The Relationship Market Openness and Democracy

by Alex D'Agostino

    Does market openness directly affect democracy? Political scientists and economists have long been divided over this question. Milton Friedman famously claimed that “History suggests only that capitalism is a necessary condition for political freedom.”[1] In speaking of his work with Pinochet’s Chile, Friedman stated that “In Chile, the drive for political freedom, that was generated by economic freedom and the resulting economic success, ultimately resulted in a referendum that introduced political democracy.”[2] Other economists have disputed this notion, claiming that the relationship is coincidental. Thomas Piketty writes that “Economic and technological rationality at times has nothing to do with democratic rationality. The former stems from the Enlightenment, and people have all too commonly assumed that the latter would somehow naturally derive from it, as if by magic.”[3] The debate over the effect of market openness on democracy is not purely academic, it is a major driver of policy, especially the implementation of market-oriented reforms. Some claim that those reforms are a prerequisite for democracy, while some maintain that such reforms create inequality which poses a danger to democracy.

                In order to answer the question of whether market openness effects democracy, I am using the Quality of Governance Index, and measuring the effect of market openness (1-10 scale) on democracy status (1-10 scale). Below is the distribution of democracy scores:

Figure 1, below, shows the relationship between market openness and democracy. This graph indicates that market economy status is positively correlated with democracy status.

To examine the relationship further, control variables are incorporated into the regression. The first is GDP per capita, which has a strong effect on democracy; it is well-established that wealthier countries generally are democracies.[4] GDP per capita is logged, as is standard practice. The second is agricultural share of the economy; countries that are less industrially developed and more agricultural are less likely to be democracies. Due to the distribution of this variable (most countries have less than 20% agricultural share of GDP), it was also logged. The third variable is years of schooling; one would expect that higher levels of education are necessary for democracies.[5] I expect to find that, conditioning on the three above-mentioned variables, I will see a strong correlation between market openness and democracy.

Table 2 shows the conditional estimates between market economy status and the democracy score.  Below is the regression table.

Table 2. Determinants of Democracy

Democracy Status	(1)	(2)
Market Openness	0.92***	1.49***
	(11.00)	(11.39)
Agricultural Share of		0.56**
Economy (logged)		(3.24)
Average Schooling		0.05
Years, Female and Male (25+)		(0.81)
GDP per capita (logged)		-0.77**
		(-3.32)
Constant	0.63	1.86
	(1.22)	(1.02)
R-Squared	0.561	0.681
Observations	95	95

Note: OLS estimates with t statistics in parentheses ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001

The first model indicates that market openness is positively correlated with democracy status at a significant level, with a 1 unit increase in market openness corresponding to a 0.92 unit increase in democracy status. With the addition of three controlling variables the coefficient increases, showing a stronger correlation between market openness and democracy status. For every 1 unit increase in market openness, our regression predicts a 1.49 unit increase in democracy status, controlling for agricultural level, average schooling level, and GDP per capita. One surprise is that schooling is not significant in our model.[6]

These findings are significant, but they do not establish a causal link between market openness and democracy status. The most glaring issue is reverse causality; it is perfectly plausible that the drive for market openness was generated by democracy (the opposite of what Friedman said). This is exacerbated by the question of which came first; it is possible that economic openness and democracy are mutually reinforcing, with incremental gains in each one serving to reinforce gains in the other. This study would be greatly aided by qualitative case studies, including countries like Chile that created open economies and then transitioned to democracies, as well as countries like Singapore (or a territory like Hong Kong) that have the most open economies in the world, but remain authoritarian.

Our findings are interesting in that they show a strong correlation between market openness and democracy, yet this fails to address Piketty’s claim that democratic and economic rationalism do not stem from each other, but have  merely coincidentally risen at the same time. A time-series analysis of market openness and transitions to democracy might provide a clearer answer to this question. Nonetheless, the empirical results do indeed establish that there is a clear correlation between market openness and democratic status.

Appendix

Table 1. Summary Statistics

	Mean	Std. Dev.	Min.	Max.	Description
Democracy Status	6.09	2.06	2.80	9.90	Status of Democracy (0=Min, 10=Max)
Market Economy Status	5.95	1.68	2.07	9.50	Market Openness (0=Min, 10 = Max)
GDP per capita (logged)	8.18	1.02	5.52	10.24	GDP per capita (log scale)
Agr. Share of Economy (logged)	2.35	1.10	-3.18	4.57	Agriculture’s share of economy (log scale)
Avg. Schooling	7.20	2.75	1.20	12.32	Average years of schooling for males and females
N	95

---

[1] Friedman, Milton. Capitalism and Freedom. (University of Chicago Press: 1962) p. 3

[2] Friedman, Milton. “Economic Freedom, Human Freedom, Political Freedom.” Speech at the Smith Center (Nov. 1, 1991)

[3] Piketty, Thomas. Capital. (Belknap Press: 2014) p.530

[4] The relationship is so strong, in fact, that many political scientists have suggested that once a country becomes middle-income (about $6,000 GDP per capita) it will generally transition to democracy, and once it becomes higher income (~$10,000 GDP per capita) democracy will become permanent. As an example, in 2011 Tunisia broke the $6,000 GDP per capita threshold, and almost immediately underwent a democratic revolution. Of course, there are notable exceptions, especially India, which became a democracy with a GDP per capita of only $800. (http://blogs.reuters.com/macroscope/2011/06/22/give-me-liberty-and-give-me-cash/)

[5] Thomas Jefferson famously wrote, “An educated citizenry is a vital requisite for our survival as a free people.” Jefferson’s Papers. (http://www.monticello.org/site/jefferson/educated-citizenry-vital-requisite-our-survival-free-people-quotation)

[6] A better model might not control for level of education, which surprisingly does not seem to have an effect, and instead replace it with another control variable that better explains democracy.