The question of why some countries join the developed world while others remain in poverty has vexed economists for decades. What makes it so hard to answer?
In 2019 there were about 648 million people living in extreme poverty, subsisting on the equivalent of $2.15 per day or less. Those 648 million people made up 8.4% of world population — representing an improvement over 1990, when 35.9% of people lived on that little. Yet even though extreme poverty has fallen, in 2018 about 80% of the world population still had material living standards less than one-third of that in the United States.
One of the most frustrating things about the persistence of global poverty is that it is possible to eliminate it — at least within a country — in the space of a generation. In 1953, South Korea emerged from the Korean War desperately poor. It was almost entirely agrarian, and whatever infrastructure the Japanese had built during their occupation between 1910 and 1945 had been destroyed. In 1960 GDP per capita in South Korea was only around $1,200, lower than in Bangladesh, Nigeria or Bolivia, and about 6% of the GDP per capita in the United States.
Shortly thereafter, everything started to change. In 1968 the growth rate of GDP per capita in South Korea topped 10%. Throughout the 1970s, per capita GDP grew nearly 9% each year on average, slowing only slightly through the 1980s and 1990s. By 1995, South Korean GDP per capita had eclipsed Portugal’s. By 2008, it was ahead of New Zealand’s and just behind Spain’s. In 2020, GDP per capita in South Korea was nearly equal to that in the U.K. Not only is South Korea no longer developing; in many areas, it leads among developed nations.
What happened in South Korea offers proof that fundamental transformations of living standards are possible in a few decades. South Korea’s experience, and similar growth trajectories in Taiwan and Singapore, have often been referred to as “economic miracles.” But what if South Korea’s economic growth wasn’t something mysterious or unpredictable, but rather something that we could comprehend and, most importantly, replicate? At current rates of growth, living standards in the poorest countries in the world will eventually catch up to the United States — in about 700 years.
If we could identify what caused South Korea’s takeoff, we might be able to make the miraculous seem routine, and see more countries catch up over decades and not centuries.
Economists have been engaged in research for decades to understand what happened in South Korea and other countries that left extreme poverty behind. It turns out to be one of the trickiest questions in economics. On the surface, it seems like the answer should be obvious: “Do whatever South Korea did.” Or, more broadly, “Do whatever countries that grew rapidly did.” But what, exactly, did South Korea do? And if we know, is it plausible to replicate it?
Scratching the Surface
Some of the first attempts to explain what happened in places like South Korea examined the role of “factors of production,” as economists like to call them. Those factors include physical capital — tangible products like buildings, infrastructure and manufacturing equipment — and human capital — skills and education embodied in workers. In a famous and widely cited study, Greg Mankiw, David Romer and David Weil looked at how the accumulation of both factors was associated with economic growth.
Countries that allocated a large share of GDP toward producing new physical capital or had high levels of secondary school enrollment tended to grow faster than others. In addition, countries with lower population growth rates tended to grow faster, as they were able to equip each worker with more physical capital, raising their productivity.
Mankiw, Romer and Weil studied a broad set of nearly 100 countries from a very high level. Alwyn Young took a similar approach but narrowed his focus to four East Asian economies — Taiwan, South Korea, Hong Kong and Singapore — that had all experienced rapid economic growth.
What he found corroborated Mankiw, Romer and Weil’s findings on physical capital to some extent. Young, however, attributed even more power to the changes in human capital. In each of the four countries, he found that families were having fewer children and investing more in their education. Increases in educational attainment created a more skilled workforce — an impact which Young was able to track in more detail than Mankiw, Romer and Weil. Their slower population growth was associated with increased labor force participation by women and an increase in the share of the population that was of working age.
Research like this established how economic growth was able to accelerate in some countries, but it does not tell us why those changes took place in the first place. Why did capital formation speed up in South Korea or Taiwan (and not in Bangladesh or Nigeria)? Why did families start to have fewer, better-educated children in those same places?
What we are after is a deeper set of fundamental characteristics, policies and events that created the circumstances under which rapid economic growth occurred.
Institutions as Fundamentals
The hunt for the fundamental whys of rapid economic growth arguably defines the study of economics. Adam Smith was concerned with exactly this question in The Wealth of Nations. While that hunt has always been near the core of the discipline, there was an eruption of research on the subject in the decades following the studies by Young and Mankiw, Romer and Weil.
Within that literature, economists have tended to group those fundamentals of economic growth into three broad categories: culture (e.g., the willingness to trust and engage in trade with strangers), geography (e.g., ease of transportation) and institutions (e.g., security of property rights). Of the three categories, institutions have received the most attention. This is in part because they tend to be more legible to economists than issues of geography or culture, and in part because they would appear to be more amenable to change.
But what exactly is an institution? Douglass North, the Nobel Prize winner credited with originating the study of institutions as a driver of long-run growth, has defined them as “humanly devised constraints that structure political, economic, and social interactions.”
That is so broad it offers little chance of identifying real policies or changes that countries could pursue. Researchers who took North’s ideas and ran with them contributed in part by being more specific. In early work, Daron Acemoglu, Simon Johnson and James Robinson, responsible for initiating detailed empirical research into institutions, focused on the security of private property rights, measured by either the risk of expropriation (based on assessments by investors) or the legal constraints on government executives (based on assessments by political scientists).
Work by Acemoglu, Johnson and Robinson, and those that followed, looked across a wide set of countries, searching for common institutional elements that existed in all the countries that experienced rapid economic growth (or that were absent in those that did not). These studies focused at first on measurements of institutions and growth during the 20th century, but soon incorporated data from even earlier. The same three authors (along with Davide Cantoni) studied the importance of an institution we could call “equality before the law” by examining the effect of Napoleonic reforms made in Germany at the turn of the 19th century on subsequent development.
In other work, they estimated that European countries with more representative institutions, like Britain and the Netherlands, were able to grow more quickly in response to the opening of trans-Atlantic trade routes than absolute monarchies like Spain and Portugal.
These authors and the literature that followed them tended to find that things like robust property rights for individuals and governments with clear restraints on executive power, democratic political processes and a lack of government corruption were all associated with economic growth.
Those institutions certainly sound “right.” They are things we’d associate with almost any major developed country like the U.S., France or Germany. But, at heart, most of these studies share the same fundamental issue as those that looked at capital accumulation: Just because certain institutions are present in places that had rapid economic growth, that doesn’t mean they were necessary for the miracle to occur. Perhaps things like property rights and a lack of corruption are “luxury goods” that rich countries can afford to indulge in but are not, in fact, the reason those countries became rich?
The problem gets even thornier when researchers try to pin down how to even measure an “institution” in the first place.
A concrete example: The World Bank has a set of “Governance Indicators” it collects from each country. Those indicators include a measure of the “control of corruption” that a country has. For example, in 2020 Eritrea had a “control of corruption” indicator of −1.33, quite low. Mauritius had a 0.47, which is around the middle of the pack, and Denmark had a 2.27, among the highest. In terms of absolute ranking, it is probably correct that Eritrea is more corrupt than Mauritius and that both are more corrupt than Denmark.
But do the numbers themselves mean something? Is Denmark exactly 4.8 times less corrupt than Mauritius? If Eritrea managed to raise their index to −1, would that imply the same change in corruption as Mauritius moving to 0.80? The answer to both questions is obviously no. At best the numbers let us rank countries on these dimensions of governance, but there is no sense that 2.27 means anything in practice.
The statistical analysis that establishes the link between control of corruption and economic growth assumes, however, that the corruption index has a precise numerical meaning.
It’s not that the statistical analysis is wrong — it’s that it has no practical interpretation. The control-of-corruption index, like other World Bank governance indicators, is based on survey data. But people in rich countries are more likely to give their institutions high ratings. In one striking case, Edward Glaeser et al. pointed out that Singapore has historically scored highly on measures like constraint on executive power — even when it was ruled by Lee Kuan Yew, a dictator who had no constraints on his power but did happen to respect property rights.
Ideally, economists would try to control for confounding variables like wealth or education, but the fact that there are only about 50 to 70 countries with available data makes that impossible. As a result, the measures are circular: They tell us that Denmark is better governed than Mauritius or Eritrea, but not much else.
This isn’t a problem unique to measuring the degree of corruption. Every index of institutional quality is subject to this critique, because every index is attempting to assign numbers to something that is not inherently quantifiable: the degree of democracy, the rule of law, government effectiveness, respect for property rights, etc. In each case, the research might indicate that “being like Denmark” is a good thing, without any practical way of expressing what that means.
Experimenting With History
The picture I painted of cross-country research on economic growth is bleak, but those issues are not lost on researchers. Knowing these issues, scholars have tried to establish better evidence for which institutions matter for economic growth.
Much of this research is based on an examination of historical or natural experiments. Once again, South Korea is a useful example. After World War II, the Korean peninsula was, of course, partitioned between South and North Korea. The two countries share similar geography, so the miracle in South Korea and the utter lack of one in North Korea cannot be attributed to their endowment of minerals or physical access to foreign markets. They have a shared language and culture, so it is hard to say that there was something unique about the South Korean culture or history that prompted the miracle there (or halted it in North Korea). They both were left devastated and poor by the Korean War.
What’s left as an explanation is that the set of institutions governing economic activity in the two countries were distinct after 1953. The North adopted a communist ideology and built a set of economic institutions around it. We can see the results of that today. North Korea has failed, by any plausible metric, to advance economically. In addition to the lack of individual freedom, living standards are among the worst in the world, and North Korea continues to suffer from recurring issues such as famine that advanced economies like South Korea left behind years ago.
This example is useful in that it tells us institutions matter for economic growth, and unlike other research can more clearly eliminate other options like geography or culture. It also doesn’t require us to assign an artificial index to the institutions of South Korea or North Korea. We know they’re different, and that’s enough.
What that case study lacks, of course, is a clear answer to which institutions were the relevant ones making South Korea an economic miracle. Was it the subsidization of the “chaebol” — conglomerates like Samsung, Hyundai or LG — with cheap credit? Was it, uncomfortably, the lack of real democracy until 1988? Was it the promotion of exports versus domestic consumption? We can’t know from this simple comparison.
Research has thus continued to search for more historical natural experiments where the nature of a particular institution is much more apparent. The experiments the authors rely on are often quite clever. Melissa Dell compared areas of Peru subject to a Spanish forced-labor requirement called the “mita” to areas that were not and found that they have lower living standards centuries later.
Lakshmi Iyer found that areas of India subject to direct British rule (as opposed to those ruled through proxies) have lower investments in schooling and health today.
Stelios Michalapoulos and Elias Papaioannou compared areas of sub-Saharan Africa that had historically more sophisticated political structures prior to colonization continue to be richer today than areas that were less organized.
In each case, a very specific institution — a forced labor regime, direct British rule, precolonial political structure — was found to have a significant effect on contemporary economic outcomes.
The empirical work here is on more solid ground, and the authors avoid the measurement issues mentioned above. But these studies, by narrowing their focus to specific historical experiments and individual institutions, have their own limitations. These studies don’t tell us about the immediate effect of any of these institutions. The British Raj ended decades ago, the Spanish forced-labor system in Peru ended over two centuries ago, and the historical political organization of sub-Saharan Africa are just that — historical. What we learn from these studies is that institutions can have persistent effects well after the institution disappears, implying that countries or regions can get stuck in a poverty trap. Once the region is impoverished, it’s more likely to stay poor.
These papers work as cautionary tales; they tell us what won’t work, but not what will work. And while they don’t provide any silver bullets for generating economic growth, they remain valuable contributions to the study of development. This work is eliminating bad options from the menu of institutional choices that countries could make.
Negotiating for Growth
Alongside the literature on what not to do, there is recent work that attempts to be more constructive. Acemoglu and Robinson, who helped initiate the empirical study of institutions, are among the leaders in this new line of inquiry as well.
The key here is a change in the question. Rather than asking what the right institutions are to promote growth, they ask why failed institutions persist. For them, countries stagnate at low levels of development because there is a stalemate among interest groups; despite the aggregate benefit, no group is willing to implement an improved set of institutions.
What their research suggests is that breaking out of that stalemate requires a fundamental expansion of the distribution of economic and political power within a country. By incorporating more people in economic and political decision-making, they argue, a country is better able to negotiate a set of economic institutions that promote economic development.
This sounds promising, but can we see it in the data? These authors and others have made progress and are beginning to provide supportive empirical work. What sets them apart from earlier work is that they have the benefit of knowing that mistakes were made in the past. A good example is from Acemoglu and Robinson along with coauthors Suresh Naidu and Pascual Restrepo.
They show that the transition to democracy leads to higher economic growth in the future, finding GDP per capita is around 20% higher in a democracy compared to an otherwise identical nondemocracy. What they see is that countries that democratize invest significantly more in public health and education, consistent with the initial work that Mankiw, Romer and Weil and Alwyn Young did on economic growth.
They explicitly take on all of the empirical issues I complained about above. They do not try to quantify “democracy” along some arbitrary scale (e.g., North Korea is a one, the U.S. is a seven, etc.). They instead focus on a simple comparison of places that clearly democratized versus those that did not. They use several methods to try to assure themselves, and us, that their results are coming from the causal effect of democracy on growth, and not the other way around. This includes a sort of natural experiment where democratization is more likely to occur when more neighboring countries are democracies.
Some counterexamples may immediately come to mind. South Korea, whose economy took off in the ’60s, did not democratize until 1988, and China has undergone impressive economic growth without democratizing at all. But once Acemoglu, Naidu, Restrepo and Robinson make the comparison across all countries, it turns out that their experiences are something of an outlier, not the norm. And in both, there were events that led to a widespread expansion of the distribution of economic power, even though it was not accompanied by political power: the massive redistribution of land in South Korea following World War II and the market reforms in the 1970s and ’80s in China that gave more people rights over their land and assets.
This result is exciting, in part because it suggests that something inherently positive — wider representation and democracy — is also conducive to economic growth. But it doesn’t mean we’ve cracked the code and are capable of generating economic miracles at will. Countries that do expand the distribution of political and economic power still have to negotiate the institutions supporting growth. This is where our expanding knowledge of which institutions don’t work becomes valuable, helping eliminate dead ends.
Making Modest Conclusions
At this point the situation may seem rather grim. Can we say, with any confidence, that we know the set of policies or institutions that can create the rapid economic growth seen in South Korea and others? The frank answer is no.
But this does not mean we are at a complete loss. Do not dismiss the power of the cautionary tales I mentioned. While the Korean “experiment” didn’t tell us what exactly South Korea did right, it continues to provide a vivid lesson that the North Korean centrally-planned authoritarian regime was not a viable economic path to take. Documenting which institutions don’t work is slow, but it is progress nonetheless. Furthermore, recent results regarding the importance of the distribution of economic and political power mean we understand more about the conditions that can cause good institutions to arise.
Can we make an economic miracle? No. Do we understand what might make economic miracles more likely? To some extent, yes. That wishy-washy answer doesn’t sound very inspiring, but it represents a tremendous amount of progress. The series of critiques and incremental improvements I’ve described is an example of the research process at work. Given the stakes, the slow pace is frustrating, but we are headed in the right direction.
Based on GDP per capita. My calculations from Robert C. Feenstra, Robert Inklaar, and Marcel P. Timmer, Penn World Table, V10.0 (updated June 18, 2021), distributed by Groningen Growth Development Centre.
GDP per capita data from the Penn World Table is adjusted for inflation and differences in the cost of living between countries. Other methods of estimation report somewhat different figures.
Calculation based on results from Dev Patel, Justin Sandefur, and Arvind Subramanian, “The New Era of Unconditional Convergence,” Journal of Development Economics 152 (September 2021): 1–18, 102687.
Implicitly, there is a fourth fundamental to add to this list: luck. It might be that some of what explains growth in South Korea or other economic successes is a fortunate set of contingent circumstances, and there isn’t any way to make a miracle. Even if I had a complete physical and psychological profile of Serena Williams, I probably cannot make complete sense of her dominance, which at times might have benefited from a favorable draw or a lucky bounce.
Douglass C. North, “Institutions,” Journal of Economic Perspectives 5, no. 1 (Winter 1991): 97. The work summarized in that article originates in Douglass C. North and Robert Paul Thomas, The Rise of the Western World: A New Economic History (Cambridge: Cambridge University Press, 1973), and Douglass C. North, Structure and Change in Economic History (New York: Norton, 1981).
Daron Acemoglu, Simon Johnson, and James A. Robinson, “The Colonial Origins of Comparative Development: An Empirical Investigation,” American Economic Review 91, no. 5 (December 2001): 1369–1401. An accessible introduction to their body of work is Daron Acemoglu and James A. Robinson, Why Nations Fail: The Origins of Power, Prosperity, and Poverty (New York: Crown Publishers, 2012).
Daron Acemoglu, Suresh Naidu, Pascual Restrepo, and James A. Robinson, “Democracy Does Cause Growth,” Journal of Political Economy 127, no. 1 (February 2019): 47–100.
Consistent with the Acemoglu, Naidu, Restrepo, and Robinson (2019) findings, South Korea did eventually democratize in 1988 and now enjoys living standards roughly equal to those in Western Europe. China, on the other hand, has failed to expand political representation and its own growth miracle is already showing signs of slowing down well short of reaching that level of GDP per capita.
Published November 2022
Dietrich Vollrath is a professor and chair of the department of economics at the University of Houston. His work focuses on economic growth. He blogs at growthecon.com.
By highlighting text and “starring” your selection, you can create a personal marker to a passage.
What you save is stored only on your specific browser locally, and is never sent to the server. Other visitors will not see your highlights, and you will not see your previously saved highlights when visiting the site through a different browser.
To add a highlight: after selecting a passage, click the star . It will add a quick-access bookmark.
To remove a highlight: after hovering over a previously saved highlight, click the cross . It will remove the bookmark.
To remove all saved highlights throughout the site, you can click here to completely clear your cache. All selections have been cleared.