Jobs, labour markets & shared growth Trends and issues

A DFID practice paper Briefing June 08 Jobs, labour markets & shared growth Trends and issues This briefing note from PRD s Growth Team is the first of a pair for DFID staff and partner governments on the links between growth and employment. It explores the evidence behind increasing concern about the nature of jobs and trends in job creation in the developing world. It concludes that growth is indeed generating employment and raising real wages for many of the poorest workers. But it also finds that there are some emerging trends rising youth unemployment, growing informality of employment, a widening skills premium and increased labour market volatility which require a policy response. The second briefing note draws out the policy implications. Introduction Economic growth is the single most powerful way of pulling people out of poverty 1. It creates the resources to fund social goods such as health and education. But more importantly, and more directly, it creates a demand for labour, the main often the sole - asset of the poor. Strong growth in the global economy over the past ten years means that the majority (60%) of the world s working-age population is now in employment. Positive growth and employment trends are set to continue, with the good performance of developing countries 2 driving the global outcome. However, of current developing country workers, over one-third in South Asia and over one-half in Sub-Saharan Africa earn less than $1 per day 3. For these people, a job has not brought with it a release from poverty. Furthermore, these low-pay, lowskilled workers have seen a worsening of their wages relative to skilled workers part of a general trend towards widening global inequality observed since 1980 4. Such trends have prompted many to conclude that economic growth alone is not enough to improve employment prospects for all and that the sometimes adverse impact of globalization makes clear the need for specific strategies and policies to create employment and to promote decent work 5 The following note explores the evidence for this assertion and for indications of what the appropriate policies are likely to be. 1 Making Governance Work for the Poor DFID White Paper, 2006. 2 Global Economic Prospects, 2007. 3 ILO Global Employment Trends, 2006 4 Global Economic Prospects, 2007. Chapter 3. 5 UN Commission for Social Development 45 th session, February, 2007. 1

1. The Story so far: there are some important positive trends for jobs 1.1 Globalisation and international trade are generating growth - and jobs. Developing countries growth picked up in the early 1990s as they began to open their economies and pursue export-led strategies. While their recovery from the recession of the 1980s was slower than the rest of the world, their average 6 growth rate is now higher, and projected to remain so. 1.2 Growth is generating jobs. Over the same period, global employment rose by over 400 million. While it was predictably China and India which accounted for most, almost all new jobs created were in developing countries. 1.3 The wages of the unskilled are rising worldwide. For those occupations broadly defined as unskilled, real wages have increased with GDP growth worldwide, indicating that the poorest workers have indeed benefited from the increase in global trade and growth. Initial fears that greater global integration and ever more footloose international investors would reduce labour s bargaining power and push down wages in absolute terms appear not to have materialised. Indeed, looking at working conditions more broadly, evidence on FDI suggests that firms are attracted to countries with higher, not lower, labour standards 7. From: What Africa needs to do to spur growth and create well-paid jobs F. Teal, CSAE, 2006. Compiled from ILO surveys by Freeman & Oostendorp, 2000 6 Though see the qualification to this regarding Africa below. 7 From GEP 2007, box 4.5. 2

These global conclusions are weaker for Africa. We discuss the particular situation of sub-saharan Africa below. 2. But these positive trends are not always strong 2.1 Trend growth rates in many developing countries are still not high enough to absorb their growing labour force. As figure 2 for global employment shows, while the number of jobs has been rising, so too has the working population. The result is that the share of the working population in employment has stagnated. This is being driven by two main demographic trends: (i) children born during the last two decades of high fertility rates now entering the labour force, and (ii) a rise in women s labour force participation rates. Death rates in developing countries declined very rapidly in the 1950s and 1960s, while fertility rates remained high. This generated the so-called population explosion of the 1960s, the children of which are now entering adulthood. Again, there are distinct regional trends, depending on the combined time and speed of each region s fertility and mortality rate changes. For example, China, Brazil and India (although absolute numbers of new young people are still large) are already past their youth bulges 8. Sub-Saharan Africa s youth population, on the other hand, - already more than four times its 1950s level - is projected to continue growing rapidly for the foreseeable future 9. As a result, developing countries need to increase employment by nearly 50 million jobs per year to keep up with working-age population growth, even without any further rises in participation rates 10. The largest needs are in the largest countries, with China and India requiring 8 10 million jobs each year. Such a target may be achieveable in these rapidly growing economies but for Sub-Saharan Africa, which must create close to 10 million additional jobs each year, the task will be much more difficult. From Good Jobs, Bad Jobs, No Jobs, WB C 2007 8 Ie, when the proportion of youth relative to the population as a whole is at a peak. 9 Development and the Next Generation, World Development Report, 2007, World Bank. 10 Global Economic Prospects, 2007. 3

2.2 Establishing growth is still a major challenge for Africa. Although economic growth across the continent has moved from negative in the 1980s to positive in the 1990s, the average over the period was less than 0.2% a year. Since the 1960s, Sub-Saharan African economies growth rate has been only one-tenth of the world average and only one-twentieth of China s recent performance. 11 It has also been highly variable and episodic. There is still much to do to embed current positive growth trends, to underpin a sustained growth in demand for labour and to encourage foreign investment in Africa still heavily concentrated in the minerals sector to diversify. Diversification is a key challenge for Africa. 12 Despite economic reforms, more than half of all FDI in Africa remains in the minerals sector. One-third of all African economies are judged to be resource-rich. Their potential to create jobs is limited not only because their fortunes are closely tied to unstable world markets and because the leading growth sector is capital intensive, but because natural resource dependence is linked to a tendency towards conflict and persistent poor governance 13. The slow growth of the larger, more populous African countries (all resource-rich) is also important here: countries such as Nigeria, Ethiopia, and the Democratic Republic of Congo will have to grow more rapidly and on a more sustained basis to improve the job prospects of the typical African - and to generate positive spillover effects for regional neighbours 14. 3. And there are some new labour market trends emerging which require a policy response 3.1 There has been an apparent recent decline in the rate at which growth has produced new jobs. In recent years there has been a weakening relationship between growth in the economy and measured growth in jobs: between 1999 and 2003, for every 1 percentage point of additional GDP growth, total global employment grew by 0.30 percentage points - a drop from 0.38 for 1995-1999 15. This global figure should be treated with particular caution it reflects cyclical variations in economies and obscures large regional differences 16 - but it is clear that there is a particular problem 11 What Africa needs to do to spur growth and create well-paid jobs, F. Teal, CSAE Oxford, 2006. 12 ibid. 13 The Bottom Billion, P. Colllier, 2007. 14 Challenges of African Growth: Opportunities, Constraints and Strategic Directions. Benno Ndulu, World Bank, 2007. 15 Employment Intensity of Growth: trends and macro-determinants, S. Kapsos, ILO, 2005. 16 for example, East Asia, ECA, LAC and OECD countries will require a larger labour force (or higher productivity) to maintain their output growth 4

emerging for the Middle East, South Asia and Sub-Saharan Africa, where current rates of job creation are not high enough to absorb their growing workforce. From Good Jobs, Bad Jobs, No Jobs, World Bank, 2007 Empirical evidence 17 of the determinants of the rate at which employment expands with growth suggests that they are influenced by macro-economic factors such as low inflation, export orientation (though only for women) and low labour taxes (especially for women 18 ). Structural factors, such as the share of services in the economy, are also important. 3.2 Youth unemployment is rising In a growing economy, a growing workforce suitably skilled - keeps the cost of labour low and encourages more labour-intensive growth. When there are constraints on growth, as is the case of low-income countries, an increase in the labour supply manifests itself primarily in low wages and high unemployment, especially for young workers. Youngsters lack of on-the-job experience makes them less productive than the average prime-age worker and thus more liable to be unemployed during periods of recession. In every region of the world, the difficulty youth face in entering the labour market is evident in higher than average unemployment rates: they make up 25% of the working population worldwide but 47% of the unemployed 19. The current cohort of young people (12-24 year-olds) in developing countries is the largest the world has ever seen 1.3 billion projected to peak at 1.5 billion in 2035. Gaining employment appears to be increasingly difficult for young people - the estimated global unemployment rate for youth has risen from 11.7% in 1993 to 14.4 % in 2003. This may be additional evidence of an increasing skills bias in growth, which is discussed further below. Youth employment can be demonstrated to be a key part of the overall employment problem, particularly in Africa. Projections of current (positive) growth rates and employment intensities of growth matched against known labour supply growth show that the latter outpaces job creation just for so long as the youth bulge is predicted to last, with supply and demand finally catching up some time after 2050. 17 Employment Intensity of Growth: trends and macro-determinants, S. Kapsos, ILO, 2005. 18 Although no relationship is found between employment protection and job intensity of growth. This may be due to lack of effective enforcement. 19 Ibid. 5

Africa Population & Jobs Projections 2005-2050 (billions) 1.2 1 0.8 0.6 0.4 0.2 0 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 Working population (15-59) All jobs Sources: UN World Population Prospects, 2004 Revisions (median variant), various ILO & GEP estimates. 3.3 Informal sector employment is growing, and with it problems of low productivity and perceptions of increasing job insecurity. The growth in informal sector employment in recent years is almost universal. In Latin America (where the data are most reliable) it was the primary generator of jobs in the 1990s: it grew at almost double the rate of the formal sector, with sixty percent of new jobs created by micro-enterprises, own-account workers and domestic services. In Africa, if rural and agricultural sectors are included, the figure is closer to 90%. The share of informal employment is closely associated with low levels of development and high levels of market regulation, together creating poor conditions for private sector growth and making it costly for firms to grow beyond some threshold size 20. These constraints keep productivity and therefore wage and employment growth low. The problem of informality is not that it exists; it lies in the apparent paradox of informality rising, rather than falling, with positive growth and despite market-oriented reforms in many developing countries throughout the 1980s and early 1990s. Traditionally 21, the informal employment has been understood to be involuntary, a sector where surplus workers scratch a living while queuing for a limited number of more desirable formal sector jobs. More recently, it has been argued 22 that informality is better understood not as the disadvantaged element of a segmented labour market but as an unregulated micro-entrepreneurial sector. From this newer 20 The Regulation of Labour, J. C. Botero et al, 2004. 21 The traditional (1954) Lewis model is one of an unlimited supply of rural workers migrating into a limited urban labour market. 22 Informality Revisited, W. F. Maloney, World Development, 2004. 6

perspective informality is seen as a voluntary response by firms driven by competition to reduce legislated or union-induced rigidities and high labour costs, particularly by subcontracting production out to unprotected workers. It is likely that both explanations would hold simultaneously in an economy, with a entrepreneurial tier co-existing alongside a poorer, often waged labour, tier; the former the result of a voluntary exit from the formal sector and the limited benefits offered by the State, the latter an involuntary exclusion 23. The relative balance of, and changes in, each will explain why informal employment varies from one country to another, and helping determine how much of a long-term problem it is for growth and poverty. 3.4 Wage inequality is widening, mainly driven by a rising skills premium. Trade and investment liberalization worldwide has coincided with a trend towards increasing returns to education, as seen in a widening of the wage differential between skilled and unskilled workers. Much of the evidence is from high-income countries but there are also signs that earnings differentials between skilled and unskilled workers have widened in a number of developing countries 24. The fact that the changes have occurred while the supply of skilled workers appears to be growing in most countries (both developed and developing) suggests that the demand for skilled labour has been increasing even faster. World Bank projections suggest that the skills premium (the ratio of skilled to unskilled wages) in developing countries will rise from an average 3.5 in 2001 to 4.2 in 2030. In India the premium rises from 4.3 to 4.9 by 2030 while in China the increase is even larger, from 5.4 to 7.7. Trends in Sub-Saharan Africa are similar, with a rise from 5.1 to 6.8 25. Investment Climate Surveys tend to confirm this: more than a fifth of developing country surveys ranging through Algeria, Bangladesh, China, Zambia show firms rating inadequate skills and education of workers as a major or severe obstacle to their operations. 26 These findings run counter to what would be expected on the basis of simple trade theory, which predicts that globalisation will increase the relative returns to abundant unskilled labour in poor countries. Instead, technological change appears to be an equal force with trade, combining with low education and skills levels to make poor (unskilled) people in poor countries relatively poorer, even as they become better off in absolute terms. 23 Informality: Exit and Exclusion World Bank, May 2007. 24 Globalization, Growth, and Poverty, 2002, World Bank 25 New Pressures in Labour Markets, Chapter 4, Global Economic Prospects, World Bank, 2007. 26 Development and the Next Generation, World Development Report, 2007, World Bank. 7

This notion of skill-biased technological change that the adoption of new technologies, which are increasingly prevalent in all sectors, is associated with a shift in labour demand towards higher-skilled workers is widely accepted as a key factor in the increasing returns to education that have been observed in many countries, including some in Africa. The evidence is strong. In the technology literature, microeconomic case studies have identified the critical role of educated workers in the innovation process, and industry-level studies have found new technology to be complementary with the education of the workforce. Human capital studies have also shown that educated farmers and workers are more productive in a rapidly changing environment, and thus earn higher incomes. And studies of endogenous growth, which stress the importance of human capital investments as a driver of economic growth, show that school enrollment rates are important variables in differences in growth across countries 27. However, there are alternative explanations of the function of labour markets and education those emphasising the signalling and screening functions of qualifications - which argue that the structure and level of employment is fixed by exogenous demand factors and that education simply determines which workers get selected for a limited number of jobs 28. Such explanations may be indeed be more appropriate for developing countries,where there is evidence of formal/informal sector segmentation and strong demand side constraints. Recent work on the sources of rising lifetime incomes in Ghana and Tanzania 29 supports this, concluding that the combination of higher unemployment observed among the better educated, increasing returns to higher levels of education and higher wages (by 50-80%) in larger firms and the public sector is consistent 27 Investment Climate, Capabilities and Firm Performance: Evidence from the World Business Environment Survey, G. Batra and A. H. W. Stone, World Bank 2004. 28 Labour Markets in pro-poor Growth: An Analysis of Fourteen Country Papers, Gary S. Fields World Bank 2005. 29 Sources of Rising Lifetime Incomes in Poor Countries Final Report for DFID Project R8308, F. Teal, CSAE, 2007. 8

with a queuing system in which young people wait longer for better jobs to turn up, the more educated they are. Again, these apparent alternative explanations need not be contradictory. Instead, they suggest that trade liberalisation and investment climate improvements to stimulate the demand-side of the economy are necessary - but may not be sufficient. In the light of technological change, the assumption that the ready supply of unemployed unskilled labour in developing countries would mean that there was no need, at least in the medium term, to go beyond a policy of universal primary education may need to be reviewed. 3.5 And greater labour market volatility and vulnerability is in prospect. Another explanation of the increase in wage disparities lies in the growth of new forms of casual employment. While these new types of contract can serve as a stepping stone into the labour market, they are often associated with prolonged employment in low-quality and low-paid jobs 30. A recent IMF report 31 concludes that growing wage inequality in South Korea is mainly explained by the growth of fixedterm and self-employment contracts, now accounting for 37 and 30 per cent of the labour force respectively. Labour turnover is high in many countries (up to 40% in some) fuelling perceptions of insecurity. This is linked not only to the nature of employment contracts; a significant part of labour turnover (30-50%) can be traced to the creation and death of firms, which is often at least in developed countries - related to positive output and productivity growth. Some studies of OECD countries also find that increased international trade is linked to more labour churning 32. This is likely to be increasingly apparent in developing countries, as they continue to grow and to integrate into the world economy. Falling communications costs are providing new opportunities for developing countries to participate in global production chains by providing specific activities and simple tasks in which to rapidly specialise. This is a positive trend, allowing countries the possibility of a short-cut growth strategy based on a niche sub-sector. However, discrete activities are more likely to be more footloose than whole sectors, bringing with better job prospects more volatility 33. The proliferation of short-term employment contracts, allied to labour market volatility, is heightening insecurity among workers throughout the world. Surveys 34 indicate individuals are indeed concerned not only with the level but also with the security of their earnings. Workers consider job security to be a prime definition of a good job. This is important not simply as a matter of perception, nor solely a problem of social cohesion to be managed through a political process. Faster worker turnover (from the firm s perspective) and greater job insecurity (for workers) both act to reduce 30 Wages Around the World: Developments and Challenges, ILO March 2007. 31 Republic of Korea: 2006 Article IV Consultation. Staff Report, IMF, 2006 32 "Trade and Employment: Stylized Facts and Research Findings" World Bank Policy Research Working Paper No. 3676 B. Hoekman & A..L. Winters,2005. 33 New Pressures in Labour Markets, Chapter 4, Global Economic Prospects, World Bank, 2007. 34 Voices of the Poor World Bank, 2005. 9

incentives on each side to invest privately in the education and training required to realise the potential benefits of increased international trade and technological advance. Means by which governments can fix this market failure have been sought before in developed countries and are by no means easy to find; however, a solution in developing countries may be the key to more and better jobs. Briefing note 2 sets out some of the emerging policy implications of these findings. 10