Monday, April 28, 2014

Are culture and economic freedom substitutes or complements in facilitating economic development?

A couple of articles by Claudia Williamson and Rachel Mathers provide a good place to start to think about this question.

The authors construct an economic culture index by identifying four categories of culture likely to encourage economic interaction and hence promote economic development. The categories are: trust (percentage who say most people can be trusted); respect (percentage who choose tolerance and respect for other people as one of the most desirable qualities for children to learn at home); individual self-determination (average rating on a scale of 1 to 10 where the highest rating is given where respondents claims a great deal of control over the way their lives turn out and the lowest rating where they claim no control); and individualism (percentage who choose obedience as one of the most desirable qualities for children to learn is viewed as a negative factor). The data comes from World Values Surveys and culture index scores for each country are obtained using principal component analysis.

The economic freedom data used in the studies comes from the Fraser Institute. This economic freedom index combines economic and institutional variables relating to size of government, monetary policy and price stability, legal structure and security of private ownership, freedom to trade with foreigners, and regulation of credit, labour and business.

The studies use regression analysis in an attempt to disentangle the impact of economic culture and economic freedom on growth in per capita incomes. There are some differences in the conclusions of the two articles.
The first article, ‘Economic Freedom, Culture and Growth’, for which Claudia Williamson is lead author, was published in Public Choice (2011). The authors found that economic freedom is “relatively more important for growth than culture” and suggested that “culture and economic freedom may best be described as substitutes”. They suggested that trust and respect may become less important from an economic growth perspective as economic institutions are established to enforce property rights and contracts. The authors also acknowledged that culture may have important indirect effects on economic growth by promoting establishment of economic freedom.

The second article, ‘Cultural Context: Explaining the Productivity of Capitalism’ for which Rachel Mathers is lead author, was published in Kyklos (2011). This article extends the line of analysis by including an interaction term (the product of the culture and freedom indexes). The authors find that “such interaction does demonstrate a significant and positive effect on economic growth”. They conclude that if economic culture is deficient, “economic freedom alone may not possess the necessary binding constraints to be as effective as theory predicts”.

After reading these articles again I wanted to get a better feel for the data which the authors use. The first chart has been constructed using the summary data provided in Appendix 2 of the Public Choice article. (The culture index has been rebased from a 0-10 scale to 1-10 scale to assist subsequent analysis) The size of the balls in the chart (it would not be appropriate to refer to them as bubbles) represents the per capita income level for each country in 2000 (That year is chosen for consistency with other data used in the studies. The relevant variable is rgdpl Penn World Tables version 6.2).

What the chart shows is, of course, consistent with some substitution between culture and economic freedom – for example, both Singapore and Sweden manage to have relatively high per capita income levels. It suggests, however, that substitution possibilities are limited. The countries that rate highly in terms of culture also tend to rate highly in economic freedom, and vice versa.

The finding in the Kyklos article concerning the importance of interaction between culture and freedom suggested to me that it might make sense to use a Cobb Douglas production function in further regression analysis. The isoquants in Chart 2 have been constructed using the coefficients of such an analysis. The isoquants show the differing combinations of culture and freedom that are estimated to produce a given level of average income. The per capita income level of the US in 2000 was about 4.25 times that of Mexico, which in turn was a bit more than 4.25 times that of Ghana.

 (For those with an interest in such esoteric matters, the estimated coefficients of the regression are as follows:
Intercept         2.44 (0.64)
Ln Culture       0.66 (0.21)
Ln Freedom     3.20 (0.47)
Adj R2= 0.68

If the isoquants are interpreted as reflecting a production function, the Chart suggests that in the absence of a supportive culture, high levels of economic freedom cannot produce high average income levels. However, it might be more appropriate for the isoquants to be interpreted as being determined simultaneously by interaction between culture, economic freedom and development levels. Under that interpretation, a high level of economic freedom is unlikely to be sustainable in the absence of some cultural support because law and order problems would constitute a major threat to lives and property.

It seems to me that in trying to put together a plausible story of economic development, it makes sense to speculate that interactions between culture and economic freedom facilitate market exchanges, which in turn provide incentives for participants to gain reputations as being worthy of trust. As people become more trustworthy and trusting, and more respectful of the rights other people, they could be expected to support greater economic freedom. We may thus observe a virtuous cycle where economic freedom promotes economic development and economic development promotes a culture supporting greater economic freedom.

However, I have yet to provide a satisfactory answer the question raised in my last past of whether emancipative values (as defined by Christian Welzel) support economic freedom.

No comments: