MAURITIUS Earnings Mobility and Inequality of Opportunity in the Labor Market OVERVIEW OV E RV I E W 1 This booklet contains the overview of the report “Mauritius: Earnings Mobility and Inequality of Opportunity in the Labor Market” as well as main findings from a review of youth employment policies in Mauritius. A PDF of the full-length report is available at https://openknowledge.worldbank.org/. Please use the full version of the report for citation, reproduction, and adaptation purposes. This work is a product of the staff of the International Bank for Reconstruction and Development/The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Photography: Arne Hoel/World Bank. Cover Design: Circle Graphics, Inc. Design and Layout: Circle Graphics, Inc. Additional material relating to this report can be found on the World Bank Mauritius website (www.worldbank.org/mauritius). The material includes a fact sheet. © 2018 International Bank for Reconstruction and Development / International Development Association or The World Bank Group 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 WWW.WORLDBANKGROUP.ORG Acknowledgments T his work has been prepared by a core team comprising Marco Ranzani (TTL, Economist, GPV01) and Giuseppe Grasso (Consultant, GPV01). The analysis has benefited from discussions held in Mauritius with government officials and representa- tives of the private sector and of academia. Staff at Statistics Mauritius were instrumental in the realization of the report. The team would like to thank particularly Ms. Aimée Li Fa Cheung Kai Suet (Director), Ms. Yasmin Cassimally (Deputy Director), Ms. Set Fong Cheung Tung Shing (Deputy Director), Ms. Naseem Ramjan (Statistician, Continuous Multi-Purpose Household Survey Unit), and Ms. Taslimah Joomun (Statistician, Tourism/Publication Unit) for their excellent support and collaboration. The team gratefully acknowledges the support of Pierella Paci (Practice Manager, GPV01), Mark Lundell (Country Director, AFCS2), Carolin Geginat (Program Leader, AFCS2), Erik von Uexkull (Resident Representative - Mauritius), Alex Sienaert (Senior Economist, CROCR), Mariella Beugue (Program Assistant, AFMMU), Rachel Ng Cheong (Temporary, AFMMU), and Martin Buchara (Program Assistant, GPV01). The peer reviewers of the report are Tom Bundervoet (Senior Economist, GPV01) and Sara Troiano (Young Professional, GSP08). OV E RV I E W 1 “If income mobility were very high, the “Higher income inequality would be degree of inequality in any given year less of a concern if low-income earners would be unimportant, because the became high-income earners at some distribution of lifetime income would point in their career, or if children of be very even. . . . An increase in income low-income parents had a good chance mobility tends to make the distribution of climbing up the income scales when of lifetime income more equal. ” they grow up. In other words, if we had a high degree of income mobility, we would PAUL KRUGMAN (1992, 28) be less concerned about the degree of inequality in any given year.” ALAN KRUEGER (2012, 3) Introduction O ver the last decade, steady economic growth has placed Mauritius solidly in the upper middle income group and has contributed to further reducing poverty. Over the last 10 years, the Mauritian economy posted an average annual per capita growth of about 3.6 percent as it continued a process of structural transformation from traditional and low-skill sectors, such as agricultural and textile, towards services. Per capita GDP of $22,309 (measured in current international dollars) in 2017 is the 3rd highest in Africa and places Mauritius solidly in the upper middle income category. Measured against the $5.5 per day 2011 PPP line, poverty is estimated at 18.1 percent in 2012, well below the average of 34.4 percent among upper middle income countries and has declined from 20.3 percent in 2006/07. Economic growth has been accompanied by an increase in income inequality, despite redistribution efforts by the Mauritian government. As measured by the Gini coefficient, inequality in Mauritius is comparable to the level of inequality in countries at a similar level of economic development (0.41) and moderate compared with the most unequal countries in the world, such as South Africa (0.63), Botswana (0.61) and Namibia (0.59).1 However, the inequality in Mauritius has widened substantially over the last 15 years. The World Bank (2017) shows that household income inequality has widened significantly from 0.36 in 2001 to 0.42 in 2015, particularly in the aftermath of the global economic downturn and terms-of-trade shock that hit Mauritius between 2008 and 2015. Rising inequality in household income from labor has been the main culprit behind the growth in overall income inequality, while the government’s efforts to redistribute the benefits of growth have helped reduce the pace of the increase. A divergence in labor incomes due to skills shortages driven by structural transformation has been the single most important contributor to increasing inequality. The economy has experienced a progressive shift from traditional and low-skill sectors to services, notably professional, real estate, and financial services. This transformation has generated a considerable rise in the demand for skilled workers that has not been matched by an equally rapid increase in the supply of skilled workers, notwithstanding the substantial improvement in educational attainment 1 Estimates are based on the latest available surveys from upper-middle-income countries (World Bank, Poverty and Equity database). OV E RV I E W 5 among the population. As a consequence, high-skilled workers benefitted from considerably larger increase in wages compared with low-skilled workers. This report sheds light on the extent to which earnings mobility and inequality of opportunity in access to the labor market have contributed to the increase in earnings inequality in Mauritius. Among the most important concerns about rising inequality is a situation where people become trapped in low-paying jobs and do not have the opportunity to improve their welfare through their own efforts. For this reason, this report takes a closer look at the extent and nature of earnings mobility and inequality of opportunity in the Mauritian labor market with the objective of addressing the following questions. Do low-paid workers catch up in earnings with high-paid workers who have the same characteristics? Which individual characteristics foster earnings mobility? To what extent do circumstances at birth affect the ability of an individual to access certain good job opportunities? The study finds that there is convergence in earnings’ growth over a period of 16 months. In other words, workers initially at the bottom of the earnings distribution enjoy a larger growth in earnings than workers initially at the top. The key to reconciling this finding with the growing inequality in earnings is that workers move along the earnings ladder over time. Workers initially at the bottom are not found at the bottom after one year because they move slightly upward and the opposite occurs for workers initially at the top. In addition, and interestingly from a policy perspective, the individual characteristics that are the main driver of the rise in inequality also hinder individual earnings mobility. Women and workers with low educational attainments not only face lower initial earnings, they also face more difficulty in catching up with high-paid workers. On the positive side, unequal access to opportunities in the labor market is moderate and has declined over time, particularly thanks to improvements in educational attainments. For these reasons, the role of inequality of opportunity in explaining the rise in earnings inequality is limited. However, women and youth are still considerably disadvantaged in accessing jobs and full-time jobs, in particular. 6 MAURITIUS Part I Earnings Mobility B etween 2005 to 2015, earnings’ growth FIGURE 1  Percentage Changes in Earnings by Initial of low-paid workers converged towards Earnings Quintile, 2005–15 that of the high-paid, suggesting the 30.0 presence of earnings mobility. Data from the panel component of the Continuous Multipurpose 20.0 Household Survey (CMPHS 2005–15) suggest that, 10.0 Percent between 2005 and 2015, workers at the bottom of the earnings distribution saw their income rise faster than 0.0 those who started off with higher earnings.2 Workers –10.0 starting with earnings in the first quintile of the distribution posted an average growth in earnings of –20.0 1 2 3 4 5 about 24 percent or about Rupees 772 (Figure 1). By Quintile - Reported Initial Earnings contrast, workers starting at the top (the 5th quintile) Source: Based on data of the Continuous Multipurpose Household Survey, on average faced a reduction in earnings of about Statistics Mauritius. −15 percent, which corresponds to a median loss of Rupees 2,200 (Figure 1).3 2 The survey consists of an initial interview and three follow-ups: for example, a household interviewed in the first quarter of 2005 is then re-interviewed in the second quarter of 2005 and again in the first and second quarter of 2006. This allows to collect information about individuals over a maximum period of 16 months. 3 Earnings captured at one point in time might be affected by measurement error or transitory shocks. For this reason, the study also makes use of more permanent measures of earnings. First, an average of the earnings of individuals using information reported during the last three interviews is used instead of earnings reported in the first interview only. Second, predicted earnings are obtained taking advantage of variables that are predictors of a permanent advantage—such as age, educational attainment, gender, and household consumption—as a proxy for welfare. Estimates based on this adjusted measure of earnings indicate a median gain of about Rupees 473 for workers in the first quintile compared with a median loss of Rupees 515 for workers at the top. This corresponds to a change in earnings of about 9 percent at the bottom (the 1st quintile) and of –2 percent at the top (the 5th quintile). OV E RV I E W 7 FIGURE 2  Percentage Changes in Earnings by Gender and Educational Attainment, 2005–15 a. By gender b. By educational attainment 30 25 20 25 15 20 Percent 10 Percent 15 5 0 10 –5 5 –10 Up to Lower Upper Post- 0 complete secondary secondary secondary/ Female Male primary tertiary Source: Based on data of the Continuous Multipurpose Household Survey, Statistics Mauritius. Note: Columns represent the predicted percentage change in earnings among men and women with tertiary education (panel a) and among workers at different educational levels (panel b) assuming that they have average characteristics in terms of initial earnings, age, district of residence, initial employment category, sector, occupation, and transitions across sectors between the initial and final period. Catching up with the earnings of high-paid workers is more difficult for women. Among the individual characteristics for which data is available, gender and educational attainment are the most important drivers of earnings mobility for Mauritian workers.4 Conditional on the initial level of earnings, women experience less mobility than men with similar characteris- tics (Figure 2, panel a). This could be ascribable to unequal treatment but might also be the result of traditional models of housework and family care where women bear more of these responsibilities. According to data collected by AfroBarome- ter in 2017, over 7 in 10 Mauritians report that it is better for a family if a woman has the main responsibility of home and children care. Family-related activities compete with the labor market for women’s time and may lead women to seek less 4 The analysis controls for a number of individual characteristics, including initial earnings, age, gender, educational attainment, district of residence, initial employment category, sector, occupation, and transitions across sectors between the initial and final period. 8 MAURITIUS competitive and less remunerative career paths in exchange for greater employment flexibility. On average, working women devote three times more of their day than working men to household chores and childcare: 3.8 hours per day among working women compared with 1.2 hours per day among working men (CSO 2005). Lower education also results in lower mobility. Conditional on the initial level of earnings, workers with lower educational attainments are found to catch up less with the earnings of better educated workers with similar characteristics (Figure 2, panel b). For example, workers with post-secondary or tertiary education have a larger percentage change in earnings (+26 percent) than identical workers with up to completed primary education. Despite the convergence in earnings’ growth, inequality in earnings has continued to increase. The rise in earnings’ inequality observed over the last decade suggests that earnings at the bottom of the distribution have grown less than earnings at the top (Figure 3, red columns). At the same time, tracking workers over a period of 16 months shows that workers initially at the bottom of the distribution have experienced a larger increase in earnings than those initially at the top (Figure 3, blue columns). The key to understand the difference between these two findings is that workers at different points of the earnings distribution change over time, as workers move along the distribution. FIGURE 3  Median Changes in Earnings by Quintile: Workers initially at the bottom are not the same Anonymous and Panel Workers, 2005–15 workers that are found at the bottom at the end of the 1,500 period (16 months later) because they have moved 857 1,000 slightly upward. The opposite is observed in the case 772 702 314 345 of workers initially at the top. Therefore, inequality, 500 Change in earnings 142 25 74 which is a comparative static measure of the earnings - gap between high and low-paid workers observed at (500) (474) each point in time, might be widening, while tracking (1,000) individuals over time indicates that initially low-paid (1,500) workers earn larger gains in earnings relative to the (2,000) initially high-paid. (2,500) (2,200) The development of skills in line with the changing Bottom 2 3 4 Top economic landscape can further enhance mobility Quintile Panel Anonymous while reversing the increase in inequality. Workers with skills that are most demanded in the labor market Source: Based on data of the Continuous Multipurpose Household Survey, Statistics Mauritius. not only start off with higher earnings, but they also Note: Earnings are expressed in 2015 prices. OV E RV I E W 9 experience faster earnings growth. Therefore, policies targeted at developing skills that are in high demand have the potential to mitigate the increase in inequality and promote mobility. As a first step, a comprehensive assessment of cur- rent and future needs of firms in terms of skills is needed and can help inform education curricula for the coming genera- tion of workers. Training systems can help those who are already working adapt to the new demands of the labor market. Family-friendly social policies are also key. While Mauritius has made remarkable progress in terms of closing gender education gaps, the returns to human capital investments will not materialize unless progress can be made in closing the gender pay gap in the labor market. This starts with helping women access the labor market and re-enter once they have children. Affordable childcare and eldercare as well as working-time regulations that promote flexibility can help in this regard. Awareness campaigns can help to shift norms on the employment of women in high-paying positions. Investing in career promotion and leadership development programs for women could help them prepare for private sector jobs. Tackling wage discrimination is another priority. Currently, the Mauritian public sector – where a modest wage discrim- ination plays in favor of women – is most successful in attracting qualified women. If the private sector wants to attract a greater share of the growing number of well educated women, it needs to address wage discrimination. 10 MAURITIUS Part II Inequality of  Opportunity in the Labor Market E mployment rates do not always tell the full story on the availability of labor market opportunities in a country. Individuals might have a job but no labor market opportunities. For example, some individuals are willing to take any job even if the job profile does not match their skills because they cannot afford to be without one. Such individuals are recorded as employed in labor statistics, yet, in reality their labor market opportunities could be more constrained relative to individuals who are voluntarily unemployed. For example, individuals may choose to stay unemployed because they wait for better job opportunities to come along. They may do so because they can rely on other income sources and are not under pressure to find a job today. How- ever, standard labor force surveys do not allow to identify voluntarily unemployment. In addition to a general employment indicator, based whether or not someone worked for at least 1 hour in the seven days preceding the interview, this study identifies the following labor market opportunities: employment full time (whether someone worked in the previous seven days for at least 40 hours), and wage employment (whether someone worked in the previous seven days for a wage or salary). The unemployed and those employed in jobs that do not have the desired characteristics are without a labor market opportunity. Inequality in access to labor market opportunities is measured by the extent to which such opportunities are available primarily to those born in privileged circumstances rather than being allocated to individuals based on skills and effort. This part of the study is based on data from the 2000 and 2011 Population Census because the census best captures variables that can be used as circumstances at birth. Equality in access to labor market opportunities improved over time, largely thanks to better educational attainments across the population. Between 2000 and 2011, inequality in access to job opportunities declined for overall employment and wage employment (Figure 4, panel a and c). By contrast, such inequality increased moderately in the access to full-time employment from 5.1 to 6.2 (Figure 4, panel b). Such improvements are OV E RV I E W 11 FIGURE 4  Inequality in Access to Labor Market Opportunities, 2000 and 2011 a. Employed in the previous week b. Employed full time 90.5 64.6 Coverage Coverage 92.3 66.5 4.2 5.1 Dissimilarity Dissimilarity 3.3 6.2 86.7 61.3 IAC IAC 89.2 62.3 0 20 40 60 80 100 0 20 40 60 80 2000 c. Employed for a wage or salary 2011 76.1 Coverage 75.8 4.4 Dissimilarity 3.1 72.8 IAC 73.4 0 20 40 60 80 Source: Based on data of the Housing and Population Census, Statistics Mauritius. ascribable to a growing proportion of society entering the retirement age and important investments in education that have created greater equality of opportunity for the next generation. These improvements in access to labor market opportunities are to be welcomed. However, women and youth continue to suffer disproportionally from less access to jobs and to full-time jobs in particular. Age and education explain the largest share of inequality in the access to job opportunities (Figure 5). Men are more likely to access employment in general and full-time employment in particular (Figure 5). In 2016, the participation rate for men was as high as 89 percent versus 56 percent among women. Age is key to explaining differences in access to employment in general. This is in line with the high unemployment rates observed among 12 MAURITIUS FIGURE 5  Contributors to Inequality: Circumstances and Characteristics, 2000 and 2011 a. 2000 b. 2011 2.1 0.4 1.0 1.8 Gender Employed last week 3.6 87.7 6.3 Employed last week 17.1 68.8 12.2 Ethnicity 0.2 0.2 Place of residence Employed full-time Employed full-time 7.3 49.0 7.9 14.8 28.1 69.2 20.5 5 years (40 hours a week) (40 hours a week) 3.6 earlier 1.9 0.3 Employed for a Employed for a Age 24.8 38.9 19.8 wage/salary 14.6 wage/salary 10.7 27.2 5.8 56.1 Education 0 20 40 60 80 100 0 20 40 60 80 100 Percent Percent Source: Based on data of the Housing and Population Census, Statistics Mauritius. youth. At about 13 percent in 2016, youth unemployment (ages 16–34) is about twice as high as the average unemployment rate, and it reaches a peak of 20 percent among young women. Finally, educational attainment is extremely important to access wage employment because it is correlated with skills and employability, and it is an important factor affecting access to good-quality jobs. However, these have not been sufficient to mark a change in the upward trend in earnings inequality as some workers, particularly women, still suffer substantial gaps in earnings once they have access to the labor market. Youth unemployment is a complex phenomenon that requires a two-pronged strategy. As explained in World Bank (2017), in Mauritius youth unemployment is the by-product of two main factors. It is ascribable to a reluctance among low skilled youth to take up jobs in certain sectors, particularly agriculture and export-oriented enterprises, because of the working conditions as well as the social status associated with being employed in these sectors. Based on a series of surveys conducted by the Mauritian Human Resources Development Council, employers report lack of adequate attitudes toward work, unfavorable conditions with respect to other sectors, unavailability to work in shifts or overtime or use flexible time arrangements, insufficient job security, or low wages as common attitudes among their younger employees. As a result, an increasing share of work permits is issued every year to attract foreign labor, which is then overwhelmingly employed in low-skilled occupations. Employers also report difficulties in finding people with proficiency in technical skills, suggesting that the Mauritian education system does not meet labor market demands. OV E RV I E W 13 Evaluating and forecasting areas of growth and improving communica- tion channels between education and workplace actors is key to promoting human resource development in line with the country’s economic trans- formation. The capacity of evaluating and forecasting areas of growth is key to better prepare the work force of the future to changing labor demand needs. For example, the development of the ocean economy can be an expanding source of jobs and income, and, in the context of an aging population, health care and social care are areas that will continue growing in the years ahead. The provision of health care and care services is labor intensive and has therefore the potential to create jobs on a large scale that are less likely to be completely automatized as the technological revolution unfolds. Improving the communication between the education and training system and the labor market is also essential to develop human resources that possess the skills required in the economy in the future. This can help provide employers with the skills they look for and youth with what they need to take informed schooling decisions. For example, dual training systems have proved successful in facilitating the transition from school to work and in reducing youth unemploy- ment in other countries. Such systems tend to be most successful when they involve an active cooperation between public institutions and the private sector. Going forward a consolidation of employment programs and better integration with education policies will be required. Current youth employment programs are fragmented and in the absence of a central registry it is difficult to track if participants benefit from more than one scheme. This is most likely as some interventions overlap: there are, for example, several entrepreneurship schemes with similar target groups. In addition, the interventions are often disconnected from the demand. Many programs offer on-the-job or short class room training, yet they lack integration with job placement services. This calls for greater emphasis on comprehensive programming that combines on-the-job training, short classroom training, life-skills training, and labor market intermediation services and career guidance for participants. 14 MAURITIUS References Central Statistics Office. 2005. Continuous MultiPurpose Household Survey – 2003. Main Results of the Time Use Study. Ministry of Finance and Economic Development, Port Louis. Krueger, Alan B. 2012. “The Rise and Consequences of Inequality.” Presentation at the Center for American Progress, Washington, DC, January 12. http://www.americanprogress.org/events/2012/01/12/17181/ the-rise-and-consequences-of-inequality/. Krugman, Paul R. 1992. “The Rich, the Right, and the Facts: Deconstructing the Income Distribution Debate.” American Prospect 11 (Fall): 19–31. World Bank. 2017. Mauritius: Addressing Inequality through More Equitable Labor Markets. Washington, DC: World Bank. OV E RV I E W 15