The Global Dimension of Inflation – Evidence from Factor‐Augmented Phillips Curves*
| Published date | 01 February 2013 |
| Date | 01 February 2013 |
| Author | Katharina Pijnenburg,Sandra Eickmeier |
| DOI | http://doi.org/10.1111/obes.12004 |
103
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford 2012. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 75, 1 (2013) 0305-9049
doi: 10.1111/obes.12004
The Global Dimension of Inflation – Evidence from
Factor-Augmented Phillips CurvesÅ
Sandra Eickmeier† and Katharina Pijnenburg‡
†Deutsche Bundesbank, Economic Research Center, Wilhelm-Epstein-Straße 14, 60431
Frankfurt/M, Germany (e-mail: sandra.eickmeier@bundesbank.de)
‡Deutsches Institut für Wirtschaftsforschung Berlin, Mohrenstraße 58, 10117 Berlin, Germany
(e-mail: kpijnenburg@diw.de)
Abstract
We examine the global dimension of inflation in 24 OECD countries between 1980 and
2007 in a Phillips curve framework. Wedecompose output gaps and changes in unit labour
costs into common (or global) and idiosyncratic components using a factor analysis and
introduce these components separately in the regression. Wefind that the common compo-
nent of changes in unit labour costs has a notable impact on inflation. Movements in import
price inflation (not driven by oil supply) and foreign competition and global interest rate
developments also affect inflation. Policy makers need to carefully observe those variables
when assessing inflation developments.
I. Introduction
The global dimension of inflation is attracting attention from academics and policymakers
for two reasons. First, the inflation rates in many countries have been historically low in
recent years, and observers speculate whether globalization (defined as an ongoing process
of integration through trade and of financial markets and labour markets) is responsible
for low inflation rates through various channels (see IMF, 2006; Deutsche Bundesbank,
2007; Ihrig et al., 2010 for overviews). The increased entry of emerging countries in global
markets that are able to produce at low costs could have held down import price inflation.
Technological advances are spurred by international competition, which puts pressure
on firms to innovate. Open economies specialize in relatively efficient sectors, which
should lead to productivity gains. Moreover, greater international competition, together
ÅWe are grateful to J¨org Breitung, Matteo Ciccarelli, Stefan Gerlach, Heinz Herrmann, Johannes Hoffmann,
Mathias Hoffmann, Massimiliano Marcellino, Dieter Nautz, Fabrizio Zampolli, anonymous referees as well as par-
ticipants of the 10th Bundesbank Spring Conference on ‘Central Banks and Globalisation’ and of a seminar at the
Reserve Bank in New Zealand for helpful comments and discussions. The article was written while Katharina Pijn-
enburg was visiting the Bundesbank. The views expressed in this article are those of the authors’ alone and do not
necessarily reflect the views of the Deutsche Bundesbank.
JEL Classification numbers: C33, C50, E31, F41.
104 Bulletin
with mobile labour and the possibility for firms to relocate production abroad, will contain
wage claims and, hence, inflation. Second, inflation rates are strongly correlated across
countries (Ciccarelli and Mojon, 2010).
We analyze the global dimension of inflation in a traditional Phillips curve framework.
Studies using such a framework typically explain inflation with domestic demand and
supply conditions such as the output gap and cost push terms capturing productivity devel-
opments, input prices and labour market developments, but disregard that these factors
are not only determined nationally but also globally. We document the international com-
ovement of major determinants of inflation in 24 OECD countries between 1980Q3 and
2007Q1. We decompose them into global or common components (which are essentially
driven by global shocks or spillovers) and idiosyncratic components using factor analysis
and assess how much of the variance is explained by the common factors. Wethen include
(in our baseline) the common and idiosyncratic components of output gaps and unit labour
cost changes together with past inflation and import price inflation in a Phillips curve and
analyze their effects on domestic inflation.
We address the following questions: Are global variables/components able to explain
domestic inflation? How important are they relative to country-specific determinants of
inflation? Although our baseline model is a backward-looking Phillips curve, we also con-
sider inflation expectations in our model. In an extensive robustness analysis we assess, for
example, whether the influence of global (national) components or variables has increased
(decreased) over time. Finally, we look at the relevance of other international determinants
for domestic inflation such as foreign competition, global developments of interest rates,
exchange rates and oil shocks.
Our article is related to two literature strands that assess the global dimension of infla-
tion. The first strand uses Phillips curve models to investigate the role of foreign output
gaps, import price inflation and other global variables, such as changes in the terms of
trade, exchange rates, foreign inflation, foreign competition and import shares for domes-
tic inflation. Some studies apply traditional Phillips curve frameworks to panels of devel-
oped countries or individual developed countries.1Others employ micro-founded New
Keynesian models.2
Borio and Filardo (2007) find positive and significant effects of foreign output gaps
on domestic inflation between 1985 and 2005 in almost all 15 industrial countries consid-
ered (an exception being Germany) and the euro area. Their finding is confirmed for the
US and the sample period 1971 to 1999 by Gamber and Hung (2001). Ihrig et al. (2010)
and Gerlach et al. (2008) show that this result is, however, not robust with respect to the
measurement of the foreign output gap, to other variables included in the equation and to
the estimation period considered. Therefore, it is not surprising that the other articles find
either no or only a very weak influence of foreign output gaps on domestic inflation or
results that are not robust with respect to the specification. Results regarding import price
1Tootell (1998), Gamber and Hung (2001), Ball (2006), IMF (2006), Pain, Koske and Sollie (2006), Borio and
Filardo (2007), Mody and Ohnsorge (2007), Gerlach et al. (2008), Calza (2009), Ihrig et al. (2010).
2Balakrishnan and L´opez-Salido (2002), Gal´ıand L ´opez-Salido (2001), Razin andYuen (2002), Batini, Jackson
and Nickell (2005), Leith and Malley (2007), Rumler (2007), Gadzinski and Hoffmann (2008), Calza (2009) and
D´ees et al. (2009).
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford 2012
To continue reading
Request your trial