1. Hysteresis and Fiscal Policy
Philipp Engler (DIW Berlin) and Juha Tervala (University of Helsinki)

2. Motivation Model Simulations Conclusions Overview
Results without hysteresis
Results with hysteresis
Conclusions

3. Motivation Recessions may have a permanent effect on output
Ball (2014): most countries suffered from a hysteresis effect after the global recession of (23 OECD countries)
Blanchard et al. (2015): Two thirds of countries suffered from hysteresis over past 50 years (sample: 23 OECD countries)

4. Motivation

5. Motivation
Summers (2015) criticizes DSGE models and New Keynesian macroeconomics because they ignore hysteresis
Yellen (2016): hysteresis and the possibility it might be reversed can have important implications for the conduct of fiscal policy
Blanchard and Summers (1986): Hysteresis in unemployment
Fatas and Summers (2016): we need a broader concept of hysteresis that allows a temporary downturn to affect productivity and capital accumulation dynamics

6. Motivation
Anzoategui et al. (2016): in the recent productivity slowdown, labor productivity declined by 7 percentage points relative to the trend
TFP declined by roughly 5 percentage points relative to the trend
Weak demand explains 4.75 percentage points of the TFP slowdown and most of the 7 percentage points decline in labor productivity
A drop in capital intensity after 2009 mainly accounts extra the drop in labor productivity relative to TFP
This suggests that endogenous changes in TFP caused by a weak demand is the most important factor of hysteresis

7. Persistent TFP changes at the core of our analysis
Motivation
Substantial TFP declines relative to pre-recession trends
Persistent TFP changes at the core of our analysis

8. % change relative to the initial steady state
Motivation
Our approach: Endogenous TFP in otherwise standard DSGE model The production function with and
TFP
Learning by doing
% change relative to the initial steady state

9. Motivation Consequences:
A temporary recession has a long-lasting effect on output
Fiscal policy has a long-lasting effect on output through TFP
Preview of Results:
Output multipliers significantly larger
Hysteresis has profound implications for the conduct of fiscal policy

10. The model Standard New Keynesian DSGE model of two countries:
Home and Foreign countries of equal size
A 2-country model to have an entirely demand-driven source for a recession
Representative households maximizing utility
Imperfect competition in the goods market
Sticky prices, à la Calvo (1983)
Central Banks follow Taylor rules
Government spending financed by lump-sum taxes, balanced budget

11. The model The domestic household maximizes
Time preference shock
The domestic household maximizes
subject to
with
Our scenario: a foreign time preference shock causes a global recession (roughly 1% drop in domestic output)

12. The model
Firm z has profits and produces with Maximizes profits s.t. demand function and the Calvo constraint.

13. The model
Fiscal policy: The government pays public consumption with lump-sum taxes: Monetary Policy:
Balanced budget
Public spending: an exogenous AR (1) process
a pure inflation targeting rule with interest rate smoothing

14. The model: Parameters

15. The model: Parameters
A foreign shock drives the home economy into a recession and the home government implements a fiscal expansion of 0.5% of GDP
Rawdanowicz et al. (2014): the hysteresis parameter, the effect of one percentage point of the negative output gap on reducing potential output, is 0.1 for the U.S. and 0.3 for the euro area
DeLong and Summers (2012): the hysteresis parameter, a proportional reduction in potential output from a temporary downturn, is 0-0.2
Our model: the proportional reduction in output in the 20th period to first-period output is 0.13

16. Simulations: Results without hysteresis
Model mechanics
A foreign recession reduces domestic exports and output
A temporary demand shock, a temporary recession
Economies return to the initial steady state
Fiscal expansion limits the size of a recession

17. Simulations: Results without hysteresis

18. Simulations: Results WITH hysteresis
Model mechanics:
Same as above + a long-lasting TFP decline → a long-lasting fall in potential and actual output
Fiscal policy:
Much more effective!
Additional channel: public spending ↑: employment today ↑: TFP in the next period ↑: output ↑: lifetime consumption ↑

19. Simulations: Results WITH hysteresis

20. Fiscal output multipliers
Cumulative multiplier (16 quarters): Net present value multiplier (a long-term view): Fiscal output multipliers measure the increase in output caused by fiscal policy, over the change of fiscal policy
WITHOUT Fiscal Expansion

21. Welfare multipliers Welfare measure:
The welfare multiplier of fiscal policy as the change in aggregate welfare, in consumption equivalent terms, for a one unit change in public spending
If the welfare multiplier is 0.5
One euro spent by the government raises domestic welfare by the equivalent of 0.5 euros of private consumption
Households are willing to pay 0.5 euros for a one euro rise in public spending

22. Fiscal multipliers Without hysteresis Output multipliers are small
Welfare multipliers are always negative: No reason for a fiscal expansion in a recession

23. Fiscal multipliers Hysteresis changes the effects of fiscal policy
Output multipliers rise significantly
Welfare multipliers are positive

24. Hysteresis in the Euro area recession of 2012-2013

25. Fiscal consolidation measures in the euro area in 2011-2013
Data: TFP fell in the recession of the 2010s and it was 1.1% below trend in 2013
Data: Output was 4.8% below the trend in 2013
Gechert et al. (2015): cumulative discretionary fiscal consolidation in the euro area in were 3.9% of output
Our results: this reduced output and TFP by 4.0% and 0.9%, respectively, relative to the no-consolidation baseline by 2013
Fiscal consolidation may explain the majority of TFP's and output's deviation from the trend in 2013
Hysteresis implies that the damage of fiscal consolidation is not limited to the short term because of its substantial medium- and long-term effects.
Fiscal consolidation may have been a mistake, but mono-causal explanations that austerity can be blamed for everything is wrong

26. Conclusions
Yellen (2016): hysteresis and the possibility it might be reversed can have important implications for the conduct of monetary and fiscal policy
Reifschneider et al. (2015): optimal monetary policy becomes more accommodative in the presence of hysteresis
Our findings:
Accommodative fiscal policy is not desirable in the absence of hysteresis
Accommodative fiscal policy becomes desirable in the presence of hysteresis
Hysteresis has more profound implications for the conduct of fiscal policy than to monetary policy