Lies, Damned Lies, and Statistics (33): The Omitted Variable Bias, Ctd.

I discussed the so-called Omitted Variable Bias before on this blog (here and here). So I suppose I can mention this other example: guess what is the correlation, on a country level, between per capita smoking rates and life expectancy rates? High smoking rates equal low life expectancy rates, right? And vice versa?

Actually, and surprisingly, the correlation goes the other way: the higher smoking rates – the more people smoke in a certain country – the longer the citizens of that country live, on average.

Why is that the case? Smoking is unhealthy and should therefore make life shorter, on average. However, people in rich countries smoke more; in poor countries they can’t afford it. And people in rich countries live longer. But they obviously don’t live longer because they smoke more but because of the simple fact they have the good luck to live in a rich country, which tends to be a country with better healthcare and the lot. If they would smoke less they would live even longer.

Why is this important? Not because I’m particularly interested in smoking rates. It’s important because it shows how easily we are fooled by simple correlations, how we imagine what correlations should be like, and how we can’t see beyond the two elements of a correlation when we’re confronted with one that goes against our intuitions. We usually assume that, in a correlation, one element should cause the other. And apart from the common mistake of switching the direction of the causation, we often forget that there can be a third element causing the two elements in the correlation (in this example, the prosperity of a country causing both high smoking rates and high life expectancy), rather than one element in the correlation causing the other.

More posts in this series are here.

Why Do Countries Become/Remain Democracies? Or Don’t? (9): The Resource Curse

If we value democracy, then it’s interesting to know

  • how societies have achieved the transition from authoritarian forms of government to more democratic ones
  • why other societies have failed
  • and how democracies have avoided the opposite transition.

This knowledge will help us to promote and sustain democracy in the future. Something we already know is that this isn’t simple. There are a huge number of factors at play and there’s no silver bullet. Some of the most widely discussed factors are economic development, levels of education, and religion and culture.

I’ll bracket two important issues here: what kind of democracy are we talking about, and how do we measure transition or development towards democracy? If you want to know what promotes or inhibits democracy and act on this knowledge in order to further the cause of democracy, you can’t avoid these questions, but discussing them here would take us too far.

What I want to focus on here is the so-called resource curse. This curse is believed to be a phenomenon that blocks countries’ development towards democracy. Promoting democracy means lifting the curse. Now, what is this curse, and is it real or just another simplistic explanation of the course of history?

Countries which own lots of natural resources such as diamonds, oil or other valuables that are found in the ground, are often relatively poor, badly governed, violent and suffering from gross violations of human rights. Resource wealth can trigger corruption and grabbing, can give autocrats the means to retain power by buying off opposition or building a repressive state apparatus, or can tempt democratically elected leaders to cling to highly beneficial positions of power.

This sounds good but even a cursory glance at reality reveals some counter-indications. There are many resource rich countries that are governed very well and are pinnacles of democracy (take Norway). Still, that may only disprove part of the resource curse. It may be the case that democracies benefit from resources and are able to solidify themselves, while non-democracies are doomed to remain as they are because of resource abundance. Resources then only create a curse when democratic institutions are absent. So we shouldn’t worry about democracies failing because of resources, but about autocracies failing to transform because of them.

However, there’s an article here claiming that

resource wealth is positively associated with both economic growth and institutional quality.

Much depends, it seems, on how to measure resource abundance. There also is a reversal of the direction of causation, a common mistake in statistics:

There is no evidence that resource-dependent countries end up with slow growth and bad institutions. Rather, countries with bad institutions attract little investment, and as a result they grow more slowly and remain dependent on exports of commodities.

Why Do Countries Become/Remain Democracies? Or Don’t? (7): Education and Prosperity

There aren’t many questions in political science that are more important than this one: which are the factors that determine whether a country becomes or doesn’t become a democracy, and that determine the degree to which a country is democratic. There are two reasons why this question is important:

  • Democracy is an important good. Hence it’s important to know what facilitates or hinders the realization of this good.
  • Countries act on this statement in their foreign policy. For example, part of the rationale for invading Iraq was the conviction held by the U.S. administration of the time that promoting democracy in Iraq was both an intrinsic good and in the interest of the U.S.

I gave a short and non-exhaustive list of possible factors promoting/undermining the development/survival of democracy here. In the current post I want to focus on two of them: education levels and income or prosperity levels.

1. Education

This graph compares the Polity IV Democracy Index scores for the countries of the world (average scores during the 1960-2000 period), with the average years of schooling of the adult population in 1960. And there’s obviously a correlation, and the quote below gives an indication about the direction of correlation:

The chart above shows the 77 percent correlation between education levels in 1960 (measured by the average years of schooling in a country as estimated by Robert Barro and Jong-Wha Lee), and the subsequent 40-year average of the Polity IV democracy index. That democracy index runs from zero to 10, where countries with index values less than three don’t look remotely democratic and countries with index values of about seven are reasonably well-functioning democracies.

One way to read the graph is that there are basically no countries with very low levels of education that have managed to be democratic over the long term, and almost every country with a high level of education has remained a stable democracy.

Thomas Jefferson wrote that “if a nation expects to be ignorant and free, in a state of civilization, it expects what never was and never will be.” In 1960, 36 nations had less than 1.74 years of schooling (which happens to be the level that Afghanistan has today). Of those 36 countries, only two — India and Botswana — managed to have average democracy scores above 4.2.

Out of the 19 countries in this sample with more than 5.3 years of schooling (the current level in Iran) in 1960, 17 have average democracy scores above 7.9. Fifteen of these have been perfectly democratic, at least by the standards of Polity IV. Only Poland and Hungary were dictatorships, and one can certainly argue that those places would have been democracies in 1960s if it were not for Soviet troops.

But in the middle ranges of education, between two and five years on average, almost anything goes. Some places, like Costa Rica and Italy, have been extremely democratic, while others, like Kuwait and Paraguay, have not. Iraq falls into this category today, which suggests a fair amount of uncertainty about that country’s political future.

Why do I think that the chain of causality runs from education to democracy rather than the reverse? Democracy in 1960 is essentially uncorrelated with subsequent growth in the levels of education. Education in 1960, on the other hand, does an extremely good job of predicting increases in democracy.

Why is there a connection between human capital and freedom? Giacomo Ponzetto, Andrei Shleifer and I have argued that the connection reflects the ability of educated people to organize and fight collaboratively. Dictators provide strong incentives for the ruling clique; democracies provide more modest benefits for everyone else. For democracy to beat dictatorship, the dispersed population needs to have the skills and motivation to work collaboratively to defeat dictatorial coups and executive aggrandizement.

Education teaches skills, like reading and writing, that enable people to work collaboratively. At younger grades, teachers spend a lot of time teaching children how to get along. In the United States, education is strongly linked to civic engagement and membership in social groups. The ability to work together enables the defense of democracy. Edward L. Glaeser (source)

2. Income

There’s an interesting paper here examining the causal relation between democracy and income. The authors find that

the level of national income provides the most important factor explaining inter-country variations in the degree of democracy with the consequence that low income is the most important barrier to democracy.

They first present the correlation between income and democracy, using not the Polity IV index but the Gastil/Freedom House index.

The authors have two reasons to believe that the causal link goes from income to democracy rather than the other way around:

  • Initial income in 1971 correlates with average democracy scores during the 1972-2005 period. This approach is similar to the one above in the case of education and democracy.
  • And – simultaneously – there doesn’t seem to be a very strong causal link going from democracy to income. Barro has concluded that the degree of democracy is only a minor variable explaining income levels. So there is only a weak causal link going from democracy to income. This means that the strong correlation shown in the graph above must be explained by a causal link going from income to democracy.

Why do higher levels of income promote the development of democracy? I gave an overview of the reasons here but some of the more important ones are:

  • Higher education levels in a population means a higher probability of contestation. Following the Maslow hierarchy of needs it’s natural to expect the appearance of political needs once more basic needs have been secured.
  • More income means more complex production. This in turn means that governments find it harder to impose central control over their economies.

Obviously, income is just one of many factors determining the development of democracy. It’s an important one, but clearly not sufficient. The graph above shows the Muslim countries separately. As you can see, all non-Muslim countries with high income levels are in the “high level of democracy” range. Affluent Muslim countries, however, aren’t. This indicates that affluence in itself promotes but doesn’t determine the development of democracy. Other factors are also in play. Culture and religion are perhaps some of them. It’s often argued that Islam is incompatible with democracy, or at least slows down the development of or transition to democracy. I’ll come back to this controversial topic another time.

Lies, Damned Lies, and Statistics (24): Mistakes in the Direction of Causation

Suppose you find a correlation between two phenomena. And you’re tempted to conclude that there’s a causal relation as well. The problem is that this causal relation – if it exists at all – can go either way. It’s a common mistake – or a case of fraud, as it happens – to choose one direction of causation and forget that the real causal link can go the other way, or both ways at the same time.

An example. We often think that people who play violent video games are more likely to show violent behavior because they are incited by the games to copy the violence in real life. But can it not be that people who are more prone to violence are more fond of violent video games? We choose a direction of causation that fits with our pre-existing beliefs.

Another widely shared belief is that uninformed and uneducated voters will destroy democracy, or at least diminish its value (see here and here). No one seems to ask the question whether it’s not a diminished form of democracy that renders citizens apathetic and uninformed. Maybe a full or deep democracy can encourage citizens to participate and become more knowledgeable through participation.

A classic example is the correlation between education levels and GDP. Do countries with higher education levels experience more economic growth because of the education levels of their citizens? Or is it that richer countries can afford to spend more on education and hence have better educated citizens? Probably both.