!(../../../../../../../../../../var/folders/34/zq18d8kx7kbgby0j06p_j6t40000gn/T/TemporaryItems/NSIRD_screencaptureui_EM2XPo/Screenshot 2022-01-04 at 17.01.16.png)
- Inflation !(../../../../../../../../../../var/folders/34/zq18d8kx7kbgby0j06p_j6t40000gn/T/TemporaryItems/NSIRD_screencaptureui_EM2XPo/Screenshot 2022-01-04 at 17.01.16.png)
- Monetary Policy
Most industrialised countries have been trying to cut public borrowing without impeding recovery from the Great Recession. Central banks have attempted to square this circle by loosening monetary policy. For example, UK finance minister George Osborne has stated that “theory and evidence suggest that tight fiscal policy and loose monetary policy is the right macroeconomic mix” for countries with excessive private and public debt (Mansion House speech 2012).
A number of recent studies have found that fiscal policy is particularly powerful in recessions – tax hikes and spending cuts harm growth more when the economy is already weak (Auerbach and Gorodnichenko 2012, Jordà and Taylor 2013). But if monetary policy is still effective, these big negative effects could in principle be offset by lower interest rates. In our new paper (Tenreyro and Thwaites 2013) we find that, at least in the US, this is not the case: official interest rates have no discernible effect on the economy during recessions. This means a crucial ingredient – the ability to stimulate a recession-hit economy by cutting policy rates – may be missing from the prevailing policy mix.
Auerbach and Gorodnichenko (2012) estimate the varying impact of tax and spending shocks over the business cycle using a ‘smooth transition local projection model’. They find that fiscal policy is more powerful in bad times than in good. We use the same framework to test whether the impact of shocks to the Federal Funds rate (identified by Romer and Romer 2004) varies similarly over the cycle. A local projection model essentially involves regressing the response variable on a shock lagged a certain number of periods. The regression coefficient on the shock is the level of the impulse response at that horizon. Estimating a family of these regressions with varying lags, one can trace out a normal impulse response function.
The average impulse response functions of the levels of real GDP and the quarterly annualised inflation rate to a one percentage point innovation in the Fed Funds rate are shown as the green lines in Figures 1 and 2. We find that unexpected changes in interest rates have the textbook effect on the US economy on average: a rise in interest rates first reduces spending, especially on durable goods, and then inflation.
Figure 1. Impulse response of level of real GDP to a one percentage point increase in the federal funds rate
Figure 2. Impulse response of the quarterly annualised inflation rate of GDP deflator to a one percentage point increase in the federal funds rate
The 'smooth transition' comes when allowing the impact of monetary policy to vary over the business cycle. This method estimates two sets of coefficients. When the economy is expanding strongly and the economy is hit by a policy shock, the model gathers information about the ‘good times’ coefficients. When it is contracting, we get information about the ‘bad times’ coefficients. When the economy is at neither extreme, the data informs estimates of responses in both booms and recessions. Local projection methods are well suited to studying how the impact of shocks varies over the cycle because the only thing that matters is the state of the economy when the shock hits. In contrast, standard methods such as VARs assume that the propagation of an old shock only depends on how the economy is doing later on.
The headline results are shown in Figures 1 and 2 – the red line is the impulse response in a boom, while the blue line is the impact in a recession. The difference between these lines is statistically significant at standard levels.
What could be driving these results? We do not find evidence that fiscal policy tends to counteract monetary policy more in recessions. Nor do we find the responses of credit spreads or quantities to policy shocks to be magnified more in booms. In line with another recent paper we do find that policy tightenings are more powerful than loosenings (Angrist et al 2013). This provides another reason to doubt the efficacy of a “tight fiscal, loose monetary” policy mix in current conditions. But it does not explain our results, because the past incidence of unanticipated increases in the policy rate is no higher in booms than in recessions.
So, having ruled out a number of plausible candidates, we are left with a puzzle as to the underlying economic reasons for our findings. If our findings are correct, recent signs of economic recovery are there in spite of the current policy mix, not because of it. And if the world economy slips back into recession, we cannot rely on conventional monetary policy to get us out.References
Angrist, Joshua D, Òscar Jordà and Guido Kuersteiner (2013), "Semiparametric Estimates of Monetary Policy Effects: String Theory Revisited," NBER Working Papers 19355, National Bureau of Economic Research.
Auerbach, Alan J, and Yuriy Gorodnichenko (2012), "Measuring the Output Responses to Fiscal Policy," American Economic Journal: Economic Policy, 4(2): 1-27.
Jordà, Òscar (2005), "Estimation and Inference of Impulse Responses by Local Projections," The American Economic Review, American Economic Association, vol. 95(1), pages 161-182, March.
Jordà, Òscar and Alan M Taylor, (2013) "The Time for Austerity: Estimating the Average Treatment Effect of Fiscal Policy," NBER Working Papers 19414, National Bureau of Economic Research.
Romer, Christina D, and David H Romer (2004), "A New Measure of Monetary Shocks: Derivation and Implications" The American Economic Review, 94(4): 1055-1084.
Greg Thwaites and Silvana Tenreyro (2013), ‘Pushing on a string: US monetary policy is less powerful in recessions’, CEP Discussion Paper 1218.