Photo: Aniruddha Chowdhury/Mint
Photo: Aniruddha Chowdhury/Mint

Convergence: A marathon, not a sprint

Emerging economies are losing steam in their bid to catch up with the developed ones in terms of per capita income

Have emerging economies stopped catching up with advanced economies, measured in terms of gross domestic product (GDP) per capita?

The World Bank’s most recent Global Economic Prospects report, published in June, suggests that this worrying scenario may now be a reality. Interested readers, including those lacking the appetite to digest a 200-page official document, may also consult a succinct report in the Financial Times.

The World Bank’s research suggests that this year, for the first time since the start of the new millennium, 47%—or less than half—of the 114 emerging and developing economies studied will be poised to converge in terms of GDP per capita with the US and other advanced economies. That ratio was at a high point of 83% on track for convergence at the onset of the global financial crisis in 2007.

How times have changed. As recently as in a 2013 op-ed, economist Arvind Subramanian, then at the Peterson Institute, a Washington DC think tank, and India’s current chief economic adviser, could write: “What we are seeing today, despite the crises, is convergence with a vengeance. An unequal world is becoming less so."

In the same 2013 op-ed, Subramanian further opined that convergence implied a “structural" rather than “cyclical" decoupling between emerging and advanced economies. As he wrote, “in the medium to long term, the rise in living standards (of emerging economies) relative to that of the rich world depends mostly on what developing countries themselves do and less on the external environment". This optimism appears to have been belied by the latest news, which suggests that commodity prices, driven as much by the global business cycle as anything emerging economies have control over in terms of policy levers, has a great deal to do with convergence, at least in the short and perhaps even in the medium run.

Meanwhile, another way to slice the same data is to observe that the average time it is expected to take emerging economies to converge to advanced economies has increased. Thus, in the period immediately preceding the crisis, the average convergence time was a little more than 40 years, or a couple of working generations. That has now shot up to almost 70 years, or beyond the lifespan of the current working generation.

It should also be noted that slower convergence in income per person between emerging and advanced economies also has knock-on effects on global inequality and poverty. This is because, as has been well documented throughout the emerging world, including India, a sustained and high rate of growth is associated with reductions in poverty and additions to the global middle class contributing to a reduction of global (although not necessarily within-country) income inequality. On the latter point, see, for instance, this recent research paper by economist Branko Milanovic, a noted expert on trends in global inequality.

The fact that fewer emerging economies are now likely to converge to the advanced economies in the foreseeable future reflects a relative drop-off in the growth rates of the former as compared to the latter group. As the World Bank report notes, it is driven by the fall in growth rates among commodity-exporting countries. Thus, while the world as a whole is expected to grow at 2.4% this coming year, commodity exporters are expected to grow at only 0.4%, which is an even lower growth rate than that of advanced economies such as the US.

India, of course, is an exception to this gloomy picture, as it is expected to grow at around 7.5% or so, or very much in the group of countries that are on track to converge. However, it must be noted that India is starting from a relatively low base of income per person compared to the US, so our catch-up is measured in decades, not years, and a lot can happen between now and a putative convergence date in the distant future.

Having said which, the gloom might also be lifted to some extent if we look at the data in a different way and over a longer time horizon.

Thus, yet another way to convey the same information on convergence is to track over time the percentage of US GDP accounted for by the emerging economies: the higher this percentage, the closer the date of convergence. As reported recently in the Financial Times, drawing on research by Capital Economics, a consultancy, the data reveal a generally upward trend in this percentage, starting from as low as around 10% in 1960 and climbing to just below 30% this year. So, while it is true that this percentage might be ticking down in the short term, which is a cause for concern, the overall trend is clearly rising. The research also reconfirms the fact that convergence is a reality in the data, although it might not be occurring as rapidly as we would wish.

To aid our understanding, it is worth stepping back from the current debate and casting a sidelong glance at the economics of growth literature, which is where the convergence prediction comes from in the first place. In a nutshell, so-called neoclassical models of economic growth developed by economists such as Nobel Prize winner Robert Solow embed the assumption of diminishing returns to capital (or, more broadly, to all reproducible inputs). In other words, capital is highly productive, at the margin, when its stock is low, and this marginal productivity diminishes as the capital stock grows. Since economies that have low levels of income also typically have small capital stocks, compared to richer countries, the implication is that poorer economies will grow more rapidly than richer economies.

The early empirical research found a fair degree of support for the convergence theory, especially using regional data sets. See, for instance, this 1990 research paper by economists Robert Barro and Xavier Sala-i-Martin. As they show, evidence for convergence among US states is quite compelling. In the case of India, for instance, economists Paul Cashin and Ratna Sahay, in a 1996 research paper, find some support for the notion that there is convergence among Indian states. It should be noted, however, that the convergence claim for Indian states has not held up so well in more recent research.

See, for instance, this 2012 research paper by economists Arvind Subramanian and Utsav Kumar, who document divergence among states during 2001-09. This pattern of inconsistent convergence is confirmed in a 2013 research paper by economists Chetan Ghate and Stephen Wright. These more recent studies do not disprove the possibility of convergence, but show that it operates more imperfectly and fitfully than had been thought in the 1990s.

In sum, empirical research in economics is at least broadly, if not uniformly, supportive of the convergence theory. However, it is important to note that it ought not to induce complacency among emerging economy policymakers and be seen, incorrectly, as some sort of iron law. This is where the academic research must meet the policy research we have looked at earlier and both must meet common sense.

The correct lesson is that good economic policy can accelerate the convergence process while poor economic policy can retard it. Indeed, the research by Capital Economics cited earlier suggests that India has been three or four percentage points below its potential level of growth in 2000-14, given its starting position. This could reflect a range of factors, including bad luck, but it is hard to rule out the possibility that an insufficient attention to pursuing the economic reforms agenda, especially over the latter part of this period, was a contributing factor.

In other words, convergence is not inscribed in stone by the laws of economics. It can, and must, be abetted by wise policy.

Economics Express runs weekly, and features interesting reads from the world of economics and finance.

Comments are welcome at feedback@livemint.com

Close