Democracy and Economic Growth

In their book, Why Nations Fail, Daron Acemoglu and James Robinson argue that poverty is produced by extractive political and economic institutions and that prosperity results from having inclusive institutions. Their book is replete with examples of autocratic governments pursuing policies that impoverish large parts of the population to benefit small elites. While far from being the worst case (and, believe me, there is some stiff competition for that distinction), Uzbekistan is a clear-cut example of what they are talking about.

Indeed, the authors featured Uzbekistan in their first blog post, perhaps because the nation offered the spectacle of a modern government forcing millions of school children into the fields to pick cotton (which is what the Uzbek child in the photo to the left is doing).

In a subsequent blog post, Acemoglu and Robinson give a thumbnail sketch of the situation that produced this forced labor:

Uzbekistan gained its independence when the Soviet Union collapsed in 1991. Islam Karimov, previously first secretary for Uzbekistan of the Soviet Communist Party, declared himself an Uzbek nationalist and became, and since then has remained, president through fraudulent elections and repression.

After independence, farmland that was previously under the control of state-owned firms was distributed to farmers. But they weren’t suddenly free to plant and sell what they wished. The government introduced regulations that determined what they should plant and how much they should sell it for. For cotton, that meant they would receive a tiny fraction of the world market price. For many, it wouldn’t make sense to grow cotton at these prices. But the government dictated that they had to. Before independence, much of the cotton was picked by combine harvesters. Yet given these rewards, farmers stopped investing in or maintaining farm machinery. So coerced child labor was Karimov’s cost-effective method of picking cotton.

Naturally, child labor on this scale could not go unnoticed and an international effort known as the Cotton Campaign was launched after the scope of the problem was documented by a London University School of Oriental and African Studies 2008 report. However, NYT reported that Uzbekistan now drafts people from the public sector to pick the cotton. So, instead of school children picking the cotton, over a million doctors, nurses, school teachers, bureaucrats and small business employees "volunteer" to pick the cotton in order to avoid being arrested or fired.

While it is morally satisfying that younger children are now exempt, the move to using otherwise skilled labor to do unskilled work (which should be mechanized in the first place) is even more economically inefficient than using school children. After all, the school kids weren't producing much in class, so the opportunity costs of sending them to the fields was low (which, of course, is why the government choose them in the first place). Now they are taking some of the more productive workers in their economy away from their jobs to do the work while receiving their normal wages. While more palatable from a PR point of view, this is more economically absurd while remaining almost as morally objectionable.

Though Acemoglu and Robinson haven't addressed this recent change, it is a good example of an argument they make in their book. Namely that when poor or undesirable performance results from extractive political and economic institutions, one cannot engineer a fix without reforming the underlying institutions. That is to say that if the system is set up to transfer wealth and income from one part of the population to another, the system will only find another way to do this when pressured to change a particular policy.

In Uzbekistan, international pressure to stop using child labor forced Karimov's regime to adopt a more costly means of enriching itself. However, that cost almost entirely falls on the population in the form of the shortage of public services resulting from having healthcare workers, teachers and government employees out in the field (not to mention the cost to the people being forced to pick the cotton). If further international pressure is applied to end the forced labor of adults, without any changes in the underlying political and economic institutions, the government will undoubtedly channel its efforts into finding a way to continue its extraction in a manner that evades criticism. Because disregard to the costs to the general economy and the populace is inherent in the institutions, whatever response the government devises will probably just make a different set of powerless people worse off.

However, if examples like Uzbekistan make the case for the link between poor economic performance and extractive institutions, is it true that more inclusive institutions necessarily produce better growth and prosperity. China would appear to be a case of an authoritarian system that has produced both remarkable growth and reductions in poverty growth, and there are certainly many cases of new democracies having economic troubles.

In a recent scholarly paper, Acemoglu, Naidu, Restrepo, and Robinson look at the data on the economic performance of democracies versus non-democracies. Acemoglu and Robinson summarize their findings in a blog post entitled Democracy, What is it good for? They chose this question for the title because much of the previous empirical research has questioned the value of democracy in producing economic growth (and they wanted to riff off the old Edwin Starr song, you know "War, what is it good for?").

Acemoglu and Robinson argue that previous negative findings reflect the complicated relationship between transitions to democracy and economic performance that is revealed when one looks at GDP growth in the years preceding and following a nation's democratization . They find that authoritarian regimes usually democratize as the result of a economic crisis and, therefore, newly democratized nations tend to be dealing with unusually unfavorable economic conditions. This may explain why new democracies have lower GDP growth in comparison to non-democracies  in the first 5 years following democratization.   In the next 5 years (5-10 years after democratization), the performance of newly democratized economies tends to improve, but even then only matches the growth of non-democracies. However, after the 10 year mark, the democratized economies clearly outperform the economies of non-democracies. They use the graph below to illustrate their findings:

As to why democracies are able to achieve these results, though there are many theoretical arguments as to why this would be the case, the authors do not have conclusive empirical evidence to answer this question. They do note that democracies appear to be better at implementing economic reforms, providing education, and providing other public goods.

It should be noted, that while Acemoglu, Naidu, Restrepo, and Robinson find that democracy is good for economic growth, in other research they have found that  it is not so good at reducing income inequality. While it does shift resources away from elites, democracy does not have as much impact on the societies overall income equality as one might expect. However, they admit that the data here is somewhat problematic and they can't determine whether the inequality they observe is the result of market induced inequality (such as one would expect if technological change was increasing the difference in wages between skilled and unskilled workers) or politically induced (such as one would expect if middle classes where capturing the system and directing income towards themselves).