Ryan Murphy on government effectiveness and growth
By Scott Sumner
A decade ago I wrote a paper that looked at several definitions of neoliberalism, and found that what I called “egalitarian neoliberalism” was especially closely correlated with civic virtue. This model was based on the various indices of economic freedom, with the sign on size of government inverted (so that bigger government was a plus, not a minus as in the typical economic freedom indices). For example, the (high trust) Nordic countries gravitate toward models that combine free markets and large government.
Ryan Murphy has a very interesting new paper that explores these ideas in much more depth. He constructs an index of “State Economic Modernity” (SEM) by subtracting size of government in the Fraser Index of Economic Freedom of the World (EFW) from the component that measures rule of law and property rights. Again, the highest values of this SEM index tend to occur in the Nordic countries. He rightly points out that this measure makes more sense than “state capacity”, which doesn’t tell us what governments are actually doing:
This paper constructs a measure, the SEM index, which can be thought of as a measure of state building or state economic power, and is related to the concept known as state capacity. In contrast to state capacity, rather than asking the hypothetical question of what is in a government’s capacity to do, it measures the extent to which a government exerts itself within an economy, and how well it provides the most basic public goods.
Murphy also points out that economic freedom may be easier to obtain that SEM:
The countries of Georgia and Libya occupy approximately the same level of SEM, but they have very different levels of economic freedom. With political will, Georgia was essentially able to pull itself up from the current spot Libya finds itself in economic freedom to the very high level it is today (see Burakova and Lawson 2014).
Compare this to countries with the same level of economic freedom, but very different scores in SEM. Guatemala has about the same level of economic freedom as the Nordic countries, but the opposite score in SEM. The idea that public officials could push Guatemala rightwards on the graph to meet the European social democracies in their degree of state building and state capacity is almost unimaginable. It would require high degrees of political will, skill, and luck for a matter of decades in investing in the country’s social capital and human capital, if it is possible at all. In this sense, the SEM index may be more “deeply” institutional than measures of economic freedom, which may relate more to “policy.”
Murphy found that while economic freedom is correlated with growth, SEM is not:
Table 12 provides regression results for economic growth. For these regressions, initial level of economic output is included as a control variable. All independent variables correspond to year t, while the independent variable corresponds to growth from year t to t+10. The most recent data points correspond to growth occurring from 2000 to 2010. A similar pattern to Table 11 emerges. The SEM index is significant in regressions that do not include country fixed effects, but loses significance when they are included. In fact, when they are included, its point estimate is negative. In the final specification, Regression (36), economic freedom retains statistical and economic significance; the coefficient actually implies that a one standard deviation increase in economic freedom corresponds to a 0.246 standard deviation increase in growth rate. Considering economists’ collective inability to predict future growth rates very effectively, as noted by Easterly (2013: 215-238), the magnitude of the effect of economic freedom is quite large, while nothing is found for the SEM index.
For anyone interested in comparative economic systems, I strongly recommend you take a look at Murphy’s paper.
Here’s how I think about the growth findings. Both Sweden and Switzerland are essentially utilitarian economic models. But the Swedes assume that big government best promotes utilitarian goals while the Swiss assume that a smaller government is better able to achieve those goals. Because government of Switzerland is smaller, per capita income is higher. On the other hand, many poor countries have even smaller government. Their poverty reflects a lack of SEM; the benefit of small government is offset by a lack of good governance in other areas such as property rights and rule of law. Note that cultural differences affect GDP in two ways, by impacting governance and also by impacting the productivity of individual citizens. But culture is not the entire story, as the two Koreas demonstrate.
Poor countries should first focus on getting richer, which means more economic freedom. In the long run, that freedom and prosperity will lead to better culture, which will allow them to choose between the Swiss and the Swedish models. But right now they don’t have that choice because (as Murphy points out) they lack the cultural prerequisites needed for the Swedish model. Even greater economic freedom is not easy to achieve, but it’s not as difficult as state economic modernity. When you are incompetent, it’s easier to do nothing than to do something.
Here is the correlation that Ryan found between SEM and EFW: