Uncertainty around economic policy tends to go up sharply in the months leading up to an election, a new working paper published by the US National Bureau of Economic Research shows. The risk of such spikes in uncertainty ahead of elections have become even more magnified in recent times due to increased polarization and the coronavirus crisis.
Academic Scott R. Baker of Northwestern University and others use newspaper coverage to gauge uncertainty in economic policy in over 20 countries across different election periods. They find that during an election month and the month before, uncertainty is on average 13% higher than in other months of the same election cycle, across countries.
This is similar to stock market volatility, which also tends to increase before elections.
However, the level of economic policy uncertainty varies from election to election. The authors find that it is typically higher when there is greater polarization and when the election is more closely fought. Past data from the US suggests a 28% increase in economic policy uncertainty in the month of the presidential elections that were close and polarized, as compared to elections that were neither.
Economic plans inevitably dominate the political conversation in every national election. This is more so in times of a pandemic, as government policy will increasingly matter in investment decisions of firms and individuals, the authors say.
In fact, earlier this year, economic policy uncertainty in the US reached high levels due to the “uneven response” of the Donald Trump administration to the covid-19 pandemic months ahead of the presidential election, the study finds.
Using their findings, the authors predict larger spikes in uncertainty around future elections in countries that have experienced rising polarization in recent years.
Snap Fact features new and interesting reads from the world of research.
Catch all the Business News , Breaking News Events and Latest News Updates on Live Mint. Download The Mint News App to get Daily Market Updates.
MoreLess