Available now: USAvailable now: UK

Response to Fukuyama's Review

We are generally very happy with the reactions that Why Nations Fail is receiving from reviewers in newspapers, magazines and blogs. (A few of those can be found here). As it may be expected, some have been positive and some less so. But one review, by Francis Fukuyama in the American Interest blog, stands out in its critical tone and content. 

Our original reaction was to ignore it. But we have been asked about our reactions to this review a number of times, so we have decided that perhaps it is worth responding to some of its main claims.

Here, we would like to focus on some of the main criticisms that Fukuyama levels at Why Nations Fail.

First, Fukuyama writes “They are making almost the identical point to that one made in the 2009 book by Douglass North, John Wallis, and Barry Weingast”.

Our work in general, and Why Nations Fail in particular, built on North’s seminal ideas (for example, those in Structure and Change in Economic History and on North and Weingast’s seminal paper in 1989 Journal of Economic History as well as on North, Wallis and Weingast (NWW). But there are also major differences — which would have been really obvious to Fukuyama. The most important one is about the role of politics. In our book, we argue that a framework built on the primacy of politics — political institutions being forged and changing as a result of conflict and in turn shaping economic institutions and then innovation and investment — will go a long way. Our framework also clarifies that what we call extractive institutions are there by design. NWW put much more emphasis on economic and social factors. For example, their concept of “open access order,” far from being similar to “inclusive political institutions” as Fukuyama argues, is defined as a “social order” similar to Seymore Martin Lipset’s “development complex” (page 3). In particular, NWW emphasize the existence of business and other non-government organizations, impersonal exchange and beliefs supporting cooperation in society as the foundations of “open access orders”.

Though our work owes much to North and his co-authors, it is disingenious to imply that we have just re-created their argument under a different guise. In fact, NWW are themselves very explicit that when they talk of political change, they build on our past work, in particular our earlier book, Economic Origins of Dictatorship and Democracy as well as our earlier work (see, for example, their Chapter 5).

Overall, it should be clear that our work, though it builds on NWW, takes a different approach, and the two works not only build on each other but are quite complementary.

Second, Fukuyama argues that our distinction between inclusive and extractive is “sharply bifurcated”. This is clearly meant as a criticism, but the exact meaning is not quite clear. If by this Fukuyama means that inclusive and extractive political (and economic) institutions are defined as extreme cases, this is exactly right. As we state in the book, most countries are in shades of gray, but it is most useful to start with the black and the white to understand what’s going on. If, on the other hand, by this Fukuyama means that we think societies in reality are fully inclusive or fully extractive, this is obviously not true as most of the book is about degrees of extraction and small steps towards inclusive institutions.

Our discussion of China, for example, which Fukuyama further comments on, is all about this: we argue China did not start growing by becoming fully inclusive, but by taking some important — and yet limited — steps towards inclusive economic institutions.

Third, Fukuyama accuses us of not unpacking the constituent parts of inclusive and extractive institutions. He even writes: “There is for example a large literature comparing the separate impacts of a modern state, rule of law, and democracy on growth, which tends to show that the first two of these factors have a far greater influence on outcomes than democracy.” Now there are two problems with Fukuyama’s statement. First, the literature he seems to have in mind is the one on cross-country growth empirics, which sometimes runs kitchen-sink regressions including several indices at the same time. But few economists would take such regressions seriously given the endogeneity and reverse causality concerns. Second, a large part of our book is in fact devoted to explaining why many extractive (and inclusive) aspects will travel together — because of the feedbacks between these different aspects.

In fact, this is an important point which is perhaps somewhat easy to miss (or at the very least Fukuyama has missed it): there is a very good reason for broad concepts such as “extractive institutions,” because extraction is often undertaken using one of many specific institutions that are on the whole different ways of “skinning a cat”. So, for example, when after the Civil War, slavery was abolished and black Americans were enfranchised, this did not put a conclusive end to extractive institutions in the US South, but opened the way to the emergence of a different complex of specific institutions to achieve the same objective. Slavery went, but then came Jim Crow, “separate and equal” schooling, vagrancy laws, convict labor, and a whole gamut of restrictions against the movement and labor market freedoms of freed slaves. Former slaves were given the right to vote, but literacy tests for voting and the Ku Klux Klan made sure that they were effectively disenfranchised. Just focusing on a specific institution of extraction would miss the continuity in the nature of extractive institutions in the US South. It would also miss the similarities in the nature of extractive institutions across the world, which we tried to illustrate, for example, in Chapter 13 of Why Nations Fail.

Fourth, Fukuyama claims that China’s rapid growth invalidates our theory. Why this is not the case is discussed both in Chapter 5 and Chapter 15 of Why Nations Fail, so it’s a little surprising to see him make this claim. We also discuss the spectacular economic performance of the Soviet Union until the 1970s. We don’t think Fukuyama would argue that the rapid growth of Soviet Union for over four decades invalidates the theory that markets are better than central planning in generating innovation and economic growth. It doesn’t — for the same reason that Chinese growth so far doesn’t invalidate our theory. As a matter of fact, China is still much poorer relative to the United States than the Soviet Union was in 1970. If China is able to reach levels of income per capita comparable to those of Spain or Portugal based on its extractive political institutions, then this would invalidate our theory. But, as we explain in detail in the book (and we have also discussed elsewhere, for example here and here), growth under extractive institutions is part of our theory, so it is certainly not inconsistent with our framework. It is just that our theory suggests that such growth will not be sustained.

Our theory does not explain all sources of success and failure of nations. Nor do we claim to be the final word on this huge question. There are thus both many new and interesting approaches to be developed and many intelligent critiques of our work to be written. We look forward to such future approaches and critiques.


Will Burma succeed? 

Tunisia, Egypt, and Libya (and hopefully Syria) chart one path for the fall of a repressive extractive regime: a revolution deposing the regime — peacefully or violently, largely depending on the reaction of the regime.

Burma shows another: the regime itself gradually letting go of the reins.

In September 2010 in Foreign Policy, we argued that this “soft landing” strategy, though far short of meaningful political reform and flawed in many ways, could be the first step of a longer process of political opening in Burma.

We have had an opportunity to answer some more questions about Burma in a recent interview.

The gradual process of reform in Burma has indeed continued, with Aung San Suu Kyi being allowed to run for and being elected to parliament but boycotting the opening session (over the wording of oath of office imposed by the junta). The European Union deemed this political opening to be significant enough to suspend most of the sanctions against Burma for one year.

Both the Arab and the Burmese paths towards more inclusive institutions come with serious pitfalls.

The Arab path opens the way to another group coming to power and re-creating yet another extractive regime as has been the case with many of the post-colonial governments that overthrew the colonial regimes after World War II or the Bolshevik Revolution that kicked out the czarist regime in Russia — both the military and the Muslim Brotherhood our candidate would-be usurpers in Egypt, for example.

The Burmese path risks creating the appearance of change without real change, and is also vulnerable to a reversal down the road if the new cadre of the military, still completely and forcefully dominating Burmese institutions, wishes to change direction.

Unfortunately, there is no easy way out of extractive institutions.


American resilience

There are clear and present dangers for US institutions: increasing economic inequality; increasing political inequality; super PACs and all that; failing educational institutions; huge incarceration rates, especially for African-Americans; and erosion of civil liberties.

Hard not to be pessimistic.

But we have also argued that there are reasons for optimism: we have been here before, and rebounded. The most direct parallel is with the Gilded Age, when despite the huge economic and political power of the elite at the time, US institutions turned out to be more open and resilient than most feared (see here ). The major reforms of the Civil Rights era, which ended the disenfranchisement of a large fraction of the population, are also grounds for believing that the US can rebound from the challenges it is facing today.

A very interesting recent paper finds hope about the resilience of the US institutions in an unlikely place — in the corruption of US politicians. Pablo Querubín and James Snyder study whether elected US congressmen have been able to enrich themselves between 1850 and 1880. (Full disclosure: Pablo was a student of ours, and James Snyder is a former colleague of Daron and a current colleague of James). Used as we are to revolving door lobbyists, staffers and politicians, it would appear obvious that the answer should be yes.

Corruption was in fact in the air. Historians describe the periods both before and after the Civil War as apogees of corruption. For example, Mark Summers writes in The Plundering Generation: Corruption and the Crisis of the Union, 1849-1861:

In every way the decade before the Civil War was corrupt. The 1850s were as depraved as any other age, and, at least from the evidence available to historians, far more depraved than the 1840s.

But the striking result in Querubín and Snyder’s paper is that both before and after the Civil War, US congressmen were not able to enrich themselves.

Whether politicians have been able to use office for personal gain is not an easy question to answer empirically. It would not be sufficient to compare the wealth of a politician before and after office, as many entered Congress at the peak of their careers, so it would be only natural that their wealth would increase during that time. It would also not work to compare congressmen to others in similar professions, since they would have proved to be different by their choice to enter politics. And it would also not be sufficient to compare them to all candidates who run for the same offices and did not win, because we might expect that the winners, the actual congressmen, are more talented than those they had defeated.

Querubín and Snyder, instead, use a regression discontinuity approach, comparing congressmen that barely won election (with a small vote margin) to those who have run but barely lost. This approach thus zeroes in on a group of candidates who are very similar, at least in the eyes of the voters — a group mainly distinguished by winning just above and below the vote margin necessary for election. (And what makes the paper a phenomenal amount of work is the collection of wealth data for all of these candidates).

Once they do all this work, Querubín and Snyder find that both before the Civil War and immediately after it, congressmen who just won election accumulate no more wealth than the candidates who just lost.

The second striking result in the paper, and the one that receives more emphasis from the authors, is that during the Civil War, a very different pattern emerges: now politicians seem to be able to use their office for personal enrichment, and do so quite brazenly. Querubín and Snyder find that those winning election with a close margin accumulated about 40% more wealth in the decade straddling the Civil War than candidates losing with a close margin — a very substantial effect.

Why was the Civil War era different? Querubín and Snyder show that this was most likely for two reasons: first, there was a huge spike in government spending associated with the war, and it was precisely the congressmen who had their fingers in that much bigger pie (e.g., those from states with significant military mobilization and those serving in committees responsible for military appropriations) who made up like bandits. Second, the media, perhaps the most important monitor of politician malfeasance, was preoccupied with something else — the war.

Not good news for those who still believe in the integrity of our politicians — if any of them are left. But actually surprisingly good news about how well US institutions used to work during regular times — so much so that they were able to control these thieving incentives fairly well.

This research raises two substantial questions:

First, how was it exactly that US political institutions checked the behavior of congressmen? Querubín and Snyder suggest media might have played a role, but there is no direct evidence of this. One might also suspect that it wasn’t just media. If so, which were the features of US political institutions that played a role in taming politician malfeasance?

Second, whatever US institutions had in 1850, do they still have it today? Can they regain it?


American failure

There is much talk of failure of the US economy and institutions — much of it with good reason. But tangible signs of that failure are few. Some would point out to the increase in inequality as clear evidence. But there is debate as to whether this increase has been caused by non-political factors such as skill-biased technological change and globalization, or political factors; we have argued in a previous blog post that the non-political factors are unlikely to account for the behavior of the share of the top 1% in national income, but this is not convincing to those who come with different priors.

Here are some less controversial signs of American failure.

The next figure shows high school graduation rates of males and females and for both genders combined by birth cohort (year of birth). For males, high school graduation rates were highest for the cohorts born in the early 1950s; they have peaked in the late 1960s at less than 85%, and have slightly declined since then. For women, they have increased slightly. Well into the 21st century, almost 15% of the US population is not graduating from high school, even though the economic consequences of this are dire. (Moreover, this is despite the fact that recent numbers include those obtaining Graduate Equivalency Degrees, GEDs, the “second chance” high school degree for those who drop out, and the economic returns from a GED are far less clear than a regular high school degree; see, e.g., this paper).

The next figure shows that the pattern is very similar for college graduation rates. The peak for males is again in the late 1960s (for cohorts born at the end of the 1940s). Then it’s downhill, only recovering in the late 1980s and very partially — the overall college graduation rate recovers and increases thanks to the higher rate for women. (See Claudia Goldin and Lawrence Katz’s excellent book, The Race between Education and Technology and its discussion by Daron Acemoglu and David Autor here for more details on these patterns).

The next figure shows the evolution of the incarceration rates of African-American, Hispanic and white males. The US is currently locking up almost 5 out of every 100 African-American men, and most of it for non-violence crimes. Few should be able to argue that this is not a uniquely American failure.



What’s the Matter with Argentina?

One hundred years ago Argentina was one of the richest countries in the world. For fifty years it rode a wave of favorable market opportunities and experienced one of the most successful periods of extractive growth in world history (sharing these honors with the Soviet Union and contemporary China or Sinagpore).

But extractive growth can’t last and in Argentina it didn’t. When you have a combination of extractive political institutions but somewhat inclusive economic institutions as Argentina did (and China does now) there are two ways to go, but only one leads to sustained economic growth. You can open the political system and try to move towards an inclusive society, or you can go in reverse and clamp down on the inclusivity in economic institutions giving up prosperity for power.

Argentina did the latter. Around the time of the First World War the traditional elites in Argentina made hesitant steps towards opening the political system, but they quickly decided they did not want to run the risk of losing power. In 1930 came the first military coup. Opening the political system was off the table. Then came a series of coups, and the one in 1943 brought a military officer named Juan Domingo Perón to power as Minister of Labor. Perón used this position to build an extraordinary political machine which has dominated Argentina politics more or less ever since. There is only one effective political party today in Argentina, the Peronists (official name: Partido Justicialista). The Peronists were not the traditional elites. Quite the contrary. But they seized control of the society and its economic institutions, and have been every bit as extractive as the traditional elites.

This is a pattern we called, following the sociologist Robert Michels, the Iron Law of Oligarchy. Extractive economic institutions always create struggles for control, and they tend to attract would-be elites bent on extraction, and whoever comes to power takes over a system without checks on their power. The outcome is to re-creation of extractive institutions under a different guise.

In Argentina in the 1940s, the Peronists won the struggle for power and are still on top. And the extraction still continues (even if it often takes the form of distorting economic institutions to hold on to power rather than just for personal enrichment).

Latest evidence: this week the Peronist government under Cristina Fernández de Kirchner seized a 51% share in the oil company YPF, previously majority owned by the Spanish energy company, Repsol YPF. The government immediately ousted the Chairman Sebastián Eskenazi and replaced him with the Peronist Minister of Planning. A tribunal will be formed to decide on the level of compensation!

                                       “I’m a head of state and not a hoodlum.”

Insecure property rights are actually a well known phenomenon in Argentina. Grabbing the oil company is actually small potatoes compared to “El Corralito” in 2001 when the government effectively expropriated 75% of people’s savings in banks (as we discuss in Chapter 13 of Why Nations Fail).

The failure to make the transition towards inclusive economic institutions condemned Argentina to a century of economic stagnation.

Will the same thing happen in China? Singapore?