1. Recent achievements
Advances in poverty reduction in Latin America over the last decade and a half have been remarkable. With a $4 a day poverty line, the region’s population living in poverty fell from 45 to 25 percent between 2000 and 2014; and with a stricter poverty line—$2.5 per day—from 28 to 14 percent. Three more facts show the depth of this improvement. First, with some variation, poverty fell in every country in the region; second, poverty did not increase between 2008 and 2010, the critical years of the global financial crisis; and third, despite a substantial deceleration in the region’s growth rate since 2011, poverty has not increased either.
What explains these improvements? The underlying story has two parts. First, there is growth. The initial decade of this century was beneficial for Latin America as a result of unusually favorable conditions in international capital markets and high commodity prices. This contributed to doubling the region’s growth rate to over 5 percent a year, compared to the 1990s, when it averaged between 2.5 and 3 percent.
But growth is not all. As mentioned, as of 2011 growth has decelerated; in fact, since then each year’s growth rate has been lower than the previous year, and in 2015 was negative (with a similar result expected in 2016). The second part of the story is associated with improved public policies. In particular, starting in the late 1990s, governments have transformed all sorts of targeted and generalized consumption subsidies into monetary transfers to the poor conditional upon investments in their human capital; an approach that Latin America pioneered. Multiple evaluations have shown that these new programs, generally known as conditional cash transfers (CCTs), have made a considerable difference in the lives of the poor. Food is more plentiful at the table and families are enjoying more diversified diets, more children and youngsters are attending school (particularly girls), and families are periodically receiving basic health services. Today, governments in the region directly transfer between 0.5 and 1 percent of GDP to the first two to three poorest deciles of the income distribution. And even though there is considerable heterogeneity across countries in terms of coverage, conditions and generosity of transfers, they are occurring in almost all countries in the region.
2. Limitations of poverty programs
Despite these advancements, three shortcomings still need to be addressed. First, in some countries there are significant portions of the poor not being reached by these programs (while non-poor groups benefit from them). In principle exclusion errors could be resolved via expanded coverage, but in the tighter fiscal context faced by the region, this is more difficult. Looking forward, exclusion errors will need to be corrected through the politically more costly route of reducing the number of non-poor households that are currently being covered.
Second, these programs have changed the allocation of poor household time, inducing more time to consume health and educational services. However, deeper indicators of human capital accumulation, like learning, show fewer improvements. In hindsight, it is clear that CCTs stimulated the demand for these services, but that governments failed in parallel to expand supply with appropriate quality, particularly in rural areas. Put differently, distributing money was relatively easy for governments, but it has proven to be significantly more complicated to set strong incentives for quality to (mostly public) providers of health and educational services in poor localities, and monitor their performance. As a result, although poor households are consuming more health and educational services than before, they are accumulating less human capital than they could.
Finally, the original design of interventions underestimated the complexities of early child development, convinced back then that better nutrition and health were sufficient to improve children’s well-being before they started school. Today we have a better understanding that issues like language acquisition and socio-emotional stimulation in the first three or four years of life are equally critical. As a result, more poor children are attending school than before, but as a result of lags in their pre-school development they are approximately two years behind in language use and understanding compared to non-poor children of the same age. These and related results indicate that it is urgent to complement traditional CCTs with enhanced interventions in early child development in order to fully achieve the objective of increasing the human capital of the poor. In this context, it is important to note that while there is evidence of specific interventions that have worked on a small-scale (a few hundred or, at best, thousands of children), we do not know enough about cost-effective interventions that can be scaled up similar to that of CCTs (covering millions of children).
In short, the public policy agenda for poverty programs needs to focus on improving targeting and extending coverage to the truly needy that are still left out; on reforming incentives to providers of health and educational services to noticeably increase quality; and on designing cost-effective and scalable interventions that can address dimensions of early child development beyond nutrition and basic health. Thus, while the balance of poverty policies over the last decade and a half is clearly positive, there is ample room for improvement.
3. The productivity imperative
If poverty policy is to help poor households to gradually escape poverty with their own efforts, the poor need to earn higher incomes. At the end of the day, the best poverty policy is one that is eventually not needed. This is the meaning of breaking the intergenerational transmission of poverty. If this is not achieved, poverty policy would, de facto, validate a situation where poor households live permanently from public welfare. Thus, CCTs and related transfer mechanisms may need to be complemented with further measures. To identify those measures, it is first useful to address a critical question: Will the accumulation of more human capital among poor households automatically translate into higher earnings, net of the transfers received from CCTs and similar programs?
Unfortunately, as yet few studies have identified whether poor household incomes are increasing as a consequence of the human capital acquired through CCTs. In fact, the preliminary evidence emerging from some countries points in the opposite direction: Younger cohorts of poor workers have more years of schooling and better health than their elder peers, but are not earning higher incomes. However, these results are based on preliminary data and a small sample of countries and, therefore, need to be interpreted with care.
Nonresident Senior Fellow - Global Economy and Development, Brookings Economic and Social Policy in Latin America Initiative
Almost by definition, the poor have few productive assets beyond their own labor. Thus, their earnings can be increased only by raising their labor productivity. More human capital is necessary but not sufficient for this to happen. Aside from being healthier and more educated, poor workers also have to find better jobs. Thus, critically, the challenge of breaking the intergenerational transmission of poverty is inevitably associated with the performance of the region’s labor markets and, in particular, with the productivity of the jobs that these markets generate.
Unfortunately, in many Latin American countries labor markets are highly dysfunctional, reflected by the fact that over half of the region’s labor force is informally employed. In parallel, since at least the early 1990s the region’s productivity performance has been quite disappointing. Indeed, this is one of the characteristics that most distinguishes Latin America from other emerging regions in the world. Dysfunctional labor markets and stagnant productivity growth are matters of great concern for economic growth generally and, for the reasons stated above, for the poor in particular.
Moreover, from the perspective of poverty alleviation, increasing productivity growth is more important today than 15 years ago, when there was more room to increase the coverage of CCTs and associated transfers. Ignoring the subset of poor households that today are not covered by these programs, today we are at the point where it is probably counter-productive to attempt further reductions in poverty through yet more transfers, as opposed to more productive employment. Even if the fiscal space to do so was there, it is important to consider the impact of ever increasing transfers on labor-leisure choices. It is also important to consider the social implications of having increasingly educated poor youth that end up with the same low earnings that their parents have. And it is important to consider as well the long term implications for growth of having between a quarter and a third of a country’s labor force permanently engaged in low productivity activities. For all these reasons, increasing productivity needs to be at the center of poverty policy.
4. The road ahead
Latin America faces a complex international environment: Monetary normalization in the United States and slower growth in China and elsewhere imply that the factors that were tailwinds before the global financial crisis are now acting as headwinds. In addition, with perhaps one or two exceptions, Latin American economies have exhausted the space for counter-cyclical fiscal spending; in fact, many are now engaged in fiscal consolidation. Therefore, faster growth will not come from the rest of the world, or from stimulating public spending. At the same time, the region’s demographic transition implies that a previous growth factor—a labor force that increases more rapidly than the population—is declining in some countries and will soon decline in others. Barring an unexpected and lasting positive external shock, growth in Latin America will have to depend more on faster capital accumulation and faster productivity growth for the foreseeable future. This is quite a challenge, as these two factors have long been the region’s Achilles heel.
On the other hand, despite the gains over the last 15 years, poverty is still high and may still increase if low or even negative growth rates persist for some years. These gains need to be preserved, but further advances are needed. In a tighter fiscal context and more complex growth scenario, this will only occur if there is a clear understanding of what objectives poverty policy should pursue.
As Section 2 noted, there is substantial room for improvements to the operation of programs directly focused on the poor: better targeting, higher quality services and attention to early child development. These tasks may not be glamorous, but the welfare of the poor could increase noticeably if governments sharply focused their efforts on them. But aside from these improvements, the majority of the efforts required to reduce poverty are not associated with poverty programs. They are instead directly connected with policies that can increase productivity growth and, most critically, improve the functioning of the region’s labor markets.
Many factors, with variations from country to country, simultaneously account for stagnant productivity and dysfunctional labor markets in the region. Trying to identify “the” factor is like trying to identify which of the five bullets in the heart killed the victim. In some countries it will be necessary to revise the balance between taxation of labor versus income or consumption, most likely lowering the former. In other countries, a sober assessment is needed of the effects that regulations on severance pay and minimum wages, or the combination of “contributory” and “non-contributory” health and pension programs, have on the formal-informal composition of firms and employment. Further, many countries will have to revisit tax regulations that affect the size of firms (like simplified regimes or compliance costs) as well as credit regulations (like subsidies to small firms or micro-credits for undertakings that have no potential for higher productivity). And yet some countries may need to consider whether credits from public development banks are inadvertently allowing unproductive firms to survive. But in all cases, the same question must be asked: Is this policy contributing to or hindering productivity growth?
Asking this question and, more importantly, effecting the required policy changes, does not mean that countries must give up on the redistribution efforts that are at times the underlying motivation for these policies. But it does mean that redistribution efforts must be carried out through other instruments that are not so costly in terms of productivity.
The policy changes needed to accelerate productivity growth are technically complex, but the real challenge is political and, in a region that has a long tradition of attempting redistribution through labor market interventions, ideological. This makes change very difficult indeed. However, from the point of view of the poor, a glimmer of hope could come from the realization that, in addition to further improvements to targeted programs, the policies needed to allow them to escape poverty with their own efforts are by and large the same ones that are needed to accelerate growth.
Homi Kharas delivered the keynote address at IFPRI’s annual staff retreat on September 12, 2018. He explored the evolving development agenda and its implications for policy research.