The relationship between unemployment, NAIRU and investment: microfundations for incomplete nominal adjustment

This paper proposes a simple method to estimate a macro shock-specific Okun elasticity: it measures by how much the unemployment rate falls over a certain horizon when output increases by one percentage point over the same horizon because of a specific macroeconomic shock. Inference is based on simple instrumental variable regressions of cumulative unemployment on cumulative output. Using data for the Republic of Moldova I consider government spending, tax, monetary policy, financial, technology, and oil shocks. We obtain eight key results: • At medium horizons (2-3 years), Okun elasticities are largely stable across different kinds of shocks. • At shorter horizons, differences are more pronounced. The speed at which unemployment adjusts relative to output depends on the shock driving fluctuations. This highlights the importance to consider longer horizons. Otherwise, one could incorrectly conclude that the elasticity breaks down for some cycles. • The elasticity is larger for financial shocks. Importantly, it is larger than for monetary policy and government spending shocks. • The largest elasticity is for technological shocks followed by oil shocks. • An increase/decrease in unemployment by (0.14 p.p.) caused an increase/decrease in GDP by 1 p.p. period 2011--2015. • An increase/decrease in unemployment by (1.79 p.p.) caused an increase/decrease in money supply by 1 p.p. period 2011--2015. • An increase/decrease in unemployment by (0.17 p.p.) caused an increase/decrease in GDP by 1 p.p. period 2007--2011. • An increase/decrease in unemployment by (0.06 p.p.) caused an increase/decrease in money supply by 1 p.p. period 2007--2011.