Daron Acemoglu is a Professor of Economics at the Massachusetts Institute of Technology. James A. Robinson is an economist too; he taught for some years at Harvard and now holds a professorship at Chicago. Both men have published widely, sometimes together. In 2012 they published a hefty tome titled Why Nations Fail: The Origins of Power, Prosperity, and Poverty.
Acemoglu and Robinson’s thesis is as follows: A nation’s success or failure depends on its institutions. A country will fail if they are extractive; that is to say, an elite has access to resources, extracts them for its own purposes, and permits no competition. Such a society will not develop, and in time will face senescence. Where enterprise and political change is possible, however, there will be progress. They draw upon a wealth of historical examples to show how some societies have progressed and others have not; those that do, have institutions that force the elites to allow change, even when their the economic interests are threatened – by, for example permitting the creative destruction of technology, rather than suppressing inventions that bring it about. They use the manufacturing and textile innovations of the 18th century as a key example, and see the erosion of absolute authority in England in the 17th and18th centuries as key to their argument; it made way for the Industrial Revolution. In particular they stress the “Glorious Revolution” of 1688, when absolutism was forever banished and the country sprang ahead of its neighbours.
In a sense, of course, this isn’t a new idea. As John Stuart Mill put it in On Liberty (1859), the prevailing opinion on any subject is rarely the whole truth and “it is only by the collision of adverse opinions [that it] has any chance of being supplied.” Acemoglu and Robinson, however, underscore their thesis with the conviction that a society’s status is path-determined; that is to say, once a society has extractive institutions, the elite that controls them will have no incentive to surrender control, and every reason not to – and the country will remain poor. But where a process of change has begun from below, it will be in an elite’s interest to accommodate it lest the rule of law as a whole be threatened, leaving them exposed to whatever chaos comes next.
The book went viral; the paperback is still in the top few thousand in Amazon’s sales rankings, seven years after it was published. That’s unusual for an academic work. The fact that the authors promoted the book energetically themselves will have helped. But also, this book has been written for public consumption; the reader won’t be troubled by econometrics, multiple regression, p-values, footnotes or in-line citations. It helps, too, that the text is accessible and, for the most part, well-written. That does not mean the ideas in the book are oversimplified; they are not, and are most absorbing. As Paul Collier put it in The Observer (March 10 2012): “Mostly, such people write only for other academics. In this book, they have done you the courtesy of writing a book that while at the intellectual cutting edge is not just readable but engrossing.”
But not everyone has bought the ideas in this book. Some have pointed out that China has advanced and prospered, despite not being a democracy. Others have been critical of the authors’ refusal to accept that natural-resource endowments, climate, and even dumb luck play a part in a country’s success or failure. In fact some of those who take the latter view have been quite harsh in their criticism, perhaps seeing Acemoglu and Robinson as passing judgement on countries that are poor for no fault of their own. Given the book’s impact, it seems worth examining the arguments.
First, this book is as much history as it is economics.
The historical perspective
As economists – and social scientists – the authors might be expected to marshal evidence from contemporary data and use it to tell us where we should go next. They do not do that. Instead, they use a series of historical examples. They begin with the town of Nogales, which is bisected by the US-Mexican border. The south, they say, is relatively poor; the northern half is not.
The reason that Nogales, Arizona, is much richer than Nogales, Sonora, is simple: it is because of the very different institutions on the two sides of the border, which create very different incentives for the inhabitants of Nogales, Arizona, versus Nogales, Sonora. To explain this, we are taken to the colonial experience. The Spanish Empire was extractive, gorging itself on mineral wealth procured for it by abundant slave labour, and the institutions of Latin America developed accordingly; an elite was afforded access to the spoils, this access being in the gift of colonial rulers who used it to keep control by proxy. There was no incentive to surrender control, and it remained absolute.
North America was different. Acemoglu and Robinson argue that the lack of gold or silver meant that, in Virginia, there was less for the English colonisers to extract; moreover they lacked a defeated population to enslave in order to extract it. There were no resources to control and thus no patronage to give; the colonists were forced to make what they would of the land themselves, and the colonial authority had no alternative but to let them. Thus a different polity developed. This fed back into the home country, where a process of change had already begun with the Black Death several hundred years earlier; this had broken the absolute power of feudalism, for with less labour available, what was left could make greater demands on the ruling class.
The country came to what the authors see as a major climacteric with the Glorious Revolution in 1688, when James II, who had tried to rule as an absolute monarch, was replaced with William and Mary, who agreed to rule with Parliament as a condition of the throne given to them. From that time, extractive institutions would decline. The growth of civil rights saw the enforceable patent that encouraged inventions, while elite interests were no longer able to prevent their adoption. Crucially for the authors, this led to the creative destruction by which new technology breaks old, outmoded institutions and drives the future.
The authors draw on a number of other historical examples to support their argument that inclusive institutions drive progress, while extractive ones cause a society to stagnate and even to collapse. Some are fascinating. There is the decline of the Mayan civilization. There are the differing paths of nations and polities in Africa – a chapter on South Africa is especially interesting, suggesting as it does that black African farmers were progressive and invested heavily in the late 19th century, but that colonial institutional constraints then frustrated them. Other evidence is drawn from the Spice Islands, Axum and Nubia. In Africa, the authors draw on Sierra Leone and Zimbabwe as examples of countries that have had extractive governance, and have failed; and Botswana as a country that evaded the worst excesses of colonialism and has prospered, with the help of wise leadership (in particular that of Seretse Khama, its first leader).
It is not unusual to construct a global thesis in this way, reaching back into the past in search of rules that may guide us into the future. H.G. Wells foresaw the utility of using history to light one’s path in his 1902 lecture The Discovery of the Future. One cannot predict the future of one man, he argued, but he is but one grain of sand; you can’t predict what a single grain will do but you can probably say, from experience, what a large quantity will behave when you tip it on the ground. In recent years several authors have been successful in this respect; Jared Diamond is one, while Joseph Tainter’s work on the collapse of complex societies has sought to illuminate the reasons why some civilizations have vanished.
But for some reason historians themselves appear to have abandoned the field. As Jo Guldi and David Armitage argue in their 2014 book The History Manifesto, they have tended to retreat into silos and specialise on a minute scale, immured in the minutiae of historiography and immune to the broad sweep of human experience. Thus a modern historian will have an encyclopedic knowledge of the parish registers of some obscure village in the 16th century, but could never write on the scale of historians of the past such as Gibbon, Toynbee or A.L. Rowse. Guldi and Armitage think this unfortunate, but I wonder if that much has been lost; after all, other disciplines have taken their place. Tainter trained as an anthropologist. Diamond is a biophysical scientist. And Acemoglu and Robinson are economists. There is nothing to regret in this. But it holds traps, not least that it demands a multidisciplinary mind; what if the clue you seek is in a body of literature or a journal that you would not have encountered? In my view Acemoglu and Robinson do fall into that trap in places – but more of that below.
For now, their argument may be summed up as follows: History shows that extractive institutions make everyone poor. Inclusive institutions, in which property rights and intellectual property are secure from seizure, and the way is not blocked by elite monopolies, will make us all rich.
One might argue that the time to say it was in the 16th century; these are lessons we have long learned. I am not so sure. But first, what has been the critical reaction to this book, and what are its perceived weaknesses?
On publication, Why Nations Fail was mostly welcomed but did have its critics. Bill Gates, no less, piled in on the book on his blog (Good Ideas, but Missing Analysis, February 26 2013). The book was, he said, “a major disappointment. I found the authors’ analysis vague and simplistic. Beyond their 'inclusive vs. extractive' view of political and economic institutions, they largely dismiss all other factors—history and logic notwithstanding.” He went on to berate the authors for dismissing (for example) the influence of weather and water on the Mayan collapse. He quoted numerous examples of countries, especially China and other Far Eastern economies, whose people had prospered under undemocratic institutions. “The incredible economic transition in China [has] occurred because the leadership embraced capitalistic economics, including private property, markets, and investing in infrastructure and education,” said Gates. “This points to the most obvious theory about growth, which is that it is strongly correlated with embracing capitalistic economics—independent of the political system.”
One wonders why the book upset Gates so much, and why he felt obliged to give it so much attention. After all, his own fortune was built on the sort of creative destruction wrought by technology that the authors see as essential to progress, and impossible under the wrong institutions. Still, the authors gave as good as they got in a piece by Acemoglu titled What Bill Gates Got Wrong About Why Nations Fail (Foreign Policy, March 12 2013). “Did the Microsoft founder even read our book before he criticized it?” he asked, slating Gates’s “inability to understand even the most rudimentary parts of our thesis”. Ouch. No scholar, they said, had ever argued that the Mayan collapse was due to the weather. (They are probably right, though the Mayan collapse does have many scholars; I suppose it is hard to resist the allure of a nation once ruled over by a king called ‘18 Rabbit’.) The authors also attacked Gates for his sweeping support for ‘capitalistic’ markets. “What about South Africa under apartheid, based on private enterprise by whites, but disempowering and exploiting the majority blacks?” There is more. Suffice to say that Gates gave the book a good kicking but got a bigger one back.
|Gates: Not impressed|
Tropical diseases cause a skilled worker, who completes professional training by age thirty, to look forward to, on the average, just ten years of economic productivity in Zambia before dying at an average life span of around forty, but to be economically productive for thirty-five years until retiring at age sixty-five in the US, Europe, and Japan (average life span around eighty). Even while they are still alive, workers in the tropics are often sick and unable to work.
The authors themselves have said that the fiercest criticisms of their book have come from those who feel they have ignored the importance of natural-resource endowments. This is understandable. The allocation of resources is not equitable, and if a country is poor because it lacks them, that is scarcely its fault. Acemoglu and Robinson, by contrast, seem to be arguing that a country’s fate is its own fault, since its institutions are extractive and closed. So it is easy to see why some critics, especially on the left, might be angry with the book for this reason. But those critics may have oversimplified the authors' argument; they see countries as locked into an institutional heritage that has often been imposed upon them, often by colonialism. There is no judgemental element here.
Even so, others besides Diamond were quite critical of their failure to take geographical and other elements into account. Jeffrey Sachs, in a review in Foreign Affairs, September/October 2012 (available on his website here), also pointed out that authoritarian elites are quite capable of modernising their countries, especially if they face an outside threat; he cites the Meiji Restoration in Japan. In fact, that is a case that Acemoglu and Robinson do cite but from which they derive a different message, seeing it as a move away from extractive institutions. Indeed, different observers may draw different conclusions from the same point in history, as Joseph Tainter shows very clearly in The Collapse of Complex Societies (1987).
I had my own problems with Why Nations Fail. Not in the sense that Gates did; I found his criticisms sweeping and dismissive. Acemoglu and Robinson present a lot of evidence for their thesis. They made a case, and Gates did not disprove it. But Diamond and Sachs were closer to doing so, for the authors’ dismissal of biophysical factors in geography is itself sweeping. This is not just about tropical diseases. Tropical soils can be very fertile, but in a hot climate, soil organic matter mineralises very quickly once the land is cleared for agriculture. Farmers know this and often use manure or crop residues, or long-fallow rotation, but the high food production per hectare common in Europe and North America is not always possible. This does constrain the growth of a productive labour force that has moved off the land. Not all economic factors are about institutions.
I had some other problems with this book.
First, as stated earlier, the authors see a society’s institutions as path-determined; once extractive or inclusive, they will not easily change. As we have seen, they state that where extractive institutions are in place, they will persist because the elites that control them have too much to lose; where extractive institutions are weaker, and fundamental change is possible, elites will permit reform for fear of something worse. But there is no reason why the second case might not apply to the first scenario. A country can be flipped on its head in an instant by war or revolution. Acemoglu and Robinson acknowledge this but say that a country cannot switch from extractive to inclusive institutions so quickly; the mentality will not be there, and the liberators will by default become replacement oppressors. This seems too neat a view, and does not explain the evolution of southern Europe in the 1970s or of eastern Europe 20 years later.
A more solid case for path-determination was made many years ago in Making Democracy Work: Civic Traditions in Modern Italy, by Robert Putnam (again, not an historian but a political scientist; the historians are nowhere to be seen in this debate). In 1970 authority was devolved to new local governments in Italy. This enabled Putnam and his associates to compare the performance of these institutions, all of which were starting, on paper, from the same base – and see which ones did better, and why. The book was a labour of love, taking over 20 years. Putnam and his colleagues used a number of ruses to test government performance – for example, writing to the local administration to ask how one could be reimbursed for medical costs incurred abroad, and comparing the speed and quality of the reply. Taken together with the public’s own perception of their region’s efficiency, there was a startling difference between the wealthy north and relatively poor south. Clearly, the richer a region was, the better its government was likely to be; but why?
Putnam found a correlation not just with per capita income, but also with ‘civicness’ – a tendency to associate in groups. The range of groups he accepted for this analysis was very broad, including for example choral societies. The number of such groups varied from one per 1,050 inhabitants in Trentino-Alto Adige to one per 13,100 in Sardinia. Newspaper readership was also much lower in impoverished regions. So Putnam did not conclude that wealth alone created civil society. The evidence seemed to suggest a more complex link than historical wealth; in earlier centuries the south had been, if anything, more industrialized and wealthier than the North. ‘Civicness’ earlier in the century seemed to indicate more economic development now. “Economics does not predict civics, but civics does predict economics,” Putnam wrote.
It is of course one thing to find a purely correlative relationship, which of itself proves nothing. It is quite another thing to find causality. But Putnam claimed he had. His argument, briefly stated, was this: in the Middle Ages, the collapse of existing power structures all over Italy led the north and south to diverge. The South and Sicily found themselves part of a strong Norman empire: “As the centuries passed, the steep social hierarchy came to be ever more dominated by a landed aristocracy endowed with feudal powers,” wrote Putnam. In the north, however, no-one imposed order; rather, there was a dark chaos against which townspeople were forced to unite: “The solution… was quite different, relying less on vertical hierarchy and more on horizontal collaboration. …The extent of popular participation in government affairs was extraordinary.”
To Putnam, what was lacking in the south was social capital. This is an amorphous beast, much discussed but hard to define. Broadly, it is the existence of a relationship with or trust in others that lowers transaction costs in such a way as to make economic or other interaction much easier. In an influential 1988 paper, Social Capital in the Creation of Human Capital, sociologist James Coleman cited the community of Jewish diamond merchants in New York, who can lend each other gems for inspection without huge investment in security and insurance; and traders in Cairo’s Khan el-Khalili market, who cooperate so closely that unrelated enterprises function in effect as a huge department store. From my own observation, the great souk of Aleppo in Syria functioned in a similar way before the awful current conflict. Moreover traders in similar commodities tended to work in the same alley, and may have shared mosques and baths. Access to such social capital may have a profound effect on living standards, although much research remains to be done on why.
|Putnam: Social capital is the key|
This sceptred isle
There is another key area where I have doubts about Acemoglu and Robinson’s argument that institutions are all-important. This is in their treatment of Britain (which they also refer to, wrongly, as England; even before 1707 this was not completely accurate). The authors put great store on the 1688 Glorious Revolution, which they see as fundamentally setting Britain on the path to inclusive institutions, and thus to the extraordinary leap forward that we now call the Industrial Revolution.
There is nothing wrong with seeing 1688 as a waypoint in this process, but it may be nothing more. The winds of change that wrought the workshop of the world were many and various. They also go back much farther. Acemoglu and Robinson rightly acknowledge that they began with the Black Death, which slashed the supply of labour and in so doing broke feudalism, giving birth (albeit protracted) to a more autonomous labour force that would, in time, demand institutional change. Yet they nowhere mention the Enclosures, the process by which the more powerful seized common lands and in so doing displaced much of the rural population.
This process long predated 1688; indeed it reaches back before Tudor times, and the resulting displacement was already much in evidence in the 16th century. A.L Rowse, mentioned earlier, highlights the demography of the time in his magisterial The England of Elizabeth. He explains that, thanks in part to the work of the 19th-century pioneer of demography John Rickman (1771-1840), we know that 16th-century rural parishes often had a high surplus of births over deaths – but that town parishes did not; mortality was high. This allows us to see that there was a surplus rural population, and also to see where it went – the towns – and that it died there. When the technological changes of the 18th century arrived (steam, the spinning jenny), there was a workforce ready to work, instead of to die. One could argue that institutional change itself drove this, but not at all in the way that Acemoglu and Robinson would have you believe; enclosure was an extractive institution par excellence.
A slightly different theory is offered by the work of J.D. Chambers (1898-1970), Professor of Economic History at Nottingham. Chambers grew up in rural Nottinghamshire and was the younger brother of Jessie Chambers, the first girlfriend of D.H. Lawrence and the presumed model for Miriam Leivers in Sons and Lovers. An outstanding economic historian, he drew much of his evidence from his home county. In 1967 he gave a series of lectures that were published, after his death, as Population, Economy and Society in Pre-Industrial England (1972). In it Chambers posits that population increase was already underway before 1750 and may have provided the economic opportunities that led to the Industrial Revolution, rather than being driven by it. Indeed Chambers seems to have seen population growth in England as an almost autonomous factor that created the economy as much as it was defined by it. At the time, not everyone bought Chambers’s conclusions – indeed one or two reviewers were quite rude, which was perhaps easier since he died just before publication. Moreover the later work of Robert Fogel suggests that there was indeed a variable – the reduction of malnutrition, caused by greater availability of food as agriculture became more productive – a theory expounded in his seminal 1990 paper The Conquest of High Mortality. Again, there was now a workforce ready to work rather than die. But the point here is that there are factors at work in the 18th century that Acemoglu and Robinson do not consider. While the Glorious Revolution of 1688 could be seen as pivotal, it could also be seen as incidental to the forces swirling around it.
|The Industral Revolution: Glasgow in 1831 (D.O. Hill)|
A partial explanation
All of the above is germane to Acemoglu and Robinson’s argument. Does it invalidate it?
The short answer is no. They have laid out a well-researched, interesting and combative argument for the importance of institutions in human affairs. To them, politics creates economics, not vice-versa, and they have presented a lot of evidence to that effect. But there are some problems here.
First of all, their argument is all-embracing; institutions, we are told, define success and failure, and there is no sense that they are one of a series of phenomena that will define the course of a country’s history – including biophysical factors, which they consider but dismiss too readily. Second, their approach is insufficiently multidisciplinary. Their sources are not confined to economics, but they are biased that way. They knew of Putnam, but do not really seem to have absorbed his work, or that of others who have worked on social capital. They may have known of A.L Rowse – Robinson especially would do, as he is British-born – but will not have searched his work for evidence, and it seems unlikely that they would have known of Chambers.
But there is no question that Acemoglu and Robinson were onto something. They may attach too much importance to institutions, but they are surely right to regard them as crucial enablers. Moreover this book has been timely. I said earlier that its lessons appeared to be more relevant to the 16th century than to today, when its lessons have – supposedly – long been learned; but they haven’t. Acemoglu and Robinson say little or nothing about the current state of the Western world. Their concern is why some countries (e.g. the developing world) remain poor. But in recent years we have, in fact, seen key institutions of the West go backwards, as the control of monopolies has become less important and Wall Street has regained much of its power. It could be argued that much of Teddy Roosevelt’s work in controlling extractive institutions and their monopolies has been undone, as has that of his kinsman FDR in bringing Wall Street to heel after the 1929 crash. It was a party on Wall Street that caused that crash; likewise, it was the uncontrolled misuse of assets by an elite that brought the 2008 meltdown. (Ironically, Why Nations Fail was shortlisted for the Financial Times and Goldman Sachs Business Book of the Year Award.)
Meanwhile the Internet’s creative destruction has delivered us all into the hands of big players such as Facebook, Amazon and Google. These may not be extractive institutions in quite the same way as the colonial silver mines of Bolivia or the conflict diamond mines of Sierra Leone, both of which the authors discuss. Neither Jeff Bezos or Mark Zuckerberg are in the slavery business. But there is a whiff of the monopolistic and extractive about them.
This lurch back towards extractive institutions has had a concrete effect on living standards. Robert Reich, Labor Secretary in Bill Clinton’s first administration, pointed out in his 2014 book Aftershock: The Next Economy and America’s Future that the process of wealth concentration had been going on for years before 2008. “The wages of the typical American hardly increased in the three decades leading up to the Crash of 2008, considering inflation. In the 2000s, they actually dropped,” says Reich, and goes on to say that the economy has grown so much over that period that, had the benefits been divided equally, the typical person would be 60% better off.
Acemoglu and Robinson actually say little about the dangers of extractive, elite-controlled institutions re-establishing themsleves. They do include an interesting passage on medieval Venice, where they see a process of elite capture as a trigger for decline. And Thomas Friedman (in the New York Times, March 31 2012) quotes Acemoglu as saying that the growth in inequality could be a threat to the US’s institutions. “The real problem is that economic inequality, when it becomes this large, translates into political inequality,” In the book, however, Acemoglu and Robinson generally see a society that has freed itself of extractive institutions as remaining so. The history of the present suggests they are wrong. One of those who saw the book as irrelevant to the current situation, it seems, was Bill Gates. “I don’t think even these authors would suggest that the Great Depression... or the global financial crisis of the last few years came about because of a decline in inclusiveness,” he wrote. Are we sure? This may need further thought.
Why Nations Fail: The Origins of Power, Prosperity, and Poverty is a remarkable book, well-presented, readable and based on serious scholarship; and anyone with a serious interest in how nations do succeed or fail should read it. There is no question that it demonstrates the importance of institutions, and at least partly proves that their influence can persist over centuries. But there is a lot here that does not quite fly – not least because those institutions can go backwards, and seem right now to be doing just that.
Maybe what this book shows us, above all, is that an all-encompassing theory of history is rarely correct; history is just too big, long and messy for that. It is a point Jeffrey Sachs made well in his own review, pointing to the uncertainties wrought by climate change and technological advances. “In such a complicated world, explanations of growth that center on a single variable will become even less useful,” he wrote. But as long as that warning is borne in mind, Acemoglu and Robinson have much to tell us. And their book may contain a more urgent warning than they themselves would claim.