Hysteresis in Potential Output and Monetary Policy
| Date | 01 September 2014 |
| DOI | http://doi.org/10.1111/sjpe.12050 |
| Published date | 01 September 2014 |
| Author | Daniel Kienzler,Kai D. Schmid |
HYSTERESIS IN POTENTIAL OUTPUT
AND MONETARY POLICY
Daniel Kienzler* and Kai D. Schmid**
ABSTRACT
We show that actively stabilizing economic activity plays a more prominent role
in the conduct of monetary policy when potential output is subject to hysteresis.
We augment a basic New Keynesian model by hysteresis in potential output and
contrast simulation outcomes of this extended model to the standard model. We
find that considering hysteresis allows for a more realistic amplification of mac-
roeconomic shocks and persistent movements in output after monetary shocks.
Our central policy implication of active output gap stabilization arises from sta-
bility analyses and welfare considerations.
II
NTRODUCTION
Due to the Great Recession, the political and academic debate has experi-
enced a revival of the so-called hysteresis phenomenon on a grand scale. In
the face of the severe crisis, a large number of major economic institutions
and think tanks, such as for example the European Commission (2009), Fur-
ceri and Mourougane (2009), the International Monetary Fund (2009) or Pi-
sani-Ferry and van Pottelsberghe (2009), have addressed the negative
implications of the economic downturn for the development of potential out-
put. From this debate, two important questions arise: Does hysteresis have
economic policy implications? If yes, how should economic policy react to
hysteresis?
Generally, hysteresis means that pronounced changes in aggregate demand
exhibit procyclical, persistent real supply-side effects. While several facets of
hysteresis have been documented in the macroeconomic literature since the
early 1970s (see section II for a short summary of this literature), the question
of how to consider such effects in terms of monetary or fiscal policy strategies
has hardly been addressed by popular models. Specifically, currently used
standard models designed for monetary policy research fade out the stimulus
of demand-determined actual output on an economy’s potential output. This
holds for the logic of the New Keynesian model
a la Woodford (2003) or Gal
ı
(2008) as well as for rather pragmatic models within the inflation targeting
*University of St.Gallen
**Macroeconomic Policy Institute
The author acknowledges generous financial support from the Swiss National Science Foun-
dation (PBSGP1-146934) for part of the work on this paper.
Scottish Journal of Political Economy, DOI: 10.1111/sjpe.12050, Vol. 61, No. 4, September 2014
©2014 Scottish Economic Society.
371
context such as Svensson (2000).
1
However, as Orphanides et al. (2000) argue,
ignoring hysteresis effects may involve substantial misjudgement for the con-
duct of fiscal or monetary policies.
We address this shortcoming by examining the consequences of hysteresis
in potential output for monetary policy. To this end, we extend Gal
ı (2008)’s
basic New Keynesian model by hysteresis, i.e. by allowing the path of poten-
tial output to be influenced by lagged actual output. To work out the rele-
vance of hysteresis for monetary policy, we contrast simulation outcomes of
the extended model with Gal
ı (2008)’s benchmark model and with empirical
second moments. Moreover, we examine the implications for the conduct of
monetary policy with respect to stability and welfare considerations.
We find that the extended model traces more realistic adjustment patterns
than the standard New Keynesian model after monetary shocks hit the econ-
omy. Specifically, hysteresis helps to reproduce empirically well-documented
persistence patterns of output. Furthermore, our model exhibits a number of
features that assign a more important role to active output gap stabilization if
the economy is subject to hysteresis. First, if the central bank applies a mone-
tary policy rule and the degree of hysteresis is large enough, achieving a
unique stable equilibrium requires a reaction to the output gap for certain
ranges of the reaction parameter for inflation. Second, if a welfare loss crite-
rion based on the variability of inflation and the output gap is applied, react-
ing to the output gap in the monetary policy rule yields sizeable welfare loss
reductions beyond those that would arise without hysteresis effects. The rea-
son for these results lies in the dynamics of the output gap. If the central bank
wants to fight inflation, the procyclical behaviour of potential output requires
a balancing reaction to the output gap to maintain downward pressure on
inflation. At the same time, this downward pressure helps to reduce inflation
variability.
As a robustness check we also show that, depending on the degree of hys-
teresis, actively stabilizing the output gap can still be welfare improving even
when there is measurement uncertainty with respect to the output gap. Fur-
thermore, if hysteresis is in effect, shifting the focus on output instead of the
output gap in the monetary policy rule is not necessarily detrimental to wel-
fare as is the case in the basic New Keynesian model.
Although hysteresis has not played a meaningful role in standard models so
far, its relevance for stabilization policy has been addressed by several authors
such as Ball (1999), DeGrauwe and Costa Storti (2007), DeLong and Sum-
mers (2012), Lavoie (2004) or Solow (2000). More specifically, there are even
some approaches that explicitly model such endogenous supply-side adjust-
ment. For example, Mankiw (2001), Fritsche and Gottschalk (2006), and
Kapadia (2005) basically share a common reduced form specification for hys-
teretic adjustment that was originally proposed by Hargreaves Heap (1980).
Our model also refers to this specification.
1
A notable exception to this is Woodford (2003)’s chapter 5, where he extends the basic
model set up by capital investment and illustrates that productive capacity as well as the
equilibrium real rate of interest are affected by monetary policy.
372 DANIEL KIENZLER AND KAI D. SCHMID
Scottish Journal of Political Economy
©2014 Scottish Economic Society
To continue reading
Request your trial