Skip to content

Solving the mystery of the benevolent autocrat


Step 1: You’re right, almost all of the biggest growth success stories are autocracies!

Step 2: Wait a second, all of the worst growth failures are also autocracies!

Step 3: Solving mystery: autocracies have much higher variance of growth rates, so they have both best and worst growth rates

Scatter plot for per capita growth (vertical axis) and democracy (horizontal axis), 1960-2008

Wonky Moral of the story: focusing on success only in Step 1 created a selection bias that led to the erroneous conclusion that autocracy was good for growth.

Plain English Moral of the Story: autocracy is extremely risky: it could result in high growth, but it could just as well result in a growth collapse — for every Lee Kuan Yew, there is a Jean-Bédel Bokassa.

Extra credit question: why would arguing that the autocrats under Step 1 are benevolent while Step 2 autocrats are malevolent be logically fallacious?

To answer some questions, the growth rate is the geometric average 1960-2008 of per capita growth per annum. The source is WDI.

The source for the democracy data is Polity IV, which has some problems, but is enough for the kind of illustration here.

The point about causality is well taken, I am just making a point about how what for these data is actually a POSITIVE and SIGNIFICANT correlation between democracy and growth is turned into an apparent NEGATIVE association in Step 1, which is where the “benevolent autocrat” discussion usually stops.

(FOR WONKS ONLY) When I write this up more fully in an eventual paper, I will explain also some exploration of different functional forms for transforming the original POLITY index from -10 to 10, which is an arbitrary scale (autocracy being the negative direction). To illustrate the strongest possible POSITIVE correlation, I chose from 3 alternatives the one with the strongest statistical significance , which was the following function: POLITY/(11-POLITY). I would normally NOT like this kind of “data mining” among several different functions, but again the point of the exercise is to show the fallacy by which a STRONG POSITIVE association appears to be a STRONG NEGATIVE ASSOCIATION.

This entry was posted in Academic research, Democracy and freedom and tagged , . Bookmark the permalink. Follow any comments here with the RSS feed for this post. Both comments and trackbacks are currently closed.


  1. przemek wrote:

    Arguing that autocrats in fastest-growing economies are benevolent while those in fastest-shrinking economies are malevolent is, without additional data, circular: economies under benevolent dictatorship grow fast, and the reason we know that a dictator is benevolent is because their economy grows fast.

    What is the democracy measure that you are using?

    Posted September 28, 2010 at 10:44 am | Permalink
  2. Hiromi.S wrote:

    Where did you get the “democracy measure” from? Freedom house? If not, where else?

    Posted September 28, 2010 at 10:47 am | Permalink
  3. Wafa wrote:

    I am truly surprised at the United Arab Emirates. Declining growth there does not match up with my personal experience of the place, which is that of a kind of Wild West boomtown in the Middle East. Is it just the global financial crisis causing this?

    Posted September 28, 2010 at 10:56 am | Permalink
  4. Jim wrote:

    Are those annual growth rates? Because something looks completely out of whack with your Kuwait data.

    Apart from that, one could argue that the first group are generally more stable and less war-torn than the latter. It’s not that surprising that war torn countries are more likely to have both negative growth and autocracy, so there is lots of potential for endogeneity and confounding here. The basic point about always checking for variance is a good one, however.

    Posted September 28, 2010 at 11:31 am | Permalink
  5. Matt wrote:

    Yikes – maybe most countries make a slow transition from autocracy to democracy at the same time their growth converges to the long term (read: lower) rate. We can’t get too far in life with scatterplots, can we?

    Plus: Liberia and Zambia = autocracies… really?

    Posted September 28, 2010 at 11:31 am | Permalink
  6. Orlando R wrote:

    Bill, I must be losing it. I can’t answer your extra-credit question. It may be fallacious because you don’t have data on “benevolence,” but it can’t be logically wrong.

    Posted September 28, 2010 at 12:17 pm | Permalink
  7. VRE wrote:

    Is growth the arithmetic or geometric mean for 1960-2008?

    Posted September 28, 2010 at 12:20 pm | Permalink
  8. Stephen Jones wrote:

    The Emirates and Kuwaiti per capita figures are explained by the vast number of low paid Asian laborers who bring the figure way down. The Emirati or Kuwaiti citizen is much better off.

    Posted September 28, 2010 at 12:39 pm | Permalink
  9. Kemi wrote:

    Correlation is not causation Bill. I wonder whether the countries you lump together as “autocracies” actually have radically different attitudes and legislation towards entrepreneurs, starting up new businesses, tax and so on.

    Political Liberty and Economic Liberty are two different things. Only one, I suspect, is related to growth.

    Posted September 28, 2010 at 2:18 pm | Permalink
  10. Pablo Kuri wrote:

    This conundrum is somewhat explained in Mancur Olson´s ¨Power and Prosperity: outgrowing communist and capitalist dictatorships¨ ( The discussion about autocrats, democracy and profit motives is explained in chapter 1: The Logic of Power.

    This book has great insights into democracies and economic growth. The mere idea that some, not all, autocrats profit from a better performing economy comes from the logic of a provision of welfare enhancing goods and services which maximizes the autocrats as well as the common persons life. This explians where the ¨benevolence¨ comes from.

    Posted September 28, 2010 at 3:02 pm | Permalink
  11. edawg wrote:

    One side point that often gets lost in this kind of analysis is that there’s a fallacy in thinking of democracy as merely instrumental (or not) to economic grown rather than as something with intrinsic value, a development goal in and of itself.

    If you frame the questions as, “what are the conditions that lead to economic growth and do they include Democracy?”, you arrive at a different set of conclusions and implications than if you ask the question, “is democracy a fundamental piece of this nebulous goal of ‘development’?”

    Posted September 28, 2010 at 4:46 pm | Permalink
  12. Fernando wrote:

    Completely agree with a previous commenter.

    This debate – though important – subtracts from democracy’s more intrinsic value, like freedom from arbitrary arrest, due process, rule of law, checks and balances, etc…

    People do not die fighting dictators so that they can have a more stable growth rate. Or at least that is not the impression I get.

    PS I would also look at consumption growth as well as income.

    Posted September 28, 2010 at 5:24 pm | Permalink
  13. Tobi wrote:

    According to Paul Collier:

    “In a celebrated study, economists Benjamin Jones and Benjamin Olken looked at whether the death of a country’s leader altered economic growth. It did, sometimes for better and sometimes for worse. Recently, an Oxford colleague, Anke Hoeffler, and I sifted through their results again, distinguishing this time between democrats and autocrats. We found that in democracies, changing the leader does not change growth — all leaders are disciplined to perform tolerably. But in autocracies, the growth rates are as unpredictably varied as the leaders’ personalities. Here lies the difference between good leaders and great ones: Good leaders put right the policy catastrophes of bad leaders; great leaders, like the men who shaped the U.S. Constitution, build the democratic checks and balances that make good leaders redundant.”


    I think he has a good point on the differences between democracies and autocracies.

    Posted September 28, 2010 at 5:48 pm | Permalink
  14. David Youngberg wrote:

    This looks a lot like the old claim that smaller classes lead to the best schools when it was really the smaller classes increase the variance of success. The diagram above looks exactly the same as the one for class size and test scores.

    Posted September 28, 2010 at 10:03 pm | Permalink
  15. Omar wrote:

    More is needed to support the variance argument– there are clearly many more countries with a democracy_measure near zero than near ten…so, even if the two populations (near zero, near ten) have the same variance, the largest/smallest values are most likely to show up in the larger population (here, near zero).

    Posted September 28, 2010 at 11:29 pm | Permalink
  16. David Zetland wrote:

    From my experience, the Kew:Bokassa ratio is more like 1:5.

    You’re paper sounds good, but I hope that it doesn’t result in another rash of macro-development papers. Too much alchemy.

    Posted September 29, 2010 at 12:22 am | Permalink
  17. Jim wrote:

    The basic methodological point about comparing groups of different size and variance is sound, but Prof Easterly seems to be making a wider argument that there is no significant pattern at all distinguishing ‘succesful’ from ‘unsuccesful’ autocracies. But if you look at the high growth autocracies they look much more likely to be temperate, peaceful and with low levels of ethno-linguistic fractionalization, while with the other group it’s the other way around. There are a couple of exceptions in each group but there does still appear to be a ‘pattern’. Hopefully the forthcoming paper will properly control for these other variables to try and isolate any real effect of autocracy in countries that really are similar in terms of security and physical and human geography.

    Posted September 29, 2010 at 3:19 am | Permalink
  18. Jezebella wrote:

    Really, no matter the model of governance, the most important is who’s making the decisions. A country can be ruled by the worst jerk in the world but he can also happen to be the most effective governor, making his country and people rich. A country can be ruled by the nicest guy in the world but he may also happen to be disastrous when it comes to governing.

    Posted September 29, 2010 at 4:25 am | Permalink
  19. Geoff wrote:

    …and isn’t Botswana one of the freeest democracies in Africa?

    Posted September 29, 2010 at 4:33 am | Permalink
  20. Rune wrote:

    Not much news here. Rodrik showed a decade ago that the variance of growth rates across democracies is signifcantly smaller than the variance across autocracies. The immediate explanation is also well known – for every XX, there’s a YY – as is the most likely mechanism (although Collier might be the first to test it). There’s a paper in Economics & Politics, also about a decade ago, analysing political regimes and inequality that has the fact that democracies have institutionalised mechanisms for choosing leaders as its starting point. What’s lacking in the discussion here is the fact that Bill’s extra credit question only makes sense if his data points are treated as random draws from two different populations of countries. But the leaders of the East Asian countries were, if not exactly benevolent, inclined to maximise the growth rates of their countries for various specific historical reasons (such as having threatening communist neighbours). So the most interesting issue here – from an intellectual point of view, if not policy – is why some autocracies becomes bad and others good for growth.

    Posted September 29, 2010 at 5:14 am | Permalink
  21. Metamorf wrote:

    It would be interesting to know just how the Polity measurements were assigned, and how they relate to the notion of individual freedom. E.g., “The Polity data include information only on the institutions of the central government and on political groups acting, or reacting, within the scope of that authority.” Could an individual be more free, for example, under certain autocracies than under certain democracies? And might such freedom correlate better with economic growth?

    Posted September 29, 2010 at 11:42 am | Permalink
  22. J P wrote:

    How are you computing the democracy measure? I find it hard to believe that South Korea and Singapore have the same level of democracy as the UAE and the Dem. Rep. of Congo (formerly Zaire).

    Posted September 30, 2010 at 6:37 pm | Permalink

10 Trackbacks

  1. […] Solving the mystery of the benevolent autocrat […]

  2. […] via Aid Watch | Solving the mystery of the benevolent autocrat. […]

  3. […] This post was mentioned on Twitter by William Easterly, William Easterly, William Easterly, Catherine Rampell, Naadir Jeewa and others. Naadir Jeewa said: Reading: Solving the mystery of the benevolent autocrat: Step 1: You’re right, almost all of the biggest growth su… […]

  4. […] Easterly solves the mystery of the benevolent autocrat: Both the best economic growth stories and the worst ones are in autocracies. Maybe there’s […]

  5. […] of a recent post, Bill Easterly points out that while non-democratic regimes have some of the world’s fastest growth rates, […]

  6. […] back in the mainstream, William Easterly has a good post on the economic consequences of dictatorship and democracy. While Lane Kenworthy reviews the Spirit Level, and Henry Farrell provides possibly the best ever […]

  7. […] Does autocracy lead to higher growth? Yes, and also to lower growth – autocracies have much higher variance of growth rates, so they have both the best and the worst growth rates […]

  8. […] previous post on the myth of the benevolent autocrat. (To be fair to the New Yorker writer,  he mentions briefly at the very end of the article Dani […]

  9. By A Maior Janela Partida do Mundo « O Insurgente on October 10, 2010 at 9:45 am

    […] este tema vale a pena ler estes dois posts do Bill […]

  10. […] Hacen los dictadores buenos crecer la renta per capita. […]

  • About Aid Watch

    The Aid Watch blog is a project of New York University's Development Research Institute (DRI). This blog is principally written by William Easterly, author of "The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics" and "The White Man's Burden: Why the West's Efforts to Aid the Rest Have Done So Much Ill and So Little Good," and Professor of Economics at NYU. It is co-written by Laura Freschi and by occasional guest bloggers. Our work is based on the idea that more aid will reach the poor the more people are watching aid.

    "Conscience is the inner voice that warns us somebody may be looking." - H.L. Mencken

  • Archives