Democracy and growth

Singapore celebrated the semi-centennial of its independence over the weekend. Under the leadership of its founder, Lee Kuan Yew, who passed away in March, it rose from being a poor country with no resources to one of the wealthiest and most developed states in the world.

I found many things to admire when I visited the city state back in June – but not democracy: A paper I reviewed back in March argues that LKY was the first of modern-day “soft dictators,” and a model to many including Turkey’s Recep Tayyip Erdoğan – which had really perplexed me since “the Gentle-Man” is not an autocrat but a democrat, and Turkey not an autocracy but an advanced democracy. Even though there are now free elections in Singapore, the degree of gerrymandering would make even Erdoğan green with envy.

Singapore’s stellar growth, along with similar success stories of South Korea, China and more recently Vietnam, have led some to argue that democracy is detrimental to growth. A well-known 1994 paper by Torsten Persson and Guido Tabellini argues that politicians waste resources to win elections in democracies. However, Turkish economist Daron Acemoğlu, who was recently named the world’s most influential economist, argued in a 2008 paper that the economy is controlled by an inefficient oligarchy in autocracies; exactly what I observed in Azerbaijan back in 2013. He and others have pointed to several such policies in autocracies detrimental to growth.

Without a definite answer from theory, we should turn to data for help. Even though the average “free” country, as denoted by Freedom House, has a GDP per capita of $17,000, around four times that of an “unfree” country, the latter have higher growth rates. But you would expect poorer countries, which are usually autocratic, to grow faster than richer ones, which are usually democracies. A 1994 paper by Robert Barro concluded, after controlling for this convergence effect as well as other factors, that “the effect of democracy on growth is weakly negative.”

In a recent paper, Acemoğlu and his three coauthors note that growth usually plunges during political transitions. Moreover, a weak economy may make transition more likely. Taking these factors into consideration, they show that a “permanent” democratization leads to an increase of around 20 percent in GDP during the next 25 years, with investment in education and health care as well as lower social unrest accounting for the bulk of the rise.

However, educated, rich people demand democracy, as was the case in South Korea in the 1980s. I got a similar sense from the Singaporeans I talked to in June. This is the well-known chicken or egg problem: To solve it, we would need a variable correlated with democracy, but not with growth – what economists call an instrumental variable. Acemoğlu & Co. have an ingenious one: The Arab Spring spread from Tunisia to Egypt, but Tunisian politics would not affect Egyptian growth. The authors use democratization of nearby countries as an instrument for democratization in a country – and find that “democracy does cause growth.”

But I am still wondering why all of the several cases of autocracies with high growth rates are in Southeast Asia. I am just curious if Turkey could at least make some economic gains if Erdoğan manages to become an all-powerful president at the expense of throwing Turkey and the region into turmoil – and providing more data for Acemoğlu & Co.