Can aid help countries avoid the middle-income trap?

BannerWeb1122_DiT_EN_with logo DEV

By Homi Kharas, Interim Vice President and Director – Global Economy and Development, Brookings Institution


This blog is part of an ongoing series evaluating various facets
of 
Development in Transition


Middle-income-trapMost aid agencies have tried to articulate a “middle-income” strategy for how to support client countries that are no longer poor. For example, see the Asian Development Bank Strategy 2030 and the World Bank’s approach to middle-income countries. In both cases, there is an emphasis on second-generation challenges, including those related to environmental, social and governance institutions. Failure to meet these challenges can trap countries in middle-income status.

The problem, however, is that there is no solid theoretical or empirical foundation on how to support growth in middle-income countries—the diversity of contexts and experiences is so large that robust policy conclusions are hard to draw, and useful interventions by aid agencies even harder to figure out.

Barro (2012) succinctly summarizes the limits of empirical work: “My view is that one has to accept the idea that pinpointing precisely which X variables matter for growth is impossible.” In a similar vein, Rodrik (2012) titled his paper “Why We Learn Nothing from Regressing Economic Growth on Policies.” Most researchers (see for example Jones (2016) and Kim and Park (2017)) find that middle-income growth is all about total factor productivity growth (TFP). TFP growth, in turn, is what is left over after accounting for the growth of all inputs. Jones breaks down TFP into two components: knowledge/ideas and M that he says stands for either misallocation or a measure of ignorance.

The problems are particularly acute for middle-income countries. For low-income countries, the Solow growth model offers tried-and-true formulae about the need for human and physical capital accumulation. These may not capture the whole picture of economic growth, but are certainly robust enough in terms of having some positive contribution to growth, to give comfort to policymakers.

Equally, for high-income countries, growth clearly revolves around innovation as Romer has shown. The institutions that support innovation, like the rule of law, protection of intellectual property rights, spending on research and development, university-business partnerships and the like, dominate the discourse.

Middle-income countries have to worry about both accumulation and growth strategies as well as manage a transition from one to the other. There are few theories of growth that speak plainly about these dynamics, and some policy prescriptions like “develop a world class university” are not easy to implement. Given the multiplicity of challenges that middle-income countries must seek to address, and the need to shift from stroke of the pen changes in public policy to longer-term institutional development, it is important for aid agencies to not be dogmatic in their approaches.

Consider the question of competition policy as an example of the importance of understanding local context. Standard neoclassical theory would suggest that more competition is better for growth—hence the common prescription for middle-income countries to liberalize markets. Schumpeterian economists however, sometimes argue the reverse. Which is that investment and growth are higher when large firms can command super-normal profits, to provide them with cash to invest or to generate economic rents, if an innovation is successful. Aghion and Howitt (2005) emphasize the role of market structure in defining the impact of greater competition—firms that lag too far behind will simply give up. Mathis and Sand-Zantman (2014) suggest that competition is beneficial when financial markets are deep, but not otherwise.

Another example relates to inequality. The standard prescription is to foster “inclusive” growth, but how to do this is unclear, and the empirics are far from compelling in terms of providing evidence for any particular policy. Aghion and others argue that inequality is often related to innovation-led growth and social mobility at the top (“self-made millionaires”) and so should be encouraged, not treated as a problem. The World Bank also highlights the lack of social mobility in many developing countries, attributing this in part to low levels of public spending on human capital and to fragmented labor markets. Geographic inequality has risen within many middle-income countries, but there is a vigorous debate about the cost-benefit choices in trying to improve lagging provinces versus accelerating growth in dynamic provinces.

Skeptics of foreign aid have used this complexity to argue against aid. But there is ample evidence that on average aid does accelerate growth, that it has saved millions of lives, that it fosters democracies, and builds food security. The key questions are about how to strengthen aid’s impact not about whether or not it works. Diagnostic tools and political economy thinking can guide country-specific programming in a useful way.

There are two distinct approaches for aid actors in thinking about catalyzing development: (i) think big and focus on a few specifics; and (ii) think small and scale up what works.

Thinking big is the approach followed by most traditional policy-based aid programming. With regard to middle-income country clients, more attention could be paid to local context to understand the likely impact of specific policy recommendations. Furthermore, understanding and boosting entrepreneurship in middle-income countries warrants greater focus that goes well beyond simply lifting the constraints to doing business.

Thinking small and scaling up is the fashion of the day. The small is studied through methods such as randomized controlled trials with the idea that successful interventions can then be scaled up. But scaling is itself a challenge, with limited analytical understanding of how to achieve it. Hartmann and Linn (2008) provide a framework suggesting that scaling up is only successful if the right drivers are in place and the right spaces are open for scaling. But they conclude that few aid agencies take scaling seriously.

Middle-income countries need to transform to avoid the middle-income trap. There is no standard recipe, so aid agencies find themselves struggling to develop a relevant strategy. The largest aid agency, the International Development Association, just completed its 19th replenishment with a record $82 billion over three years directed at the poorest countries; few middle-income countries are eligible. More and more, self-help is offered as an alternative to aid. Few would quibble with the idea that middle-income countries themselves have primary responsibility, but solidarity with a neighbor rarely depends only on income levels. In the European Union, countries like Romania and Poland each receive about 2 percent of national income from EU structural funds. Net Federal spending in Kentucky is 20% of that state’s median household income. Given all the complexities of middle-income country development, there is surely a role for aid to help the transformations needed to avoid a middle-income trap.