The New Keynesian Model and the Euro Area Business Cycle*
| DOI | http://doi.org/10.1111/j.1468-0084.2006.00440.x |
| Published date | 01 April 2007 |
| Author | Miguel Casares |
| Date | 01 April 2007 |
The New Keynesian Model and the Euro Area
Business Cycle*
Miguel Casares
Departamento de Economı´a, Universidad Pu´blica de Navarra, Pamplona, Spain
(e-mail: mcasares@unavarra.es)
Abstract
This paper describes a New Keynesian model incorporating transactions-facilitating
money and a time-to-build constraint into endogenous capital accumulation. The
calibrated New Keynesian model performs almost as well as the estimated vector
autoregressive model in replicating Euro area cyclical correlations between key
variables such as output and inflation, although it fares less well in predicting the
procyclical dynamics of nominal interest rates. The presence of a time-to-build
requirement in the model helps to improve its fit to Euro area data, whereas the role
of transactions-facilitating money is much less important. Impulse–response
functions and a decomposition of variance complete the analysis.
I. Introduction
In recent years, optimizing models with nominal rigidities have become increasingly
popular in dynamic macroeconomic analysis. They are usually referred to as New
Keynesian (NK) models. There are two major reasons for the upsurge in the use of
the NK methodology. First, structural equations are assumed to be independent
of monetary and fiscal policy regimes because they are obtained under any
implemented policy. Secondly, sticky prices (or wages) can help to capture the
*A first version of this paper was written while visiting the European Central Bank within the Graduate
Research Programme 2000 (ECB Working Paper No. 49). I would like to thank Bennett T. McCallum, Frank
Smets, Gunter Coenen, Vı´tor Gaspar, Ignazio Angeloni, Cruz A. Echevarrı´a, Oscar Bajo-Rubio, Christopher
Bowdler, Riccardo DiCecio, James Bullard, and two anonymous referees for helpful comments and sug-
gestions, and Deborah Roisman for her research assistance. The opinions given in this paper are exclusively
mine and do not necessarily reflect those of the European Central Bank. I also acknowledge financial support
from the the European Central Bank and the Spanish Ministerio de Educacio´ n y Ciencia (research projects
2002/00954 and SEJ2005-03470/ECON).
JEL Classification numbers: E20, E32, E52.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 69, 2 (2007) 0305-9049
doi: 10.1111/j.1468-0084.2006.00440.x
209
ÓBlackwell Publishing Ltd and the Department of Economics, University of Oxford, 2006. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
short-run real effects of these policies observed in real macroeconomic data. A
representative list of recent papers based on NK models should contain Chari, Kehoe
and McGrattan (2000), Erceg, Henderson and Levin (2000), Smets and Wouters
(2003) and Christiano, Eichenbaum and Evans (2005).
1
This paper extends the NK models considered in the existing literature and
analyses the extent to which the generalized model can explain key features of the
Euro area business cycle. Hence, the demand side of our model incorporates both
time-to-build endogenous capital and transactions-facilitating money that were not
included in other Euro area NK models such as those of Smets and Wouters (2003)
or Casares (2006b). Following the seminal paper by Kydland and Prescott (1982),
decisions regarding capital accumulation are subject to a time-to-build requirement
that obliges both households and firms to determine their optimal quantities some
periods in advance.
2
This paper shows how the incorporation of a time-to-build
constraint helps investment to increase in persistence and respond more gradually to
shocks. The role of money in the model is explained by the fact that money reduces
transaction costs. There are two ways to include transaction-facilitating money in
NK models. One is to assume that goods purchases require time resources, as
initially described by McCallum and Goodfriend (1987), and more recently used in
Chadha, Haldane and Janssen (1998), Pakko (1998) and De Fiore and Teles (2003).
The other alternative, which is taken in this paper, is to assume that transaction costs
require output usages from the household budget constraint, as in Feenstra (1986),
Sims (1994), McCallum (2000) and Lai, Chen and Shaw (2005). In any case, real
money holdings should be included in the transactions technology reflecting the
advantages of possessing the medium of exchange when conducting transactions.
This paper explores the implications for macroeconomic dynamics of both the time-
to-build requirement and the transactions-facilitating role of money.
The supply side of our model incorporates sticky prices in a monopolistically
competitive market. We initially follow the assumption found in Calvo (1983) that
prices are set optimally subject to some fixed probability. Those prices that cannot be
set optimally are decided by applying an indexation rule based partially on the
previous observation of inflation and partially on the long-run inflation rate. This
latter feature is a departure from the original Calvo setup where non-optimal prices
remain unchanged. Nevertheless, it is becoming increasingly popular in the NK
literature to allow for a longer persistence in inflation dynamics.
3
Another relevant aspect of the model is that there is endogenous potential output.
This variable will be computed as the amount of output produced if the economy
1
Earlier seminal papers that deserve to be mentioned are Kerr and King (1996), Yun (1996), King and
Wolman (1996) and Rotemberg and Woodford (1997).
2
Edge (2000) describes in her model a decentralized economy with a time-to-build requirement for
household capital accumulation decisions but without imposing any time-to-build constraint on the demand
for capital of the production sector.
3
Examples of New Keynesian models incorporating price indexation on lagged inflation can be found in
210 Bulletin
ÓBlackwell Publishing Ltd and the Department of Economics, University of Oxford 2006
were free of nominal rigidities as suggested by Friedman (1968), and extensively
used in the recent book by Woodford (2003).
The NK model presented in this paper is used for the business cycle analysis of
the Euro area. In particular, the paper shows the capacity of the model to replicate
some business cycle regularities observed in the Euro area data and how much
the transactions-facilitating money or the time-to-build constraint can help towards
this purpose. The Euro area business cycle will also be studied using a vector
autoregression (VAR) model directly estimated from the data.
The paper is structured as follows. Section II describes the NK model, while the
Euro area data and their estimated VAR are introduced in section III. Next,
calibration of the NK model for the Euro area is done in section IV. The business
cycle analysis is conducted in section V, in which Euro area statistics are compared
with those generated by the different models. In the same section, model-based
impulse–response functions and the variance decomposition are also thoroughly
discussed. Section VI concludes the paper with a review of the major findings.
II. The New Keynesian model
2.1. Households
The household sector is formed by a continuum of infinitely lived identical
households. The amount of utility that the representative household has in period tis
provided by the following utility function
Uðft;ct;ct1;ltÞ;ð1Þ
where, following standard notation, subscripts represent the time period, and fis a
preference shock, cis consumption and lis leisure. Consumption units are bundles of
many differentiated goods aggregated as in Dixit and Stiglitz (1977). The preference
shock ffollows an AR(1) process ft¼qfft1þef
twith a white noise innovation
ef
tNð0;r2
efÞ. Finally, the presence of both current and past consumption in
equation (1) reflects internal habit formation in consumption as first recommended
by Fuhrer (2000).
Let P
t
denote the Dixit–Stiglitz aggregate price level, w
t
the real wage rate, rk
tthe
real rental rate of capital, p
t
the rate of inflation defined as
pt¼Pt
Pt1
1;
and r
t
the real interest rate defined as
rt¼1þRt
1þEtptþ1
1
where R
t
is the nominal interest rate and E
t
denotes the rational expectation operator
conditional on information available in period t. Then, the budget constraint of the
ÓBlackwell Publishing Ltd and the Department of Economics, University of Oxford 2006
211The New Keynesian model and the Euro area business cycle
To continue reading
Request your trial