Skip to content

Does growth reflect good and bad dictators, or just good and bad statisticians?

Autocracy (measured on horizontal axis as near zero) and Growth Outcomes (measured on vertical axis), 1960-2008

As a previous post showed, autocracies have high variance of growth outcomes (also illustrated in the graph above). The usual interpretation is that benevolent autocrats cause good outcomes while malevolent autocrats cause bad growth outcomes.  Democracy has checks and balances that prevents malevolent people from having too much power to generate bad outcomes, but also restrains the good ones from doing what they want to achieve the great outcomes.

Unless this is completely wrong. Autocracy is only one dimension of society, after all, and is heavily correlated with other dimensions that could cause high dispersion of development outcomes, such as dependence on commodity exports, dependence on agriculture, civil wars, and … BAD STATISTICIANS (?!)

Vertical axis is per capita growth 1960-2008, Horizontal axis is grade on quality of data (Highest grade=4)

Bad statisticians make a lot of measurement mistakes. Average growth over 1960-2008 might have zero mistake ON AVERAGE, but there will randomly be some countries with a string of exaggerated growth rates. Other countries will randomly have a string of underestimated growth rates. So the variance of growth will be higher the worse the data quality — which is exactly what we see in the picture.

Of course, I am not saying China or Singapore or Taiwan have high growth (and Liberia has horrible growth) ONLY because of measurement error. Other indicators confirm the East Asian booms — but are we really sure growth was 6 percent per capita per year, instead of 4 percent per capita per year?

How bad is bad quality data? Alwyn Young at LSE has a fascinating recent paper in which he points out:

although the on-line United Nations National Accounts database provides GDP data in …constant prices for 47 sub-Saharan countries for each year from 1991 to 2004, the UN statistical office which publishes these figures had, as of mid-2006, actually only received data for just under half of these 1410 observations and had, in fact, received no constant price data, whatsoever, on any year for 15 of the countries for which the complete 1991-2004 on-line time series are published.

So for 15 African countries, “bad quality data on real GDP growth” really means “NO data on real GDP growth”.

Next time you are praising an autocrat for a glorious growth record, remember you may really just be praising an incompetent statistician.

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


  1. Norman wrote:

    This criticism cuts both ways, of course. The next time you argue that low growth is a feature of autocracies, remember you may really just be condemning an autocrat’s statisticians.

    Posted March 31, 2011 at 1:45 am | Permalink
  2. geckonomist wrote:

    For a poor country:
    It’s always a mystery how GDP numbers are calculated.
    Given the rampant smuggling, under-declaring of imports, the huge informal sectors and an agricultural sector making up typically >50% of the population, going totally unrecorded.

    The margin of error must be very large from the start.

    A further level of error, is how the “capita” number is chosen. More art than science.

    Then your mates at those international organisations, add another level of error: a dose of “purchasing power parity” calculus.

    By now, the margin of error must be more meaningful than any number one arrives at.

    I prefer to look at the growth of mercedes-benz/bmw/lexus dealerships in the country.

    Posted March 31, 2011 at 4:12 am | Permalink
  3. Tord Steiro wrote:

    Uh-oh. Perhaps this explains the combination of excellent statistics and revolution in Tunisia and Egypt?

    Posted March 31, 2011 at 4:15 am | Permalink
  4. Here’s an honest question: Just how much do people look at GDP as a true indicator? Is it given the heaviest weight in policy decisions? Given all that we know about how inaccurate it can be, does it really make sense to use it at all? And also, doesn’t relying on the GDP favor only one kind of growth? Aren’t we more likely to find “meaning” in the informal sector? I guess that was multiple questions. Still honest, though….if a bit biased.

    Posted March 31, 2011 at 4:23 am | Permalink
  5. P Baker wrote:

    Hang on, the data suggests that Taiwan is more autocratic than Venezuela? France more that Trinidad or Jamaica? Zut alors!!

    Posted March 31, 2011 at 7:05 am | Permalink
  6. BPeterson wrote:

    It’s worth noting that even measures of “democracy” like those used here are pretty much junk, just like much GDP data. I’m curious, though–where did the measure for “data grade” come from? I haven’t seen that before. This is not from the Alwyn Young paper, right? (I didn’t see it in there but have not read it all yet).

    Posted March 31, 2011 at 8:36 am | Permalink
  7. joe wrote:

    Nope – statisticians are the good guys. They don’t usually collect or interpret the data, they just try to make the best of what they have.

    The really bad stats are done by the scientists (and more regularly, economists) who haven’t the faintest idea of what they are doing, come up with some dodgy crap and then point in the other direction when anyone dares suggest that their analysis leaves a lot to be desired.

    But then I would say that – my wife is an autocratic statistician.

    Posted March 31, 2011 at 9:02 am | Permalink
  8. P Baker wrote:

    @joe: “statisticians are the good guys. They don’t usually collect or interpret the data, they just try to make the best of what they have.”
    Well ideally yes, but some statisticians I’ve known don’t listen to what you are telling them and try to fit the data to their latest pet technique; ‘autocratic’ is the mot juste.

    Posted March 31, 2011 at 9:25 am | Permalink
  9. Sometimes statisticians miss the point. Autocracies can have strong growth rates, and yes they can be exaggerated.
    But the people living in those countries don’t always have the freedom to improve their lives, which are presumed to come with higher growth (GDP) rates.

    Posted March 31, 2011 at 9:27 am | Permalink
  10. joe wrote:

    @P Baker – I’m sorry, I was being silly, of course statisticians can make mistakes like anyone else.

    I’d still be interested in how many of the mistakes highlighted above were made by people with degree level qualifications in statistics rather than economics or science. Surely it is possible that there are errors in analyses which were not made by actual statisticians?

    Posted March 31, 2011 at 9:50 am | Permalink
  11. P Baker wrote:

    @ joe: “Surely it is possible that there are errors in analyses which were not made by actual statisticians?”
    Agreed, it’s only really economists that always seem to get it wrong. What a mess is economic theory, anyone read Steve Keen’s book (Debunking Economics)? Quite incredible.

    Posted March 31, 2011 at 10:58 am | Permalink
  12. Vivek Nemana wrote:

    Does the democracy measure come from Freedom House ratings?

    Posted March 31, 2011 at 1:29 pm | Permalink
  13. BPeterson wrote:

    @Vivek, I’m guessing it is Polity. Polity has two scores: democracy score and autocracy score, which they combine for their composite score on a -10 to 10 scale. If you just consider the democracy score, it goes from 0 to 10, and I suspect that is what is being used here, since Freedom House is on a 7 or 14 point scale depending on whether you include their Civil Liberties measure to make it a composite index. I doubt Prof. Easterly went through the hassle of adjusting FH’s 7 point scale to some sort of 10 point scale just for fun, so my money is on part of the Polity IV series.

    Posted March 31, 2011 at 3:28 pm | Permalink
  14. Quicksilversurfer wrote:

    A crucial point is being made here that does not get enough attention despite quite a lot of evidence that much of the data may be very reliable. A few examples from my experience:
    1/ A few years ago, data published by some international agencies for some countries where I had traveled extensively indicated access to potable water at 98%: since I had seen through my travels where people got their water, I could only assume that this figure was based on a tautology i.e, People need water to live, the population is not dying “en masse” ergo almost everybody has access to potable water.
    2/ Even these alleged paragons of reliable data at the IMF in my time working with them spent their time in mission correcting the decimal on the estimated growth rate of the country despite the fact that a visit to the national unit in charge of national accounts had shown that their “methods”were completely deficient and that their revised fi8gures always ended up reflecting the IMF’s i.e. GIGO but in the IMF model there was no inconsistency and what had been intended to be done could be justified, damn the reality!
    3/ It reminds ma also of a time in the early 1990s when talking to a group of heads of local international agencies offices in Central America and discussing the inconsistencies between published figures and some evidence on the ground, some of them smiled and admitted that in a neighboring country the estimate of the growth rate was the result of a negotiation with the Minister of Finance.
    More than anywhere else in the development field regarding data, a healthy dose of skepticism is needed.

    Posted March 31, 2011 at 4:09 pm | Permalink
  15. William Easterly wrote:

    Sorry for the lack of documentation. The democracy measure is a transformation of the Polity index from -10 to 10 into the space of 0 to 10, averaged over 1960-2008. The per capita growth rate is averaged over the same period. The data quality grade is from the Summers and Heston Penn World Tables project.

    Posted March 31, 2011 at 6:40 pm | Permalink
  16. James Michael wrote:

    Statistics has always been an annoyance in my eyes in terms of judging a nation in any way (development, economy, freedom, etc.). I can agree with the statement that praising a benevolent autocrat may be the same as praising an incompetent statician. Just how the scenario applies with autocracy and growth, I believe the same can be said in this way: “The next time you bash bad development, don’t base it off a statician.”

    Posted April 1, 2011 at 1:06 am | Permalink
  17. Matt wrote:

    I have to say that this argument can go both ways. You can’t blame just the autocrat or the statistician. Although the comment above about how there was no recorded data for the 15 African countries leads me to think that statisticians will create data to keep a certain reputation. I don’t think creating data on the basis of other countries justifies you make a “stab in the dark”. On the other side the autocrat can’t be completely blamed for poor GDP growth. You can’t just base how an autocrat’s outcomes on GDP alone you need to look at other factors. So I feel in order to make an accurate judgement about these countries you should look at all aspects and make sure the statistics are completely viable and not make up.

    Posted April 1, 2011 at 10:19 am | Permalink
  18. Brandon J wrote:

    This type of omission should be unacceptable, because statistics like this factor into how much aid these countries receive. Not only that but it affects how the world’s perception of non-democratic governments. This can easily be solved by statisticians and economist making sure that they only present the data that is actually there and by not making up data to fill in for countries they have no research on. Also relying on just the UN’s report is another mistake, just like with any other research project/report one should evaluate more than one source before making any type of assesment of how a particular type of government affects a country’s growth.

    Posted April 1, 2011 at 1:31 pm | Permalink
  19. Gebre wrote:

    The big problem is when such indicators become synonym with wellbeing. In some countries, growth in GDP is directly related with increase in poverty and misery for a large portion of the population. It is scary that almost universally such growth is taken as a sign of good peformance and dictators make it an execuse to stay in power.

    Posted April 7, 2011 at 7:48 am | Permalink

4 Trackbacks

  1. […] reading here: Does growth reflect good and bad dictators, or just good and bad… AKPC_IDS += […]

  2. By Overcoming Bias : Are Dictators The Future? on March 31, 2011 at 9:29 am

    […] 30Mar: Easterly notes that this might just be because dictators have bad growth statistics. SHARETHIS.addEntry({ title: […]

  3. […] Link to the original site Posted in Aid SHARE THIS Twitter Facebook Delicious StumbleUpon E-mail « The Aid Contest of the Celebrity Exes No Comments Yet […]

  4. By Randoms of the past several days « Foseti on April 5, 2011 at 8:53 pm

    […] Democracy and growth. […]

  • 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