive pay configurations: the influence of national culture", Evidence-based HRM, Vol. 6 No. 2, pp. 187-202. https://doi.org/10.1108/EBHRM-12-2017-0059Download as .RIS
Emerald Publishing Limited
Copyright © 2018, Emerald Publishing Limited
The adoption and diffusion of organizational practices has long been viewed through an institutional theory lens, in which the central argument is that organizations in the same institutional environment conform to institutional pressures (Dimaggio and Powell, 1983; Scott, 2008; Kostova and Roth, 2002). National culture, the shared values, motives, and beliefs that people from a common geographic region share, is an institutional pressure that evolves from the cultural-cognitive institutional pillar (Scott, 2008) and influences adoption of organizational practice (Hofstede, 1980; House et al., 2004). This is especially true in regards to the human resource (HR) practices that organizations implement, as national culture is one of the core organizing principles that employees use to shape their expectations of work (Newman and Nollen, 1996). From the national culture literature, “management practices should fit the local culture” (Hofstede, 1993). The underlying logic of this statement is that if cultural-based expectations are too much at odds with HR practices implemented, then expected outcomes will not be achieved. The general finding is that national culture will influence organizations’ use of HR practices (Newman and Nollen, 1996; Schuler and Rogovsky, 1998; Brewster and Mayrhofer, 2012; Tosi and Greckhamer, 2004; Farndale and Sanders, 2016).
This is consistent with broader views of the institutional environment’s influence on HR practice adoption and effectiveness. Rizov and Croucher’s (2009) research used the varieties of capitalism lens and found that calculative practices (which focus on individual incentive pay schemes) are less successful in coordinated market economies (CMEs) than liberal market economies. In one country (Denmark), the use of calculative practices had a negative relationship to organizational performance. These results highlight the importance of viewing pay practices broader than the often noted individual incentive approach. They found that group-based practices (which included profit sharing and share schemes), which are different than the calculative approach, had more support in CMEs. The Gooderham et al. (2015) research on individual incentive pay practices across 26 countries found that both culture and labor regulation (laws supporting employment security and collective bargaining) influenced the adoption of incentive practices. Gooderham et al. (2015) found that both culture and labor regulation (laws supporting employment security and collective bargaining) influenced the adoption of incentive practices.
The configurational approach challenges researchers to move beyond testing one HR practice as a dependent variable separate from other related dependent variables to focus on multiple combinations (or bundles) of practices that provide synergistic effects beyond the additive effects of individual practices (Delery and Doty, 1996). The configurational approach has often focused on “ideal types” where multiple practice combinations conform to a common overall organizing principle this assumes that greater conformity to an ideal type provides synergies leading to higher performance, yet not all research supports this assertion (i.e. Verburg et al., 2007).
We find merit in focusing on empirical HRM practices-in-use bundles. Organizations use different combinations of practices for different reasons, and these choices may or may not conform to some ideal type logic. As noted by Jiang et al. (2012), “Strategic human resource management (HRM) researchers argue for a focus on the bundles of human resource (HR) practices in place, rather than individual practices, as a primary unit of analysis when examining the impact of HR systems on individual and organizational performance” (italics added, p. 73). Studies of HR practices-in-use configurations (such as high performance work systems) have often been studied as a unidimensional construct that include a configuration of innovative HR practices and have focused primarily on how this configuration influences performance, productivity, and turnover (Huselid, 1995; Jiang et al., 2012). A study within this stream of HR research found that the institutional context (i.e. economic growth) an organization is embedded in will influence it to adopt an innovative HR configuration (Lawler et al., 2011). Similarly, organizational context, such as size, industry, etc., has been found to influence the configuration an organization will implement (Toh et al., 2008). Recently, a study by Prince et al. (2016) identified a variety of incentive pay bundles and found significant country differences in incentive configuration adoption. These studies have highlighted two key points: organizations employ different bundles of HR practice, and that the institutional environment will influence HR bundle adoption. This study seeks to extend this prior work by investigating one aspect of the institutional environment, national culture.
2. National culture and incentive pay bundle adoption
2.1 Incentive pay practices
Incentive pay practices are used to attract, motivate, and retain employees (Nyberg et al., 2013) and are part of total compensation. Incentive pay practices can payout rewards for performance at the individual, group, or organizational levels (Schuler and MacMillan, 1984). An individual bonus pays an individual when they reach a specified level of performance. Group bonuses are based on performance attained by a group, team, or department within an organization and profit sharing is based on organization-level performance.
Theoretically, incentive pay practices perform multiple functions in an organization (Gerhart et al., 2009). The most common function that an incentive pay practice performs is to align employee behaviors with organizational requirements (Stajkovic and Luthans, 2001), but they also act as a sorting mechanism (Gerhart et al., 2009; Cadsby, et al., 2007). This sorting function is explained by Schneider’s (1987) attraction-selection-attrition model, where employees are attracted to organizations that fit their values and leave those that do not. HR practices are seen as signals of the broader nature of the organization (Stajkovic and Luthans, 2001). The logic of incentive pay is that employees will be attracted to organizations whose practices align with their values and will do what their organization asks because they value the contingent rewards offered.
2.2 Incentive pay practice bundles
Organizations employ a variety of incentive pay practices in their compensation system to incentivize multiple behaviors as opposed to one or two behaviors related to a single practice (Gerhart et al., 2009). Multiple practices are often used by organizations to reinforce the importance of both individual and collective behaviors necessary for the organization to achieve desired levels of performance (Long and Fang, 2012). To better understand incentive pay practices and their effect, it is important to understand the various bundles of incentive pay practices that organizations employ and factors that influence the adoption of particular bundles (Rynes et al., 2005; Delery and Doty, 1996).
Few studies have looked at incentive pay practices as configurations. Toh et al. (2008) found that context such as organizational values and structure as well as different incentive pay practices is included in HR practice bundles organizations adopt. Similarly, Prince et al. (2016) identified four incentive pay bundles-in-use and found country-level differences across countries in their comparative study of incentive pay bundles. The Prince et al.’s (2016) study found that firms tend to adopt one of four different bundles of incentive practices: incentive maximizer, contingent rewarder, profit rewarder, and incentive minimizer. The incentive maximizer bundle uses individual bonus, team bonus, and profit sharing for all employee groups. The contingent rewarder has broad use of individual and team bonus incentive practices, but not profit sharing. The profit rewarder had the opposite profile and emphasizes wide use of profit sharing while minimizing individual and team incentive schemes. The incentive minimizer exhibited lower use of all incentive practices within its workforce. While it is interesting that firms from different countries have strikingly different adoption patterns, this observation leaves unaddressed institutional forces across countries that explain these patterns. National culture represents a likely explanatory factor.
2.3 National culture
National culture comes from the cognitive-cultural pillar of the three institutional pillars that shape behavior at the institutional level, the other pillars being normative and regulatory (Scott, 2008). Culture refers to the “motives, values, beliefs, identities and interpretations of meanings” that people from a common society share and is transferred across generations (House et al., 2002, p. 5). It establishes assumptions that underlie employees’ evaluations as they interact with each other and their organization (Hofstede, 1980; Hofstede and Hofstede, 2005).
While there is evidence that context matters and the use of HR practices will vary between countries (Paauwe and Paul, 2008), there is a dearth of research on the effect of national culture on organizations’ use of incentive pay practices, and currently, no studies look at the influence of national culture on incentive pay bundles. Schuler and Rogovsky (1998) and Tosi and Greckhamer (2004) use Hofstede’s culture dimensions to investigate the organizational use of incentive pay. While Tosi and Greckhamer (2004) provided insight on how CEO incentive pay varies depending on culture, Schuler and Rogovsky (1998) provided information on how particular incentive practices are related to Hofestede’s culture dimensions. In the section below, we will develop hypotheses and provide rationale for investigating three different dimensions of national culture.
2.4 National culture and incentive pay practice bundles
National culture influences how people think, what they value, and how they act (Oyserman, 2011) and, thus, how employees in a country respond to different incentive bundles. Employees will be motivated to work harder and tend to stay or be attracted to organizations, where the incentive system matches their cognitive schema. Conversely, bundles that conflict with their schema will be seen as illegitimate and produce negative responses. When incentive pay practices are consistently used, often as a result of cognitive and normative values embedded in national culture, these practices become the established way of doing things (North, 1990). Deviation from accepted practices undermines management’s legitimacy (Lees, 1997). To avoid this, organizations maintain the status quo and implement “accepted” practices. As this process is repeated by multiple organizations overtime, norms of usage are strengthened making it harder for organizations to deviate from established practice.
Two culture dimensions used in this study, in-group collectivism and uncertainty avoidance, were used in studies that substantiated a link between national culture and incentive pay practice use (Schuler and Rogovsky, 1998; Tosi and Greckhamer, 2004). Additionally, the GLOBE study added the national culture dimension of performance orientation. Though not previously tested in relation to incentive pay, this dimension measures the extent that society values rewarding people for their work. Since incentive pay is one of the primary mechanisms that organizations use to reward employees, it is expected that performance orientation will influence organizations’ use of incentive pay. We investigate the link between incentive pay practice bundles identified by Prince et al. (2016) and national culture with three different dimensions of national culture from the GLOBE study: in-group collectivism, uncertainty avoidance, and performance orientation.
2.4.1 In-group collectivism
This is defined as the degree to which employees express pride, loyalty, and cohesiveness in their organization (House et al., 2004). This corresponds to Hofstede’s individualism scale. In contrast to the GLOBE institutional collectivism metric, House et al. (2004) observed that “the In-Group Collectivism practices scale is the most highly correlated with other well-established societal-level collectivism scales in the literature” (p. 475). Institutional collectivism focuses on strong feelings of national pride and concern for the greater whole at the societal level as opposed to organizational or group levels. The incentive pay practices that organizations implement are a strong indicator of their values (Stajkovic and Luthans, 2001). Organizations in countries with higher levels of in-group collectivism will reflect this value by having compensation practices that reward collective performance, such as compensation based on organizational-level performance (Kuhn, 2009). In low collectivist (or individualistic) cultures, compensation is more likely to follow the “model in which individuals are rewarded in direct relationship to their contribution to the success of the task” (House et al., 2004, p. 453). This is consistent with the Tosi and Greckhamer (2004) finding that a country’s level of individualism is positively related to increases in the proportion of variable pay granted to CEOs. Collectivist societies have more institutional pressure to treat employees equally and avoid differential individual rewards. Organizations in these cultures are more likely to face employee resistance and not gain the legitimacy necessary to operate (North, 1990). Profit sharing is a unique incentive in that it rewards broad organizational performance to participating employees. Both the profit rewarder and incentive minimizer are consistent, while the contingent rewarder and incentive maximizer bundles are contrary with this cultural dimension:
Organizations in countries higher on in-group collectivism will be more likely to use the profit rewarder and incentive minimizer bundles and less likely to use the contingent rewarder and incentive maximizer bundles.
2.4.2 Uncertainty avoidance
This is the degree to which societies avoid uncertainty by relying on social norms, rituals, and bureaucratic practices to alleviate unpredictability (House et al., 2004). Schuler and Rogocsky (1998) found that higher levels of uncertainty avoidance are associated with decreased use of individual incentive pay practices. Similarly, Tosi and Greckhamer (2004) found that increases in country’s uncertainty avoidance levels were related to decreases in the proportion of CEO’s variable pay relative to total compensation. Incentive pay practices present managers and employees with uncertainty and this conflicts with a societal emphasis on a “structured, predictable, orderly, and consistent” life (House et al., 2004, p. 619). When incentive pay practices are used, there is more uncertainty about pay as well as uncertainty for managers in projecting expenses. It has been found that contingent pay will undermine employee motivation and performance when employees are risk averse. Violating expectations by implementing contingent rewarder practices such as individual, or team performance bonuses in high uncertainty cultures may lead to negative motivational effects and result in discontinuing these practices. Profit sharing is generally a less salient part of total compensation and may be less likely to violate these cultural norms and create discomfort:
Organizations in countries higher on uncertainty avoidance will be less likely to use incentive pay bundles that use individual and team bonuses such as the contingent rewarder or incentive maximizer, and more likely to use the incentive minimizer configuration and profit rewarder bundles.
2.4.3 Performance orientation
This culture dimension emphasizes that societies should reward people for performance improvement and excellence (House et al., 2004) and has an obvious logical connection to incentive pay. Since this dimension is not clearly identified in Hofestede’s dimensions, there is less research to draw on. It has been noted to have a connection to Hofstede’s masculinity dimension, which was significantly related to the use of individual pay-for-performance in the Schuler and Rogovsky’s (1998) study. Members of societies high on performance orientation expect to be rewarded for achievement and performance improvement. As a result, pay-for-performance practices such as individual and group-level bonus plans will be attractive to employees in those countries and be adopted more frequently as these practices explicitly connect performance to compensation (Trevor et al., 1997) and act as a feedback mechanism (Stajkovic and Luthans, 2001). This suggests that performance orientation will be positively related to the use of the incentive maximizer and contingent rewarder bundles, as both use individual and group bonuses, and, conversely, negatively related to incentive minimizer configuration usage. Profit sharing, which rewards aggregate employee performance, is not likely to provide an explicit connection between individual behavior and pay that is acceptable in countries high on this dimension:
Organizations in countries higher on performance orientation will be more likely to use the incentive maximizer and contingent rewarder bundles, and less likely to use the incentive minimizer and profit rewarder bundles.
3. Data and methods
Incentive pay practice data are drawn from the 2009 CRANET database (Parry et al., 2011). The CRANET survey dates back to 1989 (Steinmetz et al., 2011). It was originally designed to gather data from firms larger than 100 employees across several countries. It has grown in scope and coverage of countries and now includes smaller firms in many countries and is designed to draw representative samples from each participating country (Steinmetz et al., 2011). This results in the coverage of a wide range of industries. Senior level HR officers of firms are the typical respondents. The availability of GLOBE culture metrics and the focus on extending the Prince et al.’s (2016) study on country differences dictated the use of the same 2,396 private sector firms from 14 countries included in that study.
3.1 Incentive pay bundles
The independent variables in this study are four different pay bundles identified in the Prince et al.’s (2016) study. They used confirmatory factor analysis (CFA) to identify three formal incentive systems (individual bonus, team bonus, and profit sharing), which measured the extent that each system is used in four employee groups (managers, professional/technical, clerical, and manual employees). Each incentive bundle scale ranges from 0 to 1.0 indicating the proportion of all four employee groups formally using the system. A score of 1.0 indicates that a given incentive practice (e.g. individual bonus system) is formally used in all four employee groups. The CFA analysis provided acceptable model fit indices (GFI=0.94, AGFI=0.91, NFI=0.95, CFI=0.96, and RMSEA=0.068). Each indicator has significant regression weights and load on a corresponding latent variable. Standardized α reliability coefficients ranged from 0.92 for the profit sharing scale to 0.80 for the individual bonus scale. Ward’s hierarchical cluster analyses (using the squared Euclidian distance measure) identify sub-clusters of cases similar in the pattern of incentive practice adoption. The agglomeration index indicates the degree of homogeneity within clusters and guides the identification of the number of clusters evident in the data. The K-means cluster was then used to assign cases to the specified number of clusters. Four distinct incentive pay bundles were identified: incentive maximizer firms with higher than average use of all three incentive practices, incentive minimizer firms having below average levels of all three incentive practices, contingent rewarder firms using relatively higher levels of individual and team bonus practices combined with low use of profit sharing, and profit rewarder firms utilizing lower levels of team and individual bonus practices combined with above average profit sharing adoption. These bundles used in this study use the same data to understand the influence of national culture on bundle adoption.
3.2 National culture dimensions
The GLOBE culture dimensions of in-group collectivism, uncertainty avoidance, and performance orientation serve as independent variables (House et al., 2004). We use the response bias-corrected “as is” culture practice metrics. These dimensions are more appropriate for use in this study than the “should be” practices because views of legitimacy are more likely to be based on practices-in-use as opposed to “should be” culture values.
3.3 Control variables
To control for potential spurious results, we include these organizational-level control variables:
Union intensity measures the percentage of the workforce that is unionized. This single-item ranges from 1 (0% union members) to 6 (76-100% union members).
Log of size is the log of the total number of firm employees. The log transformation makes the scale closer to a normal curve distribution.
International product market focus is a single-item indicating whether the main market for the firm’s products is the firm’s home country (coded 0) or international markets (coded 1).
Service versus manufacturing indicates the main sector of a firm’s industry. Service oriented firms are coded 0 and manufacturing firms are coded 1.
As noted in a meta-analytic review of Hofestede’s culture dimensions (Taras et al., 2010), the lack of multilevel analysis in past culture studies is a serious problem. Since firms (level 1) are nested within countries and culture dimensions are country-level variables (level 2), and the dependent variables (incentive bundles) are dichotomous, multilevel random-intercept logistic modeling is appropriate. At step 1, a null model (or intercept-only model) is used to identify the intraclass correlation coefficient (ICC-1, or ρ) for each incentive configuration. This identifies the proportion of variation in incentive bundle adoption that is at the country level (between groups). Step 2 enters firm-level control variables. Step 3 adds country-level GLOBE culture practice variables. Both the statistical significance of entering variables and reductions in country-level variation (effect size) are important considerations.
The number of countries (n=14) compared to the number of predictors in the model (four firm-level controls and three hypothesis-based culture variables) raises statistical concerns. Having too many predictors with 14 country-level observations can create unreliability. An additional set of analyses, which include only the three national culture dimensions, partially addresses this concern.
Tables I-III provide descriptive statistics. These tables present the means, standard deviations, and intercorrelation matrix for all variables used in the analyses. Table III provides a listing of the number of firms and each culture dimension score for each country included in our analysis.
Table IV presents random-intercept logistical models using each incentive bundle as the dependent variable and the three GLOBE culture independent variables included in our hypotheses. Step 1 ICC-1 coefficients (or ρ) represent the proportion of total variation that resides between (vs within) countries. These range from ρ=0.11 (incentive minimizer) to ρ=0.37 (incentive maximizer) indicating that further multilevel analysis is merited. There is a pattern of slight reductions in ρ values at step 2 as controls are entered, and more substantial reductions in ρ as statistically significant GLOBE culture variables are added at step 3.
4.1 Incentive maximizer
For the incentive maximizer bundle, entry of controls provided a 0.01 reduction in country-level variation associated with adopting this configuration (step 2 ρ=0.36 vs step 1 ρ=0.37 results in Δρ=0.01). None of the control variables are significant. In step 3, in-group collectivism and performance orientation are statistically significant. The 5.44 odds ratio for in-group collectivism indicates that a one standard deviation increase of in-group collectivism is associated with a 444 percent increase in the likelihood of an average firm being an incentive maximizer. This was not anticipated by H1. Consistent with H3, a one standard deviation increase in a country’s performance orientation (odds ratio=9.47) increases the likelihood of a firm being an incentive maximizer by 847 percent. With step 3 ρ-value of 0.20 and Δρ=0.16, we see a 44.4 percent reduction ((0.36−0.20)/0.36) in the country-level variance component when we account for these culture dimensions. The analysis without controls provides a robustness check and a check on statistical power concerns (given the level 2 sample size). These results are presented in the second data column. Increases of in-group collectivism (odds ratio=6.23) and performance orientation (odds ratio=6.61) significantly increase the likelihood of being an incentive maximizer. In this reduced model, the ρ-value dropped from 0.37 to 0.19 which is a 51.4 percent reduction ((0.37−0.19)/0.37) in country-level variation. Both of these models indicate that there is plenty of additional country-level variation to be further evaluated. But, two culture dimensions explain a significant portion of country-level variance.
4.2 Incentive minimizer
There is much less country-level variation associated with adopting the incentive minimizer bundle (step 1 ρ=0.11). Adding firm-level controls reduces this (ρ=0.10). Increases in firm size and being in the manufacturing sector (vs service) reduce the likelihood of the average firm selecting this bundle. Conversely, increased union intensity increases the likelihood of being an incentive minimizer. In step 3, two culture variables are significant. A one standard deviation increase of in-group collectivism (odds ratio=0.62, p⩽0.05) decreases the likelihood of a firm adopting this bundle by 38 percent (1.0-0.62). This is the opposite of the relationship anticipated by H1. An increase in performance orientation (odds ratio=0.31) is associated with a 69 percent (1.0-0.31) decrease in the likelihood of firms using this bundle. The drop in ρ from 0.10 to 0.02 (Δρ=0.08) represents an 80 percent reduction ((0.10−0.02)/0.10) in country-level variance and indicates that there is not sufficient country-level variation to proceed with further multilevel analysis. The reduced model confirmed these same relationships, and ρ drops to 0.03, indicating a 72.7 percent decrease ((0.11−0.03)/0.11) in total country-level variation.
4.3 Contingent rewarder
With the null model, we see 19 percent (ρ=0.19) of the variation in adopting a contingent rewarder configuration associated with country-level factors. Increases in firm size are associated with increased use of this incentive configuration. Our results indicate that a one standard deviation increase in a country’s in-group collectivism (odds ratio=0.41) decreases the likelihood of using this bundle by 59 percent (1.0-0.41). This relationship was not expected. The hypothesized positive relationship with performance orientation was not supported. The negative relationship with uncertainty avoidance anticipated by H2 was supported. The odds ratio of 0.24 means that a one standard deviation increase in a country’s uncertainty avoidance decreases the likelihood of the average firm adopting the contingent rewarder bundle by 76 percent and step 2 ρ=0.10 value means that accounting for these culture dimensions reduced country-level variation by 47.4 percent ((0.19−0.10)/0.19). The reduced model without controls found the same significant culture predictors and achieved a ρ-value of 0.13. While there remains country-level variation to be evaluated, two culture dimensions contribute to explaining a substantial amount.
4.4 Profit rewarder
The null model indicates 22 percent (ρ=0.22) country-level variation in being a profit rewarder. Accounting for firm-level controls reduces this to 20 percent (ρ=0.20). Increases in firm size (odds ratio=0.87) decreases the likelihood of using this configuration. Being in the manufacturing rather than the service sector (odds ratio=2.25) increase the likelihood of being a profit rewarder. H1, predicting a positive relationship with in-group collectivism, was not supported. Only uncertainty avoidance showed a trend-level relationship (odds ratio=2.80, p⩽0.10) suggesting that increases in this culture dimension may increase the likelihood of being a profit rewarder. The drop in explained country-level variation (ρ=0.14, Δρ=0.06) indicates a 30 percent ((0.20−0.14)/0.20) reduction in country-level variation. The model with only culture variables (no control variables) produced similar results but uncertainty avoidance failed to reach even trend-level (p=0.05⩽0.10) significance and this lessens the confidence that culture has an influence on the profit rewarder bundle adoption.
4.5 Summary of hypothesized results
These results provided a mix of support and non-support for our hypotheses. H1 argues that increases in a firm’s country-level in-group collectivism are associated with increased use of the profit rewarder and incentive minimizer bundles and a decrease in the use of the maximizer and contingent rewarder bundles. Only the negative relationship between the contingent rewarder bundle and a country’s in-group collectivism was supported. In fact, contrary to this hypothesis, the likelihood of increased adoption of the incentive maximizer and decrease use of the minimizer bundles are associated with increases of in-group collectivism. H2 contends that higher uncertainty avoidance will decrease the likelihood of firms pursuing contingent rewarder and incentive maximizer bundles of incentives, and increase the probability of being an incentive minimizer or profit rewarder. While increases in uncertainty avoidance are associated with the decreased use of the contingent rewarder approach and there was a trend-level increase in the profit rewarder (in one not both equations), it is not significantly related to the adoption of the incentive maximizer or minimizer bundles. Finally, H3 argues that increases in a country’s performance orientation will increase the likelihood of being an incentive maximizer and contingent rewarder and decreases the likelihood of being an incentive minimizer and profit rewarder. The results confirm the predicted relationships with the incentive maximizer and incentive minimizer bundles, but do not support the relationship between performance orientation and either contingent rewarder or profit rewarder adoption.
5.1 Theoretical implications
This paper helps to expand the academic conversation past the country comparison conducted by Prince et al. (2016) by focusing specifically on the effect of national culture. Culture matters, but all dimensions of culture are not created equal in terms of their influence on incentive pay bundle adoption. The results for in-group collectivism support that it is linked to decreased use of incentive minimizer and contingent rewarder bundles and increased use of the incentive maximizer. It appears that increases in a culture of collective pride, loyalty and cohesiveness at the group level support the simultaneous use of incentives at all levels (individual, group, and organizational performance) but are inconsistent with the more piecemeal and individualistic incentive structure found in the contingent rewarder. Not linking pay to performance (using the minimizer configuration) is more supported by lower in-group collectivism cultures. Historically, high individualism cultures (which include low in-group collectivism cultures) have been seen as supporting individual performance pay schemes. Our results with the contingent rewarder configuration are consistent with this thinking, but present a more nuanced view by suggesting that providing a broad package of multilevel performance-based rewards (the incentive maximizer bundle) is acceptable in higher in-group collectivism (low individualism) cultures. There is a r=−0.19 relationship between institutional collectivism and in-group collectivism for the “as is” sample among all countries in the GLOBE study (House et al., 2004) and this cautions against the conclusion that all forms of collectivism are positively associated with incentive usage. The distinction that the GLOBE framework makes between institutional collectivism and in-group collectivism (in contrast to Hofestede’s individualism-collectivism dimension) is clearly helpful in advancing our understanding. More research is needed to fully understand the unexpected finding with in-group collectivism, but clearly researchers need to move beyond the simple unidimensional individual vs collectivism logic and account for the two forms of collectivism.
Increased reliance on social norms, rules, and procedures to lessen the discomfort of future uncertainty seen in higher uncertainty avoidance cultures leads to decreased use of the individual and group bonus approach (contingent rewarder). This is likely due to managerial uncertainty in projecting expenses associated with individual and work-unit performance-based incentive practices and employee discomfort with uncertainty about compensation that these systems introduce. The adoption of other incentive bundles, with the possible exception of the profit rewarder bundle, is not influenced by cultural uncertainty avoidance levels.
Additionally, increased emphasis in competitiveness and rewards for innovation, excellence, and improvement, as seen in higher performance-oriented cultures, supports the simultaneous use of multiple incentives to payoffs for multiple types of performance (the maximizer configuration) and is contrary to not linking pay and performance (i.e. the minimizer approach). This is a natural extension of the importance placed on recognizing performance and achievement. It is surprising that the strong and consistent support for performance incentives was only seen with the adoption of a more multi-part performance-based bundle (incentive maximizer). The more limited individual and team bonus approach (contingent rewarder) or profit sharing only approach is not related to the performance orientation of the firm’s country, a distinction not previously reported in the literature. This suggests that different incentives do not substitute for one another but are additive or interactive in uniquely responding to the demands of higher performance-orientation cultures (Jiang et al., 2012).
Some researchers, such as Gerhart and Fang (2005), have questioned the utility of country culture in understanding the institutional context of HRM. This research demonstrates that, at least for the GLOBE approach, selected culture dimensions have both statistical and practical significance in explaining incentive pay configuration adoption. These results provide support for the viability of the GLOBE approach to culture. Their criticisms of the culture construct focused on limited effect sizes of Hofestede’s metrics, and this is clearly not evident in our results. We should not let Hofestede-based results sink the larger culture construct in HRM research. While more research on a broader range of HR practices and culture must be conducted, these results should encourage further research using the GLOBE approach to culture.
Finally, the results also provide a measure of support for the usefulness of a configurational approach to understanding incentive pay practices (Jiang et al., 2012; Delery and Doty, 1996). Using bundles and understanding the relationship of multi-practice bundle adoption patterns have provided a more fine-grained understanding of what cultures support these combinations. Sequentially analyzing each incentive practice (individual bonus, team bonus, and profit sharing) as single dependent variables would not have helped us understand the more complex relationship between incentive system adoption and country culture. While the identification of HR practice bundles beyond incentive pay choices is needed to advance research using a configurational perspective, this research provides support for this endeavor.
5.2 Practical implications
One of the constraints of the Prince et al.’s (2016) study that first identified these bundles is that it could only provide practical implications for the 14 countries that were a part of the study. By focusing on culture instead of a country-level comparative analysis, MNCs can apply the findings from this paper to guide the implementation of incentive pay configurations in all countries that have GLOBE ratings. For organizations to be successful, they need to craft incentive pay bundles that are appropriate for their institutional context and concurrently fits their strategic goals. Balancing strategic and contextual constraints is complex, and these results provide guidance for these decisions from a cultural perspective. Country culture gives us insight into the implicit assumptions and expectations that country-level employees tend to have and ignoring these when selecting incentives can lead to employee resistance and lack of acceptance. This undermines effectiveness. This research provides guidance as to which incentive bundles are most consistent with particular culture dimensions. For example, firms in countries high on in-group collectivism or performance orientation should appreciate that there is a cultural norm to offer the incentive maximizer configuration and avoid using the incentive minimizer configuration. The adoption patterns seen here suggest that going half-way and introducing a single incentive practice, such as an individual or team-based bonus system or a profit sharing system or avoiding these incentives, will not be positively received. Failure to understand these subtleties and not being sensitive to the institutional context can lead to a loss of legitimacy and employee resistance. Similarly, the contingent rewarder strategy appears less-viable in countries with high in-group collectivism and/or high uncertainty avoidance. Profit sharing may be the single best option for introducing incentives. It is clearly possible for firms to negotiate and overcome culture-based barriers to implementing incentive pay bundles at odds with a country’s cultural norms, but following this path must be done with more care and deliberateness than is the case for firms that conform to cultural-based pressures. This research provides a map of specific cultural contexts where care and caution are appropriate before proceeding.
While this research is unique in its use of multilevel analysis and the number of countries and world regions included, the use of only 14 (level 2) countries creates statistical power limitations. A larger sample of countries and world regions, particularly the inclusion of firms from China, India, and South America, would provide fewer statistical constraints and more confidence in generalizing these results. Response rate of most multi-country large-scale surveys and the attendant possibility of response bias is a general a concern and the CRANET survey is no exception. Not all participating countries provided response rate information, but of those that did the reported response rate is 9.8 percent. While this research is unique in its use of GLOBE dimensions (vs Hofestede metrics, which have been commonly used), there are additional viable approaches to culture and additional institutional factors beyond culture that have conceptual relevance to incentive pay. A larger level 2 (countries), sample size allows the inclusion of additional country-level factors to be jointly considered with the nine GLOBE dimensions. This would strengthen confidence that the results are not spurious in the sense that culture could simply be correlated with more fundamental causal institutional factors. We also appreciate that identifying cross-sectional adoption patterns does not offer the same strength of insight as a longitudinal design that follows the implementation of incentive bundles in multiple countries.
5.4 Future research implications
Delery and Doty (1996) are some of the earliest advocates of studying HR bundles. But there have been a few studies that have empirically investigated bundles, and this remains a nascent area of study in HR. Jiang et al. (2012) provided useful theoretical insights that can be used in understanding how HR practices will influence each other and provide an additive or interactive effect through the use of different combinations of practices. Identifying HR practice bundles across other HR functional areas, such as training and development, selection, performance evaluations, etc., that emerge as bundles-in-use and their relationship to culture is a logical next step. Incentive bundles are a starting point for that investigation.
The effect of national culture dimensions and other institutional factors on incentive and more expansive HR practice bundles will broaden our understanding of global HR practice configuration adoption and help us understand the influence of the institutional environment on these patterns. This has obvious implications to the study of high performance work systems and suggests that rather than conceiving of high performance work systems as a single continuum on which organizations fall, there may be multiple forms of high performance HR bundles that have differential adoption patterns and effectiveness across countries and institutional environments.
Additional research should move beyond adoption rate as the dependent variable. One implication of the culture influenced adoption patterns seen in this research is that different cultural factors may help some HR bundles more or less effective in achieving different organizational outcomes. Outcomes could range from economic performance to employee relationship outcomes, such as psychological contract violations, commitment, or withdrawal (Thomas et al., 2010). In this approach, culture becomes a potential moderator of an HR bundle (or practice) relationship to organizational outcomes.
Culture is one institutional factor and understanding combined influence of other factors relevant to incentive system or broader HR bundle adoption is important in understanding the context in which such managerial decisions are influenced. Economic development of the country, education levels, globalization, and varieties of capitalism are a few of the several additional factors to be included in a broader analysis of the institutional context.
While our research finds relationships between culture and HR system adoption, more work is also need to open up the black box between culture and practice adoption or effectiveness. Specifically, what are the mediating processes that connect culture and HR system adoption. Gooderham et al. (2015) found support for labor regulation as the mediator of culture’s influence, but a larger set of influence transmitters is likely at play. Research by Thomas et al. (2010) presents a more micro view in which cultures create different psychological contracts (employee expectations about the exchange relationship between employees and employers) and those different culture-influence sets of expectations will influence employee responses to different HR practices. Violated expectations result in withdrawal and lower motivation and increase the likelihood of practices not achieving their intended outcomes. Such reactions, or the expectation of such reactions, will in turn influence managerial choices. Additional research could also address the possibility that different incentive bundles could differentially attract employees that may be less-representative of the country population. The sorting effect of incentives is well recognized in the pay literature and research that addresses how representative the employees are may help us better understand firms that seem to operate in a ways counter to country’s’ culture-based expectations. From an institutional entrepreneurship perspective, firm may be able to circumvent cultural mandates by attracting counter-culture employee.
This research demonstrates the significance, in both the statistical and practical sense, of three culture dimensions on incentive pay system adoption: in-group collectivism, uncertainty avoidance, and performance orientation. The reduction in between-country configuration adoption variation (ρ) associated with the entry of culture variables ranged from as high as 80 percent for the incentive minimizer configuration to as low as 27.3 percent for the profit rewarder configuration. These results encourage continued theorizing using a configurational perspective to understand how HR practice adoption patterns and the use of GLOBE culture dimensions are related. While not a settled issue, this study counters the general negative sentiment against the use of country-level culture that has emerged in the HR literature.
|Log of size||2,039||6.33||1.49||0||13.24|
|International market focus||2,054||0.53||0.50||0||1|
|1. Incentive maximizer|
|2. Incentive minimizer||−0.33|
|3. Contingent rewarder||−0.38||−0.54|
|4. Profit rewarder||−0.17||−0.24||−0.28|
|5. Log of size||0.03||−0.11||0.14||−0.08|
|6. International market focus||0.08||−0.06||−0.05||0.06||0.04|
|7. Service vs manufacturing sector||0.08||−0.07||−0.07||0.12||−0.02||0.45|
|8. Union intensity||−0.13||0.13||0.02||−0.07||0.20||0.04||0.15|
|9. In-group collectivism||0.12||−0.19||0.09||−0.01||−0.05||−0.03||0.03||−0.32|
|10. Uncertainty avoidance||0.03||0.13||−0.20||0.09||0.07||0.14||0.05||0.30||−0.83|
|11. Performance orientation||0.20||−0.12||−0.10||0.08||0.15||0.07||0.02||−0.07||−0.33||0.51|
Notes: n=1,533. Pearson correlation coefficients in the 0.04-0.05 range are significant at the p<0.05 level, coefficients from 0.06 to 0.08 are significant at the p<0.01 level, and coefficients higher than 0.08 are significant at the p<0.001 level
Countries and societal culture scores
|Country||No. of firms||In-group collectivism||Uncertainty avoidance||Performance orientation|
Hypotheses-based random-intercept logistical models (with odds ratios)
|Incentive maximizera||Incentive minimizera||Contingent rewardera||Profit rewardera|
|Step 1b (null): ICC(1), ρ=||0||0.37||0.11||0.11||0.19||0.19||0.22||0.22|
|Step 2: controls|
|Log of size||1.042||–||0.84***||–||1.19***||–||0.87*||–|
|International market focus||0.97||–||0.85||–||1.23||–||0.92||–|
|Service vs manufacturing||1.10||–||0.76*||–||0.86||–||2.25***||–|
|Step 2: ICC(1), ρ=||0.36||–||0.10||–||0.19||–||0.20||–|
|Δρ, (Step 1 to Step 2)||0.01||–||0.01||–||0.00||–||0.02||–|
|Final step: GLOBE dimensions|
|Final step: ICC(1), ρ=||0.20||0.19||0.02||0.03||0.10||0.13||0.14||0.16|
|Δρ, (Step 2 to Step 3)||0.16||0.18||0.08||0.08||0.09||0.06||0.06||0.06|
Notes: aIn the first column under each incentive bundle, the control variables and GLOBE culture practices variables are entered in step 2 and 3, respectively. In the second column, only the GLOBE variables are added to the null model. The ρ values (ICC-1) are noted at each step and Δρ value is the change in ρ from the preceding step. For the column 1 model, the sample size in n=1,533 (and n=14 for the level 2 culture variables). For the column 2 models, n=2,065 for the level 1 variable and n=14 for the level 2 culture variables. brho (ρ) is the intraclass correlation coefficient (ICC-1) and the other coefficients presented are odds ratios. *p⩽0.05; **p⩽0.01; ***p⩽0.00; ****p⩽0.10
The 2009 CRANET survey covers a large range of industries that include: agriculture, hunting, fishing; energy and water; chemical products, extraction and processing of non-energy minerals; metal manufacturing; building and civil engineering; retail and distribution; hotels and catering; transport and communication; banking, finance, insurance, business services; personal, domestic recreational services; health services; education; social services; and public administration.
Brewster, C.J. and Mayrhofer, W. (2012), “Comparative human resource management: an introduction”, in Brewster, C.J. and Mayrhofer, W. (Eds), Handbook of Research on Comparative Human Resource Management, Edward Elgar Publishing Limited, Cheltenham, pp. 1-26.
Cadsby, C.B., Song, F. and Tapon, F. (2007), “Sorting and incentive effects of pay for performance: an experimental investigation”, Academy of Management Journal, Vol. 50 No. 2, pp. 387-405.
Delery, J.E. and Doty, D.H. (1996), “Modes of theorizing in strategic human resource management: tests of universalistic, contingency, and configurational performance predictions”, Academy of Management Journal, Vol. 39 No. 4, pp. 802-835.
Dimaggio, P. and Powell, W.W. (1983), “The iron cage revisited: collective rationality and institutional isomorphism in organizational fields”, American Sociological Review, Vol. 48 No. 2, pp. 147-160.
Farndale, E. and Sanders, K. (2016), “Conceptualizing HRM system strength through a cross-cultural lens”, International Journal of Human Resource Management, Vol. 28 No. 1, pp. 132-148.
Gerhart, B. and Fang, M.Y. (2005), “National culture and human resource management: assumptions and evidence”, International Journal of Human Resource Management, Vol. 16, pp. 971-986.
Gerhart, B., Rynes, R. and Fulmer, I. (2009), “Pay and performance: individuals, groups, and executives”, The Academy of Management Annals, Vol. 3, pp. 251-313.
Gooderham, P., Fenton-O’creevy, M., Croucher, R. and Brookes, M. (2015), “A multilevel analysis of the use of individual pay-for-performance systems”, Journal of Management, Vol. 44 No. 4, pp. 1479-1504.
Hofstede, G. (1980), Culture’s Consequences: International Differences in Work-related Values, Sage, Beverly Hills, CA.
Hofstede, G. (1993), “Cultural constraints in management theories”, The Academy of Management Executive, Vol. 7, pp. 81-94.
Hofstede, G. and Hofstede, G.J. (2005), Cultures and Organizations Software of the Mind: Intercultural Cooperation and its Importance for Survival, McGraw-Hill, New York, NY.
House, R., Javidan, M., Hanges, P. and Dorfman, P. (2002), “Understanding cultures and implicit leadership theories across the globe: an introduction to project GLOBE”, Journal of World Business, Vol. 37 No. 1, pp. 3-10.
House, R., Hanges, P., Javidan, M., Dorfman, P. and Gupta, V. (2004), Culture, Leadership, and Organizations: The GLOBE Study of 62 Societies, Sage Publications, Thousand Oaks, CA.
Huselid, M.A. (1995), “The impact of human resource management practices on turnover, productivity, and corporate financial performance”, Academy of Management Journal, Vol. 38 No. 3, pp. 635-672.
Jiang, K., Lepak, D.P., Jia, J.U. and Baer, J.C. (2012), “How does human resource management influence organizational outcomes? A meta-analytic investigation of mediating mechanisms”, Academy of Management Journal, Vol. 55 No. 6, pp. 1264-1294.
Jiang, K.F., Lepak, D.P., Han, K.J., Hong, Y., Kim, A. and Winkler, A.L. (2012), “Clarifying the construct of human resource systems: relating human resource management to employee performance”, Human Resource Management Review, Vol. 22 No. 2, pp. 73-85.
Kostova, T. and Roth, K. (2002), “Adoption of an organizational practice by subsidiaries of multinational corporations: institutional and relational effects”, Academy of Management Journal, Vol. 45 No. 1, pp. 215-233.
Kuhn, K.M. (2009), “Compensation as a signal of organizational culture: the effects of advertising individual or collective incentives”, International Journal of Human Resource Management, Vol. 20 No. 7, pp. 1634-1648.
Lawler, J.J., Chen, S.J., Wu, P.C., Bae, J. and Bai, B. (2011), “High-performance work systems in foreign subsidiaries of American multinationals: an institutional model”, Journal of International Business Studies, Vol. 42 No. 2, pp. 202-220.
Lees, S. (1997), “HRM and the legitimacy market”, International Journal of Human Resource Management, Vol. 8 No. 2, pp. 226-243.
Long, R.J. and Fang, T. (2012), “Do employees profit from profit sharing? Evidence from Canadian panel data”, Industrial & Labor Relations Review, Vol. 65 No. 4, pp. 899-927.
Newman, K.L. and Nollen, S.D. (1996), “Culture and congruence: the fit between management practices and national culture”, Journal of International Business Studies, Vol. 27 No. 4, pp. 753-779.
North, D.C. (1990), Institutions, Institutional Change and Economic Performance, Cambridge University Press, Cambridge.
Nyberg, A.J., Pieper, J.R. and Trevor, C.O. (2013), “Pay-for-performance’s effect on future employee performance: integrating psychological and economic principles toward a contingency perspective”, Journal of Management, Vol. 42 No. 7, pp. 1753-1783.
Oyserman, D. (2011), “Culture as situated cognition: cultural mindsets, cultural fluency, and meaning making”, European Review of Social Psychology, Vol. 22 No. 1, pp. 164-214.
Paauwe, J. and Paul, B. (2008), “HRM and societal embeddedness”, in Boxall, P., Purcell, J. and Wright, P. (Eds), The Oxford Handbook of Human Resource Management, Oxford University Press, pp. 166-184.
Parry, E., Stavrou-Costea, E. and Morley, M.J. (2011), “The cranet international research network on human resource management in retrospect and prospect”, Human Resource Management Review, Vol. 21 No. 1, pp. 1-4.
Prince, N., Prince, J., Skousen, B. and Kabst, R. (2016), “Incentive pay configurations: bundle options and country-level adoption”, Evidence-Based HRM: A Global Forum for Empirical Scholarship, Vol. 4 No. 1, pp. 49-66.
Rizov, M. and Croucher, R. (2009), “Human resource management and performance in European firms”, Cambridge Journal of Economics, Vol. 33 No. 2, pp. 253-272.
Rynes, S.L., Gerhart, B. and Parks, L. (2005), “Personnel psychology: performance evaluation and pay for performance”, Annual Review of Psychology, Vol. 56, pp. 571-600.
Schneider, B. (1987), “The people make the place”, Personnel Psychology, Vol. 40 No. 3, pp. 437-453.
Schuler, R.S. and Macmillan, I.C. (1984), “Gaining competitive advantage through human resource management practices”, Human Resource Management, Vol. 23 No. 3, pp. 241-255.
Schuler, R.S. and Rogovsky, N. (1998), “Understanding compensation practice variations across firms: the impact of national culture”, Journal of International Business Studies, Vol. 29 No. 1, pp. 159-177.
Scott, W.R. (2008), Institutions and Organizations: Ideas and Interests, Sage, Thousand Oaks, CA.
Stajkovic, A.D. and Luthans, F. (2001), “Differential effects of incentive motivators on work performance”, Academy of Management Journal, Vol. 44 No. 3, pp. 580-590.
Steinmetz, H., Schwens, C., Wehner, M. and Kabst, R. (2011), “Conceptual and methodological issues in comparative HRM research: the Cranet project as an example”, Human Resource Management Review, Vol. 21 No. 1, pp. 16-26.
Taras, V., Kirkman, B.L. and Steel, P. (2010), “Examining the impact of culture’s consequences: a three-decade, multi-level, meta-analytic review of Hofstede’s cultural value dimensions”, Journal of Applied Psychology, Vol. 95 No. 5, pp. 405-439.
Thomas, D.C., Fitzsimmons, S., Ravlin, E.C., Au, K., Ekelund, B. and Barzantny, C. (2010), “Psychological contracts across cultures”, Organization Studies, Vol. 31 No. 11, pp. 1437-1458.
Toh, S.M., Morgeson, F.P. and Campion, M.A. (2008), “Human resource configurations: investigating fit with the organizational context”, Journal of Applied Psychology, Vol. 93 No. 4, pp. 864-882.
Tosi, H.L. and Greckhamer, T. (2004), “Culture and CEO compensation”, Organization Science, Vol. 15 No. 6, pp. 657-670.
Trevor, C.O., Gerhart, B. and Boudreau, J.W. (1997), “Voluntary and job performance: curvilinearity and the moderating influences of salary growth and promotions”, Journal of Applied Psychology, Vol. 82 No. 1, pp. 44-61.
Verburg, R.M., Den Hartog, D. and Koopman, P.L. (2007), “Configurations of human resource management practices: a model and test of internal fit”, Journal of Human Resource Management, Vol. 18 No. 2, pp. 184-208.