Scholars have argued that democratic countries tend to experience stronger economic growth than those under dictatorships. But the drivers of this relationship are not well understood.
To analyze the connection between democracy and economic growth, assistant professor Priya Mukherjee and colleagues turned to Indonesia. They chose this country because it offers a unique opportunity to study how transitions to democracy affect economic growth in more than 240 districts, comparable to U.S. counties. Growth was measured as the productivity of manufacturing firms: the value of company-produced goods relative to the cost of labor and equipment operation.
“Firm productivity is a key driver of economic growth in most countries,” says Mukherjee. “Since Indonesia has rich data about the manufacturing sector, we analyzed how its productivity is related to district-level transitions to democracy at different points in time.”
These transitions are rooted in Indonesia’s political history. After the country became independent from the Dutch colonial government in 1945, its first two leaders were President Sukarno (1945-1967) and President Soeharto (1967-1998). Soeharto came to power with military support and ruled as a dictator, despite nominal elections involving multiple political parties.
The 1997 Asian financial crisis started a process toward ousting Soeharto that culminated in his surprise resignation in May 1998. Indonesia’s first democratic elections took place in June 1999. Mayors who had previously been Soeharto-appointed were now elected by district legislatures. Key for Mukherjee’s study was that these democratic transitions happened in a staggered manner. Economists refer to this as a “quasi-random natural experiment.”
The staggered timing had two reasons: Some mayors ended their five-year terms prematurely due to illness or death, and appointed mayors were allowed to complete their terms before being replaced by elected officials. Thus, 18% of transitions happened in 1999, 55% between 2000 and 2002 and 27% in 2003.
Mukherjee studied the relationship between political change and firm productivity with up to 283,000 observations provided by the Indonesian Statistical Office. This office conducted annual surveys of manufacturing plants with at least 20 workers. The surveys used in the study spanned the years 1993 to 2014 in 243 districts. The data prior to 1999 and after 2003 provided information about firm growth without political changes. The analysis model and the quasi-random timing of mayor transitions made it possible to estimate their causal effect on firm productivity.
The results showed that productivity increased substantially for several years after democratization before reaching a plateau. The increase ranged from 5% shortly after the change to 12% more than five years later. It was significantly higher than expected for a five-year period of political stability.
Next, the researchers examined potential reasons for the positive effect of democratization on economic growth. The first analysis used 860 district-level business regulations whose quality was assessed by an independent expert panel. The regulations were passed by district legislatures between 1999 and 2006. Most of them (77%) concerned business licensing; others covered labor, transportation, taxes and government levies.
The expert panel used 14 criteria for its quality assessment, such as healthy competition and a free flow of goods, services and persons. According to the panel, 15% of the 860 regulations passed all 14 quality criteria. Mukherjee found that districts with earlier transitions to elected mayors had greater improvements in their regulation quality. Thus, policies that make it easier to do business are one likely mechanism by which democratization supports economic growth.
The second analysis used a 2007 survey of more than 12,000 firm managers about their perception of the business environment. They were asked, for example, whether payments to local officials were needed to achieve business goals. The survey analysis showed that firm managers perceived districts with earlier transitions to elected mayors as more business-friendly. For example, policies that reduce bureaucratic hurdles help firms operate more efficiently and are more likely to be implemented by mayors who are accountable to their constituents, including firm managers. In other words, these policies may serve as electoral incentives.
Mayors appointed by dictators, however, do not feel electorally accountable. Bribing is a common mechanism for seeking their support of selected firms, at the expense of district-wide economic growth. This local finding helps explain the large-scale trend of greater economic growth in countries with democratically elected leaders, says Mukherjee.
A strength of the study is that it tracked the relationship between local governance and firm growth over time. “The longer a district was led by an appointed rather than elected mayor, the more time corrupt officials had to capture institutions,” explains Mukherjee. For example, mayors might appoint police officers and civil servants that do their bidding. If these officials make it harder for new firms to set up shop, economic growth will suffer.
Quantifying the effect of electoral accountability on economic growth provides lessons that go beyond Indonesia, says Mukherjee. “Many countries around the world are emerging market economies that continue to struggle with political corruption,” she notes. “Our findings suggest that transitions to democratic institutions might boost economic growth in these countries as well.”