Share This
Levers for Growth

Search Finance & Development

Advanced Search
About F&D


Back Issues

Write Us

Copyright Information

Free Email Notification

Receive emails when we post new items of interest to you.

Subscribe or Modify your profile

F and D logo
A quarterly magazine of the IMF
March 2006, Volume 43, Number 1

Levers for Growth
Simon Johnson, Jonathan D. Ostry, and Arvind Subramanian

Policy lessons from earlier bouts of growth in developing countries

Economists now generally agree that broad economic institutions (such as central banks and the judicial system)—which define the rules of the game in society for economic transactions—are critical building blocks for sustained long-run growth. Strong institutions imply effective property rights and mechanisms for enforcing contracts, thereby promoting investment and efficiency. But what happens when economic institutions have significant weaknesses, as in the vast majority of low-income countries? Changing these institutions is a slow and difficult process. Can other policy levers help kick-start growth when strong institutions are absent?

Here, we look at successful episodes of development that occurred in countries where broad institutions were weak—to draw lessons about which policy levers, if any, have been effective in fostering sustained growth despite adverse initial conditions. Of course, if there were no exceptions to what might be called a rigid institutional view of development, there would be little opportunity to draw relevant lessons, because there would be few episodes of sustained growth in countries with initially weak institutions. Fortunately, however, this is not the case. In particular, by examining growth experiences over the past four decades, we can find a number of countries that were able to sustain rapid growth even though their initial institutions were apparently weak.

For our study, we looked at 47 developing countries that have experienced rapid growth at some point since World War II. Of these countries, we focused on 43 that started a significant growth episode with initially weak institutions. The data suggest that countries with initially weak institutions that were able to ignite and sustain growth were also successful in upgrading the quality of their broad institutions during their growth episode. For this group, there is a kind of virtuous circle of policy levers (for example, fiscal, exchange rate, and trade policies, as well as policies relating to education and the costs of doing business) that ignite growth despite the weak institutions. Growth is then sustained and institutions are improved, possibly laying the foundation for an improvement in long-term growth prospects.

This article draws some lessons about which policy levers appear to have been particularly effective in promoting economic growth—that is, those levers present in the majority of successes but not in the countries that did not sustain growth—and draws some tentative implications for future policy design. We use a methodology that is halfway between regressions and case studies, and, hence, the findings here should be regarded as suggestive but not conclusive with regard to causation; some related work on the duration of growth spells is described in Berg and others (2006).

Sustained growth episodes

Our study built on work on growth accelerations by Ricardo Hausmann, Lant Pritchett, and Dani Rodrik (2004), who found that growth accelerations were unpredictable and that reform led to sustained growth in surprisingly few instances. In our study, we focused on two groups of countries, comparing them with each other and with our sample of all developing countries. Both experienced growth accelerations of at least 2 percentage points per capita, sustained growth of at least 3!/2 percent per capita for seven years, and achieved a higher postacceleration income level than the preacceleration peak. The focus here is on how countries managed to sustain growth despite initially weak institutions; countries that already had strong institutions in 1960 are thus excluded (the data are not perfect, but countries such as Botswana, India, Mauritius, and Sri Lanka are excluded on this basis).

Our first group—sustained growth countries (SGCs)—comprises countries whose growth per capita remains above 3 percent after seven years. In the second group, classified as unsustained growers (USGs), growth falls below 3 percent after seven years. We are interested in what distinguishes SGCs from USGs, with the common element being the generally weak institutions at the beginning of the growth spurt.

Chart 1 plots growth of the countries in the sample against the initial level of political institutions, measured by the constraint on the executive (known as the Polity IV rating). While this measure (which ranges from 1, the weakest level, to 7, the strongest) captures the quality of broad political rather than economic institutions, the available evidence suggests that both broad political and economic institutions were weak for most of these countries at the beginning of their respective growth spurts. Although comparable measures of the quality of broad economic institutions at the beginning of the SGCs' growth spurts are unavailable, the evidence suggests that, in Korea (1960s), Singapore (1960s), and China (1980s), for example, the investment climate improved dramatically during the growth period (World Bank, 1993). Available data are not good enough to support stronger conclusions, but it appears that in the majority of SGCs, the improvement in economic institutions occurred early on, say, within five years of the growth acceleration.

Chart 1. No holding them back

The chart shows that most sustained growth cases, denoted by their three-letter country codes, had weak initial political institutions (average of 2.1), with scores that are even lower than the USG cases (average 2.8), denoted by dots. While not captured in the chart, SGCs also appear to have started their growth accelerations from quite different initial income levels, although a number of them have nonetheless managed to sustain rapid growth for upwards of 35–40 years so far. Sustained growth with weak initial institutions is thus not only possible but appears to have the potential to be long lasting.

The SGCs generally attained great success in manufacturing exports. The average increase in the ratio of manufacturing exports to GDP for these countries over their growth episodes was 36 percentage points. Some countries—for example, Singapore and, to a lesser extent, Korea and Malaysia—achieved a rapid increase in manufacturing exports at a very early stage (specifically, in the first five years of their spurt). Clearly, there was two-way causality between export ratios and growth, but even to the extent that early rapid export growth was a proximate cause, it raises the question as to the underlying policy choices that facilitated this growth.

One obvious candidate is the exchange rate. Chart 2 looks at the competitiveness of exchange rates and suggests that the difference between the SGCs and the USGs is striking. Although both SGCs and USGs had modest (less than 10 percent) average misalignment (which is measured as the difference between the actual exchange rate and one that is consistent with a country's level of income) over the entire sample, these averages mask periods of very large overvaluation for the USGs, but generally not for the SGCs. This can be seen in Chart 2, which plots the maximum overvaluation episode for each country in the sample. It can be clearly seen that the incidence of severe overvaluation was much higher in the USGs (shown by dots) than in the SGCs (shown by country codes) and could well have been a critical factor for the sustainability of growth episodes centered on an expansion of manufactured exports.

Chart 2. Getting the currency right

To some extent, avoidance of overvaluation, which could have played a role in the growth of manufactured exports, may have been related to countries' success in avoiding macroeconomic instability. SGCs have, in fact, tended to have smaller governments and run more prudent fiscal policies; partly as a result, they have experienced less inflation over the sample than the USGs and have generally suffered from less macroeconomic instability (proxied by, for example, parallel market exchange rate premiums).

Another factor that may have helped export growth is trade liberalization. But liberalization is, generally speaking, much more a feature of the SGCs' experience than of the USGs'. While it is true that some SGCs did not liberalize their imports—so that trade liberalization is not a necessary condition for export growth—the weight of the evidence points in a different direction, as the examples of Indonesia, Korea, Malaysia, Singapore, and Taiwan Province of China show. Even countries with relatively illiberal import regimes, such as China and Vietnam, took important steps to create other incentives to export manufactures. In terms of the aggregate data, the average percentage of years that a country has been open (since the growth acceleration began) is 58 percent for the SGCs and less than half that for the USGs.

Another question is whether access to economic opportunities—proxied, for example, by measures of educational attainment—was a differentiating factor in the success of the SGCs. Interestingly, the SGCs have tended to have higher levels of educational attainment (both primary and secondary) and lower levels of inequality than USGs, suggesting that initially weak political institutions—a feature of all countries in the sample—may have coexisted with more equality of opportunity.

Although the SGCs began their growth episodes with weak political institutions, they seem to have benefited over time from a mutually reinforcing process through which their institutions improved alongside policy actions in the dimensions discussed above. Investment risk, which captures the risk of expropriation, ease of repatriating profits, and payment delays, and therefore proxies the quality of broad economic institutions, showed marked improvement in the SGCs between the mid-1980s—the earliest period for which data are available—and the 2000s. In turn, the improvement in economic institutions has generally led to a strengthening of political institutions in the SGCs (measured by the constraint on the executive rating), in contrast to a deterioration in the USGs. This highlights the notion that institutions are not immutable but can respond to economic and policy changes and, thus, that the quality of broad institutions is not a permanent barrier to long-term growth. A significant number of countries with sustained growth have found the policy space to improve their institutions over time, thus laying the foundation for sound growth in the medium run.

Lessons for Africa

What are the possible implications for other developing countries, particularly in Africa? The table focuses on a set of promising performers in sub-Saharan Africa and compares them with the SGCs (and with some other comparators in Africa and the developing world). This is a useful comparison to the extent that Africa today faces problems similar to those previously faced by the SGCs. In terms of economic institutions, as measured by investment risk in the first column, the average for the group of promising performers (score of 7.9) is already above that of the SGCs in the mid-1980s (score of 6.4). In terms of political institutions, as reported in the second column, many of these African countries already have a higher score than did the SGCs when their spurts began. Institutions are perhaps not strong today, but they are potentially good enough in parts of Africa, especially when compared with the SGC cases at the time they embarked on their growth acceleration. However, several countries in Africa have recently emerged from protracted civil conflicts, which makes their political situations more challenging today and, hence, less like those of the SGCs when they embarked on their growth episodes. Other African countries face health crises on a scale not previously seen in the SGCs and for which the experience of the SGCs may therefore not be relevant.

Turning to export performance, however, export-GDP ratios remain quite low in the promising African performers, and this is even more apparent for manufactured exports. Part of the reason could lie with trade restrictiveness, though this does not appear to be the case from the standard data (reported here), at least when compared with the SGC cases at the start of their growth accelerations. Another possible explanation would appear to be currency overvaluation, where the promising performers indeed do not compare favorably with the SGCs at the onset of their growth spurts. Overvaluation, moreover, does not appear to reflect excessive inflation, which seems well contained across the countries in the table, and is, in fact, quite low by historical standards.

Looking at other possible levers, educational indicators vary considerably, though enrollment rates are lower among the promising performers than in the SGC cases. Aid ratios are also relatively high compared with those of the SGCs, perhaps affording the promising performers more opportunities to expand spending on education and other productive social investments. One area that is perhaps worthy of further attention is businesses' costs of entry, where the promising African performers compare less favorably with other developing countries.

Building on success

Of the 43 countries we focused on, 12 experienced sustained growth—episodes of 15 years or longer—and the preponderance of these also saw the quality of their broad economic institutions improve during growth episodes. A key characteristic common to these episodes is rapid export growth (particularly of manufactures). The policy levers that helped achieve high export and overall growth were trade liberalization and the avoidance of exchange rate overvaluation and macroeconomic instability; in the sustained growers, there is also evidence of a relatively high average level of educational attainment at the start and less inequality compared with other developing countries. Countries with initially weak institutions that could not sustain growth encountered significant exchange rate misalignment or macroeconomic instability; these problems curtailed their expansions.

These findings are consistent with the view that economic institutions remain central for sustaining long-run growth. At the same time, however, they suggest the potential for a virtuous circle through which growth and the policy levers used to achieve growth lead to positive institutional change. Although one cannot be definitive about the underlying mechanisms, a possible explanation is that growth in manufactured exports benefits a cross-section of the population (in a way that natural resource–based growth does not), creating a constituency for improving institutions more broadly.

In terms of today's low-income countries, an emphasis on macroeconomic stability, trade liberalization, and avoidance of exchange rate overvaluation can thus potentially sustain a self-reinforcing spiral in which growth in manufactured exports creates constituencies for further reform and growth. Beyond the empirical analysis presented above, there appears to be anecdotal evidence as well as strong economic logic to suggest that efforts to improve fiscal transparency and reduce costs of doing business could be helpful not only in terms of their favorable effects on nominal stability and growth, but also through positive effects on broad institutions—the assumption being that these changes strengthen domestic constituencies that have an interest in working to improve institutions.


Acemoglu, Daron, and Simon Johnson, 2005, "Unbundling Institutions," Journal of Political Economy, Vol. 113 (October), pp. 949–95.

Berg, Andrew, Carlos Leite, Jonathan D. Ostry, and Jeromin Zettelmeyer, 2006, "What Makes Growth Sustained?" manuscript (January) (International Monetary Fund).

Hausmann, Ricardo, Lant Pritchett, and Dani Rodrik, 2004, "Growth Accelerations," NBER Working Paper 10566 (Cambridge, Massachusetts: National Bureau of Economic Research).

International Monetary Fund, 2003, World Economic Outlook, September (Washington).

Rodrik, Dani, Arvind Subramanian, and Francesco Trebbi, 2004, "Institutions Rule: The Primacy of Institutions over Geography and Integration in Economic Development," Journal of Economic Growth, Vol. 9 (June), pp. 131–65.

Sachs, Jeffrey, and Andrew Warner, 1995, "Economic Reform and the Process of Global Integration," Brooking Papers on Economic Activity, Vol. 1, pp. 1–118.

World Bank, 1993, The East Asian Miracle: Economic Growth and Public Policy (Washington).

Simon Johnson is a Professor at the Sloan School of Management at the Massachusetts Institute of Technology. Jonathan D. Ostry is a Senior Advisor and Arvind Subramanian is a Division Chief in the IMF's Research Department.

Showing promise