Daron Acemoglu and James Robinson have just published Why Nations Fail, a big book on development that will attract a lot of attention. The latest fad in development studies has been to conduct controlled randomized experiments on a host of micro-questions, such as whether co-payments for mosquito bed nets improves their uptake. Whether such studies will ever aggregate upwards into an understanding of development is highly questionable. By contrast, Acemoglu and Robinson have resolutely focused on only the largest of macro questions: how contemporary institutions were shaped by colonial ones, why it was that regions of the world that were the richest in the year 1500 were among the world’s poorest today, or how rich elites were ever persuaded to redistribute their wealth. In Why Nations Fail, Acemoglu and Robinson restate and enlarge upon earlier articles like “The Colonial Origin of Institutions” and “Reversal of Fortune,” but in contrast to their academic work, the new book has no regressions or game theory and is written in accessible English for general readers.

Acemoglu and Robinson (henceforth AR; Simon Johnson of the old AJR team dropped out of this volume) have two related insights: that institutions matter for economic growth, and that institutions are what they are because the political actors in any given society have an interest in keeping them that way. These may seem like obvious statements, but many people in the development business haven’t gotten the message. Among development specialists there is what AR term the “ignorance” hypothesis: failure to develop is the result of not knowing either what good policies are (this was the old Washington Consensus) or, now that the focus has shifted to institutions, what good institutions are or how to create them. Many development agencies act as if leaders in developing countries want to do the right thing, if only they knew how, and that development assistance should therefore consist of sending smart people from places like Washington out to teach them, perhaps accompanied by some structural adjustment arm-twisting.

By contrast AR argue that bad institutions are the product of political systems that create private gains for elites in developing countries, even if by doing so they impoverish the broader society. (Think Nigeria, which has many multimillionaires while 70 percent of the population lives below the poverty line.) Doing the “right thing” would take away the rents they receive, which is why no amount of hectoring or threats to withhold the next loan tranche has much effect on their behavior. They are making almost the identical point to the one made in the 2009 book Violence and Social Orders by Douglass North, John Wallis, and Barry Weingast (NWW), who argue that most underdeveloped societies are what they term “limited access orders” in which a rent-seeking coalition limits access to both the political and economic system. Indeed, I see no real difference between the “extractive/inclusive” distinction in AR and the “limited/open” access distinction in NWW.

This conclusion about the primacy of institutions and politics for development has important implications for policy as AR point out. If growth is a byproduct not just of good policies like trade liberalization, which can in theory be turned on like a light switch, but rather of basic institutions, then the prospects of foreign aid look dim. Bad governments can waste huge amounts of well-intentioned outside resources; indeed, the flow of aid dollars into poor countries can undermine governance by undercutting accountability, thereby leaving societies worse off than they would otherwise be. As the American nation-building efforts in Afghanistan and Iraq have indicated, moreover, foreign efforts to help construct basic institutions are an uphill struggle. Bad institutions exist because it is in the interests of powerful political forces within the poor country itself to keep things this way. Hamid Karzai understands perfectly well how clean government is supposed to work; it’s just that he has no interest in seeing that happen in Afghanistan. Unless the outsiders can figure out a way to change this political calculus, aid is largely useless.

So far, so good. AR have done a great deal over the years to focus the attention of both theorists and policymakers on institutions, and to shape the emerging consensus on the importance of politics to growth within the economics profession. It is, then, very disappointing that their more fully fleshed out book fails to go very much further than these broad conclusions, skirting critical issues of exactly what sort of institutions are necessary to promote growth, and failing to come to grips with some critical historical facts.

The first problem with their analysis is conceptual. They present a sharply bifurcated distinction between what they call good “inclusive” economic and political institutions, which are sometimes also labeled “pluralistic,” in contrast to what they call bad “extractive” or “absolutist” ones. Unfortunately, these terms are way too broad, so broad indeed that AR never provide a clear definition of everything they encompass, or how they map onto concepts already in use. “Inclusive” economic institutions, for example, seem to include formal property rights and court systems, but also have to do with social conditions that allow individuals access to the market such as education and local custom. “Inclusive” political institutions would seem to imply modern electoral democracy, but they also include an impersonal centralized state as well as access to legal institutions, and forms of political participation that fall well short of modern democracy. We find, for example, that England following the Glorious Revolution of 1688-89 was incipiently inclusive, despite the fact that well under ten percent of the population had a right to vote. When AR first used the term “extractive” in their early articles, it referred to truly extractive practices like the mines of Potosí or the sugar plantations of the Caribbean which extracted commodities out of the labor of slaves. In the current book extractive seems to mean any institution that denies any degree of participation to citizens, from tribal communities to ranchers in 19th century Argentina to the contemporary Chinese Communist Party.

Since each of these broad terms (inclusive/extractive, absolutist/pluralistic) encompasses so many possible meanings, it is very hard to come up with a clear metric of either. It also makes it hard to falsify any of their historical claims. Since more real-world societies are some combination of extractive and inclusive institutions, any given degree of growth (or its absence) can then be attributed either to inclusive or extractive qualities ex post.

The use of such broad categories and the failure to distinguish between the different components of political “inclusion” greatly diminish the book’s usefulness, because one wants to know how these components individually affect growth, and how they interact with one another. 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. There is in fact a lot of reason to think that expansion of the franchise in a very poor country may actually hurt state performance because it opens the way to clientelism and various forms of corruption. The Indian political system is so inclusive that it can’t begin major infrastructure projects because of all the lawsuits and democratic protest, especially when compared to the extractive Chinese one. Furthermore, as Samuel Huntington pointed out many years ago, expanded political participation may destabilize societies (and thereby hurt growth) if there is a failure of political institutions to develop in tandem. All of the good things in the “inclusive” basket, in other words, don’t necessarily go together, and in some cases may be at odds. You never get much hint of this in Why Nations Fail, however, since the authors seem to argue the more inclusion the better, along any of its axes.

Like many other works making use of history but written by economists, the AR volume contains some pretty problematic facts and interpretations. It makes the case, for example, that Rome shifted away from an inclusive Republic towards absolutist Empire, and that this was then responsible for Rome’s subsequent economic decline. Leave aside the fact that Rome’s power and wealth continued to increase in the two centuries after Augustus, and that its eastern wing managed to hold on remarkably until the fifteenth century. It can be argued that the shift from the narrow oligarchy of the Republic to a monarchy with highly developed legal institutions actually increased access to the political system on the part of ordinary Roman citizens, while solving the acute problem of political instability that bedeviled the late Republic.

Similarly, following on a tradition begun by Douglass North and Barry Weingast, AR point to the Glorious Revolution of 1688/89 as a critical juncture marking both the establishment of secure property rights and an “inclusive” political system. The latter point is fair enough, but English property rights were rooted in a much older tradition of common law dating from the Norman invasion, and had created a strong commercial civilization well before 1689. The Glorious Revolution was much less important in establishing the credibility of property rights per se, than of the Crown as a borrower, which explains why English public debt exploded in the century following that event.

Given their overall framework, the hardest thing for AR to explain is contemporary China. China today according to them is more inclusive than Maoist China, but still far from the standard of inclusion set by the US and Europe, and yet has been the fastest growing large country over the past three decades. The Chinese restrict access to the market, engage in financial repression, fail to secure property rights, have no Western-style rule of law, and are ruled by a non-transparent oligarchy called the Communist Party. How to explain their economic success? Rather than see this as a threat to their model (i.e., more inclusion, more growth) AR pull a slight of hand by arguing that Chinese growth won’t last and that their system will eventually come crashing down (like Rome did, after about 200 years?). I actually agree that China will eventually crash. But even if that happens, a theory of development that can’t really explain the most remarkable growth story of our time is not, it seems to me, much of a theory.

The broad conclusions of Why Nations Fail are, thus, incontrovertible and of great importance to policy (which is why, incidentally, I gave it a positive blurb). One only wishes then that the authors had made better use of basic categories long in play in other parts of the social sciences (state, rule of law, patrimonialism, clientelism, democracy, and the like) instead of inventing neologisms that obscure more than they reveal.

I never studied with Jim Wilson while getting my degree in the Harvard Government Department, though he was there at the time. My contacts with him came later, when we served together on the President’s Council on Bioethics in the early 2000s, and as fellow members of the Board of Governors of the Rand Graduate School. Jim in addition graciously agreed to serve on the editorial board of The American Interest.   Of course the way I got to really know him was through reading his books, from early ones like City Politics to the later volumes like Crime and Human Nature, which he wrote with Richard Herrnstein, and The Moral Sense. Unlike many narrow-minded political scientists, Jim was very happy to make use of new research coming from the life sciences and to apply it to contemporary social behavior.

Many of the obituaries and remembrances of Jim Wilson have focused on what was probably his most famous article, “Broken Windows,” which he co-authored with criminologist George Kelling in the Atlantic Monthly back in 1982. As many have noted, this article was responsible for the shift in policing that took place in New York City during the 1980s, that laid the groundwork for the city’s subsequent recovery from crime and decay over the following decade. I’ve always thought that it would be a uniquely satisfying experience for an academic to write an article that would actually have a concrete beneficial impact on the lives of people around you, as this one surely did.

The Wilson book that remains my favorite, however, and that I use the most often in teaching, is his 1989 volume Bureaucracy. Wilson argued that people like to blame bureaucrats for the failings of bureaucracies, but that the problem lay more in the nature of the public sector itself and structure of incentives created by the bureaucrats’ political masters. It began by giving three reasons why the public sector could never simply behave like the private sector, to which it is often unfavorably compared.

First, public sector agencies are not allowed to retain earnings, and therefore have no incentive towards economizing costs. A public agency that ends the fiscal year with a surplus because of efficient operations cannot distribute those savings to its managers and employees as incentives, but rather is likely to see its budget cut for the next year on the grounds that it was allocated too much in the first place. This explains the rush to push money out the door at the end of the fiscal year whether the spending is needed or not, and why bureaucracies are so often inefficient.

Second, public agencies are generally not permitted to reallocate factors of production like private companies. Bureaucrats are frequently subject either to civil service rules protecting them, or else backed by powerful unions that oppose firings. Those civil service protections exist, moreover, to prevent the bureaucracy from being used as a source of political patronage as it was during the 19th century.

Finally, and perhaps most important, public agencies must follow goals that are not of their own choosing. Private companies have a single bottom line which is maximization of shareholder returns. Public agencies have multiple mandates that are both confusing and often mutually contradictory. For example, public procurement agencies are expected to buy goods and services based on optimal price and performance, but they are also subject to mandates requiring bids from small-, minority-owned-, and other kinds of businesses, with an almost endless right of recusal that is ultimately driven by Congress. No private procurement officer has to operate under rules like the Federal Acquisitions Regulations. Publicly-owned utilities like the Post Office or Amtrak are expected to recover costs, but are also required to provide universal service, or to serve rural communities–again because someone in Congress demands it. Amtrak could become a very profitable railway if it were allowed to focus its operations just on the Boston-New York-Washington corridor.

Jim Wilson was known as a conservative who cherished limited government in the American tradition. But he also understood that government was needed for all sorts of functions, and that it could do its job better or worse depending on how it was organized and led. Bureaucracy begins with three cases–the German Army at the beginning of World War II, the Texas prison system, and inner-city Atlanta public schools, in which these very different public agencies achieved dramatically better results as a result of the right leadership and approaches to organization. Wilson understood the critical importance of organizational culture as the source of good bureaucratic performance, as opposed to the shifting around of boxes on an org chart that often passes for reform (e.g., the two big reforms of the 2000s, the creation of the Department of Homeland Security and the reorganization of the intelligence community).

This was all underscored by Jim’s student John DiIulio in a 1994 article entitled “Principled Agents,” which attacked the current tendency to understand organizations using economic models which assume that bureaucrats were self-interested rational agents. DiIulio’s article begins with a description of a prison riot which led a group of retired federal prison officials to jump on airplanes at their own expense to help manage the crisis. This kind of behavior is not the norm in most bureaucracies, but is characteristic of the best ones, and a standard that can be achieved if the bureaucracy builds a strong sense of common identity, and public officials understand that they are serving larger moral purposes.

There are so many other works of Jim Wilson that have enriched our understanding of politics and American government, and it seems almost belittling to pick out only one. He was a man of broad interests, who loved fast cars, scuba diving, and the California coastline. So it is thus with great sadness that we have to mark his passing this week.

I’ve promised to write about the surveillance drone that I’ve been building over the past couple of months. I have always wanted to have my own drone that could send back a live video feed. This is partly inspired by products like the AeroVironment RQ-11 Raven, which is currently in use by the US military, and which you can view in action here. The Raven is basically just a glorified RC airplane, with a sophisticated landing system that allows it to be recovered by a soldier without great pilot skills (which is one reason they cost around $35,000 each).

To get to the bottom line, my drone has taken its first flights, the results of which you can see in a video of my office at Stanford and in a local park.

When my kids were younger I looked into buying an RC helicopter for this purpose and actually tried to wire a camera on a car, but the consumer technology wasn’t up to snuff back then. Now it is.

Instead of using an RC airplane I went with a helicopter for a couple of reasons. I could test the helicopter in my back yard, while an airplane would require a runway. Helicopters are better for precise, close-in surveillance because they can hover. The big drawback is that they are very hard to fly; indeed, learning to fly an RC helicopter is the single biggest impediment to the use of this kind of drone.  (Among other reasons, they’re hard to fly because left and right switch meanings on the joystick when the helicopter is pointing toward you.)

I slowly worked my way up the hierarchy of helicopters, from a Syma S107 to a Blade CX2 to a T-Rex 450 (pictured above) to a DJI Innovations F450 quadcopter (below), which is the platform I used for the videos. It is very difficult to learn to fly a helicopter without a simulator; I’ve logged quite a number of hours on RealFlight 6. While many people use the single main-rotor T-Rexs or their clones as platforms for cameras, they are not very stable; their main purpose is acrobatic flying (they can be flown upside down, among other things, because of the main rotor collective pitch). The practical implication of this is that you have to constantly repair your crashed helicopters, which costs lots of money and takes lots of time.

The DJI F450 quadcopter on the other hand uses an extremely sophisticated Naza controller with a three-axis accelerometer that can sense where the machine is going, and automatically corrects for movement. As a result, they are stable and easy to fly. You still need basic piloting skills, and it is often hard, as with many quadcopters, to see where the nose is pointing, making it hard to maneuver properly when at the limit of visual range. But they hover beautifully and have the lifting capacity to loft a small digital recorder which I used for the videos above. I’m currently using an old Sony flip video camera.

It is extremely easy to build a drone now that can do not just surveillance but can carry rather large payloads.  If you want to see how large some of these planes get, check out this video of a model Airbus A380.  I don’t have to spell out the implications of this.  I want to have my drone before the government makes them illegal.  The US has been fighting such low-tech enemies lately that we haven’t thought through the nature of a world in which lots of people have sophisticated drones, not just other countries but private individuals.  One somewhat worrying thing is that virtually all of this equipment comes from China or Taiwan.

The next stage in this project is to equip the drone with telemetry. I’ve bought the package that includes a real time video transmitter and receiver, camera, and telemetry system that will send back GPS data on the drone’s location, heading, airspeed, etc. This requires, among other things, a ham radio license.  Stay tuned.

I have, to put it mildly, been somewhat astonished at the heated reaction that my blog post “Do Institutions Matter?” has provoked, culminating in a letter from the Hungarian State Secretary for Communication, Zoltán Kovács, to The American Interest complaining about my piece and contesting various points in it. I’m now one of the few Americans to have a web site in Hungarian devoted to my mistakes! In many ways, the vehemence of the response and the extremely uncivil comments that Hungarians have made about each other is a disturbing confirmation that something has gone badly off track with Hungarian democracy.

Let me begin by responding to the criticisms of Mr. Kovács and other commenters about my misunderstanding of the new Hungarian constitution. I readily admit that I made some factual errors, for example that term limits apply to ordinary and not Constitutional Court judges, and for misspelling Fidesz. I’m sorry for this and will be more careful in future fact-checking (this was after all just an unedited blog post).

However, it seems to me that Minister Kovács and the others completely missed the point of my article. Its bottom line was to say that, on paper, the new constitution doesn’t look that bad. As I noted in the post, a classic British Westminster system centralizes far more power in a prime minister and the majority party in Parliament than does the new Hungarian basic law. The problem, I suggested, was not in the formal allocation of powers, but rather in the way that the Orbán government was using those powers. The threat to democracy in Hungary is thus not new institutions per se, but an old political culture that is re-emerging.

(Even on the narrow point about Hungarian judges, my critics missed the more important underlying issue that the Fidesz government has been busy weakening judicial independence through its creation of a National Judicial Office controlled by parliament and hence answerable to the party. The attempt to charge a former Prime Minister with the crime of economic mismanagement is more worthy of Yanukovich’s Ukraine than Hungary. A true rule of law demands much more institutional autonomy for the judiciary that this.)

The Orbán government has undertaken a number of measures that suggest that it doesn’t really understand the norms that must underlie a healthy liberal democracy. Using its supermajority in the Diet, it has enacted not just the new Constitution but a flurry of new laws, almost all of which centralize power in its own hands. Affected institutions include the National Bank of Hungary, controlled now by a Monetary Council largely in turn loyal to  Fidesz–what’s gotten the IMF upset. The supervisory powers of formerly independent watchdogs like the Budgetary Council, parliamentary ombudsmen, and the Health Insurance Inspectorate have all been either eliminated or reduced. Local governments have lost powers to the center with regard to education, health, and disaster preparedness, while control over gymnasia in the capital has been put under a new government regulator and the autonomy of universities curtailed. The autonomy of the Hungarian Academy of Sciences, as well as public foundations in the arts and sciences, is being more closely controlled by the government.

Any one of these measures in themselves might be justified had they been the product of a prolonged and open public debate. But this huge mass of new legislation was passed hurriedly to take advantage of the government’s present supermajority, giving potential critics no time to even digest the content of the laws. In many cases, Fidesz, which won only a bare majority of the popular vote (52.7%), has embedded its own policy preferences in ways that will be very hard to undo should it lose power in the future.  A liberal democracy is not just about majority rule; its proper functioning rests (as it has in England) on the respect that majorities show towards minorities, and the ability of the society as a whole to engage in informed deliberation. (Not, by the way, something that’s in very good shape here in the US at the moment.)

I said in my earlier post that the Orbán government displays an “authoritarian thin skin” and this is something that I would doubly underline. Perhaps the most disturbing thing happening in Hungary is the centralization of power in a government-controlled Media Authority, and its intimidation of opposition media. Taking away the frequency of an opposition-aligned radio station is something right out of Hugo Chavez’s playbook.

A number of other commentators on my earlier post questioned the premise of my piece, “Do Institutions Matter?” Let me say clearly, of course they matter. Modern political order and economic prosperity rest on good institutions, which was the subject of my recent The Origins of Political Order (which, incidentally, has a whole chapter on medieval Hungary and will be published in Hungarian in the near future). The point I was trying to make is that sometimes the exact specification of democratic institutions matter much less than the informal modes of behavior of political actors. Systems like the British one with few checks and balances can nonetheless be run moderately, while others with lots of checks can behave at best decisively and at worst tyrannically. The kinds of institutional changes being made in Hungary now will matter, but what matters much more is the way that the government is using its present powers. Confiscating private pensions may be legally within the authority of many governments, but it is very unwise policy.

I first visited Budapest May 1989 while accompanying then Secretary of State James Baker while working for the US State Department as a young political appointee in the Bush administration. I remember this as a moment of incredible excitement, as Poland and Hungary seemed poised to leave the Communist camp and become genuine democracies. The last time I visited Budapest, Ferenc Gyurcsány was prime minister and I could well appreciate the degree of disgust that people felt toward the government at the time. The degree of political polarization anger evident then was very disturbing, and I could see why people wanted a change. But governments do not rebuild trust by acting the way that Orbán and Fidesz have.

Hungary has always represented two very positive things to me: first, a small country that has produced a disproportionate number of great physicists, mathematicians, composers, artists, writers, and intellectuals. Second, it was a model for the former Communist world in its rapid transition to democracy and a market economy. Indeed, in 1222, seven years after the English Magna Carta, the Hungarian King Andrew II was forced by his nobles to accept the Golden Bull (Aranybulla), which was one of first examples of constitutional limits being placed on the powers of a European monarch. (You can read more about this in my new book.) So it would be both a surprise and a very great shame if Hungary were to take the lead once again, but in the wrong direction toward the incremental dismantling of democracy and constitutional government.

My earlier blog post on Hungary’s new constitution has elicited a large and often angry response from some Hungarians, and now the Hungarian State Secretary for Communication, Zoltán Kovács, has written a critical letter to The American Interest that you can read here.   I will be responding to all of this in a few days, so stay tuned…

I’m beginning a new project at Stanford/CDDRL called “The Governance Project.” The intention is to focus on conceptualizing and measuring governance, and applying those measures to two specific countries, China and the United States.

The beginning point of the project is definition of governance that excludes the degree to which governments are either democratic or subject to a rule of law that constrains the executive. The reason for this is simple: it seems obvious to me that countries can be better or worse governed regardless of whether they are liberal democracies or not. Singapore is not Zimbabwe, despite the fact that neither is democratic. Separating the quality of the state from either the rule of law or democratic accountability is one of the foundational ideas in The Origins of Political Order.

The reason I want to make this separation is to then be able to empirically evaluate the relationship of governance to democracy and the rule of law. We Americans tend to believe that democracy is an intrinsic part of good governance and that more democracy means better quality government. This view is also embedded in many efforts by the World Bank, DfID, and USAID to promote good governance through the promotion of greater transparency, grass roots organization, and democratic accountability in general. Many existing governance measures like the World Bank Institute’s Worldwide Governance Indicators include “Voice and Accountability” (a.k.a. democracy) and Rule of Law; the union of the two is implicit in terms like “democratic governance.”

However, this postulated relationship remains just a theory that remains subject to more empirical testing. One can think of many ways in which greater democratic participation actually weakens the quality of governance. One case happened in the United States when Andrew Jackson was elected president in 1828, as a result of the broadening of the franchise in many states in that period. Jackson argued (1) that since his party won the election, he should get to appoint federal officials; and (2) that there was no job in the US government that was so difficult that any ordinary American couldn’t do it. This was the beginning of the patronage system in the US, in which the federal bureaucracy was controlled by the two political parties and in which jobs turned over with every election cycle. This began to end when President Garfield was assassinated by a frustrated office-seeker, which led to the Pendleton Act in 1883 and the establishment of the first US Civil Service Commission. For the first time, bureaucratic appointments began to rely on examinations and professional credentials, something the Chinese had come up with more than 2100 years earlier. Many of the Progressive Era reforms aimed at stopping patronage and clientelism on both a federal and municipal level involved shielding civil servants from political influence, and hence less democracy.

I have no doubt that more democratic accountability will improve governance in many poor countries in Africa, Latin America, and the Middle East. East Asia is different, however, insofar as it has a very long tradition of relatively high-quality centralized bureaucratic government. This begins with Shang Yang’s reforms in the early state of Qin, and continues to this day in the extraordinary record of the Chinese Communist Party in bringing China through one of the most complex economic transformations one can imagine. Many East Asians today wonder whether rapid democratization will in fact help or hurt the quality of governance there.  What they don’t have is either democratic accountability or rule of law.

Conversely, I would argue that the quality of governance in the US tends to be low precisely because of a continuing tradition of Jacksonian populism. Americans with their democratic roots generally do not trust elite bureaucrats to the extent that the French, Germans, British, or Japanese have in years past. This distrust leads to micromanagement by Congress through proliferating rules and complex, self-contradictory legislative mandates which make poor quality governance a self-fulfilling prophecy. The US is thus caught in a low-level equilibrium trap, in which a hobbled bureaucracy validates everyone’s view that the government can’t do anything competently. The origins of this, as Martin Shefter pointed out many years ago, is due to the fact that democracy preceded bureaucratic consolidation in contrast to European democracies that arose out of aristocratic regimes.

This is not to say that I think the quality of governance is better in China than in the US. The Chinese government’s lack of accountability allows it to make fast decisions and put massive infrastructure projects in place rapidly.  But more often than not this comes at the expense of the rights of ordinary citizens who are in response reacting with often violent social protest. The high-speed rail accident last summer, which the authorities tried immediately to cover up, reveals a very dysfunctional Railway Ministry that even the CCP has not been able to control.  China’s apparently good record today contains many time bombs that will go off in the future.  I believe that, down the road, China will have to permit downwards accountability–a.k.a. democracy–precisely if it is to maintain good governance.  But in the short run, the relationship is not so clear.

I discuss these issues at greater length in a paper on authoritarian government in Asia for the Journal of Democracy, and a draft discussion paper on how to define and measure governance for The Governance Project.