UNIVERSITY OF GRONINGEN – FACULTY OF ECONOMY & BUSINESS
Comparing the cultural dimensions of masculinity,
assertiveness, and gender equality as moderators on
the relation between management practices
subgroups and labor productivity.
Kyra Schepers S2160986
Herman Colleniusstraat 110, 9718 KZ Groningen
kyra_schepers@hotmail.com
Thesis IB&M Supervisor: O. Lindahl
d.d. 19-01-2017
2
ABSTRACT
3
TABLE OF CONTENTS
1 INTRODUCTION 4
2 THEORY AND HYPOTHESES 7
2.1 Management practices 7
2.2 Labor productivity 7
2.3 The link between management practices and labor productivity 8
2.3.1 Target practices 8
2.3.2 Incentive practices 9
2.4 Culture 10
2.5 Masculinity 11
2.6 Assertiveness and gender egalitarianism 12
2.7 Cultural dimensions as moderators 13
2.7.2 Culture and target practices 13
2.7.3 Culture and incentive practices 13
2.8 Conceptual model 14 3 METHODOLOGY 15 3.1 Variables 15 3.1.1 Independent variables 15 3.1.2 Dependent variable 16 3.1.3 Moderators 16 3.1.4 Control variables 17 3.2 Data analysis 18 4 RESULTS 20 4.1 Descriptive data 20 4.2 Correlations 20 4.3 Regression analysis 21 4.4 Moderation analysis 22
5 DISCUSSION AND CONCLUSION 25
5.1 Discussion of results 25
5.2 Implications and contributions 26
5.3 Limitations 28
5.4 Future research 29
5.5 Conclusions 29
REFERENCES 32
4
1 INTRODUCTION
Since the recent developments of internationalization and globalization, global competition is increasing. Firms are now not only competing against their neighbors but also against other firms across the world. In order to survive, firms should try to achieve a competitive
advantage. One way to enhance their competitive advantage is to increase productivity (Krugman, 1994). The level of productivity can be affected by constructs such as the quality of labor and inputs, R&D, firm structure decisions, regulations and management practices (Syverson, 2011; Bloom & Van Reenen, 2007). However, there are large productivity
differences present between firms, and most of these differences are still unexplained (Bloom & Van Reenen, 2010). Consequently, it is valuable to engage in deeper research, to examine how firms can improve their competitiveness in changing surroundings by achieving a higher level of productivity. More specifically, emphasizes will be given to how management practices can affect productivity differences.
Bloom and Van Reenen (2010) proposed eighteen management practices, divided into three subgroups; monitoring, targets, and incentives, that have a positive impact on firm-level productivity, profitability and survival rates. Therefore, they argue that ‘best managerial practices’ exists across countries. A higher level of these management practices can result in positive employee attitudes and enhanced motivation, through which productivity increases (Huselid, 1995). In this thesis, ‘management practices’ will refer to the management practices described by Bloom and Van Reenen (2007).
Nonetheless, there are still some questions to be asked regarding these topics. First, even when taken these management practices into consideration, there are still unexplained productivity differences between firms (Newman & Nollen, 1996). Second, scholars
following the divergent perspective argue that the context where firms are operating in still is from significant influence and should be taken into consideration when deciding upon
management practices (Milikic, 2009).
Accordingly, I argue that a cultural perspective can further explain productivity differences. Researchers have acknowledged the importance of the cultural environment in influencing the behaviors of individuals. Moreover, a number of scholars found that the effectiveness of management practices differs across different contexts (Newman & Nollen, 1996; Waldman, Sully de Luque & Wang, 2012; Van Hoorn, 2014).
5 (Hofstede, 1980). I propose that the cultural dimensions described by Hofstede can moderate the relation between management practices and productivity. More specifically, I will focus on the cultural dimension masculinity since it is the only dimension that explicitly takes the importance of performance into account (Hofstede, 2016). Therefore, I think the masculinity dimension is especially useful to examine in relation to productivity.
Notwithstanding, the Hofstede model received a lot of criticism and scholars argue that the dimensions and measures are outdated. The Global Leadership and Organizational Behavior Effectiveness (GLOBE) research program also identified several cultural
dimensions based on more recent data (House, Hanges, Javidan, Dorfman & Gupta, 2004). To my knowledge, these cultural dimensions are not examined before in combination with
management practices and labor productivity. The dimensions gender equality and
assertiveness are also characterized by strong male or female values and scholars claim that they are similar to the masculinity dimension of Hofstede (House et al., 2004). Hence, these dimensions are also expected to have a significant impact on the relation between the level of management practices and the level of labor productivity. To counteract individual criticism on the models and to give a better view on the impact of culture, both influential cultural models will be examined.
Accordingly, I propose the following research questions:
Do the WMS management practices groups, incentive and target management practices, have a positive influence on labor productivity? Besides, are these relations positively moderated by the cultural dimensions masculinity and assertiveness and negatively moderated by gender equality? Moreover, do these cultural dimensions directly influence labor productivity?
6 not been examined before as moderators. Furthermore, there is little research about the
consistency of the results and the comparison of the Hofstede model and the GLOBE project. This thesis will contribute to the cultural literature by employing both of the two models and comparing the results. Overall, comprehensive research covering a greater number of
differing practices in subgroups, in a large sample of firms, in diverse countries, in
combination with the cultural dimensions of two different cultural models, is to my knowing still missing.
This thesis also has practical relevance. If the expectations are confirmed and male values are indeed important for the level of productivity, managers should be aware of their national cultural values and their influence on productivity. Managers who know that their countries stand on a particular cultural dimension hampers their productivity can take appropriate actions. Moreover, with the moderating effect in mind, management practices should be altered to the specific cultures they are used in, to achieve the best results in terms of productivity.
7
2 THEORY AND HYPOTHESES
In this section, the relevant literature will be reviewed. In addition, hypotheses and a conceptual model are proposed.
2.1 Management practices
Bloom and Van Reenen (2010) identified eighteen management practices in a range of varying quality levels that were positively related to productivity and profitability. The management practices encompass three subgroups: monitoring, targets, and incentives. Monitoring practices emphasize the level on which things inside the firms are monitored and make use of continuous improvement. Target practices can be described as practices that set right targets, track them and exercise appropriate action. The incentive practices group is characterized by promotion based on performance and focus on retaining the best employees (Bloom & Van Reenen, 2010).
These management practices can be ‘good or ‘bad’. Bad management is characterized by failing to track performance, no effective targets, promotion based on tenure and no system in place to deal with low performance of employees. Good management, in contradiction, can be recognized by continuous monitoring and improvement, far-reaching and complete targets, promoting high performing employees and dealing with employees that underperform
(Bloom, Genakos, Sadun & Van Reenen, 2012).
Several determinants of management practices were identified. The product market competition, the educational level, the level of labor law and family firms, all had a
significant effect on the adoption of management practices (Bloom & Van Reenen, 2007). Different scholars found the positive relation between management practices and productivity, often focusing on some individual practices or an overall score of management practices (Bloom & Van Reenen, 2007; Huselid, 1995; Newman & Nollen, 1996).
Management practices can result in positive attitudes such as satisfaction and commitment. These attitudes lead to motivated and harder working employees what increases productivity (Huselid, 1995).
2.2 Labor productivity
8 trade openness (Wagner, 2005), and capital intensity (Oliner & Shichel, 2000) are examples of productivity determinants. However, also human capital is one of the important
determinants of productivity (UNIDO, 2007). In this thesis, the focus will be on labor
productivity since there can be expected that employees can have the biggest influence on this type of productivity. Labor productivity can be described as the output per worker (Sargent & Rodriques, 2000).
2.3 The link between management practices and labor productivity
Several studies examined and found relations between management practices and
productivity. Bloom and Van Reenen (2007) even found a relation between the overall WMS (World Management Survey) management practices and labor productivity. Nonetheless, results and argumentation per specific WMS subgroups are missing and/or unclear. Therefore, deeper insight will be given to this level in this thesis. Battisti and Iona (2009) argued that the benefits of joint clusters of complementary practices are higher than the sum of individual effects since synergistic effects are then not ignored. For this reason, an individual
management practices level analysis is also not suitable when examining the effects of adopting a set of practices on performance. Moreover, since these subgroups encompass practices that share characteristics and differ from the other subgroups it is plausible that also their effects differ. Previous literature focused more on some individual practices or an aggregated score of management practices through which subgroup effects are overlooked. Consequently, there will be elaborated on the effect of the different groups on labor
productivity. However, the monitoring practices subgroup represents the more technical aspects of management practices and is thus less important when examining the effects of employees on labor productivity and when focusing on the influence of culture. Moreover, research already repeatedly found that these types of practices have a strong significant relation to labor productivity (Agarwal, Brown, Green, Randhawa & Tan, 2014; Lewis, 2000). Therefore, the monitoring management practices subgroup will not be further examined in this thesis.
2.3.1 Target practices
The target management practices subgroup is able to contribute to productivity in several ways:
9 managers are more aware of their performance. Hence, the knowledge of employees increases and arguably they are able to perform better. Performance clarity and performance dialogue are WMS management practices that can contribute to this process and are, thus, expected to have a positive relation to labor productivity.
Second, goal setting can have positive effects on performance. Goals can namely be seen as an end state towards which employees strive to achieve and, hence, act as regulators of action (Erez & Kanfer, 1983). Conclusively, goals thus will lead to more motivated and harder worker employees. Performance tracking, performance review, and stretching targets are, for this reason, WMS practices expected to increase labor productivity.
Third, taking action when targets are not met and communicating these practices can motivate employees to work harder. People want to decrease threats such as dismissal and severe critic (Battisti & Iona, 2009). The practice of consequence management is, therefore, predicted to enhance labor productivity.
Fourth, the remaining target practices; target balance, interconnection, and target time horizon support similar goals as the other incentive practices described and deal with the feasibility of the targets. Hence, they are expected to have a positive relation to labor productivity as well. Concluding, there is expected that these practices together have a positive impact on labor productivity and, as a consequence, the following hypothesis is proposed:
H1a: The higher the use of target management practices, the higher the labor productivity.
2.3.2 Incentive practices
The incentive management practices subgroup can be expected to have a substantial influence on labor productivity for different reasons.
First, one of the most important incentive management practices is wage based on performance since employees can see a clear link between their efforts and the benefits they receive. Cozzarin and Jeffrey (2014) found this was the greatest contributor to labor
productivity. Moreover, also performance appraisal and skill-based pay have been linked to organizational performance (Huselid, 1995).
10 2013).
Third, selective staffing techniques can be used to hire employees whose abilities are compatible with the needs of the organization (Huselid, 1995). Thus, staffing techniques can contribute by acquiring employees of a higher quality level and with a greater fit with the firm.
Fourth, managers can drive productivity differences. There is also said that managers are conductors of an input orchestra (Syverson, 2011). Hence, practices focused on attracting, retaining and training managers can contribute to productivity.
Conclusively, there can be argued that incentive focused practices can have a positive influence on performance, and more specific on productivity, through the enhancement of skills and motivation of employees. The practice of rewarding high performance as described by Bloom and Van Reenen (2010) is directly linked to labor productivity in prior research and already frequently examined and will, therefore, be omitted in the incentive group in this research. However, the other incentive practices (managing, attracting and retaining human capital, removing poor performers and promoting high performers) are not directly examined in relation to labor productivity. Nevertheless, a similar relation is expected when combining them following the argumentation before. Accordingly, the following hypothesis is proposed:
H1b: The higher the use of incentive management practices, the higher the labor productivity.
2.4 Culture
Notwithstanding, the effects of the management practices still present unexplained productivity differences. Nonetheless, there are clues that in fact, results of research on management practices are more influenced by context than previously acknowledged (Newman & Nollen, 1996). There is argued that culture could contribute to a further understanding of these productivity differences. Researchers have acknowledged the importance of the socio-cultural environment in influencing behaviors of individuals and groups in firms. Hofstede (1980, p.24) described culture as, “The collective mental programming of people who live in a particular society.”
A number of scholars found that the effectiveness of management practices differs across different contexts (Newman & Nollen, 1996; Waldman et al., 2012; Van Hoorn, 2014). National culture is of great importance to the understanding of work, the approach of
11 and values of potential adopters. This will result in feelings of anxiety and emotional
discomfort (Canato, Ravasi & Phillips, 2013). If these practices are forced upon employees this may even lead to tension and resistance (Canato et al., 2013). Since culture causes variances in behaviors, attitudes, values, and efficacy, there can be argued that also
management practices across different cultures ask for variance. Furthermore, following this line of thought, different combinations of management practices and cultures could have different results on productivity.
The focus will be on the cultural dimension masculinity of Hofstede since it has a significant influence on the preferences for certain management practices (Hofstede, 2016; Newman & Nollen, 1996). Besides, masculinity has a profound effect on work situations, for example on how managers are expected to behave, how conflicts should be resolved and which work values are viewed as important (Hofstede, 2016). Furthermore, and most important, this cultural dimension explicitly takes the importance of performance into
consideration while the other cultural dimensions do not (Hofstede, 2016). Therefore, I think the masculinity dimension is especially useful to examine in relation to productivity.
Despite, the Hofstede model received a lot of criticism. Many scholars argue, for example, that cultural convergence has a profound effect on countries and, as a consequence, dimensions defined in the 1960’s are not accurate anymore. For this reason, it could be valuable to additionally examine a more recent model of cultural dimensions (the GLOBE project) described by House et al., (2004). Moreover, House et al., (2004) argued that the masculinity dimension described by Hofstede lacked face validity and existed out of items that measured a number of constructs (Shi & Wang, 2010). Furthermore, Hofstede did not measure feminine scores directly, while the GLOBE project does (Shi & Wang, 2010). Additionally, the GLOBE project also measures cultural practices next to cultural values. Practices that indicate how culture actually is presented in a country could arguably have more effect on economic constructs such as productivity, than values that indicate how people think things should be.
2.5 Masculinity
A masculine culture is characterized by a clear distinction between emotional gender roles, wherein a feminine culture emotional gender roles overlap. This dimension was first
12 living in a preferred area, security and quality of interpersonal relations and working life (Hofstede, 2016; Newman & Nollen, 1996). Moreover, failure is seen as less important (Newman & Nollen, 1996). Masculinity is also described as the need for performance, success, and competition (Wacker & Sprague, 1998). The masculinity dimension also has a significant influence on work situations in different cultures. Organizations operating in masculine cultures emphasize results and reward employees based on performance while organizations operating in feminine cultures base rewards on equality.
Therefore, there can be expected that firms in masculine cultures place a higher importance on productivity and strive for the highest possible goals with respect to
productivity where this is less vital in feminine cultures. Since they place a higher weight on productivity and exert more effort to achieve those productivity goals it can be argued that firms operating in masculine cultures, in general, have a higher level of productivity.
H2a: The more a culture emphasizes masculinity, the higher the labor productivity.
2.6 Assertiveness and gender egalitarianism
The Global Leadership and Organizational Behavior Effectiveness (GLOBE) project was published in 2004 by House and is argued to be the most comprehensive research at the present time, that empirically examined the relation between culture and leader in many countries and firms, and by using different measures and methods. The main result of this project was that leader effectiveness is indeed contextual (Hoppe, 2007). The project described nine cultural dimensions. Two of the dimensions are similar to the masculinity dimension; namely assertiveness and gender egalitarianism (House et al., 2004).
Assertiveness can be described as the level in which individuals show assertive,
confrontational, and aggressive behavior towards others. Moreover, gender egalitarianism is defined as the level in which gender inequality is minimized. High masculine cultures share thus similar characteristics with cultures that emphasize assertiveness and do not emphasize gender equality. Furthermore, high gender equality cultures are more similar to feminine cultures. There will be referred to these three cultural dimensions together as ‘cultural dimensions emphasizing male values’, by which cultures high on masculinity and
assertiveness, and low on gender equality is intended. The GLOBE study measures both the practices and values as different aspects of culture for all of the dimensions. In this thesis, separate hypotheses will be proposed for the GLOBE and Hofstede model to make
13 Following the argumentation that these cultural dimensions place emphasis on
performance, the following hypothesis is proposed:
H2b: The more a culture emphasizes assertiveness and does not emphasize gender equality, the higher the labor productivity.
2.7 Cultural dimensions as moderators
Previous research found that rewards based on performance lead to a fit between practices and the masculine culture and, hence, resulted in higher performance. In contrast, higher
performance was achieved in feminine cultures by less focus on reward based on merit (Newman & Nollen, 1996). Therefore, there can be assumed that masculinity, but also the not yet examined before cultural dimensions, assertiveness, and gender equality can also act as moderators on the relationship between the WMS management practices and labor
productivity. There will be elaborated on the effect of these moderators on the different subgroups.
2.7.1 Culture and target practices
Several target practices are defined by Bloom and Van Reenen (2007), these practices are especially focused on performance and achieving goals. Since this is line with the more masculine cultures, there can be expected that the relation between target management practices and labor productivity is stronger in masculine cultures (similar to H2a-b). Accordingly, the following hypotheses are proposed:
H3a: Cultures that emphasize masculinity positively moderate the positive relationship between the use of target management practices and labor productivity.
H3b: Cultures that emphasize assertiveness and do not emphasize gender equality positively moderate the positive relationship between the use of target management practices and labor productivity.
2.7.2 Culture and incentive practices
There can be made several arguments about cultures characterized by male values as moderator with respect to the six incentive practices described by Bloom and Van Reenen (2010).
14 H1a 22 H1b H3a H4a + + + + + + + +
higher performance if they employed practices based on merit-based rewards for both pay and promotion, while in feminine cultures the opposite was found. These practices are similar to the promoting high performers incentive practice of Bloom and Van Reenen (2010). Since the other incentive practices of Bloom and Van Reenen (2010), focusing on, for example,
removing poor performers and managing human capital, represent practices with similar goals, there is expected that also the other incentive practices described by Bloom and Van Reenen (201) are stronger related to labor productivity in masculine cultures.
Second, values of recognition and promotion are important in masculine cultures. These values can also be recognized in the incentive practices described. Consequently, there will be a better fit of these practices within masculine cultures. This leads to the following hypotheses:
H4a: Cultures that emphasize masculinity positively moderate the positive relationship between the use of incentive management practices and labor productivity.
H4b: Cultures that emphasize assertiveness and do not emphasize gender equality positively moderate the positive relationship between the use of incentive management practices and labor productivity.
2.8 Conceptual model
Following the literature review and the proposed hypotheses, the conceptual model is presented in Figure 1. FIGURE 1 Conceptual model Labor Productivity Cultures emphasizing masculinity Target practices Incentive practices H2a
Cultures emphasizing assertiveness and not emphasizing gender equality
gender H2b H4b
15
3 METHODOLOGY
In this section, there will be elaborated on the independent and dependent variables, the moderators and control variables. Moreover, the steps in the data analysis will be outlined and there will be explained why certain methods are used. At last, the assumptions checks are discussed.
3.1 Variables
Quantitative data will be collected and combined from different existing databases.
3.1.1 Independent variables
Management practices. Data on the management practices is gathered from the World
Management Survey of Bloom and Van Reenen (2007). These practices are measured from one (bad) to five (good). Data from the World Management Survey is collected by telephone interviews with managers and a double-blind technique was adopted. Only data in the
manufacturing industry is taken into consideration to constrain from industry effects. Data is available in eighteen countries of more than 4000 firms. Three groups are identified, of which two are used in this thesis. The target management practices group is the first group and an example of a question asked and scored in the interview is: ‘Is performance continually tracked and communicated to all staff?’. The second group is incentives, a question asked is for example ‘Are people promoted mainly on the basis of tenure or does the firm promote its top performers?’ (Bloom & Van Reenen, 2010, p. 206). A scorecard is used to assess the scores of certain practices. The scores can range on a Likert-scale from 1 to 5, these scores were also the minimum and maximum in the available data in each group. For both the incentive and target management practices subgroup the individual management practices are used to calculate the mean that is used as the subgroup variable. Data is available for the years 2001 to 2010.
16 also contend that it is unlikely that a set of eighteen practices fits all (Agarwal et al., 2014) and some practices are more culture-bound than others (Khatri, 2009). This level is also more suitable than the individual management practices level for examining the effects on
productivity. Complementary practices namely achieve higher benefits together than the sum of individual practices (Battisti & Iona, 2009). This is important to take into consideration when examining the effect of adopting a set of practices on productivity.
There are both advantages and disadvantages for using this database. The first considerable advantage is the size of the set; it contains data on 4623 companies across eighteen countries. Besides, the set described comprises eighteen different practices in three different subgroups, while other datasets often only cover a few practices. Moreover,
management practices are measured at several points in time. Furthermore, the double-blind technique protects for several biases through which reliable results are gathered. However, longitudinal data with measures on following years is missing and a sizable amount of companies has only one measurement at one moment in time. Besides, it covers only three dimensions of management while other important dimensions such as technology
management are not included (Agarwal et al., 2014).
3.1.2 Dependent variable
Labor productivity. Productivity will be measured by extracting various financial measures
from the online database Orbis. Gal (2013) describes that labor productivity can be measured as can be seen in Formula 1.
Formula 1 = log!"#$%& !" !"#$%&!!'!"#$% !"#"$%"
Nonetheless, data in Orbis on total revenue is very limited and, for this reason, operating revenue is used instead. The difference between them is that operating revenue also accounts for operational expenses. Since operating revenue already incorporated the costs of sales and operating expenses, it is a better measure when examining the efficiency (productivity) of firms that can be influenced by employees.
3.1.3 Moderators
Masculinity. Data on the level of masculinity per country will be derived from the Value
17 The score can range from 0 to 100. With respect to the countries in the dataset, the measures on masculinity vary from 5 to 95. The cultural model of Hofstede received considerable criticism and, therefore, this dimension will be complemented with similar dimensions of another prominent cultural model.
Assertiveness and gender equality. Data on the assertiveness and gender equality
dimension will be gathered from the Global Leadership and Organizational Behavior
Effectiveness (GLOBE) research program. The first results were published in 2004 and data was gathered from 17.300 middle managers from 951 firms, in 62 countries in different industries. The cultural dimensions of the GLOBE project were measured as practices and values. Both of those measurements are taken into account in this thesis since the
measurement of the values makes comparison with the Hofstede model easier since this model also measures values. However, practices are expected to have a stronger relation to productivity because it represents how things are really done in contrast to how people want to see things. The questionnaires were complemented by interviews, focus group discussions, and media (Shi & Wang, 2010). The questions were measured on a seven-point Likert scale. The scores in this dataset of the assertiveness practices range from 2,3 to 4,6, the assertiveness values range from 1,5 to 5,8, the gender equality practices range from 1,8 to 3,9 and the gender equality values range from 2,42 to 5,19.
3.1.4 Control variables
18 expected to be less common. As a consequence, also the employment laws index will be included as a control variable. The employment laws measures are derived from the article of Botero et al., (2004). The measures are index numbers and range from zero to one.
3.2 Data analysis
Stata will be used as the statistical program because of the greater possibilities and ease to work with longitudinal data. The dataset is restructured, variables are computed and z-scores are created to make regression with variables with different scales possible. Furthermore, a hierarchical model is employed because the firms are nested in particular industries that are again nested in specific countries. For this purpose, the company’s ID, the industry SIC code, and the country it operates in are used. By using a hierarchical model, there can be controlled for the lack of independence of the measurements.
Several steps will be undertaken in the data analysis.
First, several assumptions checks are done in advance. This is done because invalid assumptions could affect the results and, thus, also the conclusions. More specific outliers, normality, linearity, and homoscedasticity will be examined because those assumptions need to be fulfilled for hierarchical linear models. The results of the assumptions analysis can be found in Appendix 1. Five severe outliers were detected by looking at the inter-quartile range. These outliers were deleted since they can significantly influence the results. Results are presented in Table 7 and 8. To examine the normality of residues several graphs are made. A kernel density estimate plot, a standardized normality probability plot and a plot with
quantiles of the variables plotted against the quantiles of normal distribution can be found in appendix 1 as Figure 2-4. Especially the kernel density estimate plot in Figure 2 shows some deviation from normality and also the quantile plot in Figure 4 shows deviation in the upper tail. However, in general, the residues show normality. The linearity assumption is examined by making an augmented component-plus residual plot. The Figures 5 and 6 in Appendix 1 did not show much nonlinearity. Homoscedasticity is tested with the help of the White’s test. This was not significant (p > .05) and, therefore, there can be assumed that the variance is homogenous.
Second, descriptive and correlations will be derived to get more insight in the specific variables and their relations.
19 hypotheses 2a-b multiple regression analysis will be performed. After this, moderation
analysis will be conducted to test the influence of the cultural dimensions on the strength of the relation between the dependent and independent variables. The models of Hofstede and House will be examined separately to be better able to compare the results. In the moderation analysis, the moderators are examined as interaction terms.Several control variables will be taken into account to keep away from unwanted influences.
20
4 RESULTS
4.1 Descriptive data
In Table 1 the descriptive data is presented for the most important variables. TABLE 1
Descriptive data
Minimum Maximum Mean Std. Deviation N
LP -4,87 4,93 2,3450 ,43355 7565 MeanTarget 1,00 5,00 3,0840 ,75388 6133 MeanIncentive 1,00 5,00 2,8487 ,69572 6133 MAS 5,00 95,00 59,6420 10,86842 46230 Assert_P 2,28 4,55 3,9799 ,55479 44320 Assert_V 1,53 5,84 4,0876 ,97557 44320 Gender_P 1,77 3,94 3,2611 ,49561 44320 Gender_V 2,42 5,19 4,5030 ,69832 44320 4.2 Correlations
In Table 2 the correlations between the variables are presented. The correlations between gender equality practices and target management practices and incentive management practices were not significant, implying that these variables do not directly influence each other. Moreover, the relation between the cultural dimension assertiveness values and the incentive management practices showed not to be significant. All of the other variables correlated significantly and consequently change to the level of a certain variable (such as culture) will influence the other variables as well. Furthermore, the masculinity dimension of Hofstede does correlate with the GLOBE dimensions.
TABLE 2 Correlations LP Mean Target Mean Incentive
MAS Assert_P Assert_
V
Gender_P
LP 1
MeanTarget 0,27** 1
21 4.3 Regression analysis
For hypotheses 1a-b and hypotheses 2a-b regression analysis is done. This is done in a hierarchical model of companies, industries and countries.
Hypotheses 1a-b argued for a significant positive relation between the specific management practice groups (targets and incentives) and labor productivity. The results of hypotheses 1a-b are presented in Table 3. Labor productivity is significantly positively related to the target management practices subgroup (b = .06, SE = .02, p < .01). Thus, hypothesis 1a
is supported. Moreover, labor productivity is significantly positively related to the incentive
management practices subgroup as well (b = .05, SE = .02, p < .01). Accordingly, hypothesis
1b is supported as well. TABLE 3 Hypothesis 1a & 1b Hypothesis 1a Hypothesis 1b LP MeanTarget 0.061 MeanIncentive 0.053 (3.59)** (3.42)** EmployLaws -0.092 EmployLaws -0.090 (0.58) (0.56) Age -0.076 Age -0.078 (2.98)** (3.03)** Size -0.092 Size -0.090 (2.48)* (2.45)* N 1,329 1,329 * p<0.05; ** p<0.01
Hypothesis 2a argued that the more a culture emphasizes masculinity, the higher the labor productivity. In addition, hypothesis 2b argued that the more a culture emphasizes assertiveness and does not emphasize gender equality, the higher the labor productivity. In Table 4 the results are presented for both the Hofstede and GLOBE model. Masculinity is not significantly related to labor productivity (b = -.12, SE = .11, p > .05). Therefore, hypothesis
2a is not supported. The cultural dimensions assertiveness practices (b = -.40, SE = .13, p <
.01) and assertiveness values (b = -.23, SE = .10, p < .05) were both negatively related to labor productivity. This is in contrast to what was expected. Besides, the cultural dimension gender MAS 0,20** 0,07** 0,10** 1
Assert_P -0,18** 0,05** 0,08** -0.17** 1
Assert_V -0,07** -0,05** 0,00 0,29** 0,38** 1
Gender_P -0,31** 0,02 0,01 -0,36** 0,84** 0,28** 1
22 equality values appeared to be positively related to labor productivity (b = .90, SE = .13, p < .01), this also in contrast to the expected negative relation. Notwithstanding, the cultural dimension, gender equality practices, was significantly negatively related to labor
productivity (b = -.56, SE = .15, p < .01). Accordingly, hypothesis 2b is only supported in the
case of gender equality practices.
TABLE 4 Hypothesis 2a & 2b Hypothesis 2a Hypothesis 2b LP MAS -0.118 Assert_P -0.402 (1.09) (3.11)** Assert_V -0.226 (2.28)* Gender_P -0.558 (3.67)** Gender_V 0.897 (6.73)** EmployLaws -0.256 EmployLaws -0.408 (1.29) (3.31)** Age -0.040 Age -0.055 (1.85) (2.34)* Size -0.156 Size -0.086 (10.58)** (5.66)** N 7,211 6,254 * p<0.05; ** p<0.01 4.4 Moderation analysis
In the moderation analysis, the interaction variables are inspected. The results are presented in Table 5 and 6. The Hofstede and GLOBE models are separately examined.
Hypothesis 3a argued that cultures that emphasize masculinity positively moderate the positive relationship between the use of target management practices and labor productivity. In addition, hypothesis 3b claimed that cultures that emphasize assertiveness and do not emphasize gender equality positively moderate the positive relationship between the use of target management practices and labor productivity. Masculinity (b = -.04, SE = .11, p > .05) and the interaction variables with masculinity (b = -.02, SE = .01, p > .05) were not
23 equality practices (b = -.68, SE = .14, p < .01) appear to be direct predictors. Nonetheless, none of the interaction terms were significant. Thus, hypothesis 3b is not supported as well.
TABLE 5 Hypothesis 3a & 3b Hypothesis 3a Hypothesis 3b LP MeanTarget 0.059 MeanTarget 0.040 (3.49)** (1.63) MAS -0.037 Assert_V -0.197 (0.33) (2.06)* MeanTarget#MAS -0.020 MeanTarget#Assert_V -0.022 (1.89) (1.02) Assert_P -0.275 (2.24)* MeanTarget#Assert_P -0.025 (0.83) Gender_V 0.872 (7.13)** MeanTarget#Gender_V 0.004 (0.12) Gender_P -0.676 (4.99)** MeanTarget#Gender_P 0.041 (1.36) EmployLaws -0.146 EmployLaws -0.365 (0.70) (3.10)** Age -0.076 Age -0.085 (2,98)** (3.08)** Size -0.091 Size -0.100 (2.48)* (2.67)** N 1,329 1,208 * p<0.05; ** p<0.01
Hypothesis 4a contented that cultures that emphasize masculinity positively moderate the positive relationship between the use of incentive management practices and labor
productivity. Likewise, hypothesis 4b argued that cultures that emphasize assertiveness and do not emphasize gender equality positively moderate the positive relationship between the use of incentive management practices and labor productivity. Results of these hypotheses are presented in Table 6. Masculinity (b = -.04, SE = .02, p > .05) and the interaction variable (b = -.01, SE = .01, p >. 05) were not significant. Accordingly, hypothesis 4a was not supported. Nonetheless, in the GLOBE model, the cultural dimension, gender equality practices, was a significant positive moderator (b = .09, SE = .03, p < .01). However, hypothesis 4b is also not
24 cultural dimensions, assertiveness values (b = -.20, SE = .10, p < .05), assertiveness practices (b = -.30, SE = .13, p < .05), gender equality values (b = .89, SE = .13, p < .01) and gender equality practices (b = -.67, SE = .14, p < .01) were significant direct predictors.
25
5 DISCUSSION AND CONCLUSION
In this section, the results are discussed. Additionally, implications, contributions, and limitations of this thesis are reviewed. Suggestions for future research are made and this section ends with a conclusion.
5.1 Discussion of results
In Table 1 the correlations between the different variables are given. The masculinity dimension correlates with the GLOBE dimensions. Nonetheless, when comparing the
correlations of all the cultural dimensions with the other variables, the Hofstede and GLOBE dimensions do not show the expected (similar) correlations. In addition, in contradiction to what is expected, the cultural dimension assertiveness is negatively correlated to labor productivity. Moreover, the cultural dimension, gender equality practices, was positively related to labor productivity.
I will further elaborate on the results by discussing the different hypotheses. In hypothesis 1a there was claimed that target management practices are positively related to labor productivity. This hypothesis was confirmed by the results.
Hypothesis 1b proposed that incentive management practices are positively related to labor productivity. This hypothesis was supported as well. Thus, even when omitting the most important incentive practice described in literature (promoting based on performance), the incentive practices still have a considerable influence on labor productivity.
Hypothesis 2a contended that cultures emphasizing masculinity were positively related to labor productivity. This hypothesis is not supported.
In hypothesis 2b there was argued that cultures that emphasized assertiveness and did not emphasize gender equality were positively related to labor productivity. This hypothesis was only supported in the case of gender equality practices. Although the other cultural dimensions were significant, they presented relations in the opposite direction than expected. These contrasting results between the models (hypotheses 2a and 2b) can be caused by the substantial differences in their research design and in what they actually measure. House et al., (2004) for example, claimed that the masculinity dimension lacked face validity and in fact measures multiple constructs (Shi & Wang, 2010).
26 supported since no moderation effects took place.
In hypothesis 3b there was claimed that cultures that emphasize assertiveness and do not emphasize gender equality positively moderate the relation between target management practices and labor practices. This hypothesis was not confirmed.
Hypothesis 4a argued that cultures that emphasize masculinity positively moderate the relationship between incentive practices and labor productivity. This hypothesis was not confirmed.
Hypothesis 4b contended that cultures that emphasize assertiveness and do not emphasize gender equality positively moderate the relationship between incentive practices and labor productivity. The gender equality practices dimension appeared to be a positive moderator. The other cultural dimensions did not show significant moderation effects. However, the gender equality practices dimension was expected to negatively influence the relation. This because more masculine values such as recognition and promotion would strengthen the relation between incentive practices and labor productivity. This contrasting result could be explained by the inherent values of the incentive practices; these practices namely deal with attracting and retaining human capital, which is more feminine since it deals with interpersonal relations (Hofstede, 2016). Moreover, this dimension is characterized as minimizing gender equality, WMS management practices that promote based on performance, above other things, can, therefore, be seen as supporting this goal. Thus, the relation between incentive practices and labor productivity is stronger in cultures high on gender equality practices.
Remarkable are the contrasting results of the different management practice groups and culture (hypotheses 3b and 4b). Some scholars claim that particular practices are more culture sensitive than others (Khatri, 2009). It could be that practices that deal with targets are more universally applicable to achieve a higher performance. While in contrast, incentive practices, that for example focus on promoting based on tenure or on performance, are much more culture-bound.
5.2 Implications and contributions
There are several implications and contributions that result from this research. First, several contributions are made to literature.
27 subgroup are not mentioned. This thesis does both.
Moreover, the often-criticized masculinity dimension is examined as a moderator on the relation between a few management practices and performance in previous research (Newman & Nollen, 1996). Nonetheless, the less criticized GLOBE dimensions have not been examined as a moderator. This thesis contributes to the cultural literature by employing both of the two models and comparing the results. This also contributes since there is little research about the consistency of the results and the comparison of the Hofstede model and the
GLOBE project. In this research the models provide very differing results and, therefore, there could be argued the cultural dimensions are not as similar as proposed.
Furthermore, the cultural dimension gender practices appeared to be a significant moderator on the relation between incentive practices and labor productivity. This finding contributes to literature since it signals the importance of looking at the context and more specifically culture when examining management practices and productivity. Besides, maybe even more important, it signals the importance of also incorporating more recent cultural models such as the GLOBE project. Likewise, it is meaningful to not only look at cultural values but also at practices since they appear to have a significant effect.
Besides, this thesis found that incentive practices are positively moderated by gender equality and, thus, represent more feminine characteristics than previously acknowledged by research. Moreover, results show that the incentive management practices are also more influenced by cultural moderators and are, therefore, more culture sensitive. Thus, this thesis especially contributes to literature on incentive management practices.
Concluding, this research adds to the existing literature by using a large data set, examining a relatively large amount of management practices and comparing results on two cultural models, while still paying attention to the results on subgroup management practice level, through which a better overall picture is given.
Second, the results show some practical implications for managers and organizations. One of the main implications is that culture, and especially assertiveness and gender equality are directly important for the level of productivity. Managers should be aware of the values that are important in their respective country and their influence on labor productivity. Managers who know that their countries stand on a particular cultural dimension hampers their productivity can take appropriate actions.
Furthermore, the moderating effect of gender equality could have different
28 productivity. More specifically, this research found that especially incentive management practices have more feminine characteristics and are more culture sensitive. Thus, in countries that score high on the cultural dimension gender equality practices, incentive practices should be adapted to a high a level to achieve a higher level of labor productivity. Moreover, it is more important to alter the incentive management practices to the respective culture than target management practices.
Also, the target management practices subgroup appeared to have a positive impact on labor productivity when looking at all the countries. In addition, the cultural dimensions examined did not moderate this relation. Thus, these practices can be employed worldwide. Furthermore, scholars such as Bloom and Van Reenen (2007) argued that all of the eighteen management practices they described together positively influence productivity. However, results in this thesis show that the subgroups differ in their effects. With this knowledge in mind, managers can balance choices and can, for example, choose not to adopt certain practices since this costs time and money.
5.3 Limitations
Nonetheless, this research also has its limitations. A few will be outlined.
Firstly, as described before, there are numerous determinants of productivity, however, these are not all taken into account. There can be expected that there are several, for example, economic forces that are not incorporated.
Secondly, the Hofstede and GLOBE model measure culture on country level, while the productivity results and management practices are measured on company level. Moreover, scholars found that there is much inter-country difference in culture (Beugelsdijk, Maseland, Onrust & Van Hoorn, 2015). To have more valuable insight into the effects it would have been better to measure culture on company level.
Thirdly, this research is only focused on cultural dimensions representing male and female values. Both the Hofstede and GLOBE model describe other cultural dimensions that are not integrated. As a consequence, an overall picture of the effect of culture on the relation between management practices and labor productivity cannot be given.
29 5.4 Future research
Future research could give interesting insights in several ways.
First, in the future there could be given more attention to the overall contexts of firms when examining labor productivity or management practices. Other cultural dimensions can be incorporated, but also measures of institutions could be added. More research can be done with the GLOBE model since it gave more significant results in this thesis and it received less criticism. Moreover, only the measurement of cultural practices appeared to be a significant moderator. Therefore, in the future, when examining culture as moderator, deeper
examination should be done into the effects of these practices and the differences with values. Second, as suggested in the limitations, organizational culture could be measured to be able to get more precise and realistic results, through which fitting implications can be given. Third, this research is pure quantitative research. However, in the future in-depth quality research could be done to get a deeper understanding of the underlying processes. In addition, it can be interesting to compare and contrast both qualitative and quantitative outcomes such as employee motivation and financial outcomes.
Fourth, the results of this thesis gave some differing results than other research
because of the analysis on subgroup level. It would be interesting to examine these subgroups more by, for example, looking at other outcomes or looking at the effects of different
combinations of management practices.
Fifth, other important management areas can be examined, to get deeper insight into the influence of these areas on performance. The literature describes knowledge- and technology management as important areas.
Sixth, while this thesis provides support for the differing results of the two subgroups, there can be expected that also all of the different individual practices differ in their effect. Previous research namely focused on a few (similar) individual practices at the time.
Notwithstanding, it can be recommended to examine a greater number of individual practices to provide managers knowledge on which particular management practices especially
contribute to their performance.
5.5 Conclusions
30 Several scholars argue that a culture perspective could help to further explain these
differences. Hofstede (1980) proposed an influential cultural framework with five dimensions. One of those dimensions, masculinity, especially emphasizes the importance of performance and, therefore, is the focus of this thesis. Despite, Hofstede received a lot of criticism. For this reason, the more recent GLOBE model is also incorporated in this thesis. More specifically, the GLOBE project described gender equality and assertiveness as similar dimensions to masculinity.
The central research questions in this research were: ‘Do the WMS management
practices groups, incentive and target management practices, have a positive influence on labor productivity? Besides, are these relations positively moderated by the cultural dimensions masculinity and assertiveness and negatively moderated by gender equality? Moreover, do these cultural dimensions directly influence labor productivity?’
The proposed management practices by Bloom and Van Reenen (2010) are divided into three subgroups: monitoring, targets, and incentives. While they provided support that the overall of these management practices are significantly related to labor productivity, the results of the subgroups are unclear. Therefore, the influence of the separate management practice groups is examined in this thesis. The monitoring practices group was not further examined in thesis since it has been frequently examined before and is expected to be less interesting to research in combination with labor productivity and culture because of the technical nature. There was expected that the incentive and target management practices groups were positively significantly related to labor productivity. Moreover, since the dimensions emphasizing male values especially incorporate the importance of performance, there was expected that these dimensions have a direct positive influence on labor
productivity. Likewise, there was expected that masculinity and assertiveness positively moderated the relation between the incentive and target management group and labor productivity and this relation was negatively moderated by the cultural dimension gender equality.
Multiple regression and moderation analysis were performed in a hierarchical model to examine the results. The analysis was done in around 1200 companies from eighteen different countries.
31 described by House et al., (2004) have a significant direct impact on labor productivity.
Moreover, gender equality practices moderated the relation between incentive management practices and labor productivity.
One of the most important contributions of this thesis is that it contributes to the cultural literature by employing two influential cultural models and comparing the results. These results also signal the importance of looking at the practice measurements of culture. In addition, argumentation for the relation between the subgroups of practices and labor
productivity and results per subgroup are provided.
32
REFERENCES
Agarwal, R., Brown, P. J., Green, R., Randhawa, K., & Tan, H. 2014. Management practices of Australian manufacturing firms: Why are some firms more innovative? International Journal of Production Research, 52(21): 6496-6517.
Ansari, S.M., Fiss, P., & Zajac, E. 2010. Made to fit: How practices vary as they diffuse. Academy of Management Review, 35(1): 67-92.
Barro, R.J. 1991. Economic Growth in a Cross Section of Countries, Quarterly Journal of Economics, 106(2): 407-443.
Battisti, G., & Iona, A. 2009. The UK productivity gap in the service sector: Do management practices matter? International Journal of Productivity and Performance Management, Vol. 58(8): 727-747.
Beugelsdijk, S., Maseland, R., Onrust, M., van Hoorn, A., & Slangen, A. 2015. Cultural distance in international business and management: From mean-based to variance-based measures. The International Journal of Human Resource Management, Vol. 26(2): 165- 191.
Bloom, N., Genakos, C., Sadun, R., & Van Reenen, J. 2012. Management practices across firms and countries. Academy of Management Perspectives, 26(1): 12-33.
Bloom, N., & Van Reenen, J. 2007. Measuring and explaining management practices across firms and countries. Quarterly Journal of Economics,122(4): 1352-1408.
Bloom, N., & Van Reenen, J. 2010. Why do management practices differ across firms and countries? Journal of Economic Perspectives, 24(1): 203-224.
33 Canato, A., Ravasi, D., & Phillips, N. 2013. Coerced practice implementation in cases of low cultural fit: cultural change and practice adoption during the implementation of six sigma at 3m. Academy of Management Journal, Vol. 56 (6): 1724-1753.
Chowdhury, S. D., & Mahmood, M. H. 2012. Societal institutions and HRM practices: An analysis of four European multinational subsidiaries in Bangladesh. The international Journal of Human Resource Management. 23(9): 1808-1831.
Cozzarin, B. P., & Jeffrey, S. A. 2014. Human resource management practices and longitudinal workplace performance. Applied Economics Letters, 21(5): 344-349.
Earley, P.C. 1994. Self or group: Cultural effects of training on self-efficacy and performance. Administrative Science Quarterly, 39: 89-117.
Erez, M., & Kanfer, F. H. 1983. The role of goal acceptance in goal setting and task performance. Academy of Management Review, 8(3): 454-463.
Gal, P. N. 2013. Measuring total factor productivity at the firm level using OECD-ORBIS. Organisation for Economic Co-operation and Development.
Harris, R.G. 1999. Determinants of Canadian Productivity Growth: Issues and Prospects. Discussion. Paper#8., Ottawa: Industry Canada.
Hofstede, G., 1980. Culture’s Consequences: International Differences in Work-Related Values. Sage, Beverly Hills.
Hofstede, G. 2016. Masculinity at the national cultural level. In Y. J. Wong, S. R. Wester, Y. J. (. Wong, & S. R. (. Wester (Ed.), : 173-186. Washington, DC, US: American
Psychological Association.
Hofstede, G., Hofstede, G. J., & Minkov, M. 2010. Cultures and Organiations: Software of the Mind, 3rd Edition, McGraw-Hill USA.
34 House, P.J., Hanges, M., Javidan, P.W., Dorfman, V., & Gupta. 2004. Culture, Leadership, and Organizations: The GLOBE Study of 62 Societies. Thousand Oaks: Sage.
Huselid, M. 1995. The impact of human resource management practices on turnover, productivity, and corporate financial performance. Academy of Management Journal, Vol. 38(3): 635-672.
Krugman, P. 1994. The age of diminished expectations: U.S. economic policy in the 1990s. Cambridge: MIT Press.
Kull, T.J., Yan, T., Liu, Z., & Wacker, J.G. 2014. The Moderation of Lean Manufacturing Effectiveness by Dimensions of National Culture: Testing Practice-Culture Congruence Hypotheses. International Journal of Production Economics, 53:1-12.
Lewis, M. A. 2000. Lean production and sustainable competitive advantage. International Journal of Operations & Production Management, 20(8): 959.
Newman, K., & Nollen, S. 1996. Culture and congruence: the fit between management practices and national culture. Journal of International Business Studies, 27(4): 753-779. Mefford, R.N. 1986. Determinants of Productivity Differences in International Industry. Journal of International Business Studies, 17(1):63-82.
Milikic, B. B. 2009. The influence of culture on human resource management processes and practices: The propositions for Serbia. Ekonomski Anali / Economic Annals, 54(181): 93-118.
Mitchell, R., Obeidat, S., & Bray, M. 2013. The effect of strategic human resource
management on organizational performance: The mediating role of high-performance human resource practices. Human resource management, 52(6): 899-921.
35 Peretz, H., & Fried, Y. 2012. National cultures, performance appraisal practices, and
organizational absenteeism and turnover: A study across 21 countries. Journal of Applied Psychology, 97(2): 448-459.
Sargent, T.C. & Rodriquez, E.R. 2000. Labor or Total Factor Productivity: Do We Need to Choose? International Productivity Monitor, 1:41-44.
Shi, X., & Wang, J. 2010. Interpreting Hofstede model and GLOBE model: which way to go for cross-cultural research? International Journal of Business and Management, 6 (5): 93-99.
Syverson, C. 2011. What determines productivity? Journal of Economic Literature, 49(2): 326-365.
Wacker, J. G., & Sprague, L. G. 1998. Forecasting accuracy: Comparing the relative effectiveness of practices between seven developed countries. Journal of Operations Management, 16(2): 271-290.
Wagner, J. 2005. Exports and Productivity: A survey of the evidence from firm level data. HWWA Discussion Paper 319.
UNIDO. 2007. Determinants of total factor productivity: a literature review. United Nations Industrial Development Organization, Vienna.
Van Hoorn, A. 2014. Individualism and the cultural roots of management practices. Journal of Economic Behavior & Organization, Vol. 99: 53-68.
Waldman, D., & Sully de Luque, M., & Wang, D. 2012. What can we really learn about management practices across firms and countries? Academy of Management Perspectives, Vol. 26(1): 34-40.
36 Yauch, C. A., & Steudel, H. J. 2002. Cellular manufacturing for small businesses: Key
37
APPENDIX 1: ASSUMPTIONS CHECKS
TABLE 7 IQR Test Outliers
Low High Inner fences -1,95 1,93 # mild outliers 10 21 Outer fences -3,4 3,39 # severe outliers 2 3 TABLE 8
HILO identifying Outliers
R Company -7,89 2896 -5,73 3368 3,57 3397 3.62 3668 4,94 195 FIGURE 2
38 FIGURE 3
Standardized normal pot
39 FIGURE 5
Augmented component-plus-residual plot target practices
FIGURE 6
