Join the network
Policy Brief
Tasks, skills, and institutions
Earnings inequality and polarization in eleven countries
Concerns about widening income inequality within countries continue to gain prominence in public debate worldwide. In the last decade, attention to the concentration of income at the very top of the distribution (top 1%) has increased. This concentration largely originates from the accumulation of capital income by a small minority of very affluent people.
In low- and middle-income countries, job and wage polarization is much less common than in high-income countries, where it has emerged as a critical force behind earnings inequality
Occupational upgrading—transitioning from low-skilled to middle-skilled jobs—is more common, primarily due to declining shares of employment in the agricultural sector of the economy
Changes in workers’ returns to education, especially rising earnings premia associated with higher educational attainment, is also an important driver of earnings inequality in low- and middle-income countries
Labour market institutions, such as minimum-wages, can play an important role in taming earnings inequality
However, this focus on changes affecting the wealthiest tends to overshadow the component of earnings inequality that arguably remains most consequential for the other 99%: changes in employment opportunities and pay—or how workers are remunerated for their skills, education, and job tasks.
New research on trends in earnings inequality
A major research initiative studies trends in earnings inequalities in eleven countries located throughout Africa, Asia, and Latin America (Figure 1). The research addresses two core questions: (i) what are the main drivers of changes in earnings inequality across Global South countries? and (ii) is it the changing nature of work which makes labour markets more polarized and unequal in these countries, or other factors? It extends the most conventional explanations of changes in earnings inequality from labour economics—which are based on the relative scarcity or abundance of skilled versus unskilled labour—with recent theories that place the tasks performed by workers in their jobs, rather than their skill levels, at the centre of analysis.
Figure 1: Earnings inequality trends in selected Global South countries (Gini index)
Source: Authors’ presentation based on country studies.
The research yields three stylized facts about the differences in the nature of work worldwide. First, work in certain occupations is more routine-intensive in some countries than in others. Second, in low-income countries (LICs) and middle-income countries (MICs), the gross reallocation of labour away from routine work and toward non-routine work is occurring more slowly than in high-income countries (HICs). Since the early 2000s, the gaps between these country groups have grown. Third, HICs remain the primary source of non-routine employment, while LICs and MICs remain the primary source of routine employment (Figure 2).
Figure 2: The distribution of routine and non-routine occupations across country groups, expressed as shares in global employment in 2000 and 2017 (in %)
Source: authors’ estimations based on PIAAC, STEP, CULS, O*NET, World Bank, UIBE-GVC, and ILOSTAT data.
Findings are not consistent with a job and wage polarization hypothesis
New findings show that job and wage polarization in LMICs is much less common than in HICs, where it has emerged as a critical force behind earnings inequality. The labour shares and wage premia associated with the best-paid occupations have increased in many of the countries studied, which is a component of the polarization experienced in HICs. However, similar declines in mid-level occupations are rarely observed, and only in exceptional cases, such as South Africa.
Workers transitioning from low-skilled jobs to middle-skilled jobs (occupational upgrading) is more common in LMICs. In some countries, this pattern follows a typical path of structural transformation, with workers moving to manufacturing. In others, it involves rising employment in construction and services. In either case, occupational upgrading mitigates the risk of wage polarization.
Nevertheless, the pace of de-routinization of work in the Global South is slower than in OECD countries, and the related changes to occupational structure has a less significant impact on earnings inequality. At the same time, changes in workers’ returns to education, especially rising premia associated with higher educational attainment, are a key driver of earnings inequality in the countries studied. In this respect, recent developments in Global South countries resemble the changes HICs experienced in the 1970s and 1980s, when so-called skill-biased technological change widened wage inequality.
What are the policy implications?
The disparities in routine vs. non-routine employment and distinct trajectories across country contexts has significant policy implications. Cross-country variation in the content of workers’ jobs is much greater than cross-country variation in workers’ skills.
Investments in workers’ education and skills continue to be necessary for international convergence in the nature of work and income between countries. However, the findings indicate a need for interventions in other areas because the rate of technological adoption by employers and participation in global value chains are key factors driving cross-country differences in occupational task content. Increased adoption of technology should accompany increased investment in skills and education.
Moreover, education systems should teach basic skills in solving practical problems in an ICT-rich environment. It is likely that these skills will soon be required even in low-skilled jobs. Workers that lack these skills, especially with lower educational attainment, are at risk of being left behind.
Investments in education and skills are necessary for the convergence in the nature of work and income, as well as to avoid the adverse labour market effects of technology adoption in the Global South
However, these investments are unlikely to be sufficient, given that technological adoption and participation in global value chains are key factors driving cross-country differences in occupational task content
Policymakers should incentivize technology adoption by the private sector and encourage investments in workers’ skills and education
Moreover, education systems should enable students to gain basic skills in solving practical problems in an ICT-rich environment
In several countries studied, labour market institutions, such as minimum wage policies, have played an important role in taming earnings inequality. Their importance will likely increase as technology adoption accelerates and returns to skills rise. Minimum wage systems need to balance the conflicting goals of improving low incomes, ensuring compliance, and avoiding disemployment. The labour market information and public employment services should support workers in reallocating to more productive firms, which is a key mechanism to minimize disemployment effects and facilitate productivity gains from rising minimum wages.
