Do elections affect infrastructure projects?

Do governments manipulate policy to appear more competent just before elections? The temptation to use the power of incumbency to gain advantage is one of the reasons that the UK has a system of pre-election "purdah" on new policy announcements. A new study looking at road-building in India investigates whether governments can get round such rules in order to appear more competent in the eyes of the electorate. 

Infrastructure is one of the key drivers of economic growth. In developing countries such as India, large scale rural roadbuilding projects are particularly important for broad-based development.

In CAGE working paper Elections and Rural Road Construction: Evidence from India Ahana Basistha, Amrita Dhillon and Arka Roy Chaudhuri analyse the link between electoral cycles and a national road programme in India.

The team studied data covering 150,000 roads from the Pradhan Mantri Gram Sadak Yojna (PMGSY) government monitored scheme over a ten-year period. They found not only a significant increase in road construction activity in the run up to state elections, but that roads with a shorter construction time were prioritised during this period.

PMGSY is a rules-based programme introduced in December 2000. It is the world’s largest rural roads programme with built-in accountability and transparency features involving multiple decision makers at every stage. This includes federal government for funding, oversight by a national agency, and the state government responsible for overall execution and monitoring. In principle this means that local politicians cannot change roads once they’re approved in the core network.

However, using state legislative assembly election data from18 states between 1996 and 2007 the research team discovered evidence that the roll out of road building was being manipulated.

Their analysis shows that one and a half extra roads were built immediately before elections compared to the years soon after. Since voters base their decisions on roads they can observe before elections, the research indicates that politicians might target easier-to-build roads before elections.

Commenting on the findings Amrita Dhillon, CAGE Research Associate and Professor of Economics at Kings College London said:

"Using the case of a large rural road building programme in India, which is controlled by multiple levels of government, this paper shows how strong electoral incentives can still generate electoral cycles in road building. There’s a 40% increase in the number of new roads built in the penultimate year before the state election. This is likely driven by a focus on roads that have shorter stipulated times.

“Our work contributes to the literature on electoral cycles at the subnational level in developing countries. Studies have shown that electoral cycles are larger in developing countries relative to developed countries. India, as a developing country, a federation, and a democracy, is a particularly interesting context to study electoral cycles.”

Understanding the timing and motivations behind infrastructure projects can help when designing better governance and oversight mechanisms to ensure that electoral incentives do not compromise long-term development goals.

• Read the full research paper: Elections and Rural Road Construction: Evidence from India