The effect of human resource management and other factors to income differences in soes in vietnam

Nội dung text: The effect of human resource management and other factors to income differences in soes in vietnam

1. THE EFFECT OF HUMAN RESOURCE MANAGEMENT AND OTHER FACTORS TO INCOME DIFFERENCES IN SOEs IN VIETNAM PhD. Cuong Tat Do dotatcuong@gmail.com PhD. Anh Ngoc Thi Ngo ngocanhngo.npa@gmail.com Institute of Economics, Ho Chi Minh National Academy of Politics, Hanoi, Vietnam Assoc. Prof. PhD Van Anh Thi Le anhvanlt@neu.edu.vn Faculty of Management Science, National Economics University, Hanoi, Vietnam Abstract This paper utilizes labor force survey to explore the effect of human resource management and non-human resource management factors on income differences in SOEs in Vietnam. Employing multivariate regression method, this paper find out that income differences is existed in Vietnam SOEs and its origins could be come from professional skill level, experience and hours of working at micro level; at macro level income differences in Vietnam SOEs might come from the inefficient implementation of government policy. The policy implications from this study are: (i) government should focus on training labor skills; (ii) government should perform their best on labor policy implementation. Key words: income differences, human resource management, SOE, Vietnam JEL codes: D63, J31 1. Introduction This paper aims to find out the existence of income differences in Vietnam SOEs and then explains the relationship between income differences among employees who are working for SOEs in Vietnam in selected industries1 and its influenced factors grouped as HRM factors, and non-HRM factors. Therefore, survey data on Labor force conducted by Vietnam General Statistics Office in various years has been utilized in order to achieve the chapter’s goals due to its scope and relevant variables to this study. The survey includes all industries and types of companies, so a sub-sample, which is focused in SOEs, will be extracted from the survey. 1 Industries are selected based on the significant of income differences and number of observations. 541
2. This paper is organized in four main sections excluding introduction and conclusion sections. The model is presented in the section 2 where the author presents the theoretical and empirical models that will be used for empirical analysis. Section 3 will present some general statistical information of the sample used for the study. The next section will show the empirical results where the author explains the relationship between HRM and non-HRM factors and income differences in Vietnam SOEs in year 2007, 2012 and 2013. Section 5 will demonstrate further discussions draw from the empirical analysis. 2. Literature Review Checchi and Garcớa-Peủalosa (2008) propose that influences of institutions on income differences can be revised with different aspects such as legislation, tax, minimum wage and employees’ benefits, union intervention. Regarding to impacts of legislation, minimum wage and unemployment benefits policies on income distribution there are numerous researches on these themes. Stiglitz (2010) points out that government policy might cause changes in income distribution. He argues the policies implemented to stop the effects of the recession, such as the enormous rescue packages offered by the United States government to a number of financial institutions and industries, may cause even more income structures if financed by public deficit which will later oblige the government to increase tax rates. Checchi & Garcớa-Peủalosa (2008, p.608), examines relations of employment protection legislation on employees’ wage. He claims that ‘employment protection legislation reduces the risk of job loss and shifts reallocation costs from workers to employers. As a result, employers refrain from firing in downturns but also from hiring in booms and hence the overall effect on wages and employment is ambiguous. Similarly, a minimum wage need not reduce employment, which is determined by average wages; in addition, it may also reduce the monopolistic power of firms, thus raising employment among low-wage earners. There have been conflicts in findings on the relations between institutions and income differences in current literature. Nickell, Nunziata and Ochel (2005) and Bassanini and Duval (2006) find no statistical effect for employment protection legislation, while Bertola et al. (2002) obtain a positive and significant impact on unemployment. Union density has no significant effect, while both Nickell et al. and Bassanini and Duval find that greater wage coordination reduces unemployment. Moreover, the overall explanatory power of institutions is substantial, with LMIs explaining about 55% of changes in unemployment within countries, according to Nickell et al. Furthermore, Kerr and Kugler (2007) find that employment protection legislation reduces employment flows, while Lee (1999) provides evidence that changes in the minimum wage were responsible for part of the increase in wage 542
3. dispersion observed in the US during the 1980s (Checchi & Garcớa-Peủalosa, 2008 cited Kugler (2007) & Lee (1999) ). Reasons that explain this situation is empirical works have been taken on different contexts. The evidence indicates that more democratic countries, better law enforcement and greater financial development are associated with a more equal distribution of income, while a more segmented labour market is correlated with greater inequality (Barro (2000), Bourguignon and Morrisson (1998) and Li, Squire and Zou (1998)). Income differences has been investigated in the relationship with compensation at policy and practice level. At the policy level, these researches focus on investigating how minimum wage policies, labour law, health insurance policies and layoff legislation affect wages, benefits and employment (Werner & Ward 2004). However, most of the studies concentrate on effects of these above factors on wage and ignore their impacts on other components of compensation such as bonuses and benefits. Gerhart and Rynes (2003) claim that this literature gap is common and more research on other aspects of compensation would be expected. At practical level, compensation involves in wage dispersion in two major aspects: unionization and implementation of new HRM strategies of employers. Unions have aimed at a stabilization of wage within and among employers in order to enhance the cohesion of their bargaining units, to reduce labour-cost competition among employers and to narrow supervisors' scope for adjusting wages in an unfair or arbitrary manner (Freeman 1980). HRM models employ internal productivity- related incentives for workers, such as individual or group bonuses, merit-based raises and profit-sharing plans, any of which could raise wage dispersion directly. In addition, the absence of unionization may allow employers to use internal wage levels as a competitive device, which could raise internal wage variation (Groshen 1993). In most cases, new HRM strategies seek to reduce costs of firms. There have been various studies of wage gaps under the view of HRM, such as gender and wage gaps, individual determinants and wage structure, individual outcome of compensation and performance at work and pay equity and job satisfaction (Werner & Ward 2004). Werner and Ward (2004) state that factors related to HRM practices conciliating relationships between gender and wage gap are family background, education, affirmative action, training, mentoring and negotiation skills. However, these contents are segmented in many studies and almost these researches look at only employees’ wage rather than cover other aspects of compensation such as bonuses, awards and benefits. Gerhart and Rynes (2003) question if the other components of compensation are influenced by gender and race and it needs to call for more complex models to examine not only wage gap but also nonwage compensation gap. 543
4. 3. Model Actual labor income from their salary of wage package is defined as wi, or could be seen as competitive income, and their market income could be i; wi = λii – where λi is represented for the effect of institutional factors such as member of trade union or minimum wage. Institutional factors help to raise the actual income from salary or wage packet closer to the market income. In fact, if labor market is purely free market, actual wage will equal to potential income and λi will equal to 1. However, employers usually offer salary package lower than the market income. Indeed, without the intervention of government, employers might apply a very low wage scheme to employees. This situation can be captured in the following graph: Figure 1. Labour income from wage or salary package wage i wi Without effect of o institutional factors wo life time 0 To estimate the effects of HRM and non-HRM factors to income differences in SOEs in Vietnam recently, we apply the following econometric model at individual level based on suggestion of Katz (1998) and Schultz (1951) among others: 퐑퐢,퐣,퐭 = 후 + 후퐦(퐇퐑퐌 퐯퐚퐫퐢퐚퐛퐥퐞퐬)퐢,퐣,퐭 + 훜퐢,퐣,퐭 ( ) 퐑퐢,퐣,퐭 = 훂 + 훂퐦(퐇퐑퐌 퐯퐚퐫퐢퐚퐛퐥퐞퐬)퐢,퐣,퐭 + 훃퐧(퐍퐨퐧_퐇퐑퐌 퐯퐚퐫퐢퐚퐛퐥퐞퐬)퐢,퐣,퐭 + 퐮퐢,퐣,퐭 ( ) 퐑퐢,퐣,퐭: Ratio of actual income of individual i in industry j at the time t to mean income of industry j at the time t; 퐇퐑퐌 퐯퐚퐫퐢퐚퐛퐥퐞퐬퐢,퐣,퐭: This is a set of HRM variables including labours’ experience, education attainment, skills – proxied by level of professional in their career path for labour i in industry j at the time t and human resource development policy of local government of province x in time t. (퐍퐨퐧_퐇퐑퐌 퐯퐚퐫퐢퐚퐛퐥퐞퐬)퐢,퐣,퐭: This is a set capturing the non - HRM variables of labour including gender, ethnic and effect of firms’ sizes. 544
5. 훂 , 후 : are constants, while 훂퐦, 후퐦: are estimated coefficients for HRM variables; 훃퐧: are estimated coefficients for non – HRM variables; 퐮퐢,퐣,퐭, 훜퐢,퐣,퐭: are error terms. The aim of equation (1) is to explore the relationship between HRM’s variables and income differences, proxied by ratio between individual income and mean income of industry that the individual is working for. The aim of equation (2) is to explore the relationship between HRM and non-HRM variables and income differences. The two equations will provide us in-depth information about the effect of HRM’s variables themselves and the effect of HRM’s variables under the interaction with non-HRM’s variables. Through the outcome of estimation of the equation (2), we can know more about the effect of HRM’s variables on income differences. If the values of coefficients of HRM’s variables do not vary significant between the two equations, we might conclude that the effects of HRM’s variables are consistent and reliable. Therefore, we can base on these empirical results to draw the policy implication in the next part of the thesis. It also helps the author to figure out the limitation of the data in order to plan the next phase where the author will do some in-depth interview. It is expected that the dataset cannot provide all information we need to know about the relationship, but it will provide a preliminary insight to this complicated relationship. Therefore, the sign of estimated coefficients is more important than their values. For example, if education has negative sign, it means that education will reduce the income differences in the context of Vietnam; so if Vietnam government wants to reduce income differences, they should encourage employees in SOEs do some further education. The other considerable example is professional skill; it might expand the income gap under the circumstances of Vietnam. Therefore, if Vietnam government wants to reduce income differences, they should implement a set of several policies that encourage employees increase their professional skills and re- distribute appropriately income between high and low professional skills employees via income tax or other instruments of compensation policy. In this thesis, however, we do not focus on analyse the effects of government policies on income differences at general meaning, but we will try our best analyse significant and relevant policies. 4. Empirical results In this section, two empirical results of the equation (1) and equation (2) will be presented. Equation (1) is focused on only the effect of HRM variables on income differences, while equation (2) is concentrated on the effect of both HRM and non- HRM variables. The both models have been estimated separately in year 2007, 2012 and 2013 because there is no panel data for three years. Therefore, the comparison 545
6. and the trend analysis between the three years are limited. However, the value and sign of coefficients could be relied on to explain the relationship between HRM and non-HRM variables and income differences. Table 1. The effect of HRM variables on income differences in 2007 and 2012 Variables 2007 2012 2013 Dependent variable: Ratio between personal actual income and mean income at industrial level Professional skill level 0.419 0.454 0.131 (0.002) (0.003) (0.002) Human Capital policy at provincial level 0.002 0.013 -0.005 (0.001) (0.005) (0.006) Education attainment 0.003 -0.004 -0.003 (0.000) (0.005) (0.001) Experience 0.002 0.003 0.001 (0.000) (0.000) (0.000) Weekly hours of working 0.002 0.003 0.002 (0.000) (0.000) (0.000) Social insurance 0.015 -0.008 0.096 (0.007) (0.008) (0.018) Constant -0.248 -0.290 0.297 (0.017) (0.030) (0.051) Adjusted R2 0.3660 0.7106 0.6234 Observations 31,897 8355 3257 The validity of the model Yes Yes Yes BG test for heteroskedasticity Rejected Rejected Rejected Note: *, , : coefficient is statistically significant at 10%, 5% and 1% respectively. Source: Authors' estimation. As shown in table 1, the effect of each HRM variable on income differences in Vietnam’s SOEs is varied significantly among 2007, 2012 and 2013. The most significant and largest effect to income differences is professional skill levels. This is true because the higher skilled labor will more likely have better jobs and better payment. In the case of Vietnam SOEs, the estimated results show us that income differences might be caused mainly by the difference in labors’ professional skills. When a worker increases her professional skills by one unit, her income will likely 546
7. increase 0.42, 0.45 and 0.13 points to the mean of industrial incomes in year 2007, 2012 and 2013 respectively. The results imply that employees might compete each other's in order to have higher income by getting better on-the-job training or enrolling to the necessary courses provided by educational institutions to have new skills or knowledge. Education does perform as a factor preventing income differences in the case of Vietnam SOEs. The coefficient of education attainment change from positive to negative signs in the regression model is a demonstration of the complex effect of education on income differences. In 2007, education could be seen as a factor of widening income differences, but in 2012 and 2013 education might act as a cause of narrowing income differences. The complex effect of education reveals the changing role of education on income differences under the circumstance of Vietnam SOEs. This result might lead us to a thinking that employee in these SOEs will compete each other in order to have permission to do further learning from their manager or boss2 aiming to have better income with their new education certificate. It is fascinating that the current situation of studying of employees in SOEs is supporting this result3. Working experience and working hours positively influence to income differences in year 2007 and 2012. Indeed, those variables’ coefficients increase slightly between year 2007 and 2012. It implies the positive effect of experience and working time on the widening income differences4. The longer workers stay in the same industry the higher income they might earn; and the longer hour they work the higher income they will earn. If workers increase 1 year of experience or 1 hour per week, the ratio between individual income and mean income of industry that the individual is working for will increase about 0.002, 0.003 and 0.001 points in year 2007, 2012 and year 2013 respectively. Obviously, experience and normal working hour should not have much effect on labors in Vietnam’s SOEs because their wages are paid based on their wage package level and it is regulated by Vietnam Government5. Extra working hours, on the other hand, will cause mainly the different in income 2 In case of Vietnam where hierarchy and bureaucratic system is still very strong in SOEs, obtaining an approval from manager or boss for studying is a crucial step and is not easy task for almost any cases. 3 There are not official reports from Vietnam government bodies about this situation but based on the observation of the night time class at the various universities in Hanoi, Ho Chi Minh City, we can easily to understand this trend. 4 This idea, under the management view, is quite different with the explanation from economics perspectives. 5 In Vietnam, Vietnam government issues and amends fundamental wage every two or three years. Worker has been assigned with a wage ratio. Their wage will equal fundamental wage times wage ratio and plus some extra money due to their position and outcomes of business. For example, a beginning worker will be assigned a wage ratio 2.67 and fundamental wage is 1,000,000 VND (approximately $55); then her wage will equal 2.67 x 1,000,000 VND + extra money (if applicable) = 2,670,000 VND + extra money. After several years, she will be promoted to become a leader of a team and then she has got extra ratio at 0.3, her wage ratio will be assumed at 4.05 and fundamental wage will be 1,500,000 VND; her wage will equal to (4.05+0.3) x 1,500,000 + extra money (if applicable) = 6,525,000 VND + extra money. 547
8. among workers. The positive sign of coefficient of working hours per week confirms this idea. The more important thing is that all workers know that when they are allowed to work extra, they will earn more money, but not all of them will have chance to work extra. It is said that, workers who have close relationship to their managers/bosses or are managers/bosses’ relative may likely have more chance to work more. Unfortunately, the survey does not have any information test this idea, so we do need to do further exploration by doing in-depth interview. Human resource development policy of provincial government affects positively to income differences of the SOEs at provincial level. The major aim of human resource development policy is to provide better on-the-job training, skills and knowledge for local labors in order to meet the needs of employers. As mention above, skills and knowledge of labors might lead to a widening of income differences, so the policy of local government is accidently widened the income gap in SOEs6. The unexpected outcome of the policy has increased between year 2007 and 2012; in year 2007 the effect of the policy on income differences is 0.002, while in year 2012 its effect is increased to 0.013. Actually, it reflects the increase of effect of professional skill level. The reflection implies to us that the current situation of income differences is widening accidently by the relevant local government policies. Therefore, it also implies that the making public policy should be aware of such side effects in order to avoid the unexpected effects in the future. Actually, the main aim of public policy in Vietnam is to support the narrowing income gap rather expanding the income differences status but making public policy without consideration of all stakeholders or all internal and external factors might lead to unexpected outcomes7. In year 2013, the effect of this variable is negative and is not statistically significant. Therefore, it is said that the effect of human development policy of provincial government is quite complicated and need to be watched closely further in the future. It is understandable that the effect of government policy will be lagged for one or two years after the time of deployment. Thus, the effect of the policy in year 2013 might be a consequence of the implementation of the policy in year 2012 or 20118. In the case of Vietnam’s SOEs, only full-time workers have social insurance, while part-time, contracted or casual workers do not have social insurance9. Therefore, 6 The wrong doing effect of the government policy on income differences is an accident because policy making cycle has been affected badly by internal or external factors. The government is trying to eliminate the unexpected effect but sometimes the government cannot do as its desire. 7 Although human resource development policy in local level is measured by firms’ leader opinion, they show us the true valuation of local government policy implementation and also the making policy process. Planning public policy without thoughtful analysis will definitely lead to unexpected results. Additionally, the situation is happening in Vietnam quite often when policy makers do not wish to cooperate with researcher or to do evidence-based policy. 8 The effect of this policy therefore is not for sure at the moment. 9 If part-time, contracted and casual workers want to have social insurance, they have to pay themselves. 548
9. social insurance should be a positive coefficient. For year 2007, the coefficient is positive and statistically significant at 5%, while for year 2012 the coefficient is negative and statistical in-significant; in year 2013, the coefficient of the variable is not only statistical significant but also more robust than in year 2007. This implies that social insurance would be a factor that has positive contribution to income differences. However, due to the opposite signs of coefficient between the three years, it is needed to take more attention on this coefficient. Based on estimated result for social insurance in 2007 and year 2013, it is shown that the workers who have social insurance have more likely higher income than the other workers who do not have the insurance10. Table 2 shows us the full model where the author wishes to understand the total effect of HRM and non-HRM variables on income differences in SOEs in Vietnam. Under the interaction of those variables, the coefficients of HRM variables has changed very slightly in comparison with the previous model where only HRM variables have been considered. Table 2. Effect of HRM and Non-HRM variables on income differences in 2007 and 2012 Variables 2007 2012 2013 Dependent variable: Ratio between personal actual income and mean income at industrial level I. HRM variables Professional skill level 0.418 0.453 0.131 (0.002) (0.003) (0.002) Human Capital policy at provincial 0.002* 0.008 -0.006 level (0.001) (0.005) (0.006) Education attainment 0.003 -0.004 -0.004 (0.000) (0.000) (0.001) Experience 0.002 0.003 0.001* (0.000) (0.000) (0.000) Weekly hours of working 0.002 0.002 0.002 (0.000) (0.000) (0.000) Social insurance 0.015 0.011 0.094 (0.007) (0.010) (0.018) 10 It could be true because only employees who have long term contracts will have social insurance. These employees have stable job than seasonal employees. Thus, full-time contracted employees will likely have more chance to earn higher income than seasonal and part-time contracted employees 549
10. Variables 2007 2012 2013 II. Non – HRM variables Gender 0.013 -0.011* 0.011 (0.003) (0.006) (0.010) Ethnicity -0.003 -0.038 0.044 (0.006) (0.009) (0.019) Working for large company 0.036 0.056 NA (0.007) (0.024) Working for medium company 0.023 0.016 NA (0.005) (0.026) Constant -0.285 -0.234 0.272 (0.020) (0.044) (0.051) Adjusted R2 0.6345 0.7115 0.6240 Observations 31,879 8355 3257 The validity of the model Yes Yes Yes BG test for heteroskedasticity Rejected Rejected Rejected Note: *, and : Coefficient is statistically significant at 10%, 5% and 1% respectively. In year 2013, the survey does not have any information about the size of enterprises. Source: Authors' estimation. The effect of professional skill level on income differences has been confirmed in the full model where the coefficient is statistically significant in year 2007, 2012 and 2013. The coefficient increases from 0.418 in year 2007 to 0.453 in year 2012 and then reduces to 0.13 in year 2013. Skills are increasing during the time period and then income differences is also widening under the effect of professional skill level in the same time. Professional skills are very important for earning income and there is no doubt that high skill workers will likely have more chance to achieve better pay than the other. The reduction of the value of this coefficient in year 2013 in comparison with year 2007 and 2012 implies that the importance of professional skills on income differences in year 2013 may be less than in previous two years. The sign of the coefficient, however, is remained the same for three years 2007, 2012 and 2013. Additionally, the important underlying idea here is that the government should have some more sound policies to reduce income gap between high and low skilled labors. 550
11. Education has the same effect sign in the full model in comparison with the previous model. In year 2007, education attainment contributes positively to income differences, while in year 2012 and 2013 education attainment have negative effect to income differences. In year 2007, education seems to support the widening of income gap, while in 2012 and 2013 education could be seen as a factor closing the gap of income between skilled labors. Explaining the effect of education is not an easy task because education is here only an expression of quantity of education – years of schooling; information of the quality of education of employees and the quality of educational institutions do not available in this survey. It is hard to say simply that education will help to reduce or increase income differences because the effect of education to income differences needs to be placed in specific conditions. For example, a well-educated worker can work well in a plumber company but might be a bad worker in a chemical company. Therefore, education in the form of on-the-job training or further education would be good suggestions to reduce the gap of income differences in Vietnam SOEs. Effect of experience and hours of working on income differences in the full model do not change much in comparison with the previous model. The value of coefficients of experience and weekly working hour show us that better experience and longer hour will lead to higher income and then the ratio between personal income and mean income of industry will be widened. The value of coefficient of experience allows us to conclude that experience of workers in SOEs in Vietnam is supporting the expansion of income differences. Additionally, the length of working hours also supports the increase of income differences. The one more year of experience will likely increase 0.002, 0.003 and 0.001 points of ratio between actual personal income and mean income of industry in year 2007, 2012 and 2013 respectively. The one more working hour per week will likely help workers’ incomes escaping further from the mean income of industry by 0.002 points in year 2007, 2012 and 2013. The effect of social insurance variable in year 2007, 2012 and 2013 are positive. It confirms our expectation that social insurance may contribute positively to income differences in the case of Vietnam SOEs. It reflects the fact that only long term full-time contracted workers have social insurance, while the other types of contracted workers such as seasonal or part-time workers do not have social insurance. In addition, long term full-time contracted workers have stable job and enjoy the employment cycle. Among non-HRM variables, working for large company and working for medium company are the most two critical dummy variables. Coefficients of these variables show us that workers working for large and medium scale SOEs are earning 551
12. higher than workers working for small scale SOEs. The estimated results for large scale SOEs demonstrate that workers’ income in large SOEs is increasing between 2007 and 2012. Workers in these SOEs are likely earning 0.036 and 0.056 point of ratio, between personal actual income and mean of industry income, higher than the workers in small scale SOEs in year 2007 and 2012 respectively. It will lead to policy implication that the government should have a policy in order to make the income gap avoiding the effect from firms’ size. Unfortunately, in year 2013 those questions were taken out the survey, so we do not have any information about this variable in order to see the development trend. 5. Conclusion In this paper these authors wish to figure out the factors affecting to income differences in Vietnam SOEs by using regression analysis. In order to achieve the goal of this paper the author utilizes the Labor force survey dataset conducted by Vietnam General Statistics Office under the technical support from International Labor Organization in Vietnam on employment in year 2007, 2012 and 2013. The results of the regression analysis are in line with the expectation of the author. Income differences is existed in Vietnam SOEs and its origins could be come from professional skill level, experience and hours of working at micro level; at macro level income differences in Vietnam SOEs might come from the inefficient implementation of government policy. Income differences has occurred in Vietnam SOEs and the level of income differences is quite high. The source of income differences in Vietnam SOEs is somewhat different or similar to the other experience of the other countries. Vietnam SOEs have been transformed fundamentally from a central planning mechanism economy to oriented market mechanism economy. Therefore, SOEs in Vietnam have been re-organized and re-developed based on market economy, so they also have had severe problems like the other state enterprises in developed market economy such as income differences. Based on empirical result, the author find out that those factors can raise income also can raise income differences. Education, professional skills are major factors can raise income, so they are two factors can raise income differences. However, how can they raise income differences? Which processes of HRM cycle will help us to understand further this effect. It is said that, training process is a crucial step in employment cycle, but training could be a source of income differences later if we do not have appropriate policy to against it. Training process in Vietnam SOEs is a further education after 552
13. official education and it is very important to not only employees but also the SEOs. Further training will transform theoretical knowledge to practical knowledge and on- the-job training will transform practical knowledge to professional skills. Different professional skills will enable the ability to have higher income, so income differences will occur no matter what employees want or do not want. Based on the empirical result, professional skill levels have contributed consistently to income differences in Vietnam SOEs. The value of coefficient of professional skill levels in various regression strategies has shown that under the circumstances of Vietnam SOEs professional skill levels could be a major source of income differences. Consequently, the solution to reduce income differences should be started with a step of employment cycle such as training or retaining. Focusing on training and retaining will relate to a set of other tasks such as compensation policy, promotion policy or continue training policy or transfer policy. It means that the better solution to reduce income differences is the leaders of SOEs should perform an appropriate set of policies rather than just focus only one thing. References 1. Barro, R. J. (2000). Inequality and Growth in a Panel of Countries. Journal of economic growth, 5(1), 5-32. 2. Bassanini, A., & Duval, R. (2006). Employment patterns in OECD countries. 3. Bertola, G., Blau, F. D., & Kahn, L. M. (2002). Comparative analysis of employment outcomes: lessons for the United States from international labor market evidence. The roaring nineties: can full employment be sustained, 159-218. 4. Bourguignon, F., & Morrisson, C. (1998). Inequality and development: the role of dualism. Journal of development economics, 57(2), 233-257. 5. Checchi, D., & Garcớa-Peủalosa, C. (2008). Labour market institutions and income differences. Economic Policy, 23(56), 602-649. 6. David, H., Kerr, W. R., & Kugler, A. D. (2007). Does employment protection reduce productivity? Evidence from US states. The Economic Journal, 117(521). 7. Freeman, R. B. (1980). The exit-voice tradeoff in the labor market: Unionism, job tenure, quits, and separations. The quarterly journal of economics, 94(4), 643-673. 8. Gerhart, B., & Rynes, S. (2003). Compensation: Theory, evidence, and strategic implications: SAGE publications. 9. Groshen, E. L. (1993). HRM policy and increasing inequality in a salary survey: Federal Reserve Bank of Cleveland, Research Department. 10. Katz, L. F. (1999). Changes in the wage structure and earnings inequality Handbook of labor economics (Vol. 3, pp. 1463-1555): Elsevier. 553
14. 11. Li, H., Squire, L., & Zou, H. f. (1998). Explaining international and intertemporal variations in income differences. The Economic Journal, 108(446), 26-43. 12. Nickell, S., Nunziata, L., & Ochel, W. (2005). Unemployment in the OECD since the 1960s. What do we know? The Economic Journal, 115(500), 1-27. 13. Schultz, T. W. (1951). The declining economic importance of agricultural land. The Economic Journal, 61(244), 725-740. 14. Werner, S., & Ward, S. G. (2004). Recent compensation research: An eclectic review. Human Resource Management Review, 14(2), 201-227. 554