German Wage Moderation and European Imbalances: Feeding the Global VAR with Theory

• Published date: 01 March 2019
• Author: TIMO BETTENDORF,MIGUEL A. LEÓN‐LEDESMA
• Date: 01 March 2019
• DOI: http://doi.org/10.1111/jmcb.12517

DOI: 10.1111/jmcb.12517

TIMO BETTENDORF

MIGUEL A. LE ´

ON-LEDESMA

German Wage Moderation and European

Imbalances: Feeding the Global VAR with Theory

German labor market reforms in the 1990s and 2000s are generally believed

to havedriven the large increase in the dispersion of current account balances

in the Euro Area. Weinvestigate this hypothesis quantitatively.We develop a

three-regionopen economy New Keynesian model with search and matching

frictions from which we derive robust sign restrictions for wage bargaining

and matching efficiency shocks which we term wage moderation shocks.

We impose these restrictions on a Global VAR consisting of Germany and

eight EMU countries to identify a wage moderation shock in Germany.Our

results show that, although the German current account was significantly

affected by wage moderation shocks, their contribution to European current

account imbalances was negligible. We conclude that the German labor

market reforms cannot be the lone driver of European imbalances.

JEL codes: F10, F32, F41

Keywords: European imbalances, German wage moderation, DSGE,

global VAR,sign restrictions.

IT IS WIDELY ACKNOWLEDGEDTHAT internal current account im-

balances in Europe were an important factor behind the financial distress experienced

by countries in the Eurozone (see Figure 1).1What is more controversial, however,

is what the main drivers of these imbalances were. The IMF (2012) and ILO (2012)

The authors are grateful to an anonymous referee, Sandra Eickmeier, Carlo Favero,Reinhold Heinlein,

Hans-Martin Krolzig, Eric Leeper, Ron Smith, and Tong Wang as well as participants at the Deutsche

Bundesbank research seminar,the 17th International Conference on Macroeconomic Analysis and Interna-

tional Finance, the European Meeting of the Econometric Society, as well as the DIW Macro-econometric

Workshop for valuable comments and suggestions. This paper represents the authors’ personal opinions

and does not necessarily reflect the views of the Deutsche Bundesbank or its staff.

TIMO BETTENDORF is with DG-Economics, Deutsche Bundesbank (Email: Timo.Bettendorf@

bundesbank.de). MIGUEL A. LE´

ON-LEDESMA is with the School of Economics and MaGHiC, University of

Kent (Em ail: m.a.leon-ledesma@kent.ac.uk)

Received July 22, 2015; and accepted in revised form March 29, 2018.

1. See Chen, Milesi-Ferretti, and Tressel (2012), and Hobza and Zeugner (2014) for an overview of

trade and capital flows within the Eurozone.

Journal of Money, Credit and Banking, Vol. 51, Nos. 2–3 (March–April 2019)

C

2018 The Ohio State University

618 :MONEY,CREDIT AND BANKING

1993 1994 1995 1997 1998 2000 2001 2002 2004 2005 2006

Year

-10

-5

0

5

10

CA as % of GDP

Austria

Germany

Spain

Finland

France

Greece

Italy

Netherlands

Portugal

FIG. 1. Totalcurrent account balance as a percentage of GDP for nine Eurozone countries (1992Q1–2007Q2).

mention the increase in German competitiveness since the late 1990s as an important

determinant of these imbalances driven by German labor market reforms. Particu-

larly, the decline in German real wages, relativeto the Euro Area partners, is cited as

a key factor. In contrast, other commentators such as Wyplosz (2013) doubt this view

and argue that changes in competitiveness were the consequence and not the cause

of the problem.

We test the contribution of shocks to the German labor market,in the form of a re-

duction in workers’ wage bargaining powerand/or an increase in matching efficiency,

to Eurozone current account imbalances. We make use of a Global VAR (GVAR)

for nine EA countries2in order to measure the spillover effects of these shocks.

We identify these shocks in Germany by deriving minimal sign restrictions from a

three-region open economy New Keynesian DSGE model with search and matching

frictions. The model features wage bargaining and matching efficiency shocks which

we jointly term “wage moderation” shocks. These restrictions are then imposed on

the GVAR (see Eickmeier and Ng 2015) and we analyze the response of Eurozone

current accounts and quantify the contribution of these shocks to European imbal-

ances. This identification method follows Canova and Paustian (2011) and has the

advantage of being more agnostic about the model structure than estimated DSGE

models, which often requires knowledge of the exact specification of decision rules

and are prone to identification and specification problems. Our approach, in contrast,

consists of selecting a set of robust sign restrictions from the theoretical model which

are then used to identify wage moderation shocks in the structural GVAR, exploiting

the flexibility of the VAR approach. The structural GVAR approach is also best suited

for analyzing shock spillovers within the context of multicountry models.

2. We model Austria, Germany, Spain, Finland, France, Greece, Italy, the Netherlands, and Portugal.

Due to a lack of data, we do not model Belgium and Luxembourg.

TIMO BETTENDORF AND MIGUEL A. LE ´

ON-LEDESMA :619

Weshow that wage moderation shocks in Germany do generally cause an improve-

ment of the domestic current account, while foreign responses are heterogeneous.

However, they account for only a very small fraction of the current account balance

forecast error variances. Counterfactual analysis shows that the effect of these shocks

on the increasing dispersion of the Eurozone current accounts before the crisis is

essentially negligible. Weconclude that German wage moderation cannot be the lone

driver of European imbalances.

Related literature. While the role of the German wage moderation during the

late 1990s and early 2000s has been widely discussed by policy institutions (see,

e.g., IMF 2012, ILO 2012), the literature on its international effects is scarcer. As

mentioned above, the IMF and ILO as well as Bundesbank (2011) point out that the

German wage moderation has increased German competitiveness and thus translated

into high current account surpluses, while the current account balances of many

other European countries deteriorated. Similar conclusions are reached by Sabbatini

and Zollino (2010). Vogel (2011) employs a three-region version of QUEST to

investigate possible strategies for rebalancing the Euro Area. Among other strategies,

he investigates the theoretical outcome of wage moderation. His results indicate that

wage moderation should generally help to rebalance current accounts, as it affects

marginal cost of firms, which leads to competitiveness gains. This is in line with

Ivanova (2012) who, using panel regressions for 60 countries for the 1970–2009

period, finds that countries with more flexible labor markets tend to have larger

current account surpluses.

Gadatsch, St¨

ahler, and Weigert (2016) and Busl and Seymen (2013) make use of

policy simulations within a two-country monetary union DSGE model to analyze

the effect of German labor market reforms on its current account balance and that

of other member states. Their findings are in line with ours as they find a very

limited role for these reforms in driving current account imbalances. There is also

an important literature on the effects of German labor market reforms on the labor

market. Krebs and Scheffel (2013) find that the Hartz IV reform (see below) led to a

lower unemploymentrate and a higher job finding rate, which supports our assumption

that the labor market reforms increased matching efficiency. Fahr and Sunde (2009)

come to the same conclusion by analyzing the Hartz I/II and III reforms. Closer to

our approach, Kollmann et al. (2015) analyze the German current account balance

by estimating a three-region DSGE model for Germany, the Rest of the Euro Area,

and the Rest of the World.3Their results indicate that the German current account

surplus is mainly driven by shocks to the German savings rate, the rest of the world’s

demand for German exports, and supply shocks associated with labor market reforms.

An estimated DSGE framework allows for the identification of a large number of

structural shocks, which is a clear advantage over our approach. However, as noted

above, standard problems such as parameter identification and the assumption of

knowledge of the exact data generating process (DGP) of the data often make these

3. See also, for instance, Jacob and Peersman (2013) for an estimated two-country DSGE model to

analyze the dynamics of the U.S. trade balance.