Over at Marginal Revolution, Alex Tabarrok recently presented a graph showing a positive correlation between UN measures of gender development and the Fraser Institute’s Economic Freedom Index. Of course, Alex presented the usual caveats about causation and correlation, but he concluded “at a minimum the graph indicates that capitalism and gender development are compatible contrary to many radicals”
This prompted me to check out how the Economic Freedom index was calculated. The relevant data is all in a spreadsheet, and shows that the index is computed from about 20 components, all rated as scores out of 10, the first of which is general government consumption spending as a percentage of total consumption. Since the Fraser Institute assumes that government consumption is bad for economic freedom, the score out of 10 is negatively correlated with the raw data.
Looking back at Alex’s post, I thought it likely that high levels of government expenditure would be positively rather than negatively correlated with gender development, which raised the obvious question of the correlation between government consumption expenditure and economic freedom (as defined by the Fraser Institute index). Computing correlations, I found that, although it enters the index negatively, government consumption expenditure has a strong positive correlation (0.42) with economic freedom as estimated by the Fraser Institute. Conversely, the GCE component of the index is negatively correlated (0.43) with the index as a whole. By contrast, items with a strong ideological component, like labour market controls, were very weakly correlated with the aggregate index.
It immediately struck me that this could be the basis of some snarky pointscoring. But on reflection, I thought that it might be more useful to look seriously at the issues.
First, it’s necessary to explain the results. A large slab of the index consists of things like the absence of military interference in the rule of law. There’s also an entire section of the index devoted to sound money (low and stable inflation). These are both features characteristically provided by strong democratic governments, and not by weak ones, regardless of whether they are interventionist or laissez-faire. It seems pretty clear that a principal components analysis would show a dominant component associated with the characteristics of developed mixed economies, characteristics that include high levels of public expenditure, combined with relatively light-handed forms of government intervention in the private sector, compared to those characteristic of weak governments in less developed countries.
This is scarcely surprising. State capacity tends to rise with income, so in wealthy countries the state can achieve more, with less obtrusive use of power, than in poor countries. It is strong and not weak states that produce economic freedom.
The same principles that lead to support for separation of powers within the state with clearly defined roles for the judiciary, legislature and executive (including the military!), also yield support for a mixed economy against the alternatives of comprehensive central planning and unfettered corporate power. The mixed economy produces more economic freedom for the average person than does the minimal state. Despite its own presumption to the contrary, the Fraser Institute’s analysis supports this conclusion.