Identifying Government Spending Shocks and Multipliers in Korea

43 Pages Posted: 16 Sep 2019

See all articles by Kwangyong Park

Kwangyong Park

Economist, Economic Research Institute, The Bank of Korea

Eun Kyung Lee

Bank of Korea - Economic Research Institute

Date Written: September 16, 2019

Abstract

Accurately estimating the government spending multiplier is important so that fiscal policies can be used appropriately when recessions hit the economy. To fill the gap between the frontier research of calculating the government spending multiplier and current Korean research, and to estimate a more accurate multiplier in Korea, we construct a government spending news series in Korea based on Fisher and Peters (2010) by exploiting a market-weighted sum of excess stock returns of military contractors in Korea. We then use this military spending news series and estimate a structural VAR model to evaluate the effects of government spending. As a result, GDP and government spending show statistically significant responses to military spending news shocks. The accumulated government spending multiplier peaks after four quarters, and the five-year cumulative multiplier is calculated as 1.27. For a robustness check, different types of VAR models are tested and results are qualitatively similar.

Keywords: Fiscal policy, Multiplier

JEL Classification: C32, E62

Suggested Citation

Park, Kwangyong and Lee, Eun Kyung, Identifying Government Spending Shocks and Multipliers in Korea (September 16, 2019). Bank of Korea WP 2019-22. Available at SSRN: https://ssrn.com/abstract=3454291 or http://dx.doi.org/10.2139/ssrn.3454291

Kwangyong Park (Contact Author)

Economist, Economic Research Institute, The Bank of Korea ( email )

110, 3-Ga, Namdaemunno, Jung-Gu
Seoul 100-794
Korea, Republic of (South Korea)

Eun Kyung Lee

Bank of Korea - Economic Research Institute ( email )

110, 3-Ga, Namdaemunno, Jung-Gu
Seoul 100-794
Korea, Republic of (South Korea)

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
12
Abstract Views
123
PlumX Metrics