Professional Documents
Culture Documents
Interconnected
World Economy
EDITOR
Ashoka Mody
I N T E R N A T I O N A L M O N E T A R Y F U N D
Cataloging-in-Publication Data
Joint Bank-Fund Library
HC286.G47 2013
Disclaimer: The views in this book are those of the authors and should not be
reported as or attributed to the International Monetary Fund, its Executive
Board, or the governments of any of its members.
Foreword v
Christian Kastrop
Preface ix
Ashoka Mody
4 What Does the Crisis Tell Us about the German Labor Market? ..... 77
Martin Schindler
Introduction ........................................................................................................................... 77
Background ............................................................................................................................ 78
Recent Developments........................................................................................................ 80
Understanding German Labor Market Dynamics.................................................... 85
Conclusion .............................................................................................................................. 92
References .............................................................................................................................. 94
Appendix ............................................................................................................................... 96
iii
8 Discussion ..............................................................................................239
Comment on Chapters 2 and 3 ....................................................................................239
Malte Hübner
Comment on Chapter 4...................................................................................................242
Werner Eichhorst
Comment on Chapter 5...................................................................................................247
Felix Hüfner
Comment on Chapter 7...................................................................................................252
Carsten-Patrick Meier
References ............................................................................................................................256
Index ...................................................................................................................................261
Germany in its postwar history has a remarkable economic story to tell, not just
about success but also about failure, about ground won and lost. The successful
post-war recovery, although driven by a clear free-market confession, has also
been part of the German policy approach of Ordnungspolitik, which includes a
clear commitment to economic fairness, social safety nets, independent wage
negotiations between employers and employees, and strong regulations prevent-
ing the misuse of economic power in the most economically relevant areas.
Then we saw a first slump, starting in the mid-1960s, due to global develop-
ments, such as the crises in the coal and steel industries and the loss of the Bretton
Woods currency regime at the end of the decade. The slump was prolonged by
home-made problems, not least when trying to keep non-competitive sectors alive
with high subsidies. Today coal is still active in Germany despite high production
costs, and nobody knows if all plants really will be shut down as planned at the
end of this decade. Renewables, also heavily subsidized, might not do the trick
alone after the closing of all nuclear power plants soon after the Fukushima event.
The 1970s brought more trouble with the oil and wage shocks, and they saw
the failure of an anti-cyclic macro policy that was also (mis-)used as a tool for
goodies of all kind; the cyclical deficits were never paid back in the good times.
Last but not least, the decade saw the huge build-up of an over-generous social
safety net, at first supported by high growth and stable population figures but
soon becoming fragile as birth rates and potential growth rates began shrinking—
something that has continued to this day. And the issue hasn’t left us to this day:
the question of how to push potential growth is still the most relevant.
The 1970s also, as a result, had to manage a level of deficit never seen until
the late 1960s, driven by following the new gross investment “(‘golden’) fiscal
deficit rule”: that is, the Keynesian-motivated new debt rule, which replaced the
old conservative rule that required any deficit to be repaid by revenues created
through investment. The new, weakened focus of indirect repayment through
growth and taxes, while it may not be completely wrong in economic terms,
proved to be a seed of failure. It too could easily be misused, not only as an instru-
ment of political compromise of any incoming new term of legislation but also
during economic downturns, when cyclically short-term-communicated deficits
“surprisingly” turned out to be structural—the classic TTT (timely, temporary,
targeted) mistake mentioned in every economic textbook.
The 1980s started with stagflation and debt that had already accumulated
quite high and grew further in spite of the changing tide of economic thinking in
the direction of neoclassical/monetarist concepts (Phillips-Curve and rational
expectations) which—at least in communication—were reflected in the political
economy, where non-Keynesian/Ricardian effects were heralded, culminating in
statements like, “The economics driven by rational behavior needs no macro.”
v
Nevertheless, despite the strong wording, Germany did not really take up the
supply-side economy concept and almost completely missed the structural/social
reform period of the 1980s that many other European countries entered, such as
the Netherlands and Scandinavia. One cushion Germany had at that time was the
relatively favorable export side of its economy, even though the deutsche mark
slowly but steadily appreciated, taxing away part of gained competitiveness and
making structural reforms even more pressing.
Before the situation grew too serious, unification brought a big boom. It was
construction-driven, and while it did not lead to a general bubble it led to huge
overcapacities, which lowered growth in the second half of the decade. The boom
wiped the structural reform agenda off the political Top Ten list. Because of its
complex and sophisticated system of (federal) checks and balances, therefore
politically somewhat “slow,” Germany’s federal government had to spend its lim-
ited available political capital on unification, leaving no room for other issues.
Seen from outside, it might have been possible to use unification as a catalyst
for reform, but the political economy could never deliver. Political capacity was
bound by and led to short-term solutions based on already endangered social
systems and even higher debt, instead of fundamental reforms and a tax-based
financing of unification.
The mid-1990s saw Germany taking over Europe’s “red lantern” on economic
indicators. The “German Disease” was on the front pages of tabloids and The
Economist.
The late 1990s brought changes: wage developments were continuously mod-
erated, both specialized SMEs and big industries cleared their books, and profits
went up. The supply side won, the demand side lost. The nucleus of the next
export boom was born, as German SMEs became the little champions offering
emerging markets the technology they needed. Adding to that were Social
Security and labor market reforms, tax relief for corporates, and a whole program
for renovating Germany, even if, as measured against the bold proposals of
economists, the program didn’t look so convincing and was heavily blurred by
political compromise at the time of design.
But time would show that it was not just the German overperformance but
also the underperformance of competing countries (and therefore a no-longer-
appreciating currency) that did the trick for the new German economic (export)
miracle. For the average German, the crisis of 2008 and the years following—the
worst world crisis since the 1930s—was and still is something he would watch
with some surprise on television, but it never came closer. And this time Germany
was lucky, with a well designed macro-strategy that also delivered on the psycho-
logical side and was firmly anchored by the new fiscal rule, the so-called “debt
brake.” It was quite a new experience, too, one never seen in the decades before,
when the typically pessimistic but now optimistic (over-optimistic?) Germans
kick-started lagging internal demand and consumption at the height of the crisis.
This book on Germany by Ashoka Mody and current and former IMF and
internationally regarded experts is one of the most detailed reviews of recent
German economic history. It not only offers an excellent empirical analysis based
on long-term developments but frames that analysis in the context of our actual
post-crisis, globalized world. It points clearly at Germany’s successes, but it also
points out the remaining weaknesses as well as those upcoming if the reform
momentum is lost.
Germany has a role to play, not just in Europe but in the interconnected
global economy. This should not be forgotten. And that role is not only for its
national economy but for the whole region, its trading partners, and its export/
import markets. Germany’s macro- and microeconomic policy does matter, even
if measurable spillovers are small. So there is a point in sustaining its reform
momentum and not getting lost again on the macro-, structural, supply-and-
demand, and all underlying (micro-) factors.
Germany indeed should push the structural reform agenda, for example
through public investment, reforming the health and service sectors, removing
existing market barriers, speeding up research and development, and bridging the
existing market gap in founding new-tech companies. It should also adopt family-
and birth-friendly policies together with better approaches to migration. Last but
not least, Germany should engage in a full renovation of its education system,
while keeping its already well working instruments (“dual education”) that pre-
vent youth unemployment.
And, of course, there is the issue of debt and deficit. The new German debt
brake, which now is also enshrined in the European fiscal compact, might move
the countries concerned to an adequate soundness of public finance and foster
long-term sustainability and the quality and efficiency of public finance as neces-
sary complements.
The reader might use this book to build up his own picture of Germany. The
description and analysis are valuable for rethinking different policy approaches,
and for reaching the best solutions (or perhaps the second- or third-best, as poli-
ticians of all stripes will most naturally and legitimately ensure) to current and
upcoming problems, including the next “black swan” problems, in a world that
is interconnected and leveraged.
It is my hope that the authors of this book keep an open-minded eye on
Germany and continue to offer analyses and advice, since in Germany as with
other medium and large sized countries a more inward look sometimes tends to
become dominant. This is no longer affordable in a globalized, interconnected
world. Nobody can do better alone, not with the real and monetary market world
and not with the public sector. Maybe this is one of the main crisis lessons. All
the more, then, do we have to strive for credible, convincing, and communica-
ble—but most of all problem-solving—economic concepts and strategies, not
just on a national but on an international scale.
Christian Kastrop
Deputy Director-General, Economics Department, and
Director of Public Finance, Macroeconomics and Research Directorate
German Federal Ministry of Finance
The papers in this volume were written by IMF staff. But at all stages, this was
a collaborative enterprise with the German authorities. Early versions of the
papers were discussed at the Bundesbank, where critical comments and alterna-
tive viewpoints were offered. That consultative process culminated in a major—
and unprecedented—conference at the Federal Ministry of Finance during the
2011 Article IV Consultation with Germany. Christian Kastrop made that event
possible. For that initiative and for the many vigorous conversations, even on
controversial issues, I am most grateful to the German authorities.
At the conference, a number of senior German scholars joined the discussions
as commentators and session chairs. Their comments are included as part of this
book. Thanks are due in particular to the session chairs, Christoph Schmidt,
Klaus Eckhardt, Ansgar Belke, and Beatrice Weder di Mauro, who kindly took
time to moderate the exchange following the papers.
The conference closed with a lively panel discussion. Juha Kähkönen, then
Deputy Director in the IMF’s European Department, moderated that discussion.
The distinguished participants included Markus Kerber, then Head of Department
of Fiscal and Economic Policy at the Federal Ministry of Finance (now CEO and
Director General of Federation of German Industry, BDI), Thomas Mayer,
Deutsche Bank Research, and André Sapir, University of Brussels and Bruegel.
Finally, I must acknowledge with much gratitude the contributions of Fabian
Bornhorst in making this volume possible. He was legitimately a co-editor but his
modesty has prevented him from agreeing to share that credit with me. I also
recall with great affection my many colleagues who joined this venture and many
other such intellectual and operational adventures.
Ashoka Mody
ix
In the nearly 70 years since World War II ended, the German—earlier the West
German—economy has enjoyed dynamic phases, interspersed with periods of
slow, even anemic, growth. Germany’s continuing ability to maintain competitive
manufactured exports has been crucial to its resilience and dynamism, but it has
not always been sufficient. This dependence on exports as the economic driver of
the German economy helps in understanding its postwar evolution, including its
recovery following the post-Lehman collapse. It also points to the challenges that
Germany faces: the calls to do more for the European and global economy and
the risk that the rise of manufacturing capabilities in emerging markets will even-
tually wear down Germany’s stronghold. In this introductory essay, we review
Germany’s growth record over four phases and sneak a look into the future.
• For about a decade and a half after World War II, until the early 1960s, the
economy responded spectacularly by making up lost ground. German com-
panies exploited their traditional advantage in capital and durable goods
production for sale to a growing world economy and also to meet pent-up
domestic demand.
• In the next four decades—until about 2003—economic performance
gradually became less impressive. The catch-up potential necessarily waned
and the oil shocks of the 1970s raised costs and reduced global demand. By
the mid-1980s, with rising wages, West Germany was losing its competitive
edge. The unification of West and East Germany in 1990 provided a short
boost, but then the German economy again went into a swoon.
• Starting around 2004, a strong global economy, combined with reforms that
helped Germany regain competitiveness, provided a new opportunity to
German exporters. With success came a historically large current account
surplus, which was criticized for contributing to global imbalances. But the
growth dynamic was summarily interrupted in early 2008 by the onset of
the “Great Recession,” when the collapse in global trade also swept
Germany.
Fabian Bornhorst is an economist in the IMF’s European Department. Ashoka Mody was deputy
director in the European Department and then the Research Department of the IMF when this
chapter was written.
• The recovery from the collapse has once again demonstrated German resil-
ience. German companies deepened their export links to the growing
economies of Asia, and policymakers have supported stabilization and
growth through measures to maintain employment and a sizeable fiscal
stimulus.
New challenges have to be faced. Some are not unique to Germany. With its
reliance on exports, a high savings rate, and a current account surplus, Germany
shares similarities with Japan, to the extent that one country can be like another.
Looking ahead, like their Japanese counterparts, German exporters face height-
ened competition, especially from Asia but also from Emerging Europe. And as
in Japan, a rapidly aging population will imply a smaller workforce and changes
in savings and investment patterns with far-reaching consequences. But Germany’s
challenges arise also from its unique role in Europe: in particular, Germany’s
expected contribution to the European recovery and a resolution of the euro area
crisis remain controversial.
As this preview suggests, throughout the postwar era, (West) Germany has
benefited greatly from its relationship with the global economy, but there have
been lingering questions whether the German contributions to global and
European growth have been commensurate. From its early catch-up phase,
Germany’s formidable export engine has been its consistent driver of growth. But
with almost equal consistency, Germany has run current account surpluses, with
imports lagging exports. As a consequence, despite the size of its economy,
Germany’s ability to act as global locomotive has been limited. Germany has
contributed to global well-being in important ways—notably through its foreign
direct investment and, in particular, the production linkages its companies have
built throughout Europe. Nevertheless, the calls on Germany to do more remain
vigorous, especially as the European crisis has persisted.
The argument in this book is that the German economy has evolved along a
particular historical trajectory that tends to reinforce its growth patterns. The
innovative manufacturing sector has remained a consistent source of growth, but
it has been heavily dependent on external demand. The volatility of external
demand, in turn, has caused German GDP growth to be relatively volatile by the
standards of advanced countries. Such volatility has likely been a factor in keeping
domestic demand growth—especially consumption growth—at surprisingly low
levels. As a consequence, the domestically oriented segment of Germany’s econo-
my has lagged. This has been so especially in the production and delivery of ser-
vices. In turn, this has reinforced the drive for foreign markets. Looking ahead, a
more domestically driven growth dynamic will be good for Germany and for the
global economy.
The key challenge for Germany, then, is to generate new domestic sources of
growth. One recent episode—reunification—raised domestic demand, but that
boom proved unsustainable as the economic problems associated with unification
surfaced. Labor market reforms in response to the ensuing period of stagnation
were a bold response, and have proven largely successful. But much of the gain in
competitiveness translated into further growth in manufactured exports and
current account surpluses. Therefore, despite the notable service sector advances
achieved since the mid-1990s, a further broad-based impetus to services growth
is required. Absent such an effort, German growth will remain constrained, and
Germany will tend to run current account surpluses while playing only a modest
role in spurring growth elsewhere.
Just as the German economy is similar to that of Japan—dependent on
exports, running current account surpluses, and generating limited international
growth spillovers—the contrast with the United States is marked. The United
States is characterized by more reliance on domestic consumption and, as a con-
sequence, persistent current account deficits but high international growth spill-
overs. Despite popular characterization of the U.S. financial sector’s casino capi-
talism, both the U.S. GDP and its stock market have been less volatile than in
either Germany or Japan. In part, this reflects the greater reliance on domestic
consumption, which tends to be more stable than exports. But the United States
has also been diversified, with innovations in globally leading technologies giving
it an edge in productivity growth, especially in the services sector. In the future,
Germany will perhaps draw on approaches to fostering innovation practiced in
the United States, while also maintaining its greater emphasis on social safety
nets, where lessons may be available from the Nordic countries.
In this introductory chapter, we step back from the themes of immediate
policy focus to provide a broad overview of German economic growth and inter-
national connections over the past half century. The intent is to describe the
historical trajectory that has brought Germany to its present balance of strengths
and weaknesses, and to use that analysis to explore the best way forward. Much
of the work underlying this book and its principal policy conclusions was devel-
oped for the IMF’s Article IV consultation with Germany in 2011 (IMF 2011).
The book serves to present the more extensive background analysis in an inte-
grated form.
This chapter is presented in four main parts, reflecting the four relatively dis-
tinct phases of German growth. First, the immediate postwar period witnessed a
rapid catch-up with the United States and, importantly, reestablished its export
prowess, which has remained Germany’s wellspring ever since. Second, growth
slowed starting in the early 1960s, and the oil shocks of the 1970s and unification
in 1990 proved to be particularly onerous. Third, the reemergence from this
setback on the back of extended global prosperity was aided by wage moderation
and broad-based domestic reform. And fourth, the collapse triggered by the Great
Recession and the subsequent recovery have brought Germany to a new phase
with new challenges to tackle.
−2
−4
−6
1950 1960 1970 1980 1990 2000 2010
Aid flowing to Europe through the European Recovery Program (also known
as the Marshall Plan) was an important catalyst in addressing infrastructure bot-
tlenecks and reviving trade. Germany also reclaimed its historic advantages as an
exporter of capital and durable goods. Between 1948 and 1950, the pent-up
demand for consumption and investment drew in substantial imports, but
exports started growing rapidly right from the start, with exceptionally high
annual growth rates (Table 1.1). By 1950, exports had exceeded imports, and the
German trade surplus established then has persisted, with some ups and downs,
ever since. By 1960, West Germany’s shares of world exports and imports exceed-
ed those of the German Reich before World War II (Giersch, Paque, and
Schmieding, 1992).
Eichengreen (2006) describes this as a period of extensive European growth,
with Germany in the vanguard. He defines extensive growth as that which
deploys relatively well established technologies: “It is the process of raising output
by putting more people to work at familiar tasks and raising labor productivity by
building more factories along the lines of existing factories” (Eichengreen, 2006,
p. 6). Millions of refugees arrived in Germany and, in the immediate postwar
TABLE 1.1
period, internal labor mobility was also high. Nearly full employment conditions
were achieved, with the unemployment rate down to below 1 percent in 1960. In
reorganizing production and rebuilding the capital stock, efficiency gains were
significant and quickly realized.
Several factors facilitated this outcome. Not only was plentiful labor available,
the workers were also industrially literate—or were readily trainable through the
traditional vocational training systems. Such systems met the need of the moment
precisely because the challenge at hand was not to build new widgets but to build
known widgets in larger quantities with incremental technical improvements in a
learning-by-doing process. At the same time, an extensive network of relationship-
based banking systems provided the needed patient capital to finance the invest-
ments. This confluence led to modest wage demands, which allowed profitable
firms the resources and the confidence to invest in their workforce and in growth.
Germany was particularly well suited to take advantage of these conditions.
Giersch, Paque, and Schmieding (1992, p. 89) write: “Germany’s traditional
strength in the production of capital goods paid off handsomely in the 1950s, when
these goods were in particularly high demand on the world market. In addition, the
change in relative prices testifies to the improving quality and sophistication of these
exports goods.” In a similar vein, Eichengreen (2006, pp. 93–94) elaborates:
The country already possessed the relevant range of industries, from coal and steel
to transport equipment and electrical machinery…. Small and medium-sized firms
competed with legions of other small and medium-sized firms, requiring them to
price aggressively and reduce costs in order to survive. In turn this rendered German
firms highly competitive on international markets. Exports rose from 9 percent of
national income in 1950 to 19 percent in 1960. External conditions were also
propitious for German recovery. Investment demand was high throughout Europe,
aiding German firms specializing in the production of capital goods. The Korean
crisis stimulated demand for capital goods worldwide. And just when Germany’s
expanding industrial sector began diversifying into the production of consumer
goods, private consumption surged across Europe, reflecting rising incomes and in
turn helping to sustain the growth of German exports….Investment and exports
were the fast-growing components of aggregate demand, and government and pri-
vate consumption the slow-growing ones.
Thus, while all of Europe did well during this period, Germany did particu-
larly well. Other countries that grew about as rapidly included Austria, which had
close economic ties to Germany, and Italy. While Germany was leveraging off an
established industrial capability, Italian growth was due to the shift of resources
from agriculture to industry.
THE SLOWDOWN
The precise timing of the shift is difficult to pin down, but the growth benefits
of the German miracle leveled off in the 1960s. The indicators are clear. From an
average growth of about 9 percent in the early 1950s, GDP growth fell to about
4.5 percent by the mid-1960s (Figure 1.1). German growth, which had set the
pace for Europe in the immediate postwar years, now fell below the European
average growth rate (Figure 1.2). The rapid process of German catch-up with the
United States stalled, and even began a modest reversal in the 1980s (Figure 1.3).
The slowdown was to be expected. Growth could not persist at the early giddy
levels and, as Germany became richer, the convergence possibilities diminished,
while the latecomers started their own catch-up process.
8
10th–90th percentile of OECD economies
7
Germany
6 European economies (excluding Germany)
5
−1
1970 1975 1980 1985 1990 1995 2000 2005 2010
Figure 1.2 Growth in a Comparative Perspective, 1970–2010 (Five year average, percent)
Sources: IMF, World Economic Outlook; and IMF staff estimates.
Figure 1.3 Per Capita GDP Relative to the U.S., 1960–2010 (Real per capita GDP to
U.S. per capita GDP, in percent, PPP constant prices)
Sources: Organisation for Economic Co-operation and Development; and IMF, World Economic Outlook.
Note: PPP: purchasing power parity.
a
Maximum excludes Luxembourg.
1
That is, about half of the potential growth rate of the U.S. economy.
10
10th–90th percentile of OECD economies
Germany
United States
5 Japan
China
−5
−10
1970 1975 1980 1985 1990 1995 2000 2005 2010
100
Germany
90
Emerging and developing economies
80
Advanced economies
70
60
50
40
30
20
10
0
1960 1970 1980 1990 2000 2010
Figure 1.5 Trade Openness, 1960–2010 (Value of imports and exports, percent
of GDP)
Sources: IMF, World Economic Outlook; and IMF staff estimates.
By the end of the 1990s, Germany was dubbed “the sick man of Europe.”
Growth came to a standstill, and the economy underperformed for much of the
early 2000s (Figure 1.6). Despite high unemployment, the institutional responses
of the previous decades had raised wages to unsustainably high levels, and for the
first time in decades Germany lost a competitive edge (Figure 1.7). Indeed, even
as the global economy recovered in the early 2000s, Germany failed to respond,
reinforcing the view that Germany faced serious problems. Pessimism about
−2 −200
−400
−4
−600
−6 −800
−8 −1000
1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012
Figure 1.6 Growth Expecations in Germany, 1992–2012 (One year consensus forecast
vs. actual growth, percent)
Sources: Consensus Forecasts; IMF, World Economic Outlook; and IMF staff calculations.
4.0
1991–2000 2001–2010
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Austria France Italy Germany Netherlands Spain
Figure 1.7 Change in Unit Labor Costs, Germany and Selected European Countries,
1991–2000 and 2001–2010 (Average annual percentage change)
Source: Organisation for Economic Co-operation and Development.
REEMERGENCE
The “reemergence” period, from 2004 to 2008, is a relatively short one, more so
when compared to the previous period lasting nearly four decades, which we
treated above in summary fashion. Yet this short period is noteworthy because of
the rapid turnaround experienced and because, along with the post-Great
Recession recovery, it defined the prevailing sense of German dynamism. The
transformation not only reestablished the German economy’s strengths but also
brought calls from the international community for Germany to play a more
active role as a “global citizen.” However, as German growth accelerated following
years of secular decline, Germany developed large current account surpluses—
large even by its own historical standards. Thus, while it came to be regarded with
new respect, it was also caught in the storm of “global imbalances.” The essays in
this book reflect on both the dynamics of this reemergence phase and on
Germany’s role in the international economy.
We focus on four themes characterizing this period, reflecting the break and
renewed sense of confidence as well as the significant continuities. First, manu-
facturing productivity growth remained solid, and services productivity growth
picked up modestly. Second, wage restraint was widespread. Third, global growth
buoyancy was key to maintaining export growth. And, to a large extent, current
account surpluses were more a consequence of global developments and less of
“distortionary” domestic policies.
In Chapter 3 of this book, Hélène Poirson shows that productivity performance
varied greatly across sectors. Manufacturing productivity continued to grow respect-
ably, not just in absolute terms but also relative to international benchmarks.
Productivity growth was relatively low in the newly emerging sectors of communi-
cations and information technology and was also relatively low in the services sector.
These sectoral distinctions are important, because they highlight the challenge
ahead. Despite Germany’s manufacturing prowess, the share of manufacturing
value-added in German GDP has steadily declined over the past several decades
(Figure 1.8). This is not surprising. With the rise of lower-cost manufacturing in
the newly industrializing nations of East Asia, manufacturing has played a small-
er role in all advanced economies. Germany is remarkable only to the extent that
the manufacturing share of total value-added remains somewhat higher than in
Japan. Nevertheless, it is salient that Germany’s high manufacturing productivity
growth cannot be a dependable source of greater well-being in the future as
manufacturing inevitably continues to cede ground to international competitors.
In this respect, Poirson’s analysis has an optimistic note. She finds some evi-
dence of a rise in services productivity starting in the early 2000s. However, her
analysis does not establish a clear trend. And since the Great Recession created so
much dislocation, any trend may only be discernible in a few more years. Poirson
offers advice that is in line with that of other students of productivity growth:
greater use of information technology in the delivery of services and more impe-
tus to small and innovative service firms through incubation in higher centers of
learning and greater access to venture capital. More controversially, she suggests a
role for the government in procuring services with a public-good purpose, espe-
cially where that may help establish open standards.
The second theme of this period is wage restraint. Wage moderation has
become central to characterizing Germany—and hence to the policy measures it
must adopt in deference to its global commitments. It is the case that German
35
30
25
20
15
1970 1975 1980 1985 1990 1995 2000 2005 2010
Figure 1.8 Importance of the Manufacturing Sector, Germany, Japan, and the U.S.,
1970–2010 (Value added manufacturing in percent of total value added, three year
average)
Sources: Organisation for Economic Co-operation and Development; and IMF staff calculations.
wages have been relatively steady in the past several years (Figures 1.9 and 1.10).
Yet, as described above, the restraint followed years of significant wage increases.
Thus, even after the restraint, German wages are among the highest in Europe.
What is remarkable about the recent years is the rapid rise in wages elsewhere in
Europe. The Irish rise is particularly explosive. But wage increases elsewhere in the
European periphery clearly outstripped productivity growth in those economies.
There is an open question why German wage increases were modest during
this phase. One explanation is the introduction of labor market reforms, in
60 140
Average salary per worker 2010 ('000 EUR)
50 120
Percent increase, 1995–2010 (right-hand scale)
100
40
80
30
60
20
40
10 20
0 0
nd
ria
ce
ly
l
ga
nd
an
ai
ec
Ita
iu
an
la
st
Sp
rtu
rla
lg
re
Ire
Au
Fr
Be
er
Po
he
G
et
N
Figure 1.9 Wage Levels and Wage Growth, Selected European Countries, 1995–2010
Sources: Organisation for Economic Co-operation and Development; and IMF staff estimates.
60
55
50
45
1991 1996 2001 2006 2011
2
Slow German growth in the early 2000s, while the world economy was beginning to embark on a
new dynamic phase, led to a concerted policy effort to renew growth. After years of political deadlock
on key reform initiatives, the Agenda 2010 reform program agreed in the early 2000s led to wide-
ranging changes in the labor market institutions, a reorganization of the economy, and changes to the
pension system.
3
Posen is less convinced of the effectiveness of other elements of the reform package, including the
so-called active labor market policies (see also Jacobi and Kluve, 2006).
12
World market share in specialized
CHN
10 DEU
USA
product varieties
6
ITA
4 JPN FRA
NLD
2 KOR ESP
MYS IND
POL
IRL PRT
0 GRC
0 100 200 300 400 500
Figure 1.11 Product Specialization and Market Share, 2007 (Product varieties and
market shares)
Sources: IMF staff estimates based on Standard International Trade Classification 4 level trade data for 2007.
Note: CHN: China; DEU: Germany; ESP: Spain; FRA: France; GRC: Greece; IND: India; IRL: Ireland; ITA: Italy;
JPN: Japan; KOR: Korea; MYS: Malaysia; NLD: Netherlands; POL: Poland; PRT: Portugal; USA: United States.
500
World trade effect Increased market share
400
300
200
100
−100
na d ia a s y e in ly al e ia n s d
hi lan Ind ore and an eec pa Ita tug anc ays apa tate lan
C Po K erl erm Gr S r r l J S r e
h Po F
M
a d I
et G
ni
te
N U
7
EME and DCa US UK Eurozone World
6
5
4
3
2
1
0
−1
−2
1979–88 1989–98 1999–2008 2009–11
Figure 1.13 GDP Growth Rates: World and Major Regions, 1979–2011 (Average
annual percentage growth).
Sources: IMF, World Economic Outlook; and IMF staff estimates.
a
Emerging markets and developing countries.
16
12
10
0
1991–1998 1999–2008 2009–2011
Figure 1.14 Import Growth Rates, World, U.S., and Euro Area (Average annual
percentage growth)
Sources: Netherlands Bureau for Economic Policy Analysis (CPB); and IMF staff estimates.
maintain its traditional exports to Europe, where it was able to fend off
competition from Asian sources. But while it continued European imports of
intermediate goods, especially from Visegrad countries (Czech Republic, Hungary,
Poland, and Slovakia), German imports increasingly tilted toward products pro-
duced most cost-effectively by China (Figure 1.15), which became Germany’s
second most important import partner in 2011, after the Netherlands, overtaking
France. In other words, German exports stayed largely insulated from Asian and
lower-wage European competition due to Germany’s specialization, but much of
−2
−4
China Visegrad Rest of the ESP, GRC, Rest of United
countries World ITA, IRL, PRT EU-27 States
Figure 1.15 Change in German Import Sources, Selected Countries and Regions,
2000–2009 (Change in Germany’s import shares between 2000–2009)
Source: IMF staff estimates.
Note: Visegrad countries include Czech Republic, Hungary, Poland, and Slovakia.
12
05 06
10 07
04 08 98 00 99
03 97
Unemployment rate
95 85
02 96 01 87 86
8 88
09 10 11 94 84 89
93 90
6 83 92 91
82
78 77
4 81 79
80 76
75
2
74 72
73
71
0
–2 0 2 4 6
Private consumption growth rate
14
Germany
12
West Germany
10
0
1950 1960 1970 1980 1990 2000 2010
high, amplified by the uncertainty over the outcome of labor market reforms. At
the same time, the unification shock and specialization of the manufacturing sec-
tor toward external markets raised growth volatility. Growth volatility had been
lower in Germany than in the United States until the end of the 1980s, but since
then it has been typically higher—and even higher than in Japan since the crisis
(Figure 1.18 and Carare and Mody, 2010).
High GDP volatility in Germany is therefore partly the result of the growth
model Germany pursues. Income is volatile, the level of uncertainty is high,
saving ratios are higher, and consumption growth is low. These dynamics are
4.0
Germany United States Japan
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
1975–1979 1980–1984 1985–1989 1990–1994a 1995–1999 2000–2004 2005–2008 2009–2011
Figure 1.18 Output Volatility, Germany, U.S., and Japan (Average of 20 quarter
rolling standard deviation of year-to-year GDP growth)
Sources: Organisation for Economic Co-operation and Development; and IMF staff estimates.
a
German unification falls in this period.
35
Germany United States Japan
30
25
20
15
10
0
1993–1997 1998–2002 2003–2007 2008–2011
Figure 1.19 Stock Market Volatility (Five year average of the VDAX, VIX, and VNKY)
Sources: Datastream; IMF staff estimates.
Note: VDAX: volatility index for the DAX (Germany’s stock index); VIX: volatility index for Standard and Poor’s
500 index; VNKY: volatility index for Nikkei index, Japan.
12
10
0
1980–1984 1985–1990 1990–1994 1995–1999 2000–2004 2005–2009 2010–2011
Figure 1.20 Short-Term Interest Rates since 1980, Germany, U.S., and Japan, 1980–
2011 (Yield on Government paper with residual maturity of one year)
Sources: Haver Analytics; IMF staff estimates.
The Recovery
In the immediate wake of the crisis, the fall in German output was substantial,
but the recovery was impressive. The decline in output, at over 5 percent, was
greater than in the United States and in France and about the same as in the
United Kingdom (Figure 1.21). Only with respect to Japan was the fall less severe.
Note that both the United Kingdom and Japan have continued to underperform,
as if their initial contraction represented, in part, some longer-term downward
shift. German exports fell by 14 percent, and German investment also fell pre-
cipitously. However, by the end of 2011, Germany had not only recovered past
its precrisis output level but, despite its greater initial fall, the German excess over
precrisis GDP was greater than in the United States or France.
It is important to recognize that the German output recovery impresses
mainly because it compares favorably with even weaker performances elsewhere
in the advanced world. Almost four years after the start of the crisis, the German
output level is now only a few percentage points above its precrisis levels.
102
100
98
96
94
92
90
88
2008 2009 2010 2011
Figure 1.21 GDP during the Great Recession, Germany and Selected Economies (Real
GDP, 2007=100)
Sources: IMF, World Economic Outlook; and IMF staff calculations.
103
102
101
100
99
98
97
96
95
94
2008 2009 2010 2011
Figure 1.22 Employment during the Great Recession, Germany and Selected Economies
(2007=100)
Sources: IMF, World Economic Outlook; and IMF staff calculations.
4
The Sonderfonds Finanzmarktstabilisierung (SoFFin) was created in late 2008 to stabilize the financial
system by providing bank guarantees, funds for bank recapitalizations, and the transfer of risky assets
by creating “bad banks” (IMF 2010).
5
Allen (2005) argues that there has been a historical aversion in Germany to Keynesian demand-
oriented remedies in favor of a policy more driven by the goal of expanding the economy’s supply
capacity (see also Carlin and Soskice, 2009). Yet, this has also meant a significant social safety net,
which provides so-called automatic stabilizers to protect the vulnerable and thereby operates to stabi-
lize the economy. Moreover, as this crisis showed, further discretionary stimulus is pragmatically used
in Germany, even if it is often downplayed in public discourse.
0
es
ed
n
alia
rea
da
ly
m
an
pa
nc
Ita
do
tat
na
nc
Ko
str
Fra
rm
Ja
ing
dS
va
Ca
Au
Ge
Ad
dK
ite
20
Un
ite
G-
Un
Figure 1.23 Fiscal Stimulus During the Great Recession, Germany and Selected
Countries (2009–2011, cumulative, percent of GDP)
Source: IMF, Fiscal Monitor, November 2010.
50
Total exports Exports to China
40
30
20
10
−10
−20
−30
2005 2006 2007 2008 2009 2010 2011
180
Private consumption
170
Employment
160
150
140
130
120
110
100
90
80
1970 1975 1980 1985 1990 1995 2000 2005 2010
crisis hit, the government increased both the duration and the coverage of
subsidies. This allowed work to be shared and human capital to be maintained.6
Boeri and Bruecker (2011) point out that similar approaches were adopted in a
number of countries, but with less success. They argue that complementary
institutions are needed to make this policy effective, pointing especially to
employment protection legislation and collective bargaining. Schindler, in
Chapter 4, points to a second factor. He notes that the strategy and response of
German firms was important. Firms and workers had prior agreements that
they could deviate from collectively bargained work arrangements to avoid lay-
offs, and introduced work-time accounts at the firm level. For workers in the
core labor market, this was a reasonable approach: in the face uncertainty in
export markets, they opted for job security and flexibility at the firm level.
Thus, work-sharing schemes were already a part of the relationship between
German firms and their workers. Burda and Hunt (2011) note, moreover, that
because workers had credits on these accounts, firing them in the midst of the
downturn would have required compensating them for the credits. In this set-
ting, the subsidies were helpful in reinforcing the preexisting contractual rela-
tionship. Third, as Schindler concludes in Chapter 2, for Germany the Great
Recession was mainly an external export shock, not a supply shock. Despite the
uncertainty, firms largely saw this as a temporary shock and relied on accumu-
lated work time and short work schemes instead of layoffs. With the recovery,
this judgment was vindicated. Burda and Hunt (2011) argue that the precrisis
expansion of exports was viewed by firms as temporary, so firms held back their
hiring—this is consistent with the relatively low level of corporate investment
in the boom years, as Ivanova has noted. Thus, firms did not have as much of
a need to fire people.
Looking beyond the near term, Schindler in Chapter 4 suggests that
German unemployment had begun a secular downward trend prior to the cri-
sis, possibly triggered by the Hartz reforms. The evidence for this conclusion is
still preliminary, although it is plausible since these reforms increased labor
flexibility both among the core work force and also at its margins. Once again,
it helps to look beyond Germany. As noted above, other European countries
adopted similar labor market reforms in the mid-1990s. Boeri (2009) finds that
similar reforms to the Italian labor market led to favorable labor market
responses in that country: he estimates that the level of unemployment consis-
tent with non-accelerating inflation came down. However, Boeri goes on to
argue that Italian firms’ essential lack of dynamism implied that employment
did not respond in meaningful numbers. Here again, Germany’s historical
strengths and corporate-labor relationships distinguish it from other European
nations.
6
This would be of particular importance if skilled labor was in short supply. In chapter 3 of this vol-
ume, Schindler notes that the firms’ more intensive use of the core labor force, combined with an
adjustment in the temporary workforce (typically less skilled) is consistent with such considerations.
7
We have addressed above the issue of intra-European imbalances. As noted, a policy that weakens
German competitiveness may reduce German surpluses but not necessarily improve the current
account positions of the peripheral countries.
−200
−400
Jan 05 Jan 06 Jan 07 Jan 08 Jan 09 Jan 10 Jan 11 Jan 12
Figure 1.26 Target 2 Balances, Germany and Selected Euro Area Countries, 2005–
2012 (EUR billion)
Sources: IFS; National Central Banks; and staff estimates.
600
Financial flows ex target (+ outflow)
Change in target claims (+ increase)
400
Current account
200
−200
−400
2005:Q1 2006:Q1 2007:Q1 2008:Q1 2009:Q1 2010:Q1 2011:Q1 2012:Q1
Figure 1.27 Current and Financial Account of Germany, Finland, and the Netherlands
since 2005 (Billions of euros, rolling four quarter sum)
Sources: Haver Analytics; IFS; IMF staff calculations.
600
Financial flows ex target (+ outflow)
400 Change in target claims (+ increase)
Current account
200
−200
−400
−600
2005:Q1 2006:Q1 2007:Q1 2008:Q1 2009:Q1 2010:Q1 2011:Q1 2012:Q1
Figure 1.28 Current and Financial Account of Greece, Ireland, Italy, Portugal and
Spain since 2005 (Billions of euros, rolling 4 quarter sum)
Sources: Haver Analytics; IFS; IMF staff calculations.
24
Germany Japan United States
22
20
18
16
14
12
10
1990 1995 2000 2005 2010
Figure 1.29 Population Ageing, Germany, Japan, and the U.S., 1990–2010 (Popula-
tion over 65 years, in percent of total population)
Source: Organisation for Economic Co-operation and Development.
Recent research (Braun, Ikeda, and Joines, 2007) shows that the single most
important cause of the decline in Japanese savings rates is lower savings as people
get older. Yet this has not necessarily meant a consumption boom. Rather, even
with the fall in savings, Japan has experienced subdued demand and deflation.
The relationship between ageing and deflation is, of course, more tenuous, but it
is presumably related to the shrinking workforce (Figures 1.30 and 1.31). Germany
is about to go through a more intensive ageing phase, following Japan. The
40
Percent 65+ 2010 Percent 65+ 2050
30
20
10
0
United States Germany Japan OECD
Figure 1.30 Population Ageing, Germany, U.S., Japan and the OECD, 2010–2050
(Population over 65, in percent of total population)
Source: Organisation for Economic Co-operation and Development Labour Force and Demographic Database,
2010.
250
Forecast France
Japan Germany
200
UK US
150
100
50
1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050
Figure 1.32 Female Labor Force Participation, Germany and Nordic and OECD
Countries, 1970–2010 (18–64 years old, percent)
Sources: Organisation for Economic Co-operation and Development; and IMF staff estimates.
85
80
75
70
1970 1975 1980 1985 1990 1995 2000 2005 2010
Figure 1.33 Male Labor Force Participation, Germany and Nordic and OECD
Countries, 1970–2010 (18–64 years old, percent)
Sources: Organisation for Economic Co-operation and Development; and IMF staff estimates.
pressures are rising as the population ages, and migration, even by the most opti-
mistic scenarios, cannot fill the emerging gap in the working-age population. This
will call for policies, possibly following the Nordic model, directed toward
increasing the labor force participation especially of women (Figure 1.32 and
1.33). This in turn implies adjustments to effective tax rates to enhance female
labor supply along with supportive child care policy.
an industrial strength that has been tested in global competition over decades
and has proven resilient to severe setbacks. Most recently, the period after
German unification saw the economy in doldrums with seemingly intractable
high unemployment rates. Yet the German economy has continually displayed a
capacity to regenerate itself. Its emergence from the Great Recession—and espe-
cially the ability to generate jobs—has been widely, and rightly, lauded. In turn,
this resilience is due to a network of innovative firms that have adapted to
global changes, for example through increased sourcing from Eastern Europe
and exploiting new market opportunities in Asia. It is also due to the ability of
firms and labor to find common ground. Finally, it is due to a pragmatic bent
in policymaking.
Following World War II, Germany experienced a particularly robust
reconstruction phase. This was followed by an extended period of slowing
growth. First, there was a natural slowing after the rapid postwar catch-up.
Then the external shocks of rising oil prices and the collapse of the Bretton
Woods system caused further deceleration. The unification episode was
unique to Germany. After a short-lived unification boom in the early 1990s,
growth fell again. When in the 1990s the world economy began a new expan-
sion phase, the German economy appeared ill-prepared to take advantage of
the favorable environment. But by the mid-2000s, economic reform and
corporate and labor responses to the changed circumstances led to a German
reemergence. Since then, the German economy has displayed considerable
strength and maturity, not least by robustly navigating the crisis of 2008–9.
In particular, German employment levels weathered the recent crisis surpris-
ingly well.
With its success, Germany has been called on to assist the global recovery,
questions about its current account surplus have resurfaced, and Germany’s wage
moderation and proposed pace of fiscal consolidation following the recovery
from the 2008–9 crisis have, at times, been regarded with concern. However,
while proposals for more rapidly raising German wages or delaying fiscal con-
solidation have a plausibility, a closer examination suggests that they could com-
promise German strengths with dubious short-term stimulative value for other
countries.
The real German challenge is to strengthen its areas of weakness. Although
significantly down from their peak levels, unemployment rates remain elevated,
including when judged by Germany’s own past history. These high rates, along
with the emergence of relatively low-paying and temporary jobs, also act as a drag
on German consumption growth. The key is to counteract medium-term growth
constraints in a way that also supports sustainable rebalancing via higher domestic
demand growth. Meeting this challenge will require a new generation of pragma-
tism in policy decisions. A greater emphasis is needed on innovation that extends
Germany beyond its traditional manufacturing strengths and on a new model of
social safety nets, drawing possibly on the Nordic experience, to increase labor
force participation needed to counter rapid ageing. Such actions would also be
good for Europe and the global economy.
REFERENCES
Alesina, Alberto F., Edward L. Glaeser, and Bruce Sacerdote, 2006, “Work and Leisure in the
U.S. and Europe: Why So Different?” in NBER Macroeconomics Annual 2005, ed. by Mark
Gertler and Kenneth Rogoff (Cambridge, MA: MIT Press), pp. 1–100.
Allen, Christopher S., 2005, “Ordo-Liberalism Trumps Keynesianism: Economic Policy in the
Federal Republic of Germany,” In Monetary Union in Crisis: The European Union as a
Neo-Liberal Construction, ed. by Bernard Moss London (Palgrave), pp. 199–221.
Boeri, Tito, 2009, “Comments to ‘Two Italian Puzzles: Are Productivity Growth and
Competitiveness Really So depressed?’ by Lorenzo Codogno,” in Italy in EMU: The Challenges
of Adjustment and Growth, ed. by Marco Buti (London: Palgrave Macmillan).
———, and Herbert Bruecker, 2011, “Short-time Work Benefits Revisited: Ssome Lessons
from the Great Recession, Economic Policy, vol. 26, pp. 697–765.
Bornhorst, Fabian, and Ashoka Mody, 2012, “Target Imbalances: Financing the Capital-
Account Reversal in Europe,” article published online by VoxEU.org. http://voxeu.org/
article/target-imbalances-financing-capital-account-reversal-europe.
Braun, R. Anton, Daisuke Ikeda, and Douglas H. Joines, 2007, “The Saving Rate in Japan:
Why It Has Fallen and Why It Will Remain Low,” CIRJE Working Paper F-535, CIRJE,
Faculty of Economics, University of Tokyo.
Burda, Michael C., and Jennifer Hunt, 2011, “What Explains the German Labor Market
Miracle in the Great Recession?” SFB 649 Discussion Paper No. 2011-031, Humboldt
Universität Berlin.
Buti, Marco, ed., 2009, Italy in EMU: The Challenges of Adjustment and Growth (London:
Palgrave Macmillan).
Carare, Alina, and Ashoka Mody, 2010, “Spillovers of Domestic Shocks: Will They Counteract
the ‘Great Moderation?’” IMF Working Paper No. 10/78 (Washington, DC: International
Monetary Fund).
Carlin, Wendy, and David Soskice, 2009, “German Economic Performance: Disentangling the
Role of Supply-Side Reforms, Macroeconomic Policy and Coordinated Economy Institutions,”
Socio-Economic Review, Vol. 7, pp. 67–99.
Dew-Becker, Ian, and Robert J. Gordon, 2012, “The role of Labor Market Changes in the
Slowdown of European Productivity Growth,” Review of Economics and Institutions, Vol. 3,
No. 2, pp. 1–45.
Eichengreen, Barry, 2006, The European Economy Since 1945: Coordinated Capitalism and
Beyond (Princeton: Princeton University Press).
———, and Kevin O’Rourke, 2010, “A Tale of Two Depressions,” article published online by
VoxEU.org. http://voxeu.org/index.php?q=node/3421.
Giersch, Herbert, Karl-Heinz Paque, and Holger Schmieding, 1992, The Fading Miracle: Four
Decades of Market Economy in Germany (Cambridge, UK: Cambridge University Press).
International Monetary Fund, 2010, “Germany: Staff report of the 2010 Article IV
Consultation” (Washington, DC: International Monetary Fund). http://www.imf.org/
external/pubs/ft/scr/2010/cr1085.pdf.
———, 2011, “Germany: Staff report of the 2011 Article IV Consultation” (Washington, DC:
International Monetary Fund). http://www.imf.org/external/pubs/ft/scr/2011/cr11168.pdf.
———, 2012, “Germany: Staff report of the 2012 Article IV Consultation” (Washington, DC:
International Monetary Fund). http://www.imf.org/external/pubs/ft/scr/2012/cr12161.pdf.
Jacobi, Lena, and Jochen Kluve, 2006, “Before and After the Hartz Reforms: The Performance
of Active Labour Market Policy in Germany,” IZA Discussion Paper No. 2100.
Krugman, Paul, 2010, “Ever Expanding Government,” September 10 posting in “The
Conscience of a Liberal” blog, New York Times. http://krugman.blogs.nytimes.com/
2010/09/10/ever-expanding-government/
Nickell, Stephen, Luca Nunziata and Wolfgang Ochel, 2005, “Unemployment in the OECD
Since the 1960s: What Do We Know? Economic Journal Vol. 115, No. 500, pp. 1–27.
Posen, Adam, 2007, “Hartz IV Worked—As Far As It Went,” Op-ed in Die Welt, March 14.
http://www.iie.com/publications/opeds/oped.cfm?ResearchID=717.
Sinn, Hans Werner, 2006, “The Pathological Export Boom and the Bazaar Effect – How to
Solve the German Puzzle,” CESifo Working Paper No. 1708 (Munich: CESifo Group).
Weitzman, Martin L., 2008, “Subjective Expectations and Asset Return Puzzles,” American
Economic Review, Vol. 97, No. 4, pp. 1102–30.
While the German economy, with a GDP decline of 4.75 percent in 2009, was
among the hardest hit during the financial crisis, the impact of the crisis on its
potential output is estimated in this chapter to have been comparatively mild and
transitory, with virtually no potential output loss in the long term. This outcome
reflects both a more flexible economic structure, resulting from past reforms, and,
importantly, the nature of the shock—a temporary drop in external demand. The
temporary nature of the shock meant that no large-scale structural shifts in employ-
ment were necessary, so employers were inclined to retain workers and adjust hours
per worker instead. This incentive was buttressed by policy measures during the
crisis, but more importantly also by longer-standing labor market improvements
regarding hourly flexibility. With employment levels and, especially, hours, returning
to pre-crisis levels, the impact on potential GDP was limited, and the evidence sug-
gests that a medium-term convergence of potential output to its pre-crisis trend is
likely. However, there is little indication that long-term potential growth will rise
above its meager historical rate of about 1.25 percent. Increasing potential growth
in the medium and long term will require sustained efforts to reverse the declining
secular trend in Germany’s productivity growth and to raise its low total factor pro-
ductivity (TFP) growth.
INTRODUCTION
Germany’s economic structure has changed dramatically over the past four
decades. Until about the mid-1990s, the contribution of external trade to aggre-
gate growth—in either direction—was similar to that in most other advanced
The author has benefitted from comments and suggestions by Ashoka Mody, Hélène Poirson, Anna
Ivanova and participants at the Germany in an Interconnected World Conference at the Ministry of
Finance in Berlin (May 2011), especially the paper’s discussant, Malte Hübner. Roberto Garcia-Saltos,
Petar Manchev, and colleagues at the IMF’s Research Department Economic Modeling Unit provided
helpful advice on implementing the Matlab code. Susan Becker provided excellent research assistance.
35
1.5
1971–1995 1996–2010
1.2
0.9
0.6
0.3
0.0
Germany Japan United States United Euro area
Kingdom (excluding
Germany)
1
Time-series data exhibit a V-shaped pattern, with the German foreign contribution to growth more
elevated in the 1970s and 80s, though still substantially below the more recent uptick in the trade
surplus. The increase in the foreign contribution to growth is even more pronounced in relative terms,
i.e., the foreign growth contribution relative to total growth. For a more detailed analysis of Germany’s
historical growth drivers, see Vitek (2010), which also illustrates the increasing importance over time
of foreign demand in German business cycle fluctuations, especially during the Great Recession.
2
Exports of goods and services dropped by over 15 percent in real terms during that period, with the
impact on net exports somewhat offset by an import decline of just under 6 percent.
3
The theme of strong external growth drivers alongside relative moderate contributions from domestic
sources raises deeper questions regarding their causal connection: has weak domestic demand led firms
to focus on external markets? Or has a dominant external sector, e.g., through its increased volatility, led
individuals to raise (precautionary) savings and lower their consumption? Or are there unrelated causes
for the two? This is an important theme that, however, cannot be addressed in the context of this chapter.
4.0
1971–1995 1996–2010
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
United United Euro area Germany Japan
States Kingdom (excluding
Germany)
8
6 Domestic
Net exports
4
2
0
–2
–4
–6
–8
–10
France Germany Japan Sweden United United
Kingdom States
The reliance on external demand affects how growth shocks are transmitted to
potential GDP. Generally speaking, any given change in actual GDP may reflect
either a fluctuation of actual GDP around a stable potential GDP path (that is,
fluctuations in demand) or a change in potential GDP (that is, fluctuations in
supply), or both. The assessment of what type of shock is the source of the
observed fluctuations in GDP therefore has strong implications for the assess-
ment of the extent to which potential GDP may be affected.
Building on the notion that external demand has been the main driver of
German business cycles—at least since the mid-1990s—the analysis in this paper
concludes that potential growth was only minimally affected by the crisis, in
contrast to many other countries during the crisis (including the United States)
and in contrast to the effects typically associated with past financial crises.4 Thus,
the demand shock has pulled down actual GDP, but it has not set in motion firm
activities that would substantially alter Germany’s growth potential in the medi-
um term, as reflected particularly in the moderate crisis impact on capital stock
and employment. Consistent with this, a robust recovery in actual GDP has
pulled up potential growth, which has been seen to overshoot its long-term rate
in the medium term, making up for the small potential growth losses during the
crisis. The key result is that there is likely to be no permanent loss of potential
GDP relative to the pre-crisis trend.
On a more somber note, however, the underlying assumption of a low long-
term potential growth rate of only about 1¼ percent leaves Germany lagging
behind many of its peers, especially the United States, which, despite a more
severe crisis impact, is set to continue to grow at a faster rate than Germany.
Given Germany’s adverse demographic pressures, only continued structural
improvements to raise productivity and the supply of labor can lift German GDP
to a higher trajectory.
This chapter provides some background on the existing literature on potential
GDP, including its definition and selected findings; describes the methodology
and the main results; and presents these results within the context of a growth
accounting exercise that decomposes German growth into that deriving from
total factor productivity dynamics, (physical) capital accumulation, and employ-
ment growth. An appendix provides further details on the underlying model that
was used to estimate potential GDP.
4
For example, Furceri and Mourougane (2009) estimate that financial crises have on average perma-
nently reduced potential GDP by 1.5–2.4 percent in OECD economies during 1960–2007.
0
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
−2
−4
−6
Figure 2.4 Germany and the United States: GDP Growth (Percent)
Sources: IMF, World Economic Outlook; Gordon (2008); and IMF staff estimates.
shrink as the damage that the financial crisis has caused to medium-term
potential growth is likely to be more substantial in the United States than in
Germany.5
Over longer horizons, such trend-based estimates should conform closely
with what would be obtained based on the economic definition of potential
output, since on average an economy should not operate at a level substan-
tially different from its potential over long periods. Thus, over long horizons,
simple trend estimates are likely to be reasonable estimates of potential output
and can serve as inputs in the type of modeling approach used here to anchor
the estimates. However, as shown below, more substantial differences often
arise over shorter frequencies, where atheoretical methods may be unable to
sufficiently differentiate between changes in actual output and changes in
potential output.
A number of studies have examined potential output in Germany. Based on
a production function methodology, De Masi (1997) estimates potential growth
in Germany at 2.25–2.5 percent, which is consistent with the long-term growth
rate reported above, given her 1980–1997 sample. Also following a production-
function approach and considering the 1986–2003 time period, Baghli, Cahn
and Villetelle (2006) find a secular decline in potential growth rates in Germany,
although their post-unification average of about 1.9 percent is somewhat higher
than the average calculated above, reflecting the absence of post-2003 data,
especially the low growth rates until 2005 and the more recent crisis (see also
Cahn and Saint-Guilhem, 2007). That potential output growth estimates are
subject to considerable uncertainty is exemplified by Horn, Logeay, and Tober
(2007) who find, for various versions of their model, annual potential growth
rates during 2006–2010 of 2.4, 2.1 and 1.3 percent, respectively, owing to dif-
ferent assumptions on total factor productivity (TFP) dynamics and the level of
the NAIRU.6 At the lower end are both the Sachverständigenrat (2011) and
Deutsche Bank Research (2011) studies, which find potential growth to be on
the order of 1.25 percent, at least starting in 2000 (and about 1.5 percent during
the 1990s).7
5
The notion that Germany’s potential output held up well during the crisis but lags from a longer-
term perspective also holds in a broader cross-country comparison. For example, as the discussion in
Section D of this chapter indicates, German potential GDP growth from 2000 until the onset of the
crisis fell behind that of the United States, the United Kingdom, Japan, and the Euro Area as a whole.
As also shown there, TFP growth, the main driver of a country’s long-term growth potential, is low
in Germany compared to other OECD economies.
6
Also, while Horn, Logeay and Tober (2007) utilize data dating back to 1970, they do not explicitly
model a structural break at the time of reunification. (NAIRU = Non-Accelerating-Inflation Rate of
Unemployment).
7
Consistent with the main message of this chapter, both Sachverständigenrat (2011) and Deutsche
Bank Research (2011) expect potential growth to be little affected by the crisis. El-Shagi (2011) comes
to a similar conclusion.
TABLE 2.1
3.0
1.0
−1.0
−3.0
−5.0
−7.0
1997:Q1
1997:Q4
1998:Q3
1999:Q2
2000:Q1
2000:Q4
2001:Q3
2002:Q2
2003:Q1
2003:Q4
2004:Q3
2005:Q2
2006:Q1
2006:Q4
2007:Q3
2008:Q2
2009:Q1
2009:Q4
2010:Q3
Figure 2.5 Output Gaps Based on Alternative Methodologies (Percent of potential
GDP)
Sources: IMF staff estimates.
(supply) (see the appendix regarding where these variables enter the model). For
example, a prior belief that supply is more volatile than demand would lead the
model to assign much of the observed volatility of real GDP to potential GDP
fluctuations. Put differently, actual and estimated potential GDP would move in
sync, and the output gap would be less volatile. Conversely, if most GDP volatil-
ity is attributed to demand shocks, then potential GDP (supply) would remain
more stable, resulting in a more volatile output gap.
Both the model assumptions and the choice of methodology matter, affecting
the results that are obtained. Alternative methods of calculating potential GDP
(and the resulting output gaps) include simple filters that take a moving average
of actual GDP and methods based on the Hodrick-Prescott (HP) approach (see
Hodrick and Prescott, 1981), which estimates a statistically smoothed series.8
Also, within the framework used here, as discussed, different priors on the distri-
bution of shocks to GDP (demand versus supply shocks) matter.
Notably, with the exception of the fairly crude moving-average approach, all
approaches deliver fairly similar results historically—they do differ, however, in
their output gap dynamics during the current crisis (Figure 2.5). That is, all
atheoretical approaches assign most of the variation in actual GDP to supply
variations (i.e., variations in potential GDP) and thus exhibit a similar pattern as
that resulting from the multivariate approach with supply-sided priors. Given the
nature of the shock that Germany experienced during this crisis, these other
approaches are likely to be misleading. That is, not utilizing information on the
underlying economics, including especially the nature of the shock, provides a
very different view of the evolution of potential GDP than otherwise.
8
That is, the HP-filter does not use economic information in its estimation. It also suffers from end-
of-sample measurement problems.
10.0
8.0
6.0
4.0
2.0
0.0
2001:Q1
2001:Q3
2002:Q1
2002:Q3
2003:Q1
2003:Q3
2004:Q1
2004:Q3
2005:Q1
2005:Q3
2006:Q1
2006:Q3
2007:Q1
2007:Q3
2008:Q1
2008:Q3
2009:Q1
2009:Q3
2010:Q1
2010:Q3
Figure 2.6 Unemployment Rates (Percent)
Sources: IMF; World Economic Outlook; and IMF staff estimates.
9
Using a multivariate state-space model, El-Shagi (2011) reaches a very similar conclusion for
Germany and finds “that the crisis mostly opened the output gap and did not reduce potential GDP
(as suggested by other models)” (p. 737).
5
Euro Area Germany United States
3
−1
−3
−5
−7
2005
2006
2007
2008
2009
2010
2012
2013
2014
2015
2016
2011
10
Job creation is a more sluggish process than job destruction, suggesting some persistence in unem-
ployment even after a (temporary) shock has subsided; also, unemployment is often associated with
losses in human capital associated with unemployment spells. Both factors imply a reduction in effec-
tive labor supply and thus potential output. In “What Does The Crisis Tell Us About the German
Labor Market?”—Chapter 4 in this volume—I examine in more detail some of the factors behind the
moderate labor market response. In addition to the potential GDP dynamics, past reforms, including
the Hartz IV in 2005, were likely contributors.
11
The work underlying this chapter was completed in early 2011.
−2
−4
−6
−8
Consumer Intermediate Capital goods Consumer
durables goods nondurables
25
Agriculture Manufacturing Total Value Added Services
20
15
10
5
0
−5
−10
−15
−20
1991:Q1
1992:Q1
1993:Q1
1994:Q1
1995:Q1
1996:Q1
1997:Q1
1998:Q1
1999:Q1
2000:Q1
2001:Q1
2002:Q1
2003:Q1
2004:Q1
2005:Q1
2006:Q1
2007:Q1
2008:Q1
2009:Q1
2010:Q1
about 0.8 percent in 2010.12 Second-round effects on wages have also so far
remained moderate, not least reflecting a still-positive unemployment gap (i.e.,
actual unemployment above equilibrium). And lastly, aggregate capacity utiliza-
tion is set to remain substantially below precrisis levels. However, continued and
sharper commodity price pressures do represent an upside risk to core inflation.
Based on the model estimated here, the crisis impact on potential GDP is more
limited in Germany than in the United States. From a historical (post-unification)
average of about 1.3 percent, German potential growth declined moderately to
12
An increase in the output gap by one percentage point is associated with about a 2/5 percentage
point increase in inflation.
2012:Q3
2013:Q3
2014:Q3
2015:Q3
2016:Q3
2011:Q3
Figure 2.10 Germany: Gap Measures and Inflation (Percent)
Sources: IMF, World Economic Outlook; Deutsche Bundesbank; and IMF staff estimates.
about 1.0 percent in 2009. However, in the medium term, pulled up by a strong
growth recovery, potential growth is expected to move slightly above historical
rates, peaking at about 1.7 percent in 2013 before eventually converging back to
its long-term rate of 1.3 percent by 2016. By contrast, the U.S. economy is much
more closed; and the crisis there originated from the financial and real estate sec-
tors, suggesting a domestic supply shock. Thus, U.S. potential growth declined
more dramatically, to about 1.5 percent in 2009 from over 2.5 percent histori-
cally. In the medium term, potential growth in the United States is projected to
pick up to about 2¼ percent by 2016 (Figures 2.11 and 2.12).
3
Germany USA
0
2003 2005 2007 2009 2011 2013 2015
125 Germany
Germany pre-crisis trend
120
United States
115
United States pre-crisis trend
110
105
100
95
90
2003 2005 2007 2009 2011 2013 2015
Figure 2.12 Germany and the U.S.: Potential GDP Relative to Pre-crisis Trends
(2007=100)
Sources: IMF, World Economic Outlook; and IMF staff estimates.
13
Different labor measures can be used, but do not affect the estimates substantially. When labor
input is measured as the labor force rather than employment, its contribution is higher at 0.2 percent,
but TFP growth remains at about 0.5 percent annually. Estimated average TFP growth rises to a
slightly higher value of about 0.75 percent annually when labor is measured as total hours, reflecting
a secular decline in total hours worked.
TABLE 2.2
Decomposing growth
(percent, quarter-on-quarter, annualized)
GDP TFP Capital Labor
2000:Q1–2005:Q4 0.9 0.4 0.5 0.1
2006:Q1–2008:Q2 4.7 2.2 0.9 0.2
2008:Q3–2009:Q1 –5.5 –6.0 0.5 0.0
2009:Q2–2010:Q2 3.9 3.3 0.3 0.3
2010:Q3–2010:Q4 2.1 1.1 0.4 0.4
Average 1991–2010 1.3 0.1 0.6 0.1
Source: IMF staff estimates.
Note: TFP: total factor productivity.
6
5 Staff estimates EU Klems Conference Board
4 GGDC OECD
3
2
1
0
−1
−2
−3
−4
80
84
94
96
98
00
06
82
86
02
04
88
90
08
10
92
19
19
19
19
19
20
20
19
19
20
20
19
19
20
20
19
These estimates are broadly consistent with a variety of other estimates. The
EU KLEMS project, the Conference Board, the Groningen Growth and
Development Centre (GGDC) and the OECD are the main alternative sources
of TFP estimates. While these measures differ somewhat in detail, they provide
broadly similar estimates and are highly correlated.14 In particular, they all point
to a secular downward trend in TFP growth that only briefly picked up in the
run-up to the crisis. Consistent with our estimates of a limited impact of the
crisis on potential GDP, the post-crisis TFP estimates exhibit a sharp rebound15
(Figure 2.13).
14
Differences in a number of variables and parameters can affect TFP estimates, including capital and
labor shares and capital stock estimates, with the latter especially sensitive to different assumptions
(such as the initial stock and the rate of depreciation).
15
According to staff estimates, TFP declined by more than 5 percent in 2009, but grew by nearly 3
percent in 2010.
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Ki den
d tria
A m
Au tes
Be lia
m
G rm her an
an (O s
F D)
N Fr )
Ze ce
D and
C rk
n
ly
f
m ny land
ad
ai
af
Ita
do
iu
a
ra
(IM EC
n
G Ne ap
a
Sp
ni us
m
st
lg
ew a
te we
al
an
St
st
ng
en
S
te
t
d
y
er a
U
ni
U
Figure 2.14 Total Factor Productivity Growth (Average annual growth, 1992–2009,
percent)
Source: Organisation for Economic Co-operation and Development; and IMF staff estimates.
16
See also Schindler (2012) for an argument that the Hartz reforms (which lowered social transfer
payments and deregulated temporary employment) had a long-lasting effect on job creation that is
still being felt. In particular, in that view, the secular downward trend since 2005 represents the tran-
sition to a lower steady-state unemployment rate.
17
High-skilled immigration, in addition to expanding the labor supply, may also contribute to TFP
growth. See West’s (2011) case for revamping the U.S. immigration system.
120
USA UK Germany Japan Euro Area
115
110
105
100
95
90
85
80
2000 2002 2004 2006 2008 2010 2012 2014 2016
CONCLUSION
Germany has mastered the crisis well, but long-term challenges remain.
Reflecting the nature of the shock and the structure of its economy, Germany
has emerged from the crisis remarkably unblemished. However, it is set to
return to its previous trajectory of only moderate growth. Raising that growth
rate will require policy measures on many fronts, including higher labor-force
participation, higher investment rates, and—especially—stronger productivity
growth.
More recent developments in Germany’s growth recovery also pose some
downside risks to the view taken here. The medium-term growth prospects have
fallen short of most observers’ expectations, inducing downward revisions of
many analysts’ growth projections, including those by the IMF. The absence of a
permanent loss in the level of potential output depended on above-average poten-
tial growth in the near term, to compensate for the (mild) potential growth
slowdown during the crisis. To the extent that lower-than-expected growth also
reduces potential growth, potential output might not fully recover to its precrisis
trend. Such downside risks underscore the need for further reforms to accelerate
long-term potential growth in Germany.
REFERENCES
Baghli, Mustapha, Christophe Cahn, and Jean-Pierre Villetelle, 2006, “Estimating Potential
Output with a Production Function for France, Germany and Italy,” Banque de France
Working Paper No. 146 (Paris: Banque de France).
Benes, Jaromir, Kevin Clinton, Roberto Garcia-Saltos, Marianna Johnson, Douglas Laxton,
Petar Manchev, and Troy Matheson, 2010, “Estimating Potential Output with a Multivariate
Filter,” IMF Working Paper No. 10/285 (Washington, DC: International Monetary Fund).
Billmeier, Andreas, 2004, “Ghostbusting: Which Output Gap Measure Really Matters?,” IMF
Working Paper No. 04/146 (Washington, DC: International Monetary Fund).
Bosworth, Barry, and Susan M. Collins, 2003, “The Empirics of Growth: An Update”
(Washington, DC: Brookings Institution).
Cahn, Christophe, and Arthur Saint-Guilhem, 2007, “Potential Output Growth in Several
Industrialised Countries: A Comparison,” ECB Working Paper No. 828 (Frankfurt am Main:
European Central Bank).
De Masi, Paula, 1997, “IMF Estimates of Potential Output: Theory and Practice,” IMF
Working Paper No. 97/177 (Washington, DC: International Monetary Fund).
Deutsche Bank Research, 2011, “Outlook 2011: German Growth Remains Robust,” Germany:
Current Issues, February 14, 2011, downloaded from http://www.dbresearch.com.
El-Shagi, Makram, 2011, “Did the Crisis Affect Potential Output?,” Applied Economics Letters,
Vol. 18, pp. 735−38.
Furceri, Davide, and Annabelle Mourougane, 2009, “The Effect of Financial Crises on Potential
Output: New Empirical Evidence from OECD Countries,” OECD Economics Department
Working Paper No. 699 (Paris: Organisation for Economic Co-operation and Development).
Gordon, Robert J., 2008, “The Slowest Potential Output Growth in U.S. History: Measurement
and Interpretation,” Paper prepared for presentation at the CSIP Symposium, “The Outlook
for Future Productivity Growth” (San Francisco: Federal Reserve Bank of San Francisco).
Hodrick, Robert J., and Prescott, Edward C., 1981, “Postwar U. S. Business Cycles: An
Empirical Investigation.” Discussion Paper No. 451 (Minneapolis: University of Minnesota).
Horn, Gustav, Camille Logeay and Silke Tober, 2007, “Estimating Germany’s Potential
Output,” IMK Working Paper No. 2/2007 (Washington, DC: International Monetary
Fund).
Ljung, Lennart, 1999, System Identification: Theory for the User (Princeton, New Jersey: Prentice-
Hall).
Mankiw, N. Gregory, 2002, Macroeconomics, 5th edition (New York: Worth Publishers).
Sachverständigenrat (Sachverständigenrat zur Begutachtung der gesamtwirtschaftlichen
Entwicklung), 2011, “Herausforderungen des demografischen Wandels,” Expertise im Auftrag
der Bundesregierung, May 2011.
Vitek, Francis, 2010, “Output and Unemployment Dynamics During the Great Recession: A
Panel Unobserved Components Analysis,” IMF Working Paper No. 10/185 (Washington,
DC: International Monetary Fund).
West, Darrell M., 2011, “Creating a ‘Brain Gain’ for U.S. Employers: The Role of Immigration,”
Brookings Policy Brief No. 178 (Washington, DC: The Brookings Institution).
APPENDIX
This appendix briefly summarizes the key features of the model. A more detailed
description is provided in Benes and others (2010). The model is built around
three gaps—the output gap (y), the unemployment gap (u), and the capacity
utilization gap (c)—and three identifying equations:
The inflation equation relates the level and the change of the output gap to
core inflation:
π 4t = π 4t −1 + βyt + Ω( yt − yt −1 ) + εtπ 4 .
The dynamic Okun’s law defines the relationship between the current unem-
ployment rate and the output gap. Based on Okun’s law, an unemployment
equation links the unemployment gap to the output gap:
ut = x1ut −1 + x 2 yt + εtu .
Finally, the model also relies on a capacity utilization equation, on the assump-
tion that capacity utilization contains important information that can help
improve the potential output and output gap estimates. The equation takes the
following form:
ct = k1ct −1 + k2 yt + εtc .
Given the three identifying equations, the equilibrium variables are assumed
to evolve dynamically as follows. A stochastic process including transitory (level)
shocks and more persistent shocks guides the evolution of equilibrium unemploy-
ment (U t ) (the NAIRU equation):
ω λ
U t = U t −1 + GtU − yt −1 − (U t −1 − U SS ) + εUt
100 100
(long waves) from the steady-state growth rate GSSY . Similar dynamic equations
are specified for equilibrium capacity utilization.
The full model is estimated by regularized maximum likelihood (Ljung,
1999), a Bayesian methodology. This method requires the user to define prior
distributions of the parameters. While this can improve the estimation procedure
by preventing parameters from wandering into nonsensical regions, the choice of
priors has also non-negligible implications for the final estimates as the data are
uninformative about some parameters. The choice of priors matters also in the
German case.
In addition to the prior distributions of parameters, the analyst has to provide
values for θ and the steady-state (long-run) unemployment rate (U SS ) and poten-
tial GDP growth rates ( GSSY ), which were set to 0.55, 6.7 percent and 1.3 percent,
respectively. While values especially for U SS and GSSY matter conceptually, as the
(endogenous) estimates converge to these (exogenously given) values in the long
term, from a practical point of view, the dynamics over the time horizon of inter-
est are relatively little affected by the choice of the steady-state values.
Germany had the advantage of a catch-up phase through the early 1970s, but there-
after, and especially since the mid-1990s, productivity growth has been relatively slow.
This aggregate picture masks considerable intersectoral differences. This chapter reports
that while Germany has maintained an enduring strength in its traditional manufac-
turing competencies, it has lagged in services delivery and, perhaps more importantly,
in the so-called knowledge economy—both in the production of information and com-
munication technology (ICT) goods and in the use of ICT to raise productivity in
private services provision. These “new” areas have been key engines of productivity
growth since the mid-1990s in the United States and some other advanced countries.
An uptick in German services productivity growth occurred from 2005 to 2007. This
is indicative of the scope and potential for progress, but it could have been a cyclical
phenomenon. To foster more efficient production and widespread use of ICT in ser-
vices delivery will require complementary measures: (a) further development of venture
capital and private equity markets (backed by a more efficient insolvency process);
(b) increased commercial use of intellectual property rights held by university and
research institutions; (c) removal of uncertainties regarding tax treatment; and
(d) further European integration in services provision.
INTRODUCTION
In the years after World War II, Germany increased its productivity at a fast clip,
thereby rapidly shrinking the United States’ lead in income levels. However, this
productivity catch-up was interrupted in the mid-1990s. Between 1995 and
2007, Germany’s private economy experienced annual productivity growth of 1.7
percent, slightly above the European average but lagging behind the United
States, the United Kingdom, Sweden, Austria, and Finland (Inklaar, Timmer, and
van Ark, 2008, Table 1, page 142; and Molagoda and Perez, 2011, Table V.1,
page 70).
For their constructive comments, the author wishes to thank Fabian Bornhorst, Malte Hübner, Anna
Ivanova, Ashoka Mody, Martin Schindler, and participants in the Germany in an Interconnected
World Conference at the BMF in Berlin (May 2011).
55
1
Jorgenson, Ho, and Stiroh (2004) first highlighted the role of the production and use of ICT in
accounting for productivity gains in the case of the United States.
and Molagoda and Perez, 2011).2 Unlike in the United States, where total factor
productivity (TFP) growth in the services sector accelerated after 1995, in Europe
it declined (especially in distribution, finance, and business services). Several
studies have argued that low TFP growth in the service sectors in EU countries in
turn reflects the delay in investing in ICT assets and implementing the related
new technologies and processes (McKinsey Global Institute, 2010).
These delays have been attributed, among other factors, to the relatively less
attractive conditions in Europe for venture capital and other market-based
sources of financing for high-growth, high-risk projects (Box 3.2).3 In addition to
the availability of long-term financing, traditional explanations for Europe’s lag-
ging productivity growth also emphasize policy and institutional factors such as
strict regulations in product and labor markets.4 But these seem less relevant in
the case of Germany, which is one of the most deregulated advanced countries in
retail trade, according to OECD indicators, and where some reforms (e.g., trans-
port services deregulation) constitute an example of best practice in Europe. It is
true, however, that despite a recent improvement Germany remains more heavily
regulated in professional services than most other advanced countries.
This study applies the industry-based approach to Germany during the
1980–2007 period to help shed light on the historical sources of growth and
productivity trends. We use the best available data on cross-country comparisons
of the sources of productivity growth at the industry level from the EU KLEMS
database. This enables us to document a pick-up in labor productivity growth in
the private economy in Germany since 2005, led by higher TFP growth. This is
indicative of the scope and potential for progress. However, it remains too early
to conclude that a structural shift has taken place. The fact that TFP is measured
as a residual and the finding that the TFP growth acceleration was not accompa-
nied by a boom in ICT investment—unlike the U.S. experience in the late 1990s
and early 2000s—suggests that the recent improvement may be short-lived and
driven by cyclical rather than structural factors.
2
An argument is often made that productivity differentials, particularly in the services industries, are
biased or illusionary because of differences in data across countries. The results in Inklaar, Timmer,
and van Ark (2006), however, suggest that the productivity gap findings are robust to the use of vari-
ous productivity measurement models. A related argument is that the strong post-1995 productivity
growth in the U.S. services sector was illusionary because it was the result of an unsustainable boom
in consumer expenditure and household debt. While scale effects from increased demand may have
played a role in productivity improvements—notably for the distribution sector—the fact that the
acceleration in retail output and productivity occurred mainly during the late nineties and early
2000s, whereas the rise in household debt occurred later, from 2003 to 2007, makes it difficult to
attribute all the productivity improvements to the credit boom (van Ark, 2010).
3
Allard and Everaert (2010) argue that measures to develop capital markets further in Europe will not only
result in the establishment of more attractive conditions for venture capital but also create room for banks
to focus more on supporting smaller firms—in relatively large numbers in the euro area; these smaller
firms are at a higher risk of being constrained in financing but at the same time are key for innovation.
4
See Allard and Everaert (2010) for a discussion of the role of labor and service market reforms in
lifting euro area long-term growth.
a
Voss and Müller (2009) similarly find that financing aspects are a key limiting burden for young German
start-ups. Venture capital is hard to access and concentrates on ICT, medical research, medical appli-
ances, and biotech.
b
Bartelsman and others (2010) investigate the impact of EPL on ICT adoption. They find evidence that
both the share of employment and productivity levels in ICT-intensive sectors are relatively lower in
high-protection EU countries, suggesting that EPL slows the adoption of new ICT. However, they do not
investigate the direct impact of EPL on TFP growth and innovation.
5
The importance and cross-cutting nature of ICT is reflected in Germany’s high-tech strategy and,
more broadly, in Europe’s 2020 growth agenda, which also notes the link between ICT usage and
services’ productivity.
Finland
Sweden
Netherlands
Germany
Denmark
Japan
Austria
Belgium
United States
Norway
France
United Kingdom
Italy
Spain
6
See McKinsey Global Institute (2010). Recent structural reforms to support competitiveness include
the 2008 business tax reform, the Hartz IV reforms, and deregulation in transport services.
industry-level sectors, including goods, ICT, and private services, extending ear-
lier work by van Ark, O’Mahony, and Timmer (2008) through 2007. Within the
services sector, we separately analyze the drivers of growth and German-U.S.
productivity differentials in distribution services, finance and business services,
and personal services. The section also documents Germany’s relative progress in
adopting a knowledge economy, as measured by a number of variables and busi-
ness survey indicators usually viewed as related to higher ICT investments and a
favorable environment for innovation.
120
GDP per hour
100
80
GDP per capita
60
40
20
0
1950 1960 1970 1980 1990 2000
Figure 3.2 Total Economy GDP per Hour Worked and GDP per Capita (Percent of
U.S. levels)
Sources: The Conference Board Total Economy Database, September 2010; Timmer, Ypma, and van Ark (2003);
and IMF staff estimates.
TABLE 3.1
7
TFP is computed as a residual, thus it also includes measurement errors and the effects from unmea-
sured outputs and inputs, such as research and development and other intangible improvements,
including organizational improvements.
8
Lower investment in non-ICT assets and labor quality contributed 37 percent and 10 percent,
respectively, to the declining trend in productivity growth between the two periods.
TABLE 3.2
Memo item:
Labor productivity 2.6 1.6 1.8 1.7 2.7 0.8
growth
Source: The Conference Board Total Economy Database, September 2010; and IMF staff estimates.
a
Based on the difference in the log of the levels of each variable.
Jorgenson, Ho, and Stiroh (2004), Triplett and Bosworth (2006), Bosworth and
Triplett (2007), and more recently Jorgenson, Ho, and Stiroh (2008).
In 2005, a productivity catch-up resumed, mainly reflecting a U.S. slowdown.
Germany’s annual labor productivity growth accelerated only moderately (by 0.2
percent) in 2004–2007 relative to 1995–2004, as a pick-up in employment offset
higher TFP growth (by 0.5 percent) and a higher contribution of non-ICT capi-
tal investment (by 0.2 percent). However, since TFP is measured as a residual, the
recent turnaround could be largely due to unmeasured cyclical factors—including
higher capacity utilization immediately precrisis in Germany. Moreover, unlike in
the United States in the mid-1990s, the recent productivity acceleration in
Germany was not accompanied by higher investment in ICT assets, which also
suggests that cyclical factors might have been at play rather than a structural shift
to a more “knowledge-driven” economy.
The results discussed so far apply to the total economy, including both public
and private services. When public services—including health, education, and
other public services—are excluded, the productivity growth gaps between
Germany and the United States that emerged in the mid-1990s are magnified and
TFP, rather than capital accumulation, explains the bulk of the gap.9 Germany’s
labor productivity in the private economy grew at broadly the same rate during
1995–2004 as the total economy (including public services) grew, namely at 1.6
percent (Table 3.3). The acceleration in U.S. productivity growth therefore becomes
even more striking when focusing only on the private economy: productivity
9
Other excluded services are public administration and defense. Following van Ark, O’Mahony, and
Timmer (2008), we also exclude real estate (ISIC 70), because output in this industry mostly reflects
imputed housing rents rather than sales of firms.
growth in the private economy there rose by 3.1 percent annually over that peri-
od, compared to 2.7 percent for the total economy. The U.S. productivity lead is
thus especially pronounced in the private economy. Specifically, we find that:
• The annual productivity gap with the United States during 1995–2004 for
only the private economy is 1.5 percent, compared to 1.1 percent for the
total economy (including public services).
• Since 2005, the results continue to show the resumption of a productivity
catch-up when only the private economy is considered. During 2004–
2007, the German resurgence is more pronounced when public services are
excluded, and the U.S. slowdown is less extreme. Both trends suggest that a
catch-up process has resumed more recently, although as noted above it is
too early to conclude that the overall lag in German productivity growth has
reversed, given the importance of cyclical factors over this relatively short
period and the lack of a pick-up in ICT investments.
• TFP growth drives the differences between the two countries’ private-
economy productivity growth rates during 1995–2004, contributing over
two-thirds (72 percent) of the productivity gap (compared to 15 percent
for the total economy, including public services). The effect of slower
investment in ICT capital on Germany’s private economy is also an impor-
tant factor, accounting for more than a third (35 percent) of the slower
labor productivity growth in Germany relative to the United States over
that period (compared to more than half when the total economy is con-
sidered). The contribution of differences in the pace of labor skill growth
increases marginally (to 12 percent of total) when only the private econo-
my is considered.10
• When focusing on developments over time in the two countries, a U.S.
slowdown, rather than significantly higher productivity growth in Germany,
continues to account for most of the recent turnaround since 2005. Within
Germany’s private economy, the pick-up in TFP growth (by 1.2 percent)
since 2005 is more pronounced than for the total economy including public
services (by 0.5 percent). Similar to the result for the total economy, the
contribution of ICT capital accumulation to productivity growth in the
private economy fails to accelerate in 2004–2007, unlike the U.S. experi-
ence of the mid-1990s, suggesting that recent developments may reflect
10
The overall contribution of human capital would be underestimated in a growth-accounting frame-
work if human capital mainly influences labor productivity growth through its impact on TFP
growth. Higher education, in particular, arguably has a supportive role in fostering technological
improvements and enabling a fast adjustment to new technologies. Using a sample of 13 EU countries
plus the U.S., Molagoda and Perez (2011) confirm empirically the significant role of human capital
in explaining TFP differences across countries through both its direct impact on innovation by the
productivity leader and indirectly by increasing the size of knowledge spillovers. The largest impact of
human capital is found for countries at or close to the productivity frontier, possibly because innova-
tion is a relatively more skill-intensive activity than imitation. See also Vandenbussche, Aghion, and
Meghir (2006).
TABLE 3.3
AN INDUSTRY PERSPECTIVE
This section shifts from the aggregate perspective of the previous section to a
sector-level (more disaggregated) perspective, allowing us to document the contri-
butions of each sector to aggregate labor productivity growth and the key sources
of productivity growth at the sector and industry level. Based on industry-level
11
Molagoda and Perez (2011) do not provide the breakdown over 1995–2004 and 2004–07, only the
full sample average results.
200
Germany United States
180
160
140
120
100
80
60
40
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Figure 3.3 Value Added in the Services Sector (Gross value added, volume indices,
1995=100)
Sources: EU Klems database, November 2009 release; IMF staff estimates.
800
Germany United States
700
600
500
400
300
200
100
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
measures of output, inputs, and TFP from the EU KLEMS database, the results
show that productivity differences are strong across industries too. In the case of
the German-U.S. differential in productivity growth, we find that private services
and, to a lesser extent, ICT production are the main sectors accounting for the
aggregate productivity gaps in the private economy (Figures 3.3 and 3.4).12
Private services (including wholesale and retail trade, hotels and restaurants,
transport services, and financial and business services) also explain the bulk of
12
ICT production includes production of electrical machinery and telecommunication services.
Germany's private sector fell behind in the mid-nineties. Although productivity in the goods sector remained
strong. . .
4.5 Private Economy: Productivity Per Hour Worked 4.0 Goods Production: Productivity Per Hour Worked
4.0 (Five-Year Average Annual Percent Change) (Five-Year Average Annual Percent Change)
3.5
3.5
3.0
3.0
2.5
2.5
2.0
2.0
1.5 1.5
1.0 1.0
Germany Germany
0.5 United States 0.5 United States
0.0 0.0
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
. . .it lagged in information and communication
technologies (ICT). . . . . .and it was especially slow in the services sector.
14 ICT Production: Productivity Per Hour Worked 4.0 Private Services: Productivity Per Hour Worked
(Five-Year Average Annual Percent Change) 3.5 (Five-Year Average Annual Percent Change)
12
3.0
10
2.5
8
2.0
6
1.5
4 1.0
2 Germany 0.5 Germany
United States United States
0 0.0
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
Total factor productivity (TFP) led the US revival since . . .while German TFP lagged until just before the crisis.
1995. . .
United States: Private Economy Productivity Growth Germany: Private Economy Productivity Growth
3.5 Decomposition (Five-Year Average Contribution, 3.5 Decomposition (Five-Year Average Contribution,
in Percent) in Percent)
3.0 3.0
2.5 2.5
2.0 2.0
0.0 0.0
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
TABLE 3.5
extent differences in TFP growth at the sectoral level (i.e., efficiency of input use)
rather than differences in factor intensity growth (i.e., growth in both human and
physical capital inputs).
The closure of the gap vis-à-vis the United States between 2005 and 2007
mainly reflected slower productivity growth in the United States’ distribution
and financial and business services and improved performance in Germany’s
financial and business services. In Germany, the turnaround in financial and
business services reflects higher TFP growth rather than larger contributions of
capital accumulation and labor quality. However, international experience sug-
gests that sustaining this trend will require concurrent improvements in TFP and
investment—especially in ICT assets—since both are drivers of innovation
effects on services productivity.
The continuing productivity gaps in the services sector reflect in part
Germany’s lag, by advanced country standards, in the use of ICT. Compared to
some other advanced economies, Germany lags in research and use of ICT. This
is evidenced by the low share of ICT in total research and development spending,
compared to the average in OECD countries, and by the relatively low extent of
70
Germany United States OECD total
60
50
40
30
20
10
0
2000 2001 2002 2003 2004 2005 2006 2007
business and government internet use. The latter in turn seems caused by several
infrastructure and financing factors, including:
• a relatively low number of secure internet servers;
• low public investment in advanced technologies compared to some other
OECD countries; and
• a comparatively low willingness of the financial system to provide financing
for high-risk, innovative projects (Box 3.4).
Internet penetration in Germany (measured by the number of subscribers as a
share of total population) is in line with the average for OECD countries, but it
is lower than in the United States (Figure 3.6).
A limited initial public offering (IPO) market and tax obstacles appear to be
contributing to the limited development of German private equity and venture
capital markets and thus to the limited availability of risk financing. Difficulty in
achieving successful exits, lack of entrepreneurial talent, and unfavorable tax
policies (including tax obstacles to cross-border investments and double taxation)
are the main unfavorable climate factors for venture capital in Germany, accord-
ing to a 2010 survey of 516 firms in nine countries: 72 percent of respondents
cited lack of entrepreneurial talent and tax policies as the main obstacles, and 67
percent considered exit difficulties the main issue. Eliminating tax obstacles to
cross-border investments and double taxation would require coordinated action
at the EU level to remove these and other regulatory barriers (e.g., separate regis-
tration requirements), thus allowing even smaller funds to invest EU-wide more
efficiently, develop specialized sectoral expertise, and reap economies of scale.13
13
While there is a consensus among the member states on promoting mutual recognition of national
frameworks, no significant measures have yet been taken that would make fundraising and investing
across borders easier (European Commission Enterprise and Industry Directorate General, 2009).
The share of investment and research in information The number of secure servers is low relative
technologies is relatively small in Germany. to peers.
70 6.0
ICT R&D expenditure as a share of total R&D, 2007 KOR
USA
60 (Percent) 5.5 SWECAN CHE
NLD
ICT Usage (index 1–7)
D
d
en
y
s
es
re
an
an
3.0
nd
EC
ed
at
Ko
nl
m
rla
St
er
he
d
G
te
et
N
U
UK UK
DEU NLD CAN DNK DEU DNK
JPN FIN
5.0 JPN FRA
NOR FIN 5.0 FRA NOR
AUT AUS
AUT AUS BEL BEL
4.5 4.5
ESP ESP
4.0 PRT 4.0 PRT
ITA
3.5 3.5
GRC GRC
3.0 3.0
2.5 3.0 3.5 4.0 4.5 5.0 2.5 3.0 3.5 4.0 4.5 5.0
Government procurement of advanced Venture Capital Availability (survey index 1–7)
technology products (survey index 1–7)
The lack of exit opportunities in Germany reflects the small share of global IPO
listings in Deutsche Boerse.14 Reasons for a limited IPO market in Germany
compared to other key financial centers seem related to market liquidity: a shal-
lower pool of equity capital compared to the United States, the United Kingdom,
or the countries associated with the Euronext exchange, and a narrower investor
base compared to the United States or the UK. It is these factors that limit the
IPO market, rather than cost factors such as listing requirements and underwrit-
ing fees, where Germany appears very competitive (Oxera, 2006).
Another reason for Germany’s delay in adopting the Internet economy and
constraining factors for innovation is deficiency in the insolvency law, which
results in relatively less creditor-friendly legislation than in the United Kingdom
and thus less willingness of banks to lend to high-risk projects. Specifically,
German banks face costlier and lengthier proceedings relative to the United
Kingdom and thus potentially higher legislation-induced credit risk (Schmieder
and Schmieder, 2011). In response, they demand relatively more credit risk miti-
gation than U.K. and U.S. banks do, but still recover less than do U.K. banks.
Banks’ lower willingness to lend to risky projects in turn could hinder entrepre-
neurship and risk-taking. To be on par with U.K. banks, Schmieder and
Schmieder (2011) calculate that formal bankruptcy proceedings in Germany
would have to be shortened by about one half.
Relatively limited use of public procurement to provide sustained ICT
demand in Germany, as compared to other advanced countries such as Korea,
Sweden, and the United States, may be another factor explaining the compara-
tively low investment in new, high-risk technologies. The premise is that there are
positive externalities in the use of ICT caused, for example, by network effects or
complementary investments, such as organizational change, that go unmeasured,
and large fixed costs from the required infrastructure investments (servers, infra-
structure software, storage).15 Business models for ICT companies or services
companies that make heavy use of ICT are thus reliant on scale to be profitable.
A transparent procurement process that enables greater usage based on open
standards can create synergies.
Greater European integration in services provision would also help raise
investment in new, high-risk technologies, It would generate economies of scale
and raise incentives to invest in and provide risk financing for innovative applica-
tions of ICT in services. Obstacles to an integrated European market for online
and off-line services include a variety of compliance and regulatory issues,
14
The number of IPOs expected in Germany in 2011 is 20, compared to a backlog of 150 deals in
the U.S at end-February; Deutsche Boerse lagged other IPO markets in China, London, New York,
Tokyo, Mumbai, Australia, and Korea in 2010, both in number of deals and funds raised (Ernst &
Young, 2011).
15
See, for example, Stiroh (2002) and McKinsey Global Institute (2010). While the potential offered
and challenges posed in using public procurement as an instrument for innovation have been largely
ignored or downplayed, a number of empirical studies conclude that over long time periods, state
procurement triggered greater innovation impulses in more areas than did research and development
subsidies (see Edler and Georghiou, 2007, for a review of the evidence).
CONCLUSION
International experience suggests that ICT applications in traditionally low-tech
industries such as the retail and wholesale sectors lead to technology and innova-
tion effects, which in turn raise TFP growth. While large and internationally
16
See The Economist, October 28, 2010.
17
Successful internet start-ups, such as Berlin-based Wooga, illustrate the scope for exploiting the
synergies of a European user market.
18
Much of the available empirical research refers to the impact of the Internet or the “digital economy”
rather than ICT infrastructure. A recent study using data for the 48 U.S. states over 2003–05 suggests
that for every 1 percentage point increase in broadband penetration in a particular area, employment
growth is estimated to increase 0.2–0.3 percentage points per year (Crandall, Litan, and Lehr, 2007).
The study also finds that state output is positively associated with broadband use, although the impact
is not statistically significant.
REFERENCES
Allard, Céline, and Luc Everaert, 2010, “Lifing Euro Area Growth: Priorities for Structural
Reforms and Governance,” IMF Staff Position Note SPN/10/19 (Washington, DC:
International Monetary Fund).
Bartelsman, Eric, Andrea Bassanini, John Haltiwanger, Ron Jarmin, Stefano Scarpetta, and
Thorsten Schank, 2008, “The Spread of ICT and Productivity Growth: Is Europe Really
Lagging Behind in the New Economy?,” unpublished paper, OECD.
Bosworth, Barry P., and Jack E. Triplett, 2007, “The Early 21st Century Productivity Expansion
is Still in Services,” International Productivity Monitor, 14 (Spring).
The Conference Board, 2009, “Productivity, Performance, and Progress: Germany in
International Comparative Perspective.” Available online at http://www.conference-board.
org/publications/publicationdetail.cfm?publicationid=1682.
Crandall, Robert W., Robert E. Litan, and William Lehr, 2007, “The Effects of Broadband
Deployment on Output and Employment: a Cross-Sectional Analysis of U.S. Data,” Issues in
Economic Policy, No. 6 (Washington, DC: The Brookings Institution).
Deloitte, 2010, Results from the 2010 Global Venture Capital Survey, available at www.nvca.org.
The Economist, 2010, “Europe’s Need for E-Freedom,” available at http://www.economist
.com/node/17361454.
Edler, Jakob, and Luke Georghiou, 2007, “Public Procurement and Innovation—Resurrecting
the Demand Side,” Research Policy, Vol. 36, 949–63.
Ernst & Young, 2011, Global IPO Trends, available at http://www.ey.com/GL/en/Services/
Strategic-Growth-Markets/Global-IPO-trends-2011.
European Commission Enterprise and Industry Directorate General, 2009, Cross-Border Venture
Capital in the European Union: European Commission Work on Removing Obstacles (Brussels:
European Commission).
EU ICT Task Force, 2006, “Fostering the Competitiveness of Europe’s ICT Industry.” Available
online at http://ec.europa.eu/information_society/eeurope/i2010/docs/high_level_group/
ict_task_force_report_nov2006.pdf.
IDC, 2010, Worldwide and Regional Public IT Cloud Services 2010–14 Forecast.
Inklaar, Robert, Marcel P. Timmer, and Bart van Ark, 2008, “Market Services Productivity,”
Economic Policy, January, pp. 139–94.
Jorgenson, Dale W., Mun S. Ho, and Kevin J. Stiroh, 2004, “Will the U.S. Productivity
Resurgence Continue,” Current Issues in Economics and Finance, Vol. 10, No. 13 (New York:
Federal Reserve Bank of New York).
———, 2008, “A Retrospective Look at the U.S. Productivity Growth Resurgence,” Journal of
Economic Perspectives, Vol. 22, No. 1, pp. 3–24.
McKinsey Global Institute, 2010, “Beyond Austerity: A Path to Economic Growth and Renewal
in Europe” (McKinsey & Company).
———, 2011, “European Growth and Renewal: The Path from Crisis to Recovery, Updated
Research” (McKinsey & Company).
Molagoda, Nandaka, and Esther Perez, 2011, “Raising Potential Growth in Europe: Mind the
Residual,” Chapter V in Euro Area Policies—Selected Issues, SM/11/160, pp. 62–79 (Washington:
International Monetary Fund).
Nahuys, Richard, and Henry van der Wiel, 2005, “How Should Europe’s ICT Ambitions
Look Like? An Interpretative Review of the Facts,” Discussion Paper No. 05–22 (Utrecht:
Tjalling C. Koopmans Research Institute.)
Oxera Consulting Ltd, 2006, The Cost of Capital: An International Comparison). Available at
http://www.nd.edu/~carecob/May2008Conference/Papers/OxeraCostofcapitalreport
ExecSummary.pdf.
Schmieder, Christian, and Philipp Schmieder, 2011, “The Impact of Legislation on Credit
Risk—Comparative Evidence from the United States, the United Kingdom, and Germany,”
IMF Working Paper No. WP/11/55 (Washington, DC: International Monetary Fund).
Stiroh, Kevin J., 2002, “Information Technology and the U.S. Productivity Revival: What Do
the Industry Data Say?” American Economic Review, Vol. 92, No. 5, pp. 1559–76.
Timmer, Marcel P., Gerard Ypma, and Bart van der Ark, 2003, “IT in the European Union:
driving productivity divergence?” GGDC Research Memorandum 200363, Groningen
Growth and Development Centre, University of Groningen.
Triplett, Jack E., and Barry P. Bosworth, 2006, “Baumol’s Disease Has Been Cured: IT and
Multifactor Productivity in U.S. Services Industries,” in The New Economy and Beyond: Past,
Present, and Future, ed. by Dennis W. Jansen (Cheltenham, U.K.: Edgar Elgar).
van Ark, Bart, 2010, “Productivity, Sources of Growth, Potential Output in the Euro Area and
the United States,” Intereconomics, Vol. 1, pp. 17–20.
van Ark, Bart, Mary O’Mahony, and Marcel P. Timmer, 2008, “The Productivity Gap Between
Europe and the United States: Trends and Causes,” Journal of Economic Perspectives, Vol. 22,
No. 1, pp. 25–44.
Vandenbussche, Jérôme, Philippe Aghion, and Costas Meghir, 2006, “Growth, Distance to
Frontier, and Composition of Human Capital,” Journal of Economic Growth, Vol. 11, No. 2,
pp. 97–127.
Voss, Romy, and Christoph Müller, 2009, “How Are the Conditions for High-Tech Start-Ups
in Germany?,” International Journal of Entrepreneurship and Small Business, Vol. 7, No. 3,
pp. 284–311.
APPENDIX
TABLE 3A.1
Germany’s employment fell only marginally during the crisis, despite a sharp drop in
output and in contrast to historical employment patterns over the business cycle. This
paper offers new perspectives on the apparent “puzzle” of the German labor market
response. Labor market developments during the crisis are consistent with a view that
Germany experienced mainly a temporary demand shock, necessitating relatively little
sectoral reallocation and with employers therefore adjusting hours of work while
retaining workers. The labor market dynamics also provide evidence that past reforms,
including especially the Hartz reforms during the early 2000s, enabled the labor mar-
ket to react in a way that previously had not been possible to the same extent. These
reforms facilitated a shift from adjustment in the number of individuals employed to
adjustments in hours. Thus, while the employment decline was smaller than expected
based on past dynamics, the reduction in hours worked was more in line with the drop
in output. And lastly, the moderate employment decline during the crisis reflected the
longer-term structural changes in the labor market. The upward momentum in
employment stemming from the earlier reforms held back the adverse crisis impact.
INTRODUCTION
The German labor market has gone through a remarkable transformation over
the past two decades. From being labeled the “sick man of Europe” as unemploy-
ment rates steadily went up during the 1990s, and even reached double-digit
territory during the mid-2000s, to references to the “German labor market mir-
acle” during the recent financial crisis when unemployment, after a brief rise at
its onset, continued the downward trend that had started around 2005. This
transformation is particularly striking in comparative perspective: Germany was
one of the European countries that motivated Ljungqvist and Sargent’s (1998)
study of how and why the U.S. labor market performed so much better, espe-
cially in turbulent times. Now, economists are puzzling over why the German
The author has benefitted from comments and suggestions by Ashoka Mody, Helène Poirson, Fabian
Bornhorst and participants at the Germany in an Interconnected World conference at the Ministry of
Finance in Berlin (May 2011), especially the paper’s discussant, Werner Eichhorst. Susan Becker
provided excellent research assistance.
77
labor market has proven to be so much more resilient than the U.S. market, even
worrying about a jobless recovery there while German unemployment continues
to decline.
This chapter aims to provide new perspectives on the recent labor market
dynamics and to understand how “miraculous” the crisis performance actually
was. It also aims to answer the question of what the crisis can tell us about the
German labor market, and in particular whether the recent experience reflects
something more fundamental. The analysis suggests that, broadly speaking, the
labor market response has been benign, but less so than meets the eye, and that
the German labor market does indeed appear to be structurally different from
what it was even a decade ago. The chapter proceeds by providing some general
background on longer-term and more recent developments, discussing a number
of factors that help to better understand the recent labor dynamics. The appendix
provides details on the labor market simulations reported here.
BACKGROUND
Germany has long been labeled the “sick man” of Europe, owing in large part to
its ailing and sluggish labor market with its high and increasing unemployment
(e.g., The Economist, 1999). Indeed, starting from an unemployment rate of less
than one percent in 1970, German unemployment has almost continuously
trended upwards, interrupted only by brief cyclical declines in unemployment,
such as those during 1976–1980, 1985–1990, 1997–2001, and the most recent
decline since 2005 (with only a moderate uptick in 2009). In each of those cases,
the decline in unemployment was paralleled by a growth recovery in output
(Figure 4.1).
12
10
4
West Germany Unified Germany
0
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
France Italy
Japan United Kingdom
15
United States Germany
12
−3
−6
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
Figure 4.2 Unemployment in Advanced Economies (ILO definition, percent)
Source: International Labour Organization (ILO).
However, the cyclical upswings in GDP were not sufficient to turn around the
secular increase in unemployment. With the exception of the most recent (ongo-
ing) episode, each trough in unemployment was higher than the previous one.
Thus, from nearly zero unemployment in the early 1970s, German unemploy-
ment peaked in 2005 at over 11 percent. Notably, the deterioration in labor
market outcomes was somewhat unique to Germany (Figure 4.2). Along with
Japan, Germany had one of the lowest unemployment rates among the large
advanced economies in the 1970s, substantially lower than in the United States,
where unemployment was on the order of 6 percent at that time. However, while
U.S. unemployment peaked at nearly 10 percent in 1982 and declined thereafter,
German unemployment continued to rise. By the early 2000s, Germany had one
of the highest unemployment rates among its peers.
The long upward trend that began as early as the 1970s suggests that Germany’s
labor market problems were not specific to the structural challenges posted by
German unification in 1990. A broad consensus has emerged on the main under-
lying causes. Siebert (2003) views the secular step-wise increase in unemployment
as the result of weak job creation, in turn caused by the German institutional
design for wage formation, which at times has led to binding negotiated wage
increases and insufficient wage differentiation; and as the result of a high reserva-
tion wage, resulting from high unemployment and social welfare benefits.1
1
This notion of “binding wage increases” is based on the observed decline in the “wage drift” (the
difference between actual and negotiated wage increases), especially during the 1990s (Siebert, 2003,
p. 8).
With its combination of high unemployment and welfare benefits, rigid wage
determination mechanisms, and high firing costs, Germany was typical of what
many authors saw as the main differences between European and U.S. labor mar-
ket models (Siebert, 1997; Ljungqvist and Sargent, 1998). In addition, Prescott
(2004) and Rogerson (2010) cite high taxation that restricts the labor supply. In
this comparison, U.S. labor market flexibility is seen as the result of low unem-
ployment benefits, low firings costs, and low labor taxation, leading to a more
dynamic labor market that can more easily adjust to economic shocks.
The exploding cost of unemployment along with public pressure to reduce the
high levels of unemployment eventually triggered the largest social policy reform
in post-war Germany (Wunsch, 2005; see also Box 4.2). This reform agenda
limited the duration and level of unemployment benefits and aimed to improve
job creation. As argued in this chapter, these reforms led to a labor market per-
formance in Germany during the recent financial crisis that was quite different
from its historical performance as well as from that in other economies.
RECENT DEVELOPMENTS
German employment, measured as the number of employed individuals, has
performed remarkably during the crisis. Employment decreased only by a mod-
erate 0.6 percent from peak (2008:Q4) to trough (2009:Q4). By contrast, the
cumulative peak-to-trough decline in real GDP (2008:Q1–2009:Q1) reached
over 6.6 percent. The change in employment is one of the smallest among
advanced economies, even compared with those that experienced smaller GDP
declines. A comparison with the United States is particularly telling: despite a
smaller decline in real GDP, U.S. employment was about 6 percent lower at the
trough, a larger decline than that in output (Figure 4.3). These differences were
reflected also in terms of recovery—by end-2010, the United States had nearly
recovered its precrisis level of GDP, but still remained about 4 percent below its
pre-crisis employment level. Conversely, in Germany employment has moved
back to precrisis levels quickly, but with output remaining more sluggish
(Figure 4.4).
Germany’s experience stands out not only in cross-country comparison, but
also from the perspective of its historical business cycle dynamics. Despite the
unique nature of the financial crisis—both in terms of magnitude and in terms
of its global reach—most countries’ labor markets responded in a fashion consis-
tent with historical dynamics. Germany was among only a small group of coun-
tries whose labor markets responded so strikingly differently. Figure 4.5 shows
actual versus predicted unemployment patterns, based on the cross-country
analysis in the 2010 World Economic Outlook (IMF, 2010, ch. 3). Most countries
experienced increases in unemployment either close to their predicted values or,
in many cases, substantially larger. By contrast, the increase in unemployment in
Germany fell substantially short of its predicted value, unique among all countries
in the sample.
101
Germany
100
Sweden
99 United Kingdom
98 France
Japan Denmark
97
Employment
96 Finland USA
95
94
93
92
91 GIPS
90
90 91 92 93 94 95 96 97 98 99 100 101
Real GDP
An alternative version of Okun’s law can provide additional insights into the
German labor market dynamics (Box 4.1). The past pattern between employment
and growth would have predicted a much larger employment decline than
observed, given the output loss, on the order of more than one percent (Figure
4.6). In contrast to employment levels, total hours worked declined substantially,
by a cumulative 4.4 percent between their pre-crisis peak in 2008:Q2 and their
trough in 2009:Q2. When re-estimating a modified Okun’s relationship between
economic growth and total hours worked, the peak-to-trough decline implied by
8.0
7.0 Actual change Predicted change
6.0
5.0
4.0
3.0
2.0
1.0
0.0
−1.0
No ny
itz aly
Ja y
n
the nd
Be nds
Po ium
Ne Fra al
e
Sw and
Gr en
Au ce
De tria
Ca ark
ite Fin a
ing d
m
Ire A
Sp d
ain
a
pa
nc
d
d K lan
lan
US
g
do
a
rw
ee
Ne erla
ed
na
It
rtu
s
nm
rm
rla
lg
al
Ze
Ge
Sw
Un
Figure 4.5 Actual versus Predicted Change in Unemployment (Percentage points)
Source: IMF, World Economic Outlook (WEO).
Note: Based on figure 3.8 in IMF (2010). The predicted changes in unemployment are based on Okun’s law
specifications that include additional control variables, including measures of financial, housing and equity
market stress.
40.2
40.0
39.8
39.6
39.4
20 Q1
20 Q2
20 Q3
20 Q4
20 Q1
20 Q2
20 Q3
20 Q4
20 Q1
20 Q2
20 Q3
3
:Q
:Q
:Q
:Q
:
:
07
07
07
07
08
08
08
08
09
09
09
09
09
10
10
20
20
20
20
this modification is about 3.9 percent (Figure 4.7). Interpreted in this sense, the
labor market response has been less out of line with historical patterns than the
employment dynamics suggest, at least during the first stage of the crisis (i.e.,
from onset to trough). Notably, however, the dynamics in hours and (to a small-
er extent) employment were more positive than predicted during the second,
recovery stage. This is consistent with a continuation of the longer-term down-
ward trend in unemployment. As argued below, that trend would require
'Y/Y = α + β'u
where Y is real GDP and u is the unemployment rate. (Different versions of Okun’s law have
been estimated in a large literature focusing on the relationship between unemployment
and output. See, e.g., Knotek, 2007, and the references therein.) The coefficient β (“Okun’s
coefficient”) can be interpreted as the (semi-)elasticity of output growth with respect to
percentage changes in the unemployment rate. Okun (1962) estimated a value of β | 31/3
for the United States during 1947–1960, while later studies have typically found lower val-
ues on the order of 2.
While Okun’s law has received much empirical support, different versions of it are
appropriate in different contexts. In particular, in this paper an alternative version is esti-
mated, namely, the relationship between changes in total employment and real GDP
growth, as well as a second specification that considers the relationship between changes
in total hours worked and real GDP growth.
These specifications are preferable for a number of reasons. First, the particular interest
here is on the distinction between adjustment on the external margin—individuals being
hired or laid off—and adjustment on the internal margin—changes in hours worked by
each worker. The distinction between these two margins turns out to be important in
understanding recent labor market dynamics in Germany, but it cannot be made using the
traditional specification of Okun’s law that is focused on unemployment.
Secondly, and more broadly, employment is arguably the more preferable object of
interest, since the unemployment rate may change over the business cycle, not only due to
flows between employment and unemployment, but also because of flows into and out of
the labor force. Thus, employment is likely to be a more direct measure of economic well-
being than unemployment. That said, the broader result in this paper, that Germany’s his-
torical Okun’s law did not hold during the current crisis, is a finding that is not unique to any
specification. See, e.g., Figure 4.1 in this chapter and Chapter 3 in IMF (2010).
20 Q2
20 Q3
20 Q4
20 Q1
20 Q2
20 Q3
20 Q4
20 Q1
20 Q2
20 Q3
20 Q4
20 Q1
20 Q2
3
:Q
:
:
07
07
07
07
08
08
08
08
09
09
09
09
10
10
10
20
400 15.5
Hours worked per worker, left axis
390 Total hours worked (in billions), right axis
380 15
370
14.5
360
350
14
340
330 13.5
320
310 13
1991:Q1
1992:Q1
1993:Q1
1994:Q1
1995:Q1
1996:Q1
1997:Q1
1998:Q1
1999:Q1
2000:Q1
2001:Q1
2002:Q1
2003:Q1
2004:Q1
2005:Q1
2006:Q1
2007:Q1
2008:Q1
2009:Q1
2010:Q1
2
More recently, the continued labor market recovery appears to have started to shift back to the
extensive margin, with the increase in total hours worked driven, once again, by employment gains.
3
More complex models may include also the flows from outside the labor force into employment or
unemployment, and vice versa. For tractability, the model considered here abstracts from these.
4.0
1.0
2.0
0.0 0.0
−2.0
−1.0
−4.0
−6.0
−2.0
−8.0
−10.0 −3.0
1991:Q1
1992:Q1
1993:Q1
1994:Q1
1995:Q1
1996:Q1
1997:Q1
1998:Q1
1999:Q1
2000:Q1
2001:Q1
2002:Q1
2003:Q1
2004:Q1
2005:Q1
2006:Q1
2007:Q1
2008:Q1
2009:Q1
2010:Q1
Figure 4.9 Germany: Cyclical Changes in Output and Unemployment
Sources: Federal Statistical Office; and Deutsche Bundesbank.
100.0 8.0
95.0 6.0
90.0
4.0
85.0
80.0 2.0
1991:Q1
1992:Q1
1993:Q1
1994:Q1
1995:Q1
1996:Q1
1997:Q1
1998:Q1
1999:Q1
2000:Q1
2001:Q1
2002:Q1
2003:Q1
2004:Q1
2005:Q1
2006:Q1
2007:Q1
2008:Q1
2009:Q1
2010:Q1
and nonmonetary value derived from being unemployed). These factors directly
impact the extent to which an unemployed individual will actively search, as well
as which job opportunities might be taken up. Another important factor for labor
market outcomes is the matching efficiency, that is, the ease with which firms
encounter the right candidates and unemployed individuals seeking work find the
14
Unemployment rate (actual)
13 Simulated unemployment (USS = 5.25 percent)
Simulated unemployment (USS = 6.0 percent)
12
Simulated unemployment (USS = 7.0 percent)
11 Simulated unemployment (USS = 8.1 percent (historical average))
10
5
2000:Q1
2000:Q3
2001:Q1
2001:Q3
2002:Q1
2002:Q3
2003:Q1
2003:Q3
2004:Q1
2004:Q3
2005:Q1
2005:Q3
2006:Q1
2006:Q3
2007:Q1
2007:Q3
2008:Q1
2008:Q3
2009:Q1
2009:Q3
2010:Q1
2010:Q3
Figure 4.11 Actual and Simulated Unemployment (Percent)
Sources: Deutsche Bundesbank; and IMF staff estimates.
Note: Hypothetical paths are model-based transition paths under the assumption that the Hartz IV reforms
lowered the structural (steady-state) unemployment rate (USS) to various levels. See text for details.
right jobs. The appendix outlines a reduced-form version of a search model, with
a focus on the implied transition dynamics.4
Model simulations suggest that the decline in unemployment observed since
2005 is likely to continue. From an average of about 8.1 percent during 1991–
2005 and a peak of 10.6 percent in 2005, the unemployment rate subsequently
started a downward trend that was interrupted only briefly at the height of the
financial crisis. Most recently, the unemployment rate, at just under 6 percent in
2011:Q2, has reached a level not seen in nearly two decades. The model suggests
that an increase in the rate at which workers encounter the right jobs can explain
much of the precrisis trend in the unemployment rate.
More specifically, the model as described by equation (1) in the appendix
contains two key parameters: the rate at which unemployed individuals find
work, α, and the rate at which workers lose employment, δ. This equation
describes the unemployment dynamics, given that current unemployment is equal
to last period’s unemployment minus the share of unemployment that has found
employment plus the share of the employed that has lost employment.5 Different
parameterizations of the equation imply different structural (steady-state) unem-
ployment rates. To the extent that the Hartz reforms raised the job-finding rate,
these reforms imply a lower steady-state unemployment rate (Figure 4.11).
4
An important feature of this reduced-form model is that one can write down an explicit search-
model which, for the right set of parameters, can rationalize the reduced form model used here.
5
As previously mentioned, these simplifying assumptions include ignoring movements into and out
of the labor force, as well as any heterogeneity in the workforce, whereby different worker groups may
have potentially different job finding/separation rates. Another important assumption is that of con-
stant parameters over time (with the exception of discrete changes at the time of a shock).
Starting from the pre-Hartz IV unemployment rate of about 10½ percent, and
assigning various parameter values to the post-reform value of α, equation (1) can
also be used to trace out the simulated unemployment transitions to a hypo-
thetical new steady-state (Figure 4.9).6 From the peak in 2005 through the onset
of the crisis, the simulated path based on a 5.25 percent long-term unemploy-
ment rate gives an almost perfect fit with the post-2005 data. In this view, the
crisis temporarily halted the trend decline, but employment quickly recovered
and is likely to continue its downward trend.7 The simulations based on a
5.25 percent long-term rate imply that the job-finding rate would be about 50
percent higher than prior to Hartz IV, an implication that is broadly supported
by other research (Lam, 2011).
The interpretation of the post-2005 path as a transition to a lower structural
unemployment rate also affects the interpretation of the observed labor market
response. If the downward trend since 2005 is indeed a transition to a new and
lower steady-state unemployment rate, rather than a low-frequency cyclical
variation around the previous unemployment rate of more than 8 percent, then
the unemployment dynamics during the crisis are best assessed relative to the
counterfactual transition path rather than relative to precrisis levels of unemploy-
ment. Thus, compared to what unemployment might have been in the absence
of the crisis, unemployment at its crisis peak in 2009:Q3 had increased by over
one percentage point relative to the counterfactual—this is more than twice the
change calculated as the difference between the precrisis level of 7.2 percent and
the crisis peak of 7.6 percent. In this reading, then, the German labor market
response has not been quite as mild as typically understood.8
6
See the appendix for details on the construction of the hypothetical paths.
7
However, the simulations are suggestive that unemployment may have shifted to a different trend
line.
8
As a corollary, employment growth could also have been stronger absent the crisis, suggesting that
the above- noted mild employment response may underestimate the true damage from the crisis.
a
The committee was chaired by Peter Hartz, then-member of the executive board of Volkswagen
Aktiengesellschaft in charge of human resources and personnel management.
result, Germany’s growth potential was also little affected. (See Chapter 2). The
nature of the shock differed substantially from that in the United States, which
suffered a more domestic and structural shock to its economy.
9
The notion that firms rationally responded to a temporary shock by hoarding labor does not require
the assumption of perfect foresight—e.g., if firms use some form of Bayesian updating, then the low
level of external demand at the trough of the crisis would not have been perceived as the new equilib-
rium without it persisting for a sufficiently long time.
10
Germany continues to rank high according to the OECD’s employment protection legislation
(EPL) indicators, especially on regular employment (on temporary employment, Germany was below
the OECD average in 2008, the latest available data point). Consequently, the OECD has repeatedly
called on Germany for further reform in this area (OECD, 2011). In addition, Berger and Neugart
(2012) point to an additional cost of reducing employment in the form of uncertainty over labor court
decisions. In terms of labor market implications, because EPL tends to reduce both job destruction
and job creation, the impact on the level of employment is conceptually unclear, reflected in simi-
larly ambiguous empirical findings (Garibaldi and Mauro, 2002, versus Takizawa, 2003). But there is
a broader consensus that high EPL inhibits labor market dynamism and its ability to respond to
economic shocks. See Barone (2011) and Schindler (2009) for reviews of related literatures. In addi-
tion, Boeri (1999) argues that the asymmetric type of EPL, as increasingly observed in Germany (i.e.,
regular versus temporary employment) can lead to labor market dualism.
11
That is, the same level of employment protection legislation may have a larger impact on allocations
when firms have new ways of avoiding layoff costs, as in this case through more hourly flexibility. See
Ljunqvist and Sargent (1998) for a similar reasoning on the interactions of institutions and shocks.
are typically less skilled, while relying on more intensive adjustment for their
more skilled core workforce.12
However, despite much anecdotal evidence, labor market outcomes do not
fully bear out the signs of a true shortage: Brenke (2010) has noted that such a
shortage should put upward pressure on their relative wages, something that has
not been observed. Also, university enrollment rates in engineering and the natu-
ral sciences remain at a high level and have in recent years increased substantially,
suggesting that supply is relatively strong.13 While these statistics suggest that
excess demand for skilled labor in the German labor market may not be as broad-
based and severe as some may fear, there is little doubt that demographic pressures
will pose problems in the medium term.
Matching inefficiencies appear to have diminished. Even without shortages of
skilled labor, firms may opt for “labor hoarding” if they find it difficult to rehire
the workers they need in an upswing. The relationship between vacancies and
unemployment over time (the so-called Beveridge curve, first discussed by Dow
and Dicks-Mireaux, 1958) is a frequently used gauge for a country’s matching
efficiency.14 It is especially insightful in the context of search models—in such
models, labor supply (unemployed workers) and labor demand (firms’ vacant
positions) are brought together through a matching technology, or matching
function. A more efficient matching technology then implies that a given level of
vacancies translates into a lower unemployment rate than would be the case with
a less efficient matching technology. Graphically, shifts in the Beveridge curve
toward the origin suggest gains matching efficiency, while movements along the
curve are the result of equilibrium dynamics for a constant matching technology.
The plot for Germany suggests that around 2007, the matching process
started to improve, with the Beveridge curve now at a substantially lower level
(i.e., a shift toward the origin) than pre-2007 (Figure 4.12). This apparent
improvement in matching efficiency is consistent with the previously noted faster
unemployment-to-employment transition associated with the Hartz reforms,
including, importantly, the measures taken during the first two stages that facili-
tated the development of atypical employment.
All else equal, a higher matching efficiency should increase firms’ willingness
to shed labor, on the expectation that finding a good match in the future will be
relatively easy. While the shift in the Beveridge curve makes it at first sight more
difficult to understand the stable employment during the crisis, the aggregate data
hide the fact that the Hartz reforms mostly added labor supply on the lower end
of the income and skill distribution. That addition is consistent with the fact that
employment losses were larger among low-wage segments, suggesting that
12
Employment of temporary workers declined by 2 percent in Germany during 2009, compared with
an increase of 0.4 percent in total employment (IMF, 2010, Box 3.1).
13
However, enrollment rates in these fields were broadly flat during 2004–2008, suggesting that in
the coming years, graduation rates may not keep up with demand of a growing economy.
14
Vacancies and unemployment are negatively correlated, since strong hiring demand by firms (high
level of vacancies) helps reduce unemployment.
12
11
10
9
2000–07
8
7
2008–present
6
150 200 250 300 350 400 450
CONCLUSION
The apparent puzzle of the German labor market during the crisis can be
explained by a combination of factors. The German economy suffered a demand
shock, which firms presumably expected to eventually reverse. Hence, firms
adjusted mainly by taking advantage of options to reduce the hours worked, keep-
ing employment levels relatively high (i.e., relative to that predicted by Okun’s
law). At the same time, the German unemployment rate had been on a downward
path since about 2005, most likely reflecting the Hartz IV reforms. This also had
the effects of dampening the decline in employment and helping the employment
recovery.
While authors in the existing literature place emphasis on different factors
(Box 4.3), most agree that the likely explanation involves some combination of
hourly flexibility measures, including work-time accounts and short-time subsi-
dies, low precrisis hiring, and long-standing wage moderation. While these views
are consistent with the interpretation put forth in this paper, the emphasis here
is different: namely, in this analysis most of the credit belongs to the Hartz
reforms, which have put Germany in a position to deal well with a temporary
demand shock. This still leaves ample room for further reforms. The increased
flexibility of the lower-wage segment has introduced an uneven distribution of
employment and wage risk: more secure high-skill and high-wage employment
along with less secure low-wage work. Also, while job creation has been made
easier, rigidities still remain on the job termination side, resulting in part from
REFERENCES
Abel, Andrew B., and Ben Bernanke, 2005, Macroeconomics, 5th edition. Addison-Wesley.
Barone, Andrea, 2011, “Employment Protection Legislation: A Critical Review of the
Literature.” Published online at http://jpkc.ojc.zj.cn/ldfx/UploadFiles/2010928214832623.
pdf, downloaded October 18, 2011.
Berger, Helge, and Michael Neugart, 2012, “How German Labor Courts Decide—An
Econometric Case Study,” German Economic Review, Vol. 13, pp. 56–70.
Boeri, Tito, 1999, “Enforcement of Employment Security Regulations, On-the-Job Search and
Unemployment Duration,” European Economic Review, Vol. 43, pp. 65–89.
Boysen-Hogrefe, Jens, and Dominik Groll, 2010, “The German Labour Market Miracle,”
National Institute Economic Review, Vol. 214, pp. R38–R50.
Brenke, Karl, 2010, “Fachkräftemangel kurzfristig noch nicht in Sicht,“ DIW Berlin
Wochenbericht No. 46 (2010).
Burda, Michael, and Jennifer Hunt, 2011, “What Explains the German Labor Market Miracle
in the Great Recession?,” NBER Working Paper No. 17187 (Cambridge: National Bureau of
Economic Research).
Dow, J. C. R., and L. Dicks-Mireaux, 1958, “The Excess Demand for Labour: A Study of
Conditions in Great Britain, 1946–1956,” Oxford Economic Papers, No. 10, pp. 1–33.
The Economist, 1999, “The Sick Man of the Euro,” June 3rd (downloaded from http://www.
economist.com/node/209559).
Fahr, René, and Uwe Sunde, 2009, “Did the Hartz Reforms Speed Up the Matching Process?
A Macro-Evaluation Using Empirical Matching Functions,” German Economic Review,
Vol. 10, pp. 284–316.
Garibaldi, Pietro, and Paolo Mauro, 2002, “Anatomy of Employment Growth,” Economic Policy,
Vol. 17, pp. 67–113.
Gartner, Hermann, and Sabine Klinger, 2010, “Verbesserte Institutionen für den Arbeitsmarkt
in der Wirtschaftskrise,” Wirtschaftsdienst, Vol. 90, pp. 728–34.
Gartner, Hermann, and Christian Merkl, 2011, “Die ökonomische Basis des Arbeits-
marktwunders,” http://www.oekonomenstimme.org/artikel/2011/03/die-oekonomische-basis-
des-arbeitsmarktwunders/.
IMF (International Monetary Fund), 2010, World Economic Outlook: Rebalancing Growth, April
(Washington, DC: International Monetary Fund).
Klinger, Sabine, and Thomas Rothe, 2010, “The Impact of Labour Market Reforms and
Economic Performance on the Matching of Short-Term and Long-Term Unemployed,” IAB
Discussion Paper No. 13/2010 (Nuremburg: Institute for Employment Research (IAB) of the
German Federal Employment Agency (BA)).
Knotek, II, Edward S., 2007, “How Useful is Okun’s Law?,” Federal Reserve Bank of Kansas City
Economic Review, Vol. 4, pp. 73−103.
Lam, W. Raphael, 2011, “Does Labor Market Flexibility Explain Unemployment Dynamics in
the Great Recession?,” IMF Working Paper, forthcoming (Washington, DC: International
Monetary Fund).
Ljungqvist, Lars, and Thomas Sargent, 1998, “The European Unemployment Dilemma,”
Journal of Political Economy, Vol. 106, pp. 514−50.
OECD (Organisation for Economic Co-operation and Development), 2011, Economic Policy
Reforms 2011: Going for Growth (Paris: OECD).
Okun, Arthur M., 1962, “Potential GNP: Its Measurement and Significance,” Proceedings of the
Business and Economics Statistics Section, American Statistical Association, pp. 98–104.
Pissarides, Christopher A., 2000, Equilibrium Unemployment Theory, 2nd edition (Cambridge,
Massachusetts: The MIT Press).
Prescott, Edward C., 2004, “Why Do Americans Work So Much More Than Europeans?”
Federal Reserve Bank of Minneapolis Quarterly Review, Vol. 28, pp. 2–13.
Rogerson, Richard, 2010, The Impact of Labor Taxes on Labor Supply: An International Perspective
(Washington, DC: AEI Press).
Sachverständigenrat für Wirtschaft (Sachverständigenrat), 2009, “Die Zukunft nicht aufs Spiel
setzen,” Jahresgutachen 2009/10.
Schindler, Martin, 2009, “The Italian Labor Market: Recent Trends, Institutions, and Reform
Options,” IMF Working Paper No. 09/47 (Washington, DC: International Monetary Fund.
Siebert, Horst, 1997, “Labor Market Rigidities: At the Root of Unemployment in Europe,”
Journal of Economic Literature, Vol. 11, pp. 37−54.
———, 2003, “The Failure of the German Labor Market,” Kiel Working Paper No. 1169 (Kiel:
Kiel Institute for World Economics).
Takizawa, Hajime, 2003, “Job-Specific Investment and the Cost of Dismissal Restrictions—The
Case of Portugal,” IMF Working Paper No. 03/75 (Washington, DC: International Monetary
Fund).
Wunsch, Conny, 2005, “Labour Market Policy in Germany: Institutions, Instruments and Reforms
since Unification,” University of St. Gallen, Department of Economics, Discussion Paper Nr.
2005–06.
APPENDIX
A simple search model framework can help in understanding the impact of the
Hartz IV reforms on unemployment. In labor markets with search frictions, tran-
sitions from unemployment to employment are not instantaneous. Ignoring
changes to the labor force, unemployment can thus be seen to evolve over time
according to:
ut + 1 = ut(1 − α) + (1 − ut) δ (1)
where ut is unemployment at time t and α,δ are the probabilities (hazard rates) of
exiting unemployment and employment, respectively. This equation also implies
that steady-state unemployment can be expressed as
uSS = δ / (δ + α ). (2)
The Hartz IV reforms have likely lowered structural unemployment through
a higher job finding rate. The pre-Hartz IV average unemployment rate was
about 8.1 percent. Lam (2011) estimates the German job destruction rate at
about 0.5 percent. These two values imply a pre-2005 average job finding rate of
about 6.1 percent.15 Assuming δ to have remained unchanged since 2005, the
steady-state equation provides a one-to-one correspondence between estimates of
the (new) long-term steady-state and the implied (new) job finding rate α.16 The
figure in the main text plots the resulting transition paths for a variety of possible
values of the new structural unemployment rate. See also Fahr and Sunde (2009),
Klinger and Rothe (2010), and Gartner and Klinger (2011) for a similar view of
events.
15
Lam (2011) estimates job finding and destruction rates for 1970–2009 and for 1996–2009 and
finds little difference for the different time periods (for the latter period, he estimates δ = 0.6 percent).
The implied value for α corresponds to Lam’s estimate who also finds a value of also 6.1 percent dur-
ing 1970–2009.
16
For the purpose of this reduced-form exercise, we take the job arrival and destruction rates as exog-
enous. In a more general search model, both margins would be endogenous due to changes in indi-
viduals’ reservation wages. Generally speaking, a reduction in unemployment benefits would raise job
creation and lower job destruction (Pissarides, 2000)—in such models, the key channel is through
wage bargaining.
Can positive growth shocks from the faster-growing countries in Europe spill over to
the slower-growing countries, providing useful tailwinds to their recovery process? This
study investigates the potential relevance of growth spillovers in the context of the crisis
and the recovery process. Based on a Vector Autoregression (VAR) framework, our
analysis suggests that the United States and Japan remain the key sources of growth
spillovers in this recovery, with France also playing an important role for the European
crisis countries. Notwithstanding the current export-led cyclical upswing, Germany
generates relatively small outward spillovers compared to other systemic countries, but
it likely plays a key role in transmitting and amplifying external growth shocks to the
rest of Europe given its more direct exposure to foreign shocks compared to other
European countries. Positive spillovers from Spain were important prior to the 2008–
09 crisis, but Spain is generating negative spillovers in this recovery due to its depressed
domestic demand. Negative spillovers from the European crisis countries appear lim-
ited, consistent with their modest size.
INTRODUCTION
Can positive growth impulses from the faster-growing countries in Europe spill
over to the slower growing countries, providing useful tailwinds to their recovery
process? This analysis investigates the relevance of such potential growth spillovers
and seeks to identify the countries most likely to serve as “growth leaders.” In
particular, we examine the extent to which Germany’s current upswing may spill
over to other countries and accelerate recovery elsewhere.
A simple correlation between the lagged quarter-on-quarter output growth
rates of Germany and other euro area countries shows an increasing co-movement
between Germany and other countries, including the European crisis countries
(Greece, Ireland, and Portugal) in the last 20 years. A similar pattern can be
The authors thank Ashoka Mody for insightful comments and valuable suggestions. They also would
like to thank Céline Allard, Ansgar Belke, Rupa Duttagupta, Felix Huefner, Irina Tytell, and Francis
Vitek, participants in the Germany in an Interconnected World Economy conference (Berlin, 2011) and
in the Graduate Institute of International and Development Studies internal seminar (Geneva) for
useful discussions and comments; and Susan Becker for excellent research assistance.
97
observed for the correlations of the lagged growth rates of France and Italy with
the current GDP growth rate of the rest of the euro area. The correlation of the
lagged U.S. growth rate with other countries’ current GDP growth rates has also
increased markedly (Figure 5.1). While more synchronized business cycles suggest
increased growth spillovers, they do not provide a measure of the spillover effects
of individual countries. Moreover, they could also reflect the growing influence of
common factors, such as euro area monetary policy, or global factors, such as oil
price developments or financial conditions in major financial centers including
the United States and the United Kingdom.
To disentangle the independent impact of growth shocks in individual coun-
tries from the effect of common factors and measure the country-specific spillover
effects, we use a structural vector autoregression (SVAR) approach with an iden-
tification scheme similar to that proposed in Bayoumi and Swiston (2009). Using
this framework, we undertake the following analysis:
• We characterize countries by their international growth spillovers, outward
and inward. Unlike former studies, which rely on regional aggregates, we use
country-level data for a sample of 17 countries (accounting for almost 60
percent of global GDP, at market exchange rates), which includes 11 of the
euro area countries (representing 98 percent of euro area output). Focusing
on individual members rather than treating the euro area as an aggregate as
in earlier studies allows us to shed light on the country-by-country spillovers
and inter-linkages that drive growth dynamics.
• We describe the dynamics of these spillovers during the recent crisis and
recovery. This is done by applying a dynamic growth accounting calculation
to the SVAR estimation results to quantify the contribution of individual
countries to the recent decline and recovery in output growth.
• We quantify the relevance of different channels of transmission of spillovers.
We use counterfactual analysis and smaller country–by-country regressions,
which include exports as an additional variable to assess the empirical rele-
vance of various transmission channels, including trade and third-country
effects (e.g., transmission through a common trade partner).
• We analyze the determinants of the spillover size by relating outward spill-
overs to both the country’s size and the relative importance of the domestic
demand contribution to the country’s growth.
Our main findings can be summarized as follows:
• Confirming earlier results, we find evidence of significant spillover effects
from the United States to the rest of the world. Outward spillovers from
Germany are found to be surprisingly small (relative to growth impulses
emanating from other large systemic countries), and they are largely con-
fined to smaller trade partners.
• The results suggest that despite the increased correlation of Germany’s
growth rate with that of the rest of the euro area, the United States and
Japan remain the key sources of growth spillovers in this recovery, with
0.6 0.6
0.4 0.4
0.2 0.2
0 0
−0.2 −0.2
1975- 1980- 1985- 1990- 1995- 2000- 2005- 2009- 1975- 1980- 1985- 1990- 1995- 2000- 2005- 2009-
1979 1984 1989 1994 1999 2004 2008 2010 1979 1984 1989 1994 1999 2004 2008 2010
Figure 5.1 Growth Correlation with EMU and GIP Countries over Time: Selected Economies
(1975–2010)
Source: OECD, IMF staff calculations.
Note: EMU: Austria, Belgium, Finland, Greece, Ireland, Netherlands, Portugal, and Spain, weighted by their respective
share in the group; GIP: Greece, Ireland, and Portugal.
France also playing an important role for the European crisis countries.
Positive spillovers from Spain were important prior to the crisis. However,
Spain is generating negative spillovers in this recovery due to its depressed
domestic demand.
• In line with earlier VAR-based evidence on channels of spillover transmission
across regions, the results of a country-by-country estimation of a smaller
SVAR model (augmented with exports) suggest that financial and other non-
trade channels explain the biggest share of cross-border growth spillovers.
However, trade effects are particularly relevant in the case of Germany.
• Taking the analysis one step further, we find that countries that generate the
largest estimated outward spillovers are the ones where growth is largely
driven by autonomous sources of domestic demand. In contrast, countries
highly sensitive to external shocks, such as Germany, tend to have a rela-
tively small independent, aggregate impact on other countries. This result,
together with the finding that third-country effects play a significant role in
the international transmission of shocks, suggests that Germany plays an
important role as a transmitter and amplifier of growth shocks originating
in other countries.
Three main caveats should be mentioned at the outset. All three provide inter-
esting avenues for future research but are beyond the scope of this paper.
• Countries outside our sample might play an important role in global spill-
overs, either as recipients or as a source of shocks. For instance, several
Eastern European countries are highly integrated with Germany and have
become an integral part of the production chain of German products. For
these countries, empirical evidence supports the idea that Germany has
played a more prominent role in recent years (see Danninger, 2008). In
recent years, China and other large emerging countries probably play an
important role as a source of shocks as well, reflecting their size and
increased trade linkages.1
• There may be other sources of growth spillovers beyond the business-cycle
fluctuations considered in this study. For example, foreign direct investment
(FDI) outflows between advanced countries and from advanced to develop-
ing countries could generate significant spillovers over time, including
through knowledge transfer and employment creation. Quantifying such
FDI-related spillovers would require a different approach, one focused on a
long-term horizon, since such effects are likely to develop only over time.
• The analysis here is backward looking and does not allow for continuous
time-varying relationships among the countries. In addition to full sample
results, we present estimation results for a more recent sub-sample (since
1993) to account for the possibility of changed responses, in line with rap-
idly increasing cross-border trade and financial linkages. The results confirm
that the potential size of cross-border growth spillovers has generally
increased, with the notable exceptions of Japan and Germany, for which we
find no evidence of higher outward spillovers on average in the recent period.
Finally, it should be noted that this analysis is descriptive. An analysis of wel-
fare implications is beyond the scope of this chapter.
The remainder of this chapter reviews the related literature on cross-country
spillovers, outlining the main empirical approaches that have been pursued and the
key findings; discusses the empirical strategies employed in this study; and high-
lights the main findings and provides estimates of individual countries’ relevance
for other countries’ growth dynamics. The potential reasons for the relevance of
particular countries as a source of spillovers are also discussed.
1
While data availability constrains the inclusion of China and other emerging countries in the sample,
we indirectly attempted to test this hypothesis in an earlier version of the study by including exports
to China, to developing Asia, and to the world as additional control variables. Those variables were
not statistically significant and the results remained largely unchanged, suggesting that the sample
countries already capture the bulk of relevant global demand shocks for the period under consider-
ation (1975–2010).
the main sources of growth spillovers? and, What are the main channels of trans-
mission of growth spillovers?
Existing empirical studies generally confirm the existence of spillover effects,
but the empirical evidence about the size of spillovers is inconclusive, as results
are not robust to different samples and specifications. Previous studies frequently
divide the world into regions,2 the euro area being one of them, and examine
spillovers from large advanced countries to these regions.3 The general finding of
these studies is that the United States is the main source of growth spillovers.
Relatively few studies have examined the channels through which growth shocks
are transmitted to the other regions and countries. Existing findings vary across
studies, with simulation-based results suggesting a bigger role for the trade chan-
nel—perhaps due to the difficulty of empirically modeling asset price spillovers
or confidence channels—while VAR analyses, which impose less structure on the
interlinkages, point to the relative importance of financial and other nontrade
channels.
The finding that international spillovers are relatively small under standard trans-
mission channels was first established by Helbling and others (2007). Their study
finds a limited extent of U.S. growth spillovers into other regions—excluding the
euro area and Japan—and even smaller effects of spillovers from the euro area or
Japan, when controlling for possible channels of transmission including commodity
prices (terms of trade) and financial conditions (Libor interest rate). The results
obtained are similar using three alternative approaches (simple panel regressions, a
more sophisticated dynamic analysis, and model simulations). The estimated spill-
overs are moderate in magnitude: the results from annual panel regressions for 130
countries over 1970–2005 suggest that a 1 percentage point decline in U.S. growth
is associated in the long run with an average 0.16 percent drop in growth across the
sample. The findings based on a VAR approach for 46 countries, both advanced and
developing, also suggest that U.S. growth disturbances have on average moderate
dynamic effects on growth in other regions. The simulation results in Helbling and
others (2007) also suggest that the potential spillovers from a temporary, U.S.-
specific demand shock via trade channels alone are moderate, roughly of the same
magnitude as the results from the panel and VAR analyses. However, alternative
simulations, assuming correlated disturbances across countries, generate larger spill-
over effects. The authors of the study conjecture that such a higher impact of U.S.
shocks could arise, for example, if the transmission had also involved asset price
spillovers or confidence channels.
Using a long-run (five-year average) panel regression approach for 101 coun-
tries over 1960–1999, Arora and Vamvakidis (2006) find much larger spillovers.
2
See, for example, Helbling and others (2007), Arora and Vamvakidis (2006), Bayoumi and Swiston
(2009), and Swiston (2010). For an example of similar approaches applied to the case of China spill-
overs, see Arora and Vamvakidis (2010).
3
For the latter, see Bayoumi and Swiston (2009). This study examines the extent of spillovers across
industrial regions including the U.S., the euro area, Japan, and an aggregate of small industrial coun-
tries, using VARs of growth across the four regions.
and the new member countries, with 40 percent and 100 percent, respectively, of
any German growth shock transmitted after three quarters.
A related literature examines fiscal policy spillovers within Europe. Bénassy-
Quéré and Cimadomo (2006) generally find positive cross-border spillovers
from Germany, in the sense that a fiscal expansion in Germany raises GDP
abroad, at least in neighboring and smaller countries. Similarly, Beetsma,
Giuliodori, and Klaassen (2005) find that, averaged across all partner countries,
the effect on foreign GDP of a fiscal stimulus in Germany of 1 percent of GDP
is estimated to be 0.12 percent for a spending increase and 0.03 percent for a
net tax cut. In Chapter 6 of this volume, Ivanova and Weber find spillovers of
a similar order of magnitude for fiscal consolidation in Germany, France, the
United Kingdom, and the United States. The authors argue that even when
multipliers are very large, spillovers are limited, in particular from core EU
countries to peripheral countries, since trade links between the two areas are not
very strong.
Few studies have examined the transmission channels of spillovers, that is, the
major international channels through which shocks are propagated. Using model-
based simulation analysis, Helbling and others (2007) find that most of the U.S.
spillover effects are trade-related, and the effects are relatively small, roughly of
the same magnitude as identified in the panel and VAR analyses. To generate
larger effects, alternative simulations need to assume that disturbances are corre-
lated around the world. The authors posit that such correlated disturbances could
be related to increased trade and/or financial integration and could particularly
arise in times of financial crisis.
The simulation results in Bagliano and Morana (2011) also suggest a rela-
tively more important role for the trade channel as a mechanism for transmitting
U.S. economic developments to the rest of the world. Based on a large-scale open-
economy factor VAR macroeconometric model, Bagliano and Morana find no
clear-cut impact of adverse U.S. financial developments on foreign economic
activity. While increases in a U.S. credit spread index lead to an output contrac-
tion abroad, U.S. stock price dynamics do not have any relevant effect on foreign
GDP, and U.S. house price dynamics only affect the non-OECD group. Hence,
the authors of the study conclude that the trade channel appears to be the key
transmission mechanism of U.S. shocks to the rest of the world.
By contrast, Bayoumi and Swiston (2009) find that the largest estimated con-
tributions to spillovers come from financial rather than trade variables. Their
result is based on a comparison of the response of GDP growth in a basic VAR
model to that in a model augmented with a potential spillover source (either
trade, commodity prices, or financial conditions) to measure the contribution of
this source to the estimated spillovers. In particular, short-term interest rates and
financial conditions more generally (bond yields and equity prices) are found to
play an important role in the international transmission of U.S. growth shocks.
Galesi and Sgherri (2009), using a global VAR approach, also report that equity
prices are the main channel through which—in the short-run—financial shocks
are transmitted from the United States to other countries. They find that other
variables—including real credit growth, real GDP growth, and real interbank
rate—become more important over a two-year horizon.
The analysis of cross-country growth spillovers and, more generally, multicoun-
try estimations is generally hampered by dimensionality constraints. Four different
VAR-based approaches have been suggested to tackle this issue:4 Bayesian VARs,
factor model VARs, global VARs, and VARs based on regional groupings. All four
techniques also require an approach for resolving the identification issue.
• The Bayesian VAR approach tackles the problem with the use of priors
about the cross-country correlation patterns, which are subsequently updat-
ed with the data (Banbura et al., 2008, and Canova and Ciccarelli, 2006).
• Factor models, instead, collapse cross-country co-movements of several
variables into common factors that are then allowed to affect the dynamics
of the individual countries (Bénassy-Quéré and Cimadomo, 2006).
• Global VARs reduce the individual countries’ spillovers to their share in a
weighted average for the variable of interest, which then affects the individual
countries’ dynamics.5 The spillover in the global VAR therefore has a direct
interpretation, unlike the spillover in the factor VAR (Bussière, Chudik, and
Sestieri, 2009; Galesi and Sgherri, 2009; and Dees and others, 2007).
• A fourth approach focuses on a small set of countries or regions—usually
two to four—and then uses the traditional structural VAR (SVAR) approach
(Bayoumi and Swiston, 2009; and Danninger, 2008). The degrees of free-
dom are preserved by reducing the number of regressors, that is, by reducing
either the number of countries involved or the number of variables consid-
ered, or a combination of both.
Bayesian VARs and SVARs are more general than global VARs or factor VARs,
since they impose less structure on the inter-linkages. Compared to SVARs,
Bayesian VARs require making more assumptions on the data generating process
in return for more degrees of freedom, which makes the estimation feasible if the
number of regressors is high relative to the size of the available data sample. The
SVAR approach proposed by Bayoumi and Swiston (2009) requires an extensive
dataset, but has the advantage that it imposes no structure on the inter-linkages,
and thus the coefficient estimates are purely data-driven.
Our analysis takes the existing literature further in the following way:
• We confirm the finding of earlier studies, based on the assessment of spill-
overs across regions, that the United States remains the main source of
growth spillovers, using an approach that involves a larger set of 17 indi-
vidual countries.
4
Another possibility is to use model-based simulation analysis. See, for example, the analysis based on
structural estimated macro models using panel unobserved components estimation as suggested by
Vitek (2009 and 2010).
5
Cross-border trade weights are generally used to estimate the country-specific aggregate foreign vari-
able, although one study uses annual bank lending exposures over 1999–2007 (Galesi and Sgherri,
2009).
EMPIRICAL APPROACH
Given our focus on a univariate growth spillover framework, we follow Bayoumi
and Swiston (2009) and minimize the structure we impose on the data. The
resulting coefficient estimates thus capture all potential channels of transmission
of shocks, including both trade and financial channels (the latter of which may
be the most relevant in times of financial crises, when correlations between all
risky assets tend to rise). The main specification is based on a reduced-form VAR
estimation. Identification is obtained by weighting different orderings. The
results from the different orderings can then be summarized by focusing on the
average impulse response. This approach has the additional advantage that it
provides not only a measure of uncertainty regarding the coefficient estimates but
also a measure of uncertainty associated with the variation of responses across
different orderings. The following derivations are reduced to a minimum when
referring to the Bayoumi and Swiston (2009) approach.
Estimation Framework
The general model is given by the following reduced-form model for the growth
rate of output:
B( L ) yt = D( L )x t + et
where the vector y is given by stacking each country’s GDP growth rate ( yi,t):
yt = ( y1,t y i ,t y I ,t )
the oil shocks in 1979 and in 1990, respectively, and a constant term. When discuss-
ing the results we refer to the baseline regression as the regression which includes
additionally a crisis dummy that takes the value 1 from 2008:Q4–2009:Q1, to reflect
estimates of “normal times,” while the framework without the crisis dummy also
reflects the relationship across growth rates in crisis times.6
Identification is obtained using the Choleski ordering of the countries in the
sample, which provides the structural errors and coefficients:
A(l ) = A(0)−1 B(l ) F (l ) = A (0) D (l )
−1
εt = A(0)−1 et
For the ordering, we distinguish three sets of countries according to their share
in the total sample size, measured in USD PPP-adjusted GDP: Countries that
surpass the 8 percent threshold are considered large countries and can be a leading
country (U.S., Japan, and Germany). Countries that contribute less than 4 per-
cent to the total output are considered small economies (Canada, Netherlands,
Belgium, Sweden, Austria, Switzerland, Greece, Portugal, Ireland, and Finland).
These latter countries are ordered last in the order of their size. Since we focus on
shocks from the major countries to other nations, the ordering for the group of
the remaining smaller countries does not affect the results. The intermediate
group of countries comprises the medium-size countries (United Kingdom,
France, Italy, and Spain). While they are never ordered first, they are generally
ordered before the small countries.
We arrange the orderings roughly according to the respective country’s relative
size. The orderings assign a probability of 50 percent to the United States, being
the lead country (in line with Bayoumi and Swiston, 2009), that is, the country
that is not contemporaneously affected by other countries, and respectively a 25
percent probability to the United States to be ordered second or third. Japan and
Germany are treated symmetrically throughout (although Japan is somewhat
larger than Germany). Germany and Japan both have a chance of being ordered
first, second, or third of 25 percent and a 12.5 percent probability of being
ordered fourth or fifth. The United Kingdom, France, and Italy are also treated
symmetrically, given their comparable average size. Each of these three countries
has a probability of being ordered second or third of 81/3 percent, of being
ordered fourth or sixth of 25 percent, and of being ordered fifth or seventh of
162/3 percent. Spain has a probability of being ordered fifth or sixth of 25 percent
and of being ordered seventh of 50 percent. This size-based procedure results in
48 different orderings, the details of which are provided in the Appendix.
6
Alternative control variables, including the oil and non-oil commodity price indices, U.S. and
German short-term and long-term interest rates, U.S. investment grade and high-yield credit spreads,
German corporate bond spreads, U.S. and German real equity prices, world trade, and Asia trade,
were also included as a robustness check in an earlier version of the analysis. However, none of these
control variables except the U.S. credit spreads and, to a lesser extent, U.S. real equity prices was
significant, and their inclusion left the results unchanged. The impact of the U.S. credit spread, how-
ever, becomes insignificant when included in addition to the 2008–09 crisis dummy, suggesting that
this variable is essentially a proxy for the global financial crisis.
The computation of the standard errors and the ordering uncertainty follow
Bayoumi and Swiston (2009).
Step 2: Compute the moving average (MA) representation of the entire history
of each country’s growth rate.
T
yT = ∑ Mˆ (t )εT −t
t =0
Step 4: Apply the relevant compounding rule to compute the annualized con-
tribution.8
While the ordering of the small countries matters in theory for this exercise,
in practice the ordering of these countries is of minor importance for the decom-
position of the growth contribution, since growth dynamics are dominated by the
larger economies.
A similar calculation applies to the constant term, which allows decomposing
the long-run growth rate into the contributions from the individual countries in
the sample.
Transmission Channels
To gain a better understanding of the transmission channels of the shocks, we
employ a twofold strategy. The first approach uses a counterfactual analysis. We
maintain the estimation framework as outlined above, but contrast the results to
a scenario under which only the direct impact of the shock in country i is allowed
7
Note that yt = Cˆti where i is a column vector of ones, with dimension N in the absence of any
exogenous controls.
8
The rule will depend on whether the dependent variable and the shock are a level variable or a
growth rate.
to affect country j and all spillovers through third countries are prevented. This
provides us with information about the relevance of inter-linkages across coun-
tries for the transmission of shocks. In a second approach, we estimate country-
wise VARs, in which we include real exports as an explanatory variable to test for
the importance of the trade channel.
Counterfactual Analysis
In the counterfactual analysis, we constrain the structural coefficients, which are
associated with third-country effects, to equal zero. Thus, if we are interested in
looking at the direct spillover from country i’s growth shocks to country k’s
growth rate, we set all structural coefficients corresponding to the impact of coun-
try j’s growth shocks on country k equal to zero:9
α kj (l ) = 0
for all l = 0,…,L and all j ≠ i and j ≠ k, where α are the structural coefficients.
Impulse responses are then recalculated under the counterfactual assumptions.
Country-by-Country Regressions
To arrive at a parsimonious specification, we assume the sensitivity to specific
countries to be homogenous within a region and the difference in the slopes
within a group to be random. The United States, Japan, Canada, Sweden,
Switzerland and the United Kingdom comprise the non-European Monetary
Union (non-EMU, or RoW) group, and the rest of the countries constitute the
EMU group. For each country we then run the following regression:
A( L ) yt = et
9
This is likely to underestimate the relevance of the third-country effects since it is still possible that
third countries can have a positive feedback to country i and via this to country k. In practice these
are very small.
10
For all EMU members, the RoW shock is identical, while for all RoW members the EMU shock is
identical.
11
In particular, we look at the average response to an EMU and a non-EMU shock ordering
once Yi ,RoW
t
and once Yi ,EMU
t
first. This leaves the point estimates mostly unchanged. Allowing addi-
tionally for the following ordering Yt = ( EXPi ,t Yi ,t Yi ,RoW
t Yi ,EMU
t ) affects the point estimates, but
not the relative magnitude across countries, leaving the interpretations unchanged.
Thus, this model and the baseline estimation are identical under the assump-
tion that α i ,t (l ) = α i ,t −t (l ) = α i (l ) for all l = 1,.., L and all t = 1,..,T for this
model and all α ki (l ) = α k (l ) + vki with E (v ) = 0 and E (vv ) = σ v I in the baseline
model and the extension that the real exports of country k are included as an
additional variable.12
RESULTS
This section provides an overview of the key transmission channels (trade vs. finan-
cial linkages) to help interpret the results, followed by a discussion of the main
results derived from the estimation framework. The first set of results focuses on the
potential impact of spillovers from a one percent growth shock in selected countries
on other European countries. The second set of results presents the actual impact
of selected countries on all other countries in the run-up to, during, and in the
recovery from the recent financial crisis by combining the country-specific shocks
with the impulse response functions. A third and fourth set of results, respectively,
point to the relevance of the different transmission mechanisms of growth shocks
and of different potential determinants of spillover size.
Cross-Border Linkages
To interpret our results on the growth linkages, it is helpful to understand some
key facts about the relative exposures of countries in the sample to the largest
economies and regions. The most obvious channel is trade linkages: a rise in tra-
ding partners’ growth leads to an increase in their demand for imports, which
12
Unsurprisingly, this turns out not to hold, and the aggregation bias causes responses to be more
pronounced (Imbs and others, 2005). Results are discussed in more detail in the respective section.
TABLE 5.1
then contributes directly to an increase in the net exports of the home country
(Table 5.1). With growing financial integration and cross-border ownership of
assets, growth spillover effects may also be transmitted through financial linkages
(Table 5.2). Both trade and financial exposures highlight the importance of intra-
euro area transmission channels. For most European countries in our sample, the
euro area as a whole is by far the biggest export market and accounts for the larg-
est single banking sector exposure. The United States is also a key source of
financial spillover risk for European countries. Within the euro area, it is note-
worthy that for most countries (except the smaller trading partners), trade expo-
sures to Germany are relatively smaller than trade exposures to the rest of the euro
area as a whole, reflecting Germany’s relatively limited demand for imports from
other European advanced countries. Several European countries, however, have
large financial exposures to Germany.
Trade exposures follow a strong regional pattern. In particular, they suggest a
limited relevance of Asia for the sample countries (although growing in impor-
tance in the case of China). More specifically, we find that:
• Trade links within Europe are important, as is evident in Table 5.1. The euro
area is the largest export market for member countries in the sample, except
Greece. For the Netherlands and Belgium, exports to the euro area account
for about half of GDP, with a sizeable share (30 to 40 percent) directed to
Germany alone.13 Even for Ireland, which has close trade ties with the
13
This holds also true when controlling for re-exports.
TABLE 5.2
United States, the share of exports going to the euro area exceeds that going
to the United States.
• The euro area is also the largest importer for the United Kingdom and
Switzerland, and the second largest for the United States. However, it is only
the fourth export destination for Japan, which relies more on U.S., Chinese,
and other regional markets.
• Despite its size and reflecting a relatively closed economy, the United States
is generally not the major trading partner for European countries.
• Trade exposures to Japan are even more limited. Trade exposures to China
are also limited (except in the case of Japan), although the growing impor-
tance of China is underscored by the fact that China has now overtaken
Japan in its importance as importer for all sample countries (except Ireland).
• Similar to the United States and Japan, Germany’s relevance as an importer
is relatively limited—including for other eurozone countries. While
Germany is the second largest export destination for Austria and the
Netherlands, it is only at best the third largest export exposure for the other
European countries. These countries export relatively more either to the
non-German euro area as a whole or to the rest of the world (excluding the
euro area, the United States, the United Kingdom, China, and Japan). In
the case of non-euro area countries, exports to Germany account for less
than 0.5 percent of GDP.
Financial linkages single out the United States and Europe, in the aggregate,
as main sources of global spillover risks, while Germany’s importance is mostly
regional. Overall, financial exposures—proxied by bilateral bank lending expo-
sures on an ultimate risk basis—follow a broadly similar regional pattern as trade
exposures, illustrating the relative importance of intra-European linkages:
• For European countries, exposure to European developed countries is the
single largest source of spillover risks, amounting to 30 percent of GDP or
more (except in Finland and Greece).14 For the United States, exposure to
Europe also ranks as the first source of risks; however, relative to GDP many
European countries (including Switzerland, the United Kingdom, France,
Germany, and the Netherlands) have much larger exposures to the United
States than the United States does to Europe.
• Within Europe, some countries have high exposures to Germany of 10
percent of GDP or more (France, Austria, Switzerland, Netherlands,
Ireland, Sweden). By contrast, non-European countries (United States,
Canada, and Japan) have banking exposures to Germany at or below 3 per-
cent of GDP, well below their level of exposures to the United Kingdom in
the case of the United States and Canada.
• For Canada and Japan, financial linkages highlight the key role of the
United States as a source of spillover risks, with European exposure the
second largest source of risk.
• Exposures to Japan are at or below 6 percent of GDP for all countries
(except Switzerland), and exposures to China are below 1 percent of GDP
in most cases (except the United Kingdom, the Netherlands, and
Switzerland).
14
In all cases, except for Greece, Spain, Austria, and the U.K., which also have significant exposures to the
rest of the world (excluding the U.S., Japan, and Europe) and Switzerland (which has the largest single
exposure to the U.S.), European developed countries are the single largest source of banking exposures.
2.50
Response by all other countries to a growth shock in...
1.50
1.00
0.50
0.00
United States Japan Germany United Kingdom Rest of EAb GIP
2.50
Response of the GIP to a growth shock in...
1.00
0.50
0.00
−0.50
Figure 5.2 Cumulative Peak Impulse Response after 10 Quarters to a 1 Percent Growth
Shock (In percent)a
Source: IMF staff estimates.
a
GDP-weighted average response of other countries.
b
Excluding Germany.
influence on Ireland and Portugal. This is much less the case for Germany.
However, Germany tends to have a higher impact on Greece than Italy does,
but it still has less impact on Greece than France does.
• The United States has much more of an impact on Ireland than on the other
GIP countries, consistent with results in Kanda (2008) and the close trade
and financial linkages between the two countries. The United States also has
2 0.7
After 1 year After 10 quarters Large model
1.8
0.6
1.6
1.4 0.5
1.2
0.4
1
0.3
0.8
0.6 0.2
0.4
0.1
0.2
0 0.0
Germany Italy France Spain Germany Italy France Spain
EMU shock Non-EMU shock
Figure 5.3 Sensitivity to a 1 Percent Growth Shock in the Large European Countriesa
Source: IMF Staff estimates.
a
Right axis = weighted impact on growth in percentage points in response to shock under the large model;
left axis = impact on growth in percentage points in response to shock under the small model.
independent spillovers and more of a conduit for U.S. and other external shocks
to the rest of Europe. A second set of estimates from the main baseline estimation
(i.e., large model)15 yields the same rank ordering of sensitivity to external shocks
as the smaller country-specific VARs.16
Regression results for the more recent period (since 1993, based on the large
model) show that inward spillovers from the United States to the four large EMU
member states have increased, while the United Kingdom and Japan have become
less relevant (Figure 5.4). In the baseline estimation, in normal times a 1 percent
growth shock in the United States tends to increase output growth within 10 quar-
ters by about 0.3 percent in Germany, 0.4 percent in Italy and France, and 0.1
percent in Spain. These values increase to 0.4 percent in Germany and Spain, 0.5
percent in France, and 0.6 percent in Italy when not controlling for the effect of
crisis times. The respective values more than double for Germany and Spain, in the
more recent episode (estimation for 1993:Q1–2010:Q3 including a crisis dummy),
as both become more sensitive to the United States than France and Italy.
The increased sensitivity to the United States is in line with the increased cor-
relation of the EMU countries’ GDP growth with the lagged GDP growth of the
United States (Figure 5.1). This reflects rising financial linkages, which have
become more important, particularly in the latter half of the sample period. It is
consistent with similar findings by Helbling and others (2007), who find spill-
overs to be increasing in financial and trade linkages and significantly higher
spillovers from the United States to other countries in the 1987–2006 period as
compared to the entire 1970–2006 period. Since our sample includes years that
mark the peak of financial linkages (2007 and 2008), we find even higher effects.
Within the EMU, inward spillovers from Germany, Italy, and France to the
other large EMU members are relatively stable across the two sample periods and
are robust to the inclusion of the crisis dummy (i.e., likely to persist even outside
of crisis times) (Figure 5.4). Specifically, we find that:
• Germany’s effect on the other three large euro area countries ranges at the
lower end for Spain and France, causing output there to rise by 0.1–0.2
percent, although Italy’s growth rate rises by around 0.4 percent in response
to a 1 percent growth shock in Germany.
• France affects Italy and Germany roughly by the same order of magnitude,
yielding an increase in growth by 0.4 percent in normal times and above 0.5
percent in crisis times. The effect on Spain is only marginally higher.
• A 1 percent growth shock in Italy causes German and French growth to
increase by 0.4 percent and Spanish growth by 0.2 percent in the baseline
15
While estimates from the smaller model are directly obtained for “EMU” and “Non-EMU” shocks,
the corresponding values from the baseline VAR (large model) is obtained by weighting the responses
to the single countries’ shocks which constitute the EMU and the non-EMU group in the country-
specific VARs.
16
However, it should be noted that the small country-specific VARs overestimate the impact, due to
the aggregation bias which tends to increase the persistence of the shocks and thus overestimate the
response.
2.50 Inward Spillovers to Germany from Selected 2.50 Inward Spillovers to France from Selected
Large Countries Large Countries
2.00 2.00
1993Q1:2010Q3 Excl. crisis dummy 1993Q1:2010Q3 Excl. crisis dummy
1993Q1:2010Q3 1993Q1:2010Q3
1975Q1:2010Q3 Excl. crisis dummy 1975Q1:2010Q3 Excl. crisis dummy
1.50 1975Q1:2010Q3 1.50 1975Q1:2010Q3
1.00 1.00
0.50 0.50
0.00 0.00
USA JPN DEU GBR FRA ITA ESP USA JPN DEU GBR FRA ITA ESP
−0.50 −0.50
2.50 Inward Spillovers to Italy from Selected 2.50 Inward Spillovers to Spain from Selected Large
Large Countries Countries
2.00 2.00
1993Q1:2010Q3 Excl. crisis dummy 1993Q1:2010Q3 Excl. crisis dummy
1993Q1:2010Q3 1993Q1:2010Q3
1975Q1:2010Q3 Excl. crisis dummy 1975Q1:2010Q3 Excl. crisis dummy
1.50 1975Q1:2010Q3 1.50 1975Q1:2010Q3
1.00 1.00
0.50 0.50
0.00 0.00
USA JPN DEU GBR FRA ITA ESP USA JPN DEU GBR FRA ITA ESP
−0.50 −0.50
Figure 5.4. Inward Spillovers to the Four Large Euro Area Countries 1993–2010 and 1975–2010
Source: IMF Staff estimates.
TABLE 5.3
ITA 0.9 0.6 –0.2 –0.2 0.1 0.4 0.2 0.0 1.5
ESP 0.7 0.3 0.0 –0.3 0.1 –0.1 0.2 0.1 2.1
CAN 2.1 0.3 0.0 –0.2 –0.2 –0.1 0.4 0.0 2.7
NLD 1.3 0.3 –0.2 –0.2 0.0 –0.1 0.6 0.2 1.9
BEL 0.7 0.4 –0.1 –0.1 0.1 –0.2 0.5 0.2 –0.1 1.6
SWE 1.4 0.2 0.1 –0.4 0.1 –0.2 0.6 0.2 0.0 2.1
AUT 0.5 0.2 –0.2 –0.1 0.2 –0.1 0.5 0.1 –0.5 1.3
CHE 0.8 0.4 0.0 –0.2 0.2 –0.3 0.5 0.1 –0.1 1.4
GRC 0.8 0.1 –0.3 –0.1 0.3 –0.1 0.8 0.1 –0.3 1.7
POR 0.2 0.5 –0.2 –0.1 0.4 –0.2 0.9 0.2 –0.7 1.7
FIN 1.6 0.4 0.2 0.3 0.3 –0.2 0.7 0.4 0.5 2.8
IRL 2.3 0.1 –0.1 0.2 0.2 –0.2 1.1 0.4 –0.7 3.6
Source: IMF staff estimates.
Note: USA: United States; JPN: Japan; DEU: Germany; GBR: United Kingdom; FRA: France; ITA: Italy; ESP: Spain; CAN: Canada.
in recent years, consistent with the increase in trade and financial integra-
tion following EMU and euro adoption as well as with a domestic demand
and property boom in Spain, both partly fuelled by unsustainable increases
in corporate and household indebtedness.17
17
While the subsequent deflation of the property bubble and private sector deleveraging has resulted
in negative dynamic contributions of Spain to other countries’ growth during the 2008–09 global
recession, we find that on average, over the long run, Spain has been one of the major sources of
positive growth spillovers to other countries, especially in Europe (see section D). However, the poten-
tial positive impact of Spain in future episodes could be lower than suggested by historical results if
the unwinding of Spain’s imbalances is protracted and durably undermines Spain’s growth prospects.
−10
−8
−6
−4
−2
0
2
4
6
8
−10
−8
−6
−4
−2
0
2
4
6
8
20
–10
–8
–6
–4
–2
0
2
4
6
05 05
−10
−8
−6
−4
−2
0
2
4
6
8
05 05
:Q :Q
: 20 1 20 1 20 :Q1
20 Q1 05 05
05 05 :Q
: :Q
: 20 Q3 20 3 20 3
20 Q3 06 06 06
06 :Q :Q :
: 1
ITA
1 20 20 Q1
own
20
CAN
20 Q1 06 06
06 06 :Q :Q
: :Q 3 20 3
20 Q3 20 3 20
07
07
07 07
: : :
: 20 Q1 20 Q1 20 Q1
20 Q1 07 07 07
07 :Q :Q :Q
: 3 3
JPN
3 20 20
GBR
20
Italy
20 Q3 08 08
Austria
08 08 : :Q
Germany
: : 1
Long-run
20 Q1 20 Q1 20
20 Q1 08 08
08 08 :Q :Q
United States
:Q :Q 3 3
20 3 20 3 20
09
20
09
09 09 : :
: : 20 Q1 20 Q1
20 Q1 20 Q1 09 09
09 09 : :
FRA
ESP
: : 20 Q3
others
20 Q3 20 Q3
20 Q3 10 10
10 10 :Q :Q
: :Q 1 20 1
20 Q1 20 1 20
10
10 10 10
:Q :Q :Q
:Q
3 3 3
3
USA
DEU
20 20
20
20
–10
–8
–6
–4
–2
0
2
4
6
−15
−10
−5
0
5
10
−10
−8
−6
−4
−2
0
2
4
6
8
−10
−8
−6
−4
−2
0
2
4
6
8
05 05
05
: 05
Growth rate
:Q
20 Q1 20 :Q1
1
20 20 :Q1
05 05 05 05
: :Q :Q
20 Q3 20 :Q3 20 3 20 3
06 06 06 06
: : : :Q
20 Q1 20 Q1 20 Q1 20 1
06 06 06 06
: :Q :Q :Q
20 Q3 20 3 20 3 20 3
07 07 07 07
: : :
08 08 08 08
: :Q Finland :Q :Q
Netherlands
20 Q1 20 1 20 1 20 1
Japan
08 08 08 08
: :Q :Q :Q
20 Q3 20 3 20 3 20 3
09 09 09 09
: : : :
20 Q1 20 Q1 20 Q1 20 Q1
09 09 09 09
:Q : : :
20 3 20 Q3 20 Q3 20 Q3
10 10 10 10
: :Q :Q :Q
1 1
3
119
20 20
120
0
2
4
6
8
-8
-6
-4
-2
–10
–8
–6
–4
–2
0
2
4
6
–10
–8
–6
–4
–2
0
2
4
6
−10
−8
−6
−4
−2
0
2
4
6
8
-10
20 20 20
−10
−8
−6
−4
−2
0
2
4
6
8
05 05 05 05 05
:Q :Q :Q : :
20 1 20 1 20 1 20 Q1 20 Q1
05 05 05 05 05
:Q :Q : :
States.
20 3 20 3 :Q
06 06 20 3 20 Q3 20 Q3
Figure 5.5
06 06 06
:Q :Q :Q : :
20 1 20 1 20 1 20 Q1 20 Q1
06 06 06 06 06
:Q :Q :Q : :
20 3 20 3 20 3 20 Q3 20 Q3
07 07 07 07 07
:Q :Q :Q : :
20 1 20 1 20 1 20 Q1 20 Q1
07 07 07 07 07
:Q :Q :Q : :
20 3 20 3 3 20 Q3 20 Q3
(continued)
08 08 08 08 08
Greece
:Q : :
Portugal
:Q :Q
1 1
Spain
20 20 1
Sweden
20 20 Q1 20 Q1
08 08 08 08 08
:Q :Q :Q : :
United Kingdom
20 3 20 3 20 3 20 Q3 20 Q3
09 09 09 09 09
:Q :Q :Q :Q :
20 1 20 1 20 1 20 1 20 Q1
09 09 09 09 09
:Q :Q :Q : :
20 3 20 3 20 3 20 Q3 20 Q3
10 10 10 10 10
:Q :Q :Q : :
20 1 20 1 20 1 20 Q1 20 Q1
10 10 10 10 10
:Q :Q :Q :Q :Q
3 3 3 3 3
20 20 20
−10
−8
−6
−4
−2
0
2
4
6
8
20
−10
−8
−6
−4
−2
0
2
4
6
8
05
–10
–8
–6
–4
–2
0
2
4
6
05 05
–10
–8
–6
–4
–2
0
2
4
6
05
:Q :Q : :
1
Growth Spillover Dynamics: From Crisis to Recovery
20 20 1 20 Q1 20 Q1
05 05 05 05
:Q :Q : :Q
20 3 20 3 20 Q3 20 3
06 06 06 06
:Q :Q : :
20 1 20 1 20 Q1 20 Q1
06 06 06 06
:Q :Q : :
20 3 20 3 20 Q3 20 Q3
07 07 07
ITA
:Q 07 : :
own
JPN
CAN
GBR
20 1 :Q 20 Q1 20 Q1
07 20 1 07 07
07 : :
Long-run
:Q :Q
20 3 20 3 20 Q3 20 Q3
08 08 08
Ireland
:Q 08 : :
20 1 :Q 20 Q1 20 Q1
20 1 Belgium
Canada
08 08 08 08
:Q :Q :Q
Switzerland
20 3 :Q 20 3 20 3
09 20 3 09 09
09 : :
FRA
ESP
:Q
USA
DEU
1 :Q
others
20 20 1 20 Q1 20 Q1
09 09 09
:Q 09
:Q : :
20 3 20 Q3 20 Q3
Growth rate
10 20 3 10 10
:Q 10
:Q : :
20 1 1
20 Q1 20 Q1
Note: CAN: Canada; DEU: Germany; ESP: Spain; FRA: France; ITA: Italy; JPN: Japan; GBR: United Kingdom; USA: United
Poirson and Weber 121
of the growth rate from 2005:Q1 to 2010:Q3 for all countries in the sample,
splitting a country’s growth rate into its own contribution (light grey bar), the
cyclical contribution stemming from each G7 member and Spain, and the long-
run growth rate (dark grey bar). The height of the individual bars represents the
overall contribution at each moment in time.
We first examine the long-term impact of each country on the other countries
in the sample. The long-run decomposition in Table 5.3 shows that the United
States has been the largest positive contributor to long-term growth in other
countries. The United States matters particularly for the Anglo-Saxon countries
and the smaller Northern European countries. Long-run spillovers from Japan
and Spain have also been positive and relatively important, with Spain more
relevant for European countries. Canada and France provide relatively minor
long-run growth support to other countries, although spillovers from France are
particularly relevant for the GIP. By contrast, long-run external growth spillovers
from Germany are close to zero, and the United Kingdom and Italy’s long-run
spillovers have been small and negative. In the case of Italy, this reflects rela-
tively weak GDP growth over the period (similar to Germany), while the result
for the United Kingdom could reflect the United Kingdom’s dependence on
U.S. prospects.
The dynamic contribution analysis—focusing on the recent period, including
the global financial crisis—highlights for all countries the dominant contribution
of external growth shocks and the relatively low contribution of domestic shocks
to overall GDP growth. This is particularly true for the smaller open economies
such as the Netherlands, Austria, Belgium, and Finland. Even for the United
States, the results suggest that synchronized downturns in Japan and the European
advanced countries contributed to amplify the depth of the U.S. recession in
2008–09. The finding that external spillovers have been large and significant in
the recent period may be regarded as supporting the appropriateness of the
model, since it implies that the model captures most of the likely sources of
global growth spillovers and thus generates a limited idiosyncratic error compo-
nent, which in turn implies a relatively high explanatory power.18
Turning to a finer analysis of growth contributions pre-, during, and post-
crisis, the main findings are as follows:
• The boom period before the crisis is reflected in the significant domestic
contribution to each country’s GDP growth, which cannot be accounted for
completely by growth fluctuations in other countries. In terms of outward
spillovers, the United States and Spain have been major sources of positive
growth support in the precrisis period, with the United States mattering
most for Canada, the United Kingdom, and Ireland. France played an
important role for some countries in the run-up to the crisis, notably the
southern peripheral countries, as well as Belgium, Austria, and Finland.
18
The average (adjusted) R-squared value of the reduced form equations for the baseline model is
around 0.6 (0.4). Including the crisis dummy implies an increase by 0.04 in explanatory power in
both cases.
19
While the cyclical contributions of U.S. and Japan growth dynamics on other countries are gener-
ally small, both countries provide the bulk of the long-run growth rate support.
case of France and Italy. The United Kingdom in turn appears mostly relevant for
the northern European economies and for Spain.
Third-Country Transmission
In a first approach, we analyze the extent to which third-country effects are relevant
for the transmission of shocks. In particular, we identify the countries most directly
affected by global shocks versus those primarily affected by shocks through third-
country effects and that therefore tend to be affected only with a lag. Further, if third-
country effects are important, countries less relevant as a source of shocks may still
play a central role as transmitter and amplifier of shocks to other countries, through
either trade or financial linkages, and therefore they matter on a systemic level.
The relevance of third-country effects is calculated as the fraction of the inward
spillover from a shock in a given country that is transmitted through other coun-
tries after one year. The lower this indicator of third-country effects, the more
directly sensitive the country is to the impact of a given external shock. Conversely,
the higher the indicator, the more likely it is that the country will only respond
after a lag, since the entire spillover effect needs more time to trickle down through
trade or financial channels. Since the measure of third-country transmission is only
meaningful in cases where a significant impact is observed, we calculated third-
country effects only for such constellations.
20
In an earlier version of the paper, we attempted to test for the relevance of financial transmission
channels by following the approach of Bayoumi and Swiston (2009), i.e. by including global financial
variables (such as the U.S. equity prices, interest rates, and U.S. and German credit spreads) as addi-
tional control variables. The U.S. credit spread was found to be the most significant variable, with an
impact similar to that of including the crisis dummy (i.e. reducing estimated outward spillovers from
the U.S. and other large countries); however, the spread variable had not statistically significant effect
once the crisis dummy is also included, suggesting that this channel of transmission is only relevant
during times of crisis.
Germany stands out as a country that tends to respond swiftly and directly to
shocks to growth in the United States or Japan. For France, Italy, and the small core
euro area members, similar shocks are channeled to a larger extent through third
countries before they impact growth in the home economy. The relatively high
values of the third-country indicator for the non-German euro area suggest that
inter-linkages between euro area countries are highly relevant in the transmission of
shocks to Europe from outside the euro area. Combined with the relative directness
with which Germany reacts to shocks in the United States and Japan, this finding
suggests that Germany acts as an important transmitter and amplifier of shocks
originating outside the euro area to other euro area members (Figure 5.6).
When looking at the relevance of third-country effects for intra-European
shocks, the regional patterns are again noticeable. Growth shocks from Italy and
Spain affect Dutch growth primarily through third countries, while growth
shocks from Germany affect Dutch growth mostly directly and hardly at all
through third countries. Switzerland’s growth is directly affected by growth
shocks in Italy, and Belgium’s growth is most directly affected by French growth
shocks. Austrian growth is most directly affected by France and Germany, less
directly by Italy, and least by Spain. Italy’s growth is most directly affected by
Germany followed by France and then Spain (Figure 5.7).
Interestingly, Germany’s impact on Greece and Portugal appears to be less
direct than the impact of France and Italy on these two countries, which is com-
parable to the impact of Spain on Portugal. This confirms the relevance of France
and Italy for the southern peripheral countries. Finally, it is worth noting that in
120
United States
Japan
Average over United States and Japan
100
80
60
40
20
0
Germany France Italy Netherlands, Belgium,
Austria, and Finland
Figure 5.7 Relative Importance of Third-Country Effects for Spillovers from the Four Large
European Countries to Selected European Countries
Source: IMF staff estimates.
Note: AUT: Austria; BEL: Belgium; CHE: Switzerland; DEU: Germany; ESP: Spain; FIN: Finland; FRA: France;
GRC: Greece; IRL: Ireland; ITA: Italy; NLD: Netherlands; PRT: Portugal; SWE: Sweden.
crisis times, third-country effects appear to play a larger role, confirming earlier
results on the importance of confidence and asset-price spillover effects during
times of financial distress.21
21
Although the VAR modeling framework does not allow testing directly for asymmetry in the pattern
of spillovers, the importance of third-country effects during times of financial distress could explain
why negative spillovers originating during a crisis tend to be empirically larger than either positive or
negative spillovers outside of crisis times: unlike “normal” spillovers, “crisis” spillovers tend to be
amplified to a greater extent by confidence and asset price effects.
22
See for instance Rudebusch (2005) for the case of monetary policy.
23
See Vitek (2010) for model simulation-based evidence of a strong transmission of supply shocks via
non-trade channels in a monetary union.
0.2 0.2
0 0
Germany Italy France Spain Germany Italy France Spain
Figure 5.8 Relative Importance of Trade Channel for Inward Spillovers for Large EMU Members
Source: IMF staff estimates.
Note: EMU: Austria, Belgium, Finland, Greece, Ireland, Netherlands, Portugal, and Spain.
GDP) and to the existence of autonomous domestic drivers of growth. While size
is a natural a priori determinant of spillovers, we also conjecture that countries
that are relatively more sensitive to external shocks are more likely to receive large
inward spillovers but less likely to generate large outward spillovers; by contrast,
countries that rely more on domestic drivers of growth are more likely to generate
large outward spillovers.
To identify the presence of autonomous sources of domestic demand growth,
we look at two dimensions: (i) the average contribution of trade to GDP, and (ii)
the co-movement of GDP growth and net exports. The former gives us an indi-
cation of the relevance of the external sector for overall GDP growth. The latter
helps us to understand whether GDP growth and net exports move in tandem
or whether net exports and GDP growth move in opposite directions. In the
latter case, a country is a potential spillover risk (positive or negative) for other
countries, since its growth relies to a greater extent on autonomous domestic
demand fluctuations, while in the former the country is more likely to import
spillovers through trade and other links, and possibly re-export them to others
by serving as a third-country transmitter rather than acting as an independent
engine of growth.
Countries that exhibit a positive average net contribution of trade to GDP
include Japan, Germany, Netherlands, Belgium, Sweden, Austria, Switzerland,
Finland, and Ireland; those that do not include the United States, the United
Kingdom, France, Italy, Spain, Canada, Greece, and Portugal (Table 5.4).24 In
terms of relevance of external demand for overall GDP growth, Germany is the
leading country, followed by Ireland, Switzerland, Japan, Austria, Sweden,
Finland, Belgium and the Netherlands. In the last decade, exports have become
the major engine of growth for Japan, Austria, and particularly for Germany.
Regarding the co-movement of external demand and GDP growth, the coun-
tercyclical pattern of domestic and external contribution to growth indicates the
24
Note that this concept is not identical to whether a country is a net exporter or net importer, since
it refers to the change in the net trade position rather than the level.
greatest spillover risk is in the case of the United Kingdom, the United States,
Canada, Spain, Portugal, and Greece (Table 5.5). These countries have a high
negative correlation between the contribution of domestic and external demand
and a high positive correlation between domestic demand and GDP. This represents
TABLE 5.4
TABLE 5.5
a pattern under which high domestic demand drives growth and worsens the
net trade position due to increased domestic demand.
Germany, Japan, Austria, and Finland illustrate the opposite, export-driven
growth pattern (Table 5.5). These countries exhibit a positive correlation between
the external contribution to growth and GDP and a positive correlation between
the domestic contribution and the external contribution to GDP. Higher external
demand accelerates growth, which in turn stimulates domestic consumption and
investment. For the Netherlands, Sweden, France, and Belgium, the patterns
appear less pronounced according to these indicators.
Plotting the correlation between external contribution to growth and GDP
against the weighted outward spillovers from the main (baseline) regression, we
find that the link between the presence of domestic drivers of growth and the
relevance of a country as a source of growth spillovers to other countries is con-
firmed.25 While we also find evidence of a positive relationship between a coun-
try’s size and the estimated outward spillover, the presence of clear outliers such
as Canada and Spain (larger spillovers than predicted by size alone), or Germany
(smaller spillovers than expected based on size) suggest that size alone fails to
explain spillover risk (Figure 5.9).
0.60 External Contribution to Growth and Size of Growth 0.6 Economic Size and Size of Growth
Growth spillover, 1975:Q1–2010:Q3, percent
Growth spillover, 1975:Q1–2010:Q3, percent
Spillovers Spillovers
0.50 0.5
Canada Canada
0.40 USA 0.4 USA
Spain Spain
0.30 0.3
Italy Italy
Figure 5.9 Correlation of Outward Spillovers, Size and Export-Driven Growth, Selected
Countries
Source: IMF staff estimates.
25
The relationship also holds when excluding the crisis dummy. Regressing the size of the outward
spillover on a constant, the log size of the country, and the correlation between GDP growth and the
external contribution to growth yields a positive significant effect for the former and a negative sig-
nificant effect for the latter. Increasing the size of the country by 10 percent increases the outward
spillover by 0.1 percentage points, and reducing the correlation of external demand and GDP growth
from +0.5 to −0.5 increases the size of outward spillovers by 0.14 percentage points. The smallest four
counties are excluded from the graph. While Finland confirms the pattern, Greece, Portugal and
Ireland are too small to generate significant growth spillovers.
CONCLUSION
The divergent growth recovery paths in Europe in the aftermath of the 2008–09
financial crisis have brought the question of growth spillovers again to the fore-
front, since spillovers from faster growing countries could provide positive growth
impetus to the slower growing countries, whose recovery is hampered by domes-
tic demand constraints. Furthermore, against the background of renewed sover-
eign debt concerns in the euro area, negative growth shocks originating in the
crisis countries could spill over to other countries through trade linkages or
extensive cross-border asset ownership.
Our empirical analysis suggests that growth spillovers can indeed explain a
significant fraction of the variation in output growth for some euro area members,
in particular the small open economies. In terms of spillover origins, we find that
the United States remains the key source of growth spillovers in this recovery.
Despite the increased correlation of Germany’s GDP growth with several other
countries’ GDP growth, its spillover impact remains primarily confined to its
smaller neighbors, while France and Italy are more relevant for the southern
peripheral countries. Both Japan and Spain also appear to be significant sources
of potential growth spillover risk to European countries, although the positive
spillover impact of Spain in future episodes could be much smaller than indicated
by our results if domestic demand growth is persistently hampered by the ongo-
ing unwinding of longstanding imbalances. To some extent, a similar caveat
applies to the U.S. results in light of uncertainties surrounding the sustainability
of the current U.S. recovery.
Our analysis of transmission channels suggests that trade channels matter rela-
tively less than financial and other non-trade channels. Trade effects seem to mat-
ter relatively more for Germany’s inward spillovers, compared to other large euro
area countries, reflecting Germany’s relatively high trade exposures. For all euro
area members, we find that growth shocks from outside the EMU are more likely
than those originating within the EMU to be transmitted through trade; those
originating within the EMU appear to be predominantly transmitted through
monetary and financial linkages. We fail to find an increasingly important role for
Germany as an independent source of growth shocks in the euro area in recent
years, which reflects the country’s relatively high sensitivity to external shocks.
Finally, the results suggest that growth spillover risks from the European crisis
countries to the rest of Europe remain limited, although stronger effects could be
expected if the debt crisis were to spread to larger countries, such as Spain.
Our results are consistent with the premise that for countries to be an impor-
tant source of growth spillovers, their growth should rely to a greater extent on
autonomous domestic sources. Germany fails to meet these criteria, since it is
more sensitive to growth shocks in other countries than other large advanced
countries and its GDP growth tends to closely follow the performance of its
external sector. However, consistent with the view that Germany’s imports are
very sensitive to demand for German exports (due to the high import content
of German exports), our results suggest that Germany may be acting as an
REFERENCES
Arora, Vivek, and Athanasios Vamvakidis, 2006, “The Impact of U.S. Growth on the Rest of
the World: How Much Does It Matter?” Journal of Economic Integration, Vol. 21, No. 1, pp.
21–39.
———, 2010, “China’s Economic Growth: International Spillovers?” IMF Working Paper No.
WP/10/165 (Washington, DC: International Monetary Fund).
Bagliano, Fabio C., and Claudio Morana, 2011, “The Great Recession: U.S. Dynamics and
Spillovers to the World Economy,” Journal of Banking and Finance, forthcoming.
Banbura, Marta, Domenico Giannone, and Lucrezia Reichlin, 2008, “Large Bayesian VARs,”
Working Paper No. 966 (Frankfurt: European Central Bank).
Bayoumi, Tam, and Andrew Swiston, 2009, “Foreign Entanglements: Estimating the Source
and Size of Spillovers Across Industrial Countries,” IMF Staff Papers, Vol. 56, No. 2,
pp. 353–83.
Beetsma, Roel, Massimo Giuliodori, and Franc Klaasen, 2005, “Trade Spillovers of Fiscal Policy
in the European Union: A Panel Analysis,” DNB Working Paper No. 52 (Amsterdam: De
Nederlandsche Bank).
Bénassy-Quéré, Agnès, and Jacopo Cimadomo, 2006, “Changing Patterns of Domestic and
Cross-Border Fiscal Policy Multipliers in Europe and the U.S.,” Working Papers, Centre
d’Etudes Prospectives et d’Informations Internationales No. 2006–24 December.
Bussière, Matthieu, Alexander Chudik and Giulia Sestieri, 2009, “Modelling Global Trade
Flows - Results from a GVAR Model,” Working Paper Series 1087, European Central Bank.
Canova, Fabio, and Matteo Ciccarelli, 2006, “Estimating Multi-Country VAR Models”
Working Paper Series 603, European Central Bank.
Danninger, Stephan, 2008, “Growth Linkages within Europe“, in IMF Country Report No.
08/81, Germany: Selected Issues, International Monetary Fund. (Washington, DC:
International Monetary Fund).
Dees, Stephane, Filippo di Mauro, L. Vanessa Smith and M. Hashem Pesaran, 2007, “Exploring
the international linkages of the euro area: a global VAR analysis,” Journal of Applied
Econometrics, John Wiley & Sons, Ltd., vol. 22(1), pages 1–38.
Galesi, Alessandro, and Silvia Sgherri, 2009, “Regional Financial Spillovers Across Europe: A
Global VAR Analysis,” IMF Working Papers 09/23, International Monetary Fund.
Helbling, Thomas, Peter Berezin, Ayhan Kose, Michael Kumhof, Doug Laxton, and Nicola
Spatafora, 2007, “Decoupling the Train: Spillovers and Cycles in the World Economy,”
Chapter 4 in World Economic Outlook, April 2007 (Washington, DC: International Monetary
Fund).
Imbs, Jean, Haroon Mumtaz, Morten O. Ravn, Hélène Rey, 2005, ”PPP Strikes Back:
Aggregation and the Real Exchange Rate,” The Quarterly Journal of Economics, Vol. 120, No.
1 (Feb.), pp. 1–43 (Oxford: Oxford University Press).
Kanda, Daniel, 2008, “Spillovers to Ireland,” IMF Working Paper No. 08/2 (Washington, DC:
International Monetary Fund).
Lucas, Robert, 1976, “Econometric Policy Evaluation: A Critique,” Journal of Monetary
Economics, Supplementary Series 1, pp. 19–46.
Pesaran, M. Hahsem, Til Schuermann, and Scott M. Weiner, 2004, “Modeling Regional
Interdependencies Using a Global Error-Correcting Macroeconometric Model,” Journal of
Business & Economic Statistics, Vol. 22, pp. 129–62.
Rudebusch, Glenn D., 2005, “Assessing the Lucas Critique in Monetary Policy Models,” Journal
of Money, Credit, and Banking, Vol. 37, pp. 245–72.
Swiston, Andrew, 2010, “Spillovers to Central America in Light of the Crisis: What a Difference
a Year Makes,” IMF Working Paper No. WP/10/35, (Washington, DC: International
Monetary Fund).
Vitek, Francis, 2009, “Monetary Policy Analysis and Forecasting in the World Economy: A
Panel Unobserved Components Approach,” IMF Working Paper No. WP/09/238
(Washington, DC: International Monetary Fund).
———, 2010, “Monetary Policy Analysis and Forecasting in the Group of Twenty: A Panel
Unobserved Components Approach,” IMF Working Paper No. WP/10/152 (Washington,
DC: International Monetary Fund)International Monetary Fund).
APPENDIX
TABLE 5A.1
Figure 5A.1 Response to a 1 Percent Growth Shock in the United States (Percent)
– 0.4 – 0.4
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8
135
©International Monetary Fund. Not for Redistribution
136
Growth Spillover Dynamics: From Crisis to Recovery
1.6 1.6
DEU FRA 1.8 USA
1.1 1.1
1.3
0.6 0.6 0.8
– 0.4 – 0.4
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8
137
©International Monetary Fund. Not for Redistribution
138
Growth Spillover Dynamics: From Crisis to Recovery
1.6 1.6
DEU FRA 1.8 USA
1.1 1.1
1.3
0.6 0.6 0.8
Figure 5A.3 Response to a 1 Percent Growth Shock in the United Kingdom (Percent)
– 0.4 – 0.4
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8
139
©International Monetary Fund. Not for Redistribution
140
Growth Spillover Dynamics: From Crisis to Recovery
1.6 1.6
DEU FRA 1.8 USA
1.1 1.1
1.3
0.6 0.6 0.8
0.1 0.1 0.3
141
©International Monetary Fund. Not for Redistribution
142
Growth Spillover Dynamics: From Crisis to Recovery
1.6 1.6
DEU FRA 1.8 USA
1.1 1.1
1.3
0.6 0.6 0.8
0.1 0.1 0.3
– 0.4 – 0.4
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8
143
©International Monetary Fund. Not for Redistribution
144
Growth Spillover Dynamics: From Crisis to Recovery
1.6 1.6
DEU FRA 1.8 USA
1.1 1.1
1.3
0.6 0.6 0.8
0.1 0.1 0.3
145
©International Monetary Fund. Not for Redistribution
146
Growth Spillover Dynamics: From Crisis to Recovery
1.6 1.6
DEU FRA 1.8 USA
1.1 1.1
1.3
0.6 0.6 0.8
0.1 0.1 0.3
147
©International Monetary Fund. Not for Redistribution
This page intentionally left blank
This chapter assesses the impact of fiscal spillovers on growth in the context of a coor-
dinated exit from crisis management policies. We find that despite potentially sizeable
domestic effects from consolidation, aggregate negative spillovers to other countries are
likely to be contained in 2011–12 unless fiscal multipliers and/or imports elasticities
are very large. However, small and open European economies will be substantially
affected. In contrast, the coordinated exit from fiscal stimulus will have a limited
direct effect on European peripheral countries, since they are relatively closed, with the
notable exception of Ireland.
INTRODUCTION
Under normal circumstances, when business cycles and fiscal policies are unsyn-
chronized, changes in domestic fiscal stances are unlikely to have a significant
global impact because the reduction in domestic demand can be partly offset by
the increase in net exports, as documented, for example, in the 2010 World
Economic Outlook (IMF, 2010). However, the current situation is not normal.
Countries went through a global 2008–09 financial crisis and in response have
implemented synchronized fiscal stimuli, which have left substantial amounts of
public debt that now need to be reduced. For many governments, fiscal consoli-
dation has thus become a major objective, and since 2011 these governments have
embarked on ambitious fiscal consolidation plans. This implies that many coun-
tries will consolidate at the same time.
Does the ongoing synchronized fiscal consolidation have the potential to lead
to significant spillover effects? In other words, will fiscal actions in one country
convey to economic activity in other countries? Some would argue that such a risk
exists. Several considerations favor such a view: Exchange rates cannot adjust if
many countries consolidate simultaneously. Additionally, a large number of coun-
tries undertaking consolidation are in the eurozone, where the real exchange rates
can adjust only slowly anyway. Hence, the offsetting effect of adjustments in net
exports may not be feasible. Moreover, empirical evidence suggests that fiscal
The authors would like to thank Ashoka Mody for detailed suggestions and insightful comments, and
Daniel Leigh and Vladimir Klyuev for helpful discussions. They also would like to thank Emre Alper,
Bas Bakker, Xavier Debrun, Peter Doyle, Lorenzo Figliuoli, Kevin Fletcher, Doug Laxton, Ester Perez
Ruiz, Francis Vitek, and Erik de Vrijer for useful comments.
149
multipliers are likely to be higher at the time of financial stress and when interest
rates are close to the zero bound (Blanchard and others, 2009; Christiano
Eichenbaum, and Rebelo, 2009; IMF, 2010; Auerbach and Gorodnichenko,
2010; Corsetti et al. 2010b). Both aspects have thus the potential to magnify
spillover effects from fiscal consolidation.
We use a simple analytical framework to evaluate the relevance of fiscal spill-
over effects through trade channels based on estimates of fiscal multipliers and
import elasticities obtained in other studies. The methodology is applied to a
sample of 20 countries covering more than 70 percent of world GDP.1 The
approach accounts for carry-over effects from previous years’ fiscal positions and
allows differentiating between revenue and expenditure measures. The baseline
estimates of multipliers obtained in the literature are based on the premise that
monetary policy is accommodative. To reflect the current environment, in which
exchange rate adjustments to “soften the blow” may not be feasible, we perform
a series of robustness checks with higher multipliers and a range of import elas-
ticities. We also assess the sensitivity of the results to various measures of the fiscal
stance.
The results do imply that the domestic contractionary effects of fiscal consoli-
dation could be sizable. However, aggregate spillovers of these contractionary
impulses to other countries are likely to be contained in 2011–12 unless fiscal
multipliers and/or imports elasticities are significantly larger than seems reason-
able now. However, the effect will be different across countries. Small and open
European economies, including Ireland, Belgium, Austria and the Netherlands
will be substantially affected. European peripheral countries other than Ireland
will face limited direct impacts, because they are relatively closed. Ireland would
benefit from a more relaxed pace of fiscal consolidation elsewhere, but such sup-
port would be meaningful only if it were coordinated across the major economies,
including the United States and the United Kingdom. In contrast, a reduced
consolidation effort by Germany alone would have a limited impact on the
European periphery.
Our analysis consists of two parts. First, we estimate the impact of a uniform
shock (1 percent of GDP reduction in expenditure) in 20 major economies to
gauge the relative strength of the impact on growth across countries. Second, we
estimate the growth impact based on the projected fiscal position in these econo-
mies in 2011–12, which also reflects the size of the expected fiscal change for each
country under the current plans. In both cases, we quantify the potential effect of
fiscal consolidation on output growth and the trade balance and calculate the
contribution of spillovers from other countries’ consolidation plans to the respec-
tive changes.
The approach only quantifies the direct demand impact and does not
reflect credibility or other non-demand-driven effects (to the extent that they
1
The full list of countries includes Austria, Belgium, Brazil, China, France, Germany, Greece, India,
Ireland, Italy, Japan, Korea, Netherlands, Portugal, Russia, Spain, Sweden, Switzerland, United
Kingdom, and United States.
LITERATURE
The literature on economic spillovers across borders has grown in recent years.
However, there are only a few quantitative studies measuring the growth impact
of fiscal spillovers, that is, the impact of domestic fiscal changes on growth in
other countries. This is not surprising, since aggregate fiscal spillovers are negli-
gible when the fiscal cycles of countries are independent from each other, because
the sum of fiscal changes in the rest of the world is likely to be small as consolida-
tion and expansion in different countries offset each other.
But in the event of a global downturn, fiscal spending tends to become syn-
chronized as countries step up spending to bolster output during the recession.
For example, in the aftermath of the financial crisis of 2008–09, governments
simultaneously implemented fiscal stimulus packages, while now there is a global
tendency to reduce fiscal deficits.
Estimates of growth spillovers in the context of crises and synchronized fiscal
consolidation are scarce. Thus, our understanding of the international growth
impact of fiscal changes derives from studies that focus on the domestic effect of
fiscal consolidation. Since the size of the domestic effect of fiscal consolidation on
growth is rather important for evaluating the potential for cross-country spillover
effects, we also review the literature on the domestic effects of fiscal policy. We
focus on studies that investigate the difference in the effects on growth between
times of crises and ‘normal’ times.
In reviewing the literature, we reach two main conclusions. First, the existing
estimates of fiscal spillovers suggest that they are limited, although spillovers from
the United States may be relevant. In most cases, however, the analysis of spill-
overs is based on the effect of an individual country while keeping fiscal policy in
other countries unchanged. Hence, the effect of coordinated consolidation may
not be fully captured. Furthermore, the estimates of growth spillovers are based
on ‘normal times’ and simulation results often rely on forward-looking agent
models; both favor the finding that the impact of fiscal changes on growth is low.
Second, while estimates of the impact of fiscal multipliers from domestic policy
action in a single country on its own growth vary widely, the evidence suggests
that the multipliers are likely to be on the higher side in the current environment.
In particular, interest rates are close to the zero bound and cannot fall much fur-
ther to crowd in investment. Also, the shares of liquidity-constrained households
and firms are likely to be high in the aftermath of the financial crisis.
Some recent studies investigate the spillover effects of fiscal policy.2 Beetsma,
Giuliodori, and Klaassen (2006) find that the average effect of a fiscal stimulus of
1 percent of GDP in Germany is an increase of 0.23 percent in foreign GDP for
a spending increase and 0.06 percent for a net tax cut, within two years.3
Spillovers from France are found to be lower but still non-negligible. The authors
employ a two-step procedure. In the first step, they use a standard panel vector
autoregression (VAR) approach to identify fiscal shocks. In the second step, a
panel bilateral trade model is estimated to obtain the effects of changes in domes-
tic output on foreign exports. Merging the responses from the two blocks allows
them to compute the overall effect of the fiscal impulses on bilateral exports and
thereby on the output of other countries. However, their estimates do not repre-
sent the full extent of the spillovers, since they do not account for further feed-
back effects among the economies.4
Bénassy-Quéré and Cimadomo (2006) find positive cross-border spillovers
from Germany, at least in neighboring and smaller countries. The authors find
tax multipliers to be larger than spending multipliers and the effect of tax shocks
on output to be more persistent.5 They estimate a factor-augmented VAR model,
appending the GDP and the real exchange rate of one country at a time to the
German model. Their focus is on the seven biggest European Union (EU) mem-
ber countries. Germany is assumed to be contemporaneously unaffected by the
foreign variables, while German shocks can affect the country under analysis. The
estimation procedure constrains the analysis to direct effects from Germany to the
respective country while not accounting for multi-country spillovers and poten-
tial feedback loops.
Some authors have employed multi-country macro models to simulate the
extent of spillovers from fiscal policies. For instance, Gros and Hobza (2001)
provide an overview of results from four major macroeconomic models on the
2
Another study which looks at fiscal spillovers is Canova and Pappa (2007). However, the authors
focus on the effect of regional expenditure and revenue shocks on the price differentials, and not
growth, in monetary unions using the example of the U.S. states as well as nine EMU member coun-
tries. Since the authors run separate BVARs for each unit and construct average responses from these
estimates, they also cannot account explicitly for spillovers across regions.
3
German fiscal expansion has a particularly strong effect on its small neighbors. An increase in public
spending (a decrease in net taxes) by 1 percent of GDP in Germany leads to a more than 0.4 percent
(0.1 percent) normalized increase in the GDP of Austria, Belgium, Luxemburg, and the Netherlands
after two years.
4
The authors thus argue that the effects should be regarded as lower bounds and that further research
is needed on the feedback between all countries.
5
The authors find that German tax shocks have a beneficial impact on foreign GDP. However, this
effect seems to be limited to neighboring countries. Cross-border spillovers from fiscal spending
shocks are found to be low and rarely significant, except for a few countries (Belgium, Austria, and
the Netherlands).
6
The impact on German GDP in the first year amounts to a change of 0.4 to 1.2 percent. The origi-
nal paper includes results from four models including MULTIMOD (IMF), NiGEM (NIESR),
QUEST (EC), and Marmotte (CEPII). We excluded results from the latter for the discussion here,
since it is based on a multi-country framework that assumes full flexibility of output prices and
rational forward-looking agents.
7
The four models based on the New-Keynesian approach do not support a textbook Keynesian mul-
tiplier effect. The reason is the forward-looking behavior of households and firms. They anticipate
higher tax burdens and higher interest rates in the future and therefore reduce consumption and
investment. Only the ECB’s area-wide model, which largely ignores forward-looking behavior, is
found to generate government spending multipliers that are significantly above one.
8
It should be noted that the results are based on a G7 country model and on the assumption of no
fiscal change in the other countries. Thus, positive spillovers from third country effects are likely to
be underestimated (due to the country sample) and negative repercussions from the appreciation of
the euro overestimated (due to the country sample and the absence of a fiscal expansion in the other
countries).
9
For a summary of literature on multipliers see Schindler, Spilimbergo, and Symansky (2009).
10
On the other hand, the OECD (2009) argues that multipliers may be lower in the current crises,
about 0.5 percent, due to households’ higher propensity to save.
While the recent literature provides some results on the impact of fiscal
spillovers on growth, the analysis is generally conducted in an “all else being
equal” manner, that is, it looks at the direct effect of a single country’s fiscal
policy on others without taking into account indirect second-round effects
through trading partners, which could amplify the impact. We contribute to
the literature by accounting for these second-round effects. Moreover, unlike
earlier studies, we study not only the potential spillover effects but also the
spillovers that are implied by the announced global fiscal consolidation plans
for 2011–12.
FRAMEWORK
Fiscal Spillover Framework
The spillover framework is based on a representation of the national accounts and
behavioral assumptions for government spending, taxes, consumption, invest-
ment, exports, and imports. Starting from the national accounting identity, we
know that:
Yt , j = C t , j + I t , j + Gt , j + X t , j − M t , j (1.1)
Where Yt , j is the real output, I t , j is real investment, Gt , j is the real government
spending, X t , j is real exports and M t , j is real imports of country j in time t
denominated in a common currency. The single elements of output are respec-
tively given by:11
(
Ct , j = C 0 + c1 Yt , j − Tt , j ) M t , j = μ jYt , j
I
I t , j = I 0 + d1Yt , j − d 2rt , j X t , j = ∑ ω ij μ iYt ,i (1.2)
i≠ j
i =1
μ i is the marginal propensity to import of a trading partner i,12 Yt,i is the output
of a trading partner i, and ωij is the weight of imports from country j in total
imports of country i. Substituting the definitions (1.2) in (1.1) yields
I
Yt , j = ext , j + m jGt , j − m j c1Tt , j + m j ∑ ωij μ iYt ,i (1.3)
i≠ j
i =1
11
The model does not account for potential crowding-out effects. Allowing consumption and invest-
ment to react to fiscal changes (beyond the output effect) potentially reduces the contractionary effect
of fiscal consolidation. While this is clearly a simplifying assumption, it is not unreasonable in the
current economic environment.
12
In the calculations, the marginal propensity to import μi was computed as the ratio of imports to
GDP multiplied by the imports elasticity for each country.
( )
−1
Where ext , j = C 0 + I 0 − d 2rt , j and m j = 1 − c1 − d1 + μ j is the expenditure
multiplier, which is also the multiplier for exports. Taking the first difference and
dividing by real output in t-1 yields the contribution of the fiscal change to out-
put growth:
ΔYt , j ⎡ ΔGt , j ⎤ ⎡ ΔTt , j ⎤ I
ΔYi Yt −1,i
= mj ⎢ ⎥ − m j c1 ⎢ ⎥ + m j ∑ ωij μi (1.4)
Yt −1, j ⎣⎢ Yt −1, j ⎦⎥ ⎣⎢ Yt −1, j ⎦⎥ i≠ j Yt −1,i Yt −1, j
i =1
Converting expenditure and revenue ratios into nominal terms with respect to
GDP we have:13
Where Pj ,t is the price level at time t, which is measured by the GDP deflator.
Consistent with empirical findings, we allow the fiscal measures to incorporate
a current period as well as a lagged period effect from fiscal measures imple-
mented in the previous period:
⎡ ΔGtN, j Pj ,t −1 ⎤
⎥ = m j g j ,t + m j g j ,t −1
G ,1 G ,2
mj ⎢ N
Y P
⎣⎢ j ,t −1 j ,t ⎦⎥
⎡ ΔTt ,Nj Pj ,t −1 ⎤
⎥ = m j t j ,t + m j t j ,t −1
T ,1 T ,2
m j c1 ⎢ N (1.6)
Y P
⎣⎢ j ,t −1 j ,t ⎦⎥
* * N*
Yt −1,i qijYt −1,i sij Pt −1,i Yt −* 1,i sijYt −1,i Yt −N1,i
13
Note that we used the following transformation: = = = N = N
Yt −1, j Yt −1, j Pt −1, j Yt −1, j Yt −1, j Yt −1, j
where qij and sij are respectively the real and the nominal exchange rate between country i and country
j and stars denote values in foreign currency. The nominal exchange rate is assumed to be stable across
the period of analysis.
We can use the definitions in (1.2) to derive the implicit change in the real
trade balance that is caused by the change in fiscal spending. To do so, we first
compute the real change in exports and imports relative to real GDP in t-1:
ΔX t , j I
ΔYt ,i Yt −1,i
= ∑ ω ij μ i ΔM t , j ΔYt , j
Yt −1, j i≠ j Yt −1,i Yt −1, j = μj (1.9)
i =1 Yt −1, j Yt −1, j
Converting into nominal terms with respect to GDP and subtracting gives the
real change in the trade balance relative to GDP in t-1:
( )
X jN,t Pj ,t −1 / Pj ,t − X jN,t −1
−
( )
M jN,t Pj ,t −1 / Pj ,t − M jN,t −1
N N
Y j ,t −1 Y j ,t −1
(1.10)
I
YN
= ∑ ωij μi i N,t −1 yt ,i − μ j yt , j
i≠ j Y j ,t −1
i =1
14
We compared the estimates of fiscal changes based on cyclically adjusted revenue/expenditure in per-
cent of GDP with those in percent of potential GDP and found the differences to be small. We chose
to report the results for the measure scaled by GDP to facilitate comparison with the headline measure.
where Gtr is real fiscal spending at time t and Yt r−1 is real output at time t-1. Then
the change in expenditure in ratios to GDP or potential GDP can be written as
Gtr Pt Gtr−1Pt −1 Gtr
ΔG ratios = − = ΔG real
− g r
Yt r Pt Yt r−1Pt −1 Yt r Yt
where gYtr is a real GDP or potential GDP growth at time t. Therefore, provided
that the growth in GDP or potential GDP is non-zero, the differences can be
substantial for revenue/expenditure subcomponents, since the ratio of expendi-
ture/revenue to GDP is typically large. The differences for the overall balance,
however, will be small, since the differences for revenue and expenditure largely
offset each other.
While the theoretical definition of multiplier is based on a concept of real
change in revenue/expenditure, the empirical estimates of fiscal multipliers are
sometimes geared towards the measure based on ratios to GDP or potential GDP.
Therefore, none of the measures mentioned above is perfect, and the three mea-
sures capture different aspects of fiscal policy, so all three can be useful in assessing
the impact on growth.
SIMULATION RESULTS
For practical reasons we limit our discussion to 20 countries, with a focus on
European countries but a fair representation of major international actors. More
precisely, our exercise includes all nations with a ratio of domestic output to world
output above 2 percent. Given our particular interest in the euro area countries,
we also include in the sample a range of euro area members and their relevant
trading partners. The final sample includes the following 20 countries:15
Austria Germany Japan† Spain
Belgium Greece Korea† Sweden
Brazil† India† Netherlands Switzerland‡
China†‡ Ireland Portugal United Kingdom†‡
France Italy Russia† United States†‡
The sample represents more than 70 percent of world GDP and covers on aver-
age two-thirds of a country’s imports and of its exports. For the euro area members
in the sample, the values are roughly three-quarters for imports and exports.
The OECD (2009) reports revenue and spending multipliers for current peri-
ods and lagged effects for subcomponents of revenues and expenditures for a wide
sample of countries. We draw on these multipliers for specific tax and revenue
T ,1
policies to compute the respective values for the current-period revenue ( m j )
G ,1 T ,2
and expenditure ( m j ) as well as the lagged effect revenue ( m j ) and expendi-
G ,2
ture ( m j ) multiplier.16 We use each country’s share of specific revenue compo-
nents to compute an overall revenue multiplier and similarly an overall spending
multiplier. In some cases, the resulting average multipliers are adjusted in line
with country-specific estimates provided by IMF country desks.
Import elasticities are taken from Kee, Nicita, and Olarreaga (2008). The
marginal propensity to import ( μ i ) is then computed by multiplying the elastic-
ity with the respective imports-to-GDP ratio in 2009. An overview of the multi-
pliers and import elasticities is provided in Table 6A.1.
The fiscal measures are based on the IMF’s April 2011 World Economic
Outlook data. The simulation framework implies that the differences in the
impact of fiscal consolidation are a combination of the following elements:
• The country-specific revenue and expenditure multipliers
• The composition of the consolidation (revenue versus expenditure measures), and
• The trade links between countries (and thus these countries’ characteristics
for sub-points 1 to 2) and their propensity to import when income changes.
We will refer to variations in the above in the respective robustness checks.
Uniform Fiscal Shock
Baseline Multipliers and Import Elasticities
We first demonstrate an impact of a 1 percent of GDP shock to expenditures to
gauge the relative size of spillovers between countries under the baseline assumption
15
Countries marked by † account for more than 2 percent of world output, and countries marked
with ‡ are major trading partners for one or more of the euro member countries. We excluded Canada
and Mexico in favor of several smaller euro zone members. Both Mexico and Canada have negligible
effects on the European countries but are very much subject to U.S. shocks.
16
More precisely we employ the country-specific multipliers labeled “high multipliers” by the OECD (2009).
The term “high” in this context refers to the fact that the OECD’s “reference” multiplier employed in its study
is “judgmentally scaled down, by more for tax cuts than for government spending,” since the current eco-
nomic circumstances are “more likely to reduce multipliers.” Thus we use effectively the original series.
on multipliers and import elasticities. The baseline multipliers average at 0.5 for
revenue and 0.8 for expenditure after two years across the sample of 20 countries.
Consequently, the baseline multipliers are relatively high. The import elasticities
average at 1.15.
To clarify the mechanics and the intuition behind the reported results, we
present the calculation of the first-round spillover effects from a 1 percent decline
in government spending in Germany to the peripheral European countries after
one year in Table 6.1.
The first-round effect of the fiscal consolidation in Germany on growth in
Portugal can be approximated as follows. German imports from Portugal com-
prise only 0.7 percent of total German imports,17 while Germany’s marginal
propensity to import out of income is 0.5 (imports share in GDP times imports
elasticity). Therefore, out of every additional euro of income Germany will
import only 0.35 cents from Portugal, and the opposite is true for income
reduction. Since fiscal spending in Germany has a multiplier of only about 0.4
after one year, a one percent of GDP decline in fiscal spending in Germany will
reduce German GDP due to domestic consolidation by only about 0.4 percent
after one year, which would result in a decline of about 10 billion euros. As a
result, Germany will import 0.03 billion euros less from Portugal, and
Portuguese exports will decline by this amount. However, this does not trans-
late to the equivalent amount of income loss for Portugal since, for example,
some of this reduction will be compensated by lower imports. Since exports
have the same multiplier as expenditures (see equation (1.3)), which is about
0.5 for Portugal, the actual income loss for Portugal would only be about 0.015
billion euros, which corresponds to about 0.009 percent reduction in
Portuguese GDP in the first year (German GDP is 15 times larger than
Portuguese GDP).
This calculation, however, does not incorporate second-round effects, since a
reduction in German GDP will result in lower growth in other common trading
partners of Germany and Portugal. Taking into account these second-round
effects will result in a slightly greater reduction in GDP growth in Portugal,
namely 0.011 percent. The impact on Ireland is somewhat greater but on Greece
it is almost negligible.18
Consequently, the impact of Germany’s fiscal policy on the peripheral coun-
tries is likely to be rather small. As we demonstrate below, even very high multi-
pliers result in only a moderate impact from Germany alone.
The matrix of results of a coordinated 1 percent decline in fiscal spending
across all 20 countries is reported in Table 6A.2 in the Appendix. The table
17
In fact, the share is even smaller; the results in the table were rescaled by the total over the sample
of 20 countries to sum up to 1.
18
It should be noted that the calculation results are based on the pattern of trade in goods. Greece,
however, has a substantial share of trade in services, hence the estimates are biased downward.
Nonetheless, the impact is likely to be very small; even assuming that trade in services is about 50
percent of total trade in Greece and, hence, by roughly doubling the results would yield very small
spillovers from Germany.
An Example of a Simple Calculation of a Spillover Effect from Germany to the Peripheral European Countries, Baseline Multipliers
Total Spillover
Effect on
First-Round Effect on Growth after
First-Round Effect on German GDP First-Round Effect on Peripheral Exports Peripheral GDP One Year
The solution
Share of given by (1.8),
country’s Ratio of Export First round reflects
imports in German German German multiplier spillover effect indirect effects
Expenditure Expenditure total German imports marginal output to First-round effect of the on growth through other
shock in multiplier First-round effect German imports share in propensity country’s on peripheral peripheral after one year countries
Germany in Germany on German GDP imports elasticity GDP to import output exports country (percent) (percent)
reports the growth impact (percent deviation from the baseline of no fiscal
change) after two years. Countries where the fiscal shock originates are reported
in columns, while recipient countries of the growth impact are in rows. Hence,
the diagonal elements of this matrix show the growth impact of the country’s
domestic fiscal policy while the off-diagonal elements show the spillovers—the
impact on country’s growth due to the fiscal changes in other countries. The total
at the end of the row, therefore, is the total growth impact on a particular country
reported in this row from the coordinated fiscal consolidation. The PPP-weighted
average at the bottom of the table can be interpreted as an individual country’s
impact on the whole group of 20 countries—a proxy of the impact on world
growth. The PPP-weighted average, however, includes both the impact on global
growth from the changes in domestic growth of a country and through the spill-
overs from this country to other countries weighted by the PPP GDP of each
country.
The results indicate that the overall impact of a 1 percent of GDP coordinated
fiscal consolidation is sizeable, reducing growth in the 20 countries on average by
about 0.9 percent after two years (PPP-weighted basis), largely due to the impact
on growth from domestic consolidation, with only about 15 percent being
accounted for by spillovers from one country to another. The largest contribu-
tions to the PPP-weighted average growth decline come from the United States
and China (close to 0.2 percent) reflecting their large weight in the world econo-
my, followed by Japan and India (close to 0.1 percent) with Germany, France,
Brazil, Italy and Russia contributing close to 0.05 percent while United Kingdom,
Spain, and Korea contributing about 0.03 percent each.
Total inward fiscal spillovers to most countries are limited, not exceeding 0.3
percentage points over two years and averaging at 0.1 on PPP-weighted basis and
0.2 on simple average basis. However, spillovers to Ireland, Belgium, Austria, the
Netherlands and Korea are more substantial, close to ½ percentage points over
two years. (Figure 6.1) In the case of Ireland, the largest single-country contribu-
tion comes from the United States (Table 6A.2). For Austria and the Netherlands,
spillovers from Germany are particularly pronounced, while for Belgium spill-
overs from Germany and France are equally important. In Korea, spillovers from
China play an important role. However, individual country spillovers to other
individual countries are rather small, not exceeding 0.16 percentage points over
two years.
0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.6
–0.7
–0.8
Simple average PPP–weighted average
–0.9
–1
ina
d
m
nd
m
e
es
ia
ly
ain
l
ion
d K den
ia
il
dS e
Sw lic of
ga
az
nd
an
lan
nc
pa
c
Ita
Ind
str
lgiu
do
tat
ee
rla
rat
Ch
Sp
rtu
Br
rla
Fra
rm
Ja
Ire
Au
ing
ub
Gr
itze
Sw
Be
de
Po
the
Ge
ep
Fe
ite
Ne
,R
Un
ite
ian
rea
Un
ss
Ko
Ru
Figure 6.1 Inward Fiscal Spillovers (Impact on real GDP from a 1 percent of GDP
reduction in fiscal spending in other countries, baseline multipliers, cumulative after two
years, percent)
Source: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.6
–0.7
–0.8
Simple average PPP–weighted average
–0.9
–1
d
um
ia
y
e
ly
l
ain
Sw n
en
n
ia
il
dS e
es
Sw lic of
ina
az
nd
an
lan
an
nc
io
pa
c
Ita
str
Ind
do
ee
tat
ed
rtu
Sp
rat
Ch
lgi
Br
rla
rm
Fra
erl
Ja
Ire
Au
ing
ub
Gr
Be
Po
de
the
itz
Ge
ep
dK
Fe
ite
Ne
,R
Un
ite
ian
rea
Un
ss
Ko
Ru
Figure 6.2 Inward Fiscal Spillovers (Impact on real GDP from a 1 percent of GDP
reduction in fiscal spending in other countries, higher multipliers, cumulative after two
years, percent)
Sources: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
Results reported in the table above suggest that the average growth impact
increases with higher multipliers. However, what is more striking is the non-lin-
earity with which this affects the impact through spillovers. While multipliers are
increased by only about 25 percent, the overall impact of fiscal consolidation on
output growth is increased by more than 30 percent. This is primarily due to a
more than 60 percent increase of spillovers from other countries’ consolidation
efforts, while the domestic effect increased proportionally to the average increase
in multipliers. In PPP-average terms, the increase implies a reduction of GDP
growth due to spillovers by only 0.2 percentage points, while a simple average is
now close to 0.4 percentage points. However, for some countries, spillovers now
account for a sizable fraction of growth reduction, with the largest spillovers close
to 1 percentage point for Ireland and Belgium (Figure 6.2).
Import elasticities also have a magnifying effect on spillovers, although the
effect is less pronounced. Import elasticities that are higher on average by 40
percent lead to an increase in spillovers by over 50 percent, while the domestic
impact remains virtually unchanged. As a result, the overall growth impact
increases by about 10 percent compared to the baseline, since the share of spill-
overs remains relatively small (just over 20 percent on the PPP-average basis of
the overall growth impact). After two years, the average spillovers on a PPP basis
are close to 0.2 and on a simple-average basis are over 0.3 percentage points. The
list of countries substantially affected by spillovers remains unchanged
(Figure 6.3).
0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.6
–0.7
–0.8
–0.9 Simple average PPP–weighted average
–1
d
um
ia
ly
ain
l
ion
d K den
m
ia
il
dS e
es
Sw lic of
ina
ga
az
nd
an
lan
an
nc
pa
c
Ita
str
Ind
do
ee
tat
Sp
rtu
rat
Ch
lgi
Br
rla
Fra
rm
erl
Ja
Ire
Au
ing
ub
Gr
Sw
Be
Po
de
the
itz
Ge
ep
Fe
ite
Ne
,R
Un
ite
ian
rea
Un
ss
Ko
Ru
Figure 6.3 Inward Fiscal Spillovers (Impact on real GDP from a 1 percent of GDP
reduction in fiscal spending in other countries, higher imports elasticities, cumulative after
two years, percent)
Sources: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
These results suggest that, on average, the size of spillovers remains limited
under alternative assumptions on multipliers and import elasticities, with the
exception of small open economies, where the effects can be substantial.
However, even for those countries where spillovers can be substantial, the
impact of fiscal changes in a single trading partner remains contained, not
exceeding ¼ percentage point over two years. German fiscal policy, in particu-
lar, has limited implications for growth in the European periphery. Very high
multipliers would have to operate for Germany to exhibit a relatively modest
impact on these countries. For example, with expenditure multipliers equal to
the baseline plus four standard deviations (an average expenditure multiplier of
1.6 after two years), after two years a one percent of GDP fiscal expenditure
stimulus in Germany would raise the GDP growth in Ireland by only 0.3 per-
centage points, in Portugal by 0.1 percentage points, and in Greece with virtu-
ally no effect on growth. Similarly, fiscal policy changes in Germany alone have
only a small impact on the trade balance of the peripheral countries and are thus
unlikely to contribute to the reduction in the peripheral countries’ imbalances
(Figure 6.4).
0.6
0.5
0.4
0.3
0.2
0.1
0
Ireland Greece Portugal
PPP weighted average –0.5 –0.2 –0.3 0.0 –0.3 0.3 0.6 –0.8 1.4
Simple average –0.3 –0.4 0.2 0.3 –0.6 0.9 0.2 –0.6 0.8
Sources: IMF, World Economic Outlook, April 2011; and IMF staff estimates.
Note: PPP: purchasing power parity.
a
Financial sector support recorded above-the-line was excluded for Ireland (2.5 percent of GDP in 2009 and 5.3 percent of GDP in 2010) and the US (2.4 percent of GDP in 2009, 0.2 percent of GDP in 2010 and 2011,
and 0.1 percent of GDP in 2012). Financial sector support is not expected to have a significant impact on demand. For Russia only non-oil revenues are assumed to have an impact on growth.
PPP weighted average 0.9 0.8 0.1 –0.2 –0.1 0.0 –0.7 –0.6 –0.1
Simple average 0.9 0.7 0.2 –0.4 –0.4 0.0 –0.8 –0.6 –0.2
Sources: IMF, World Economic Outlook, April 2011; and IMF staff estimates.
Note: PPP: purchasing power parity.
169
a
Financial sector support recorded above-the-line was excluded for the calculation of growth impact for Ireland (2.5 percent of GDP in 2009 and 5.3 percent of GDP in 2010) and the US (2.4 percent of GDP in 2009,
0.2 percent of GDP in 2010 and 2011, and 0.1 percent of GDP in 2012). Financial sector support is not expected to have a significant impact on demand. For Russia only non-oil revenues are assumed to have
an impact on growth.
©International Monetary Fund. Not for Redistribution
170 Do Fiscal Spillovers Matter?
0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.6
–0.7
–0.8
Simple average PPP-weighted average
–0.9
–1
d
y
ium
s
ia
d
l
n
e
ain
ly
d K tion
ia
il
dS e
es
Po of
ina
ga
az
nd
an
la n
de
pa
Sw anc
ec
an
Ita
Ind
str
do
tat
lic
Sp
rtu
Ch
Br
rla
rm
g
ra
Un Gre
ian Swe
erl
Ja
Ire
Au
ing
ub
l
Fr
Be
Un ede
the
Ge
itz
ep
ite
Ne
F
,R
ite
rea
ss
Ko
Ru
Figure 6.5 Inward Fiscal Spillovers (Impact on real GDP from cyclically adjusted fiscal
changes in other countries, cumulative 2011–2012, percent)
Sources: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
fiscal relaxation, although this would require contributions from the major
economies, including the United States and the United Kingdom, both countries
where such relaxation is not in the cards (Figure 6.5).
The decomposition of spillovers by country (Table 6A.3) reveals the rela-
tively large impact on PPP-weighted average from the United States and China
followed by the United Kingdom, Spain, France, and Italy. This reflects both the
size of the country and the actual amount of consolidation. For instance, while
a uniform fiscal shock would result in a larger impact from Germany than the
United Kingdom (Table 6A.2), Germany’s consolidation plans are much more
moderate than the United Kingdom’s consolidation plans, resulting in a rela-
tively larger impact from the United Kingdom under the actual consolidation
plans.
For some countries, the overall growth effect masks the various forces that are
at work. This is evident once the effect is decomposed into the effects from cur-
rent period consolidation and the carry-over effects from the last period’s fiscal
change. For instance, in the case of the Netherlands and Belgium, the spillovers
in 2011 are negative from the current period consolidation, but there are also
small positive spillovers from the previous year’s mostly expansionary fiscal
change in relevant trading partner countries, reducing the overall negative effect
from spillovers in 2011. However, for countries that are large and not very
open (e.g., the United States), spillovers tend to be negligible in both periods
(Table 6.5).
171
©International Monetary Fund. Not for Redistribution
172 Do Fiscal Spillovers Matter?
2.5
Domestic
2.0 Spillover
Total
1.5
1.0
0.5
0.0
– 0.5
–1.0
–1.5
2010 2011 2012
Figure 6.6 Germany: Growth Contribution of Domestic Fiscal Changes and Spillover from
Fiscal Changes in Other Countries, 2010–2012 (Percentage points)
Sources: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
PPP weighted average 0.0 –0.2 0.2 0.3 –0.4 0.6 0.8 –0.8 1.6
Simple average 0.1 –0.4 0.5 0.4 –0.7 1.0 0.4 –0.6 1.0
Source: IMF, World Economic Outlook, April 2011; and IMF staff estimates.
Note: PPP: purchasing power parity.
a
Financial sector support recorded above-the-line was excluded for Ireland (2.5 percent of GDP in 2009 and 5.3 percent of GDP in 2010) and the US (2.4 percent of GDP in 2009, 0.2 percent of GDP in 2010 and
173
2011, and 0.1 percent of GDP in 2012). Financial sector support is not expected to have a significant impact on demand. For Russia only non-oil revenues are assumed to have an impact on growth.
19
Given the recent divergence in the developments in the labor market in Germany and the output
gap, possibly, reflecting structural changes in the labor market, the commonly used cyclically adjusted
measure with elasticities estimated from historical data and the output gap is likely to understate the
true degree of discretionary policy intervention. While for comparability we used a common cyclical
adjustment method for all countries, we believe that for Germany the headline measure better cap-
tures changes in the underlying fiscal position.
PPP weighted average 1.2 1.0 0.2 –0.4 –0.3 –0.1 –0.8 –0.7 –0.1
Simple average 1.4 1.1 0.3 –0.7 –0.5 –0.1 –0.9 –0.7 –0.2
Sources: IMF, World Economic Outlook, April 2011; and IMF staff estimates.
Note: PPP: purchasing power parity.
175
a
Financial sector support recorded above-the-line was excluded for the calculation of growth impact for Ireland (2.5 percent of GDP in 2009 and 5.3 percent of GDP in 2010) and the US (2.4 percent of GDP in 2009,
0.2 percent of GDP in 2010 and 2011, and 0.1 percent of GDP in 2012). Financial sector support is not expected to have a significant impact on demand. For Russia only non-oil revenues are assumed to have
an impact on growth.
©International Monetary Fund. Not for Redistribution
176 Do Fiscal Spillovers Matter?
0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.6
–0.7
–0.8
Simple average PPP–weighted average
–0.9
–1
d
ina
m
ion
ia
nd
y
s
Sw ance
zil
ce
l
ly
ain
d K den
ia
Gr n
es
of
ga
an
nd
la n
pa
Ita
str
Ind
lgiu
do
Bra
ee
tat
rla
lic
Ch
rat
Sp
rtu
rla
rm
Ja
Au
Ire
ing
ub
Fr
dS
itze
Sw
Be
de
Po
e
Ge
ep
h
Fe
ite
t
Ne
,R
Un
ite
an
rea
Un
ssi
Ko
Ru
Figure 6.7 Inward Fiscal Spillovers (Impact on real GDP from headlined fiscal changes in
other countries, cumulative 2011–2012, percent)
Sources: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
With that in mind, the results presented below are not surprising. (Table 6.8)
The changes in the overall balance using the real measure based on the CPI20 are
rather close to those obtained by using the ratios to GDP. The composition of
changes, however, is quite different. In particular, while the measure in ratios sug-
gests that consolidation in 2011 and 2012, on average, includes both revenue and
expenditure contributions, the measures based on real changes suggests that in
real terms expenditures are in fact projected to increase, so consolidation is
mainly revenue-based.
Since the multiplier on revenue is lower than that on expenditure, this leads
to a substantially smaller estimated negative impact on growth, with the contribu-
tion to growth remaining positive, on average, in 2011, partly due to the lagged
effect from 2010. However, for some countries (e.g., Greece) where growth is
projected to remain in the negative territory in 2011, fiscal impulse measured by
the real change implies a larger consolidation on the expenditure side, so the
growth impact is more negative compared to the measure in ratio to GDP
(Table 6.9). (See Table 6A.5 for detailed country-by-country estimates).
The spillovers are also correspondingly much smaller on average than in the
case of the fiscal measure in ratios. While the list of top countries affected by the
spillovers (Ireland, Belgium, Netherlands, and Austria) is unchanged, the magni-
tude of spillovers in 2011–12 has declined substantially, and for Korea spillovers
have turned positive. However, the milder negative impact on growth should be
interpreted with caution, since the empirical estimates of multipliers obtained
20
We use CPI rather than the GDP deflator because the majority of fiscal changes work through either
private consumption decisions or government consumption. The results are not substantially different
if we employ the GDP deflator.
PPP weighted average 1.2 1.3 –0.1 1.3 1.0 0.4 1.9 0.6 1.4
Simple average 1.1 0.7 0.4 1.1 0.1 0.9 1.3 0.4 0.8
Sources: IMF, World Economic Outlook, April 2011; and IMF staff estimates.
Note: PPP: purchasing power parity.
a
Financial sector support recorded above-the-line was excluded for Ireland (2.5 percent of GDP in 2009 and 5.3 percent of GDP in 2010) and the US (2.4 percent of GDP in 2009, 0.2 percent of GDP in 2010 and
177
2011, and 0.1 percent of GDP in 2012). Financial sector support is not expected to have a significant impact on demand. For Russia only non-oil revenues are assumed to have an impact on growth.
PPP weighted average 1.6 1.4 0.2 0.3 0.3 0.0 –0.2 –0.1 –0.1
Simple average 1.5 1.2 0.3 –0.2 –0.2 0.0 –0.4 –0.3 –0.1
Sources: IMF, World Economic Outlook, April 2011; and IMF staff estimates.
Note: PPP: purchasing power parity.
a
Financial sector support recorded above-the-line was excluded for the calculation of growth impact for Ireland (2.5 percent of GDP in 2009 and 5.3 percent of GDP in 2010) and the US (2.4 percent of GDP in 2009,
0.2 percent of GDP in 2010 and 2011, and 0.1 percent of GDP in 2012). Financial sector support is not expected to have a significant impact on demand. For Russia only non-oil revenues are assumed to have
an impact on growth.
0.2
0.1
–0.1
–0.2
–0.3
–0.4
Simple average PPP-weighted average
–0.5
d
ium
s
ia
l
e
ain
ly
y
ina
n
m
ion
ce
dS a
es
il
n
of
ga
az
nd
an
lan
nc
an
pa
Ita
str
Ind
do
ee
tat
d
lic
rtu
Sp
Ch
rat
Br
g
rla
Fra
rm
erl
we
Ja
Ire
Au
ing
ub
l
Gr
Be
Po
de
the
Ge
itz
ep
dK
Fe
ite
Sw
Ne
,R
Un
ite
ian
rea
Un
ss
Ko
Ru
Figure 6.8 Inward Fiscal Spillovers (Impact on real GDP from headline real fiscal changes
in other countries, cumulative 2011–2012, percent)
Source: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
from the literature may not correspond to this fiscal measure and the effects may
be underestimated (Figure 6.8).
To summarize, using real changes rather than the change in the ratio of the
fiscal position implies that the growth contribution of consolidation is generally
less negative, due to the fact that consolidation is no longer dominated by expen-
diture reductions but rather by revenue increases. While the net change in the
fiscal balance is mostly unaffected, the combination of higher revenue adjustment
and lower multipliers for revenue causes the contribution of fiscal changes to
GDP growth to fall. Consequently, using the real fiscal change as the relevant
fiscal measure leads to an even less important role of cross-country spillovers.
Alternative Scenarios
The Growth Impact of More Consolidation
Continued market pressure and rising concerns about debt levels could lead gov-
ernments to consolidate beyond the currently announced level in the forthcom-
ing two years. We thus simulate the results for a scenario in which some countries
in the euro zone (Eur I) reduce spending by an additional 0.5 percent of GDP in
2011 and 2012 or, alternatively, manage to increase structural revenues by an
additional 0.5 percent of GDP in the two years.
Under the baseline multiplier scenario, spillovers are hardly affected, in par-
ticular if tax revenues increase in selected European countries. But even in the case
0.4
Domestic Spillover
0.2
0
–0.2
–0.4
–0.6
–0.8
–1
–1.2
–1.4
–1.6
–1.8
Baseline Higher Higher Baseline Higher Higher Baseline Higher Higher
parameters multipliers imports parameters multipliers imports parameters multipliers imports
elasticities elasticities elasticities
Cyclically Headline Headline
adjusted measure measure
measure (ratios) (real)
Figure 6.9a Cumulative Domestic Growth Impact and Fiscal Spillovers in 2011–2012
Under a Range of Fiscal Multipliers and Imports Elasticities and Using Various Measures of
Fiscal Changes (PPP-weighted average)
Source: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
0.7
Domestic Spillover
0.2
–0.3
–0.8
–1.3
–1.8
–2.3
Baseline Higher Higher Baseline Higher Higher Baseline Higher Higher
parameters multipliers imports parameters multipliers imports parameters multipliers imports
elasticities elasticities elasticities
Cyclically- Headline Headline
adjusted measure measure
measure (ratios) (real)
Figure 6.9b Cumulative Domestic Growth Impact and Fiscal Spillovers in 2011–2012
Under a Range of Fiscal Multipliers and Imports Elasticities and Using Various Measures of
Fiscal Changes (Simple average)
Source: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
PPP weighted average –0.1 –0.1 –0.1 –0.1 –0.1 –0.2 –0.2 –0.2 –0.2 –0.2
Simple average –0.2 –0.2 –0.2 –0.2 –0.2 –0.3 –0.4 –0.3 –0.3 –0.3
Source: IMF staff estimates.
Note: PPP: purchasing power parity.
183
a
Less consolidation is an increase in expenditures by 0.5 percent of GDP in 2011 and 2012, while more consolidation is either a reduction in expenditures (G) or an increase in revenues (T) by 0.5 percent of GDP in
2011 and 2012. “Eur I” includes Belgium, France, Italy, Ireland, Greece, and Portugal. “Eur II” includes Austria, Germany, Netherlands, and Switzerland.
TABLE 6.11
Change in the Real Trade Balance Due to Fiscal Consolidation in 20 Selected Countries (2010–2012)a
Fiscal measure = cyclical adjusted revenue/expenditure, change in percent of GDP
Baseline multiplier High multiplier
Difference to baseline Difference to baseline
Baseline German stimulus Selected surplus country Baseline German stimulus Selected surplus country
Austria 0.0 0.1 –0.3 0.0 0.1 –0.4
Belgium –0.6 0.2 0.3 –0.5 0.2 0.4
Brazil –0.1 0.0 0.1 –0.2 0.0 0.1
China, People’s Republic of –0.1 0.0 –0.1 –0.1 0.0 –0.1
France 0.2 0.0 0.1 0.3 0.0 0.1
Germany –0.6 –0.3 –0.1 –0.7 –0.4 –0.2
Greece 1.1 0.0 0.0 1.6 0.0 0.0
India 0.1 0.0 0.0 0.1 0.0 0.1
Ireland 0.0 0.1 0.2 0.3 0.1 0.2
Italy 0.2 0.0 0.1 0.2 0.0 0.1
Japan –0.2 0.0 0.0 –0.2 0.0 –0.1
Korea, Republic of 0.1 0.0 –0.2 0.1 0.0 –0.2
Netherlands 0.0 0.2 –0.1 0.1 0.2 –0.2
Russian Federation 0.0 0.0 –0.1 0.0 0.0 –0.1
Portugal 0.6 0.0 0.1 0.8 0.0 0.1
Spain 0.7 0.0 0.1 0.9 0.0 0.1
Sweden –0.9 0.0 –0.1 –1.1 0.1 –0.2
185
Netherlands, Russia, Sweden, and Switzerland.
CONCLUSION
In a world of unsynchronized fiscal spending patterns across countries and nor-
mal interest rate levels, spillovers from fiscal policies across countries are likely to
be limited. However, since 2009 the fiscal patterns across most developed coun-
tries have been largely synchronized. While the magnitude varies, in most coun-
tries the fiscal expansion of 2009 and 2010 is set to be followed by fiscal consoli-
dation in 2011, 2012, and beyond. At the same time, the interest rates remain
low, while the output gaps have not closed yet in many advanced economies. In
such an economic environment, fiscal multipliers are likely to be above the usual
levels, and a synchronized and significant swing in fiscal policy from expansion to
consolidation is likely to magnify the role of spillovers from fiscal policy across
countries.
We find that even in this setting, aggregate negative spillovers to other coun-
tries are likely to be contained in 2011–12. Despite potentially sizeable domestic
effects from consolidation, we find that the cumulative impact on GDP over the
two years (2011 and 2012) is not likely to exceed 0.3 percentage points on a PPP-
weighted basis and ½ percentage points on simple-average basis under the various
assumptions on fiscal multipliers and import elasticities. While the absolute size
of spillovers varies depending on the measure of fiscal stance, size of multipliers,
and imports elasticities, average spillovers are invariably small compared to the
size of the impact of domestic fiscal policy.
Nevertheless, the average masks differences across countries. For small and
open European economies such as Belgium, Netherlands, and Austria, spillovers
are important. In contrast, the coordinated exit from fiscal stimulus will have a
limited direct effect on European peripheral countries, since they are relatively
closed, with the notable exception of Ireland. While the latter could benefit from
external support, such support would require contributions from the major
economies, including the United States and the United Kingdom—both coun-
tries where fiscal relaxation is not currently in the cards. Changes in the German
fiscal plan alone would have a very limited impact on the European periphery.
Under the baseline scenario, projected fiscal change for 2011 and 2012 will
help reduce external imbalances. However, the effects over these two years are
likely to be relatively small. While a stronger fiscal expansion in surplus countries
could reduce the respective countries’ surpluses, the “leakages” tend not to go to
the peripheral countries. Therefore, most of the correction in the peripheral coun-
tries’ trade balances will have to be brought about by domestic consolidation in
these countries.
Consequently, the bad news is that the countries in need cannot rely much on
other countries’ fiscal policies to stimulate their growth in the short run. The
good news, however, is that ambitious consolidation plans in the European
peripheral countries will have only limited repercussions for much of the rest of
the world.
REFERENCES
Almunia, Miguel, Agustín S. Bénétrix, Barry Eichengreen, Kevin H. O’Rourke, and Gisela Rua,
2009, “From Great Depression to Great Credit Crisis: Similarities, Differences and Lessons,”
NBER Working Paper No. 15524 (Nov.) (Cambridge, Massachusetts: National Bureau of
Economic Research).
Auerbach, Alan J., and Yuriy Gorodnichenko, 2010 “Measuring the Output Responses to Fiscal
Policy,” NBER Working Paper No. 16311 (Aug.) (Cambridge, Massachusetts: National
Bureau of Economic Research).
Beetsma, Roel, Massimo Giuliodori, and Franc Klaassen, 2006. “Trade Spill-overs of Fiscal
Policy in the European Union: A Panel Analysis,” Economic Policy, CEPR, CES, MSH, Vol.
21, No. 48, pp 639–87.
Bénassy-Quéré, Agnès and Jacopo Cimadomo, 2006, “Changing Patterns of Domestic and
Cross-Border Fiscal Policy Multipliers in Europe and the U.S.” CEPII No. 2006–24 (Dec.).
Blanchard, Olivier, Carlo Cottarelli, Antonio Spilimbergo and Steven Symansky, 2009, “Fiscal
Policy for the Crisis,” CEPR Discussion Paper No. 7130 (London: Centre for Economic and
Policy Research).
Blanchard, Olivier, and Roberto Perotti, 2002, “An Empirical Characterization of the Dynamic
Effects of Changes in Government Spending and Taxes on Output,” The Quarterly Journal of
Economics, Vol. 117, No. 4 (Nov.), pp. 1329–68.
Canova, Fabio, and Evi Papa, 2007 “Price Differentials in Monetary Unions: The Role of Fiscal
Shocks,” The Economic Journal, Vol. 117 (March), pp. 713–37.
Christiano, Lawrence, Martin Eichenbaum, and Sergio Rebelo, 2009. ”When Is the
Government Spending Multiplier Large?,” NBER Working Paper No. 15394 (Cambridge,
Massachusetts: National Bureau of Economic Research).
Cogan, John F., Tobias Cwik, John B. Taylor, and Volker Wieland, 2009, “New Keynesian
versus Old Keynesian Government Spending Multipliers,” NBER Working Paper No. 14782
(March) (Cambridge, Massachusetts: National Bureau of Economic Research).
Corsetti, Giancarlo, André Meier and Gernot J. Müller, 2010a. “Cross-Border Spillovers from
Fiscal Stimulus,” International Journal of Central Banking, Vol. 6, No. 1, pp. 5–37 (March).
———, 2010b, “What Determines Government Spending Multipliers,” Unpublished manu-
script.
Cwik, Tobias, and Volker Wieland, 2009. ”Keynesian Government Spending Multipliers and
Spillovers in the Euro Area,” CEPR Discussion Paper No. 7389 (London: Centre for
Economic and Policy Research).
Girouard, Nathalie, and Christophe André, 2005, “Measuring Cyclically-Adjusted Budget
Balances for OECD Countries,” OECD Economics Department Working Paper No. 434
(Paris: Organisation for Economic Co-operation and Development).
Gros, D., and A. Hobza, 2001, “Fiscal Policy Spillovers in the Euro Area. Where Are They?”,
CEPS Working Document No.176 (Brussels: Centre for European Policy Studies).
IMF, 2010, “Recovery, Risk and Rebalancing,” World Economic Outlook (Washington, DC:
International Monetary Fund).
———, 2011, “Slowing Growth, Rising Risks,” World Economic Outlook, Chapter 4
(Washington, DC: International Monetary Fund).
Kee, Hiau Loo, Alessandro Nicita, and Marcelo Olarreaga, 2008. “Import Demand Elasticities
and Trade Distortions,” The Review of Economics and Statistics, Vol. 90, No. 4, pp. 666–82.
Leibovici, Fernando, and Michael E. Waugh, 2011 “International Trade and Intertemporal
Substitution,” Working paper, New York University.
OECD, 2009, “The Effectiveness and Scope of Fiscal Stimulus,” OECD Economic Outlook
Interim Report (March) (Paris: Organisation for Economic Co-operation and Development).
Romer, Christina D., and David H. Romer, 2010 “The Macroeconomic Effects of Tax Changes:
Estimates Based on a New Measure of Fiscal Shocks,” American Economic Review, Vol. 100
(June 2010), pp. 763–801.
Schindler, Martin, Antonio Spilimbergo, and Steve Symansky, 2009, “Fiscal Multipliers,” IMF
Staff Position Note No. SPN/09/11 (May) (Washington, DC: International Monetary Fund).
Tagkalakis, Athanasios, 2008, “The Effects of Fiscal Policy on Consumption in Recessions and
Expansions,” Journal of Public Economics, Vol. 92 (2008), pp.1486–1508.
TABLE 6A.1
Assumptions on Fiscal Multipliers and Elasticities by Country, Baseline Multipliers (Higher values in, baseline multipliers brackets)
Revenue multiplier Expenditure multiplier Elasticities
Current year Previous year Current year Previous year Import elasticity Revenue Expenditure
Austria 0.2 (0.29) 0.6 (0.82) 0.7 (0.82) 1.1 (1.29) 1.08 (1.57) 0.90 –0.07
Belgium 0.18 (0.27) 0.48 (0.69) 0.35 (0.47) 0.74 (0.93) 1.05 (1.54) 0.99 –0.13
Brazil 0.22 (0.31) 0.42 (0.64) 0.56 (0.67) 0.82 (1.01) 1.34 (1.83) 0.77 –0.16
China 0.22 (0.31) 0.42 (0.64) 0.56 (0.67) 0.82 (1.01) 1.13 (1.62) 1.00 0.00
France 0.2 (0.29) 0.3 (0.52) 0.7 (0.82) 1.1 (1.29) 1.14 (1.63) 0.88 –0.10
Germany 0.35 (0.44) 0.74 (0.95) 0.4 (0.51) 0.8 (1) 1.14 (1.63) 0.94 –0.24
Greece 0.22 (0.31) 0.42 (0.64) 0.56 (0.67) 0.82 (1.01) 1.12 (1.61) 0.77 –0.16
India 0.22 (0.31) 0.42 (0.64) 0.56 (0.67) 0.82 (1.01) 1.33 (1.82) 1.14 –0.03
Ireland 0.2 (0.29) 0.4 (0.62) 0.4 (0.51) 0.8 (1) 1.07 (1.56) 0.90 –0.10
Italy 0.16 (0.24) 0.32 (0.54) 0.58 (0.69) 0.96 (1.15) 1.14 (1.63) 0.89 –0.23
Japan 0.35 (0.44) 0.74 (0.95) 0.4 (0.51) 0.8 (1) 1.37 (1.86) 1.00 –0.05
Korea, Republic of 0.22 (0.31) 0.42 (0.64) 0.56 (0.67) 0.82 (1.01) 1.1 (1.59) 0.99 –0.13
Netherlands 0.1 (0.18) 0.28 (0.49) 0.34 (0.46) 0.76 (0.95) 1.07 (1.56) 1.16 –0.03
Portugal 0.23 (0.31) 0.49 (0.7) 0.45 (0.57) 0.84 (1.03) 1.09 (1.58) 1.15 –0.04
Russian Federation 0.22 (0.31) 0.42 (0.64) 0.56 (0.67) 0.82 (1.01) 1.12 (1.61) 1.00 0.00
Average 0.22 (0.3) 0.45 (0.66) 0.49 (0.6) 0.83 (1.02) 1.15 (1.64) 0.98 –0.10
Sources: Organisation for Economic Cooperation and Development (2010); Kee, Nicita and Olarreaga (2008); and IMF staff calculations.
189
©International Monetary Fund. Not for Redistribution
190 Do Fiscal Spillovers Matter?
TABLE 6A.2
Impact of a 1 Percent of GDP Reduction in Fiscal Spending on Growth After Two Years,
Baseline Multipliers (Percent)
From: Austria Belgium Brazil China France Germany Greece India Ireland Italy Japan
To:
Austria –1.10 –0.01 –0.01 –0.02 –0.03 –0.16 0.00 –0.01 0.00 –0.05 –0.01
Belgium –0.01 –0.75 0.00 –0.02 –0.12 –0.12 0.00 –0.02 0.00 –0.04 –0.01
Brazil 0.00 0.00 –0.82 –0.02 0.00 –0.01 0.00 0.00 0.00 0.00 –0.01
China 0.00 0.00 –0.01 –0.82 –0.01 –0.01 0.00 –0.01 0.00 –0.01 –0.04
France 0.00 –0.01 0.00 –0.01 –1.11 –0.05 0.00 –0.01 0.00 –0.03 –0.01
Germany –0.02 –0.01 0.00 –0.02 –0.03 –0.81 0.00 –0.01 0.00 –0.02 –0.01
Greece 0.00 0.00 0.00 0.00 0.00 –0.01 –0.82 0.00 0.00 –0.01 0.00
India 0.00 0.00 0.00 –0.01 0.00 –0.01 0.00 –0.82 0.00 0.00 –0.01
Ireland 0.00 –0.04 –0.01 –0.02 –0.04 –0.05 0.00 –0.01 –0.81 –0.03 –0.04
Italy –0.01 0.00 0.00 –0.01 –0.03 –0.03 0.00 –0.01 0.00 –0.97 –0.01
Japan 0.00 0.00 0.00 –0.02 0.00 0.00 0.00 0.00 0.00 0.00 –0.81
Korea, 0.00 0.00 –0.01 –0.14 –0.01 –0.01 0.00 –0.02 0.00 –0.01 –0.05
Republic of
Netherlands –0.01 –0.03 0.00 –0.01 –0.05 –0.11 0.00 –0.01 –0.01 –0.03 –0.01
Portugal 0.00 0.00 0.00 0.00 –0.03 –0.02 0.00 0.00 0.00 –0.01 0.00
Russian 0.00 0.00 0.00 –0.02 –0.01 –0.03 0.00 –0.01 0.00 –0.02 –0.01
Federation
Spain 0.00 0.00 0.00 –0.01 –0.04 –0.03 0.00 0.00 0.00 –0.02 –0.01
Sweden 0.00 –0.01 0.00 –0.01 –0.01 –0.03 0.00 –0.01 0.00 –0.01 –0.01
Switzerland –0.01 0.00 –0.01 –0.02 –0.03 –0.06 0.00 –0.03 0.00 –0.02 –0.02
United 0.00 0.00 0.00 –0.01 –0.01 –0.02 0.00 0.00 –0.01 –0.01 –0.01
Kingdom
United States 0.00 0.00 0.00 –0.01 0.00 0.00 0.00 0.00 0.00 0.00 –0.01
PPP –0.01 –0.01 –0.04 –0.18 –0.05 –0.06 0.00 –0.07 0.00 –0.04 –0.08
weighted
average
Impact of a 1 Percent of GDP Reduction in Fiscal Spending on Growth After Two Years,
Baseline Multipliers (Percent)
Korea,
Republic Russian United United Inward
of Netherlands Portugal Federation Spain Sweden Switzerland Kingdom States Total Spillovers
–0.01 –0.01 0.00 –0.01 –0.01 –0.01 –0.02 –0.02 –0.04 –1.53 –0.43
0.00 –0.04 0.00 –0.01 –0.02 –0.01 –0.01 –0.03 –0.04 –1.26 –0.51
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 –0.02 –0.89 –0.07
–0.01 –0.01 0.00 –0.01 0.00 0.00 0.00 –0.01 –0.07 –1.03 –0.21
0.00 –0.01 0.00 –0.01 –0.02 0.00 –0.01 –0.02 –0.03 –1.34 –0.23
–0.01 –0.01 0.00 –0.01 –0.01 –0.01 –0.01 –0.01 –0.03 –1.04 –0.22
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 –0.86 –0.05
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 –0.02 –0.90 –0.08
–0.01 –0.02 0.00 –0.01 –0.03 –0.01 –0.01 –0.06 –0.14 –1.33 –0.52
0.00 –0.01 0.00 –0.01 –0.01 0.00 –0.01 –0.01 –0.02 –1.15 –0.18
–0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 –0.02 –0.88 –0.08
–0.82 0.00 0.00 –0.01 0.00 0.00 0.00 –0.01 –0.07 –1.17 –0.34
0.00 –0.76 0.00 –0.01 –0.02 –0.01 –0.01 –0.03 –0.03 –1.13 –0.37
0.00 –0.01 –0.84 0.00 –0.05 0.00 0.00 –0.01 –0.01 –1.01 –0.16
–0.01 –0.01 0.00 –0.82 –0.01 0.00 0.00 –0.01 –0.02 –0.98 –0.16
0.00 –0.01 –0.01 0.00 –1.00 0.00 0.00 –0.01 –0.01 –1.17 –0.17
0.00 –0.01 0.00 –0.01 –0.01 –0.74 0.00 –0.01 –0.02 –0.88 –0.14
0.00 0.00 0.00 –0.01 –0.01 0.00 –0.74 –0.01 –0.03 –1.00 –0.26
0.00 –0.01 0.00 0.00 –0.01 0.00 0.00 –0.65 –0.02 –0.77 –0.12
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 –0.60 –0.63 –0.03
–0.03 –0.02 0.00 –0.04 –0.03 –0.01 –0.01 –0.03 –0.20 –0.91 –0.14
Source: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
TABLE 6A.3
From: Austria Belgium Brazil China France Germany Greece India Ireland Italy Japan
To:
Austria –0.52 –0.01 0.00 –0.02 –0.04 0.00 –0.02 –0.01 0.00 –0.08 0.00
Belgium 0.00 –0.69 0.00 –0.02 –0.15 0.00 –0.02 –0.01 –0.01 –0.06 0.00
Brazil 0.00 0.00 –0.01 –0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00
China 0.00 0.00 0.00 –0.79 –0.01 0.00 0.00 –0.01 0.00 –0.01 0.00
France 0.00 –0.01 0.00 –0.01 –1.41 0.00 –0.01 0.00 0.00 –0.04 0.00
Germany –0.01 –0.01 0.00 –0.02 –0.04 0.03 –0.01 0.00 0.00 –0.04 0.00
Greece 0.00 0.00 0.00 0.00 –0.01 0.00 –3.54 0.00 0.00 –0.02 0.00
India 0.00 0.00 0.00 –0.01 0.00 0.00 0.00 –0.58 0.00 –0.01 0.00
Ireland 0.00 –0.04 0.00 –0.02 –0.06 0.00 –0.01 0.00 –1.20 –0.04 0.00
Italy 0.00 0.00 0.00 –0.01 –0.04 0.00 –0.02 0.00 0.00 –1.56 0.00
Japan 0.00 0.00 0.00 –0.02 0.00 0.00 0.00 0.00 0.00 0.00 –0.02
Korea, Republic of 0.00 0.00 0.00 –0.13 –0.01 0.00 –0.02 –0.01 0.00 –0.01 0.00
Netherlands 0.00 –0.03 0.00 –0.01 –0.06 0.00 –0.02 0.00 –0.01 –0.05 0.00
Portugal 0.00 0.00 0.00 0.00 –0.03 0.00 0.00 0.00 0.00 –0.02 0.00
Russian Federation 0.00 0.00 0.00 –0.02 –0.02 0.00 0.00 0.00 0.00 –0.02 0.00
Spain 0.00 0.00 0.00 –0.01 –0.05 0.00 –0.01 0.00 0.00 –0.03 0.00
Sweden 0.00 0.00 0.00 –0.01 –0.02 0.00 0.00 0.00 0.00 –0.01 0.00
Switzerland –0.01 0.00 0.00 –0.01 –0.03 0.00 –0.01 –0.02 0.00 –0.04 0.00
United Kingdom 0.00 0.00 0.00 –0.01 –0.02 0.00 0.00 0.00 –0.02 –0.01 0.00
United States 0.00 0.00 0.00 –0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00
PPP weighted 0.00 –0.01 0.00 –0.17 –0.07 0.00 0.00 –0.05 –0.01 –0.06 0.00
average
–0.01 –0.01 –0.01 –0.01 –0.04 0.01 0.00 –0.05 –0.04 –0.85 –0.52 –0.34
0.00 –0.06 –0.01 –0.01 –0.07 0.01 0.00 –0.11 –0.04 –1.25 –0.69 –0.57
0.00 0.00 0.00 0.00 –0.01 0.00 0.00 –0.01 –0.02 –0.09 –0.01 –0.07
–0.02 –0.01 0.00 –0.01 –0.01 0.00 0.00 –0.02 –0.07 –0.97 –0.79 –0.18
0.00 –0.01 –0.01 –0.01 –0.07 0.01 0.00 –0.05 –0.03 –1.67 –1.41 –0.27
–0.01 –0.02 –0.01 –0.01 –0.04 0.01 0.00 –0.04 –0.03 –0.24 0.03 –0.27
0.00 0.00 0.00 0.00 –0.01 0.00 0.00 –0.01 –0.01 –3.59 –3.54 –0.05
0.00 0.00 0.00 0.00 –0.01 0.00 0.00 –0.01 –0.02 –0.66 –0.58 –0.08
–0.01 –0.03 –0.01 –0.01 –0.09 0.01 0.00 –0.20 –0.15 –1.85 –1.20 –0.66
0.00 –0.01 –0.01 –0.01 –0.04 0.00 0.00 –0.03 –0.02 –1.76 –1.56 –0.20
–0.01 0.00 0.00 0.00 0.00 0.00 0.00 –0.01 –0.02 –0.10 –0.02 –0.08
–0.87 –0.01 0.00 –0.01 –0.01 0.00 0.00 –0.02 –0.07 –1.16 –0.87 –0.30
0.00 –1.10 –0.01 –0.01 –0.06 0.01 0.00 –0.09 –0.03 –1.46 –1.10 –0.36
0.00 –0.01 –2.53 0.00 –0.16 0.00 0.00 –0.03 –0.01 –2.81 –2.53 –0.27
–0.01 –0.02 0.00 –0.66 –0.02 0.01 0.00 –0.02 –0.02 –0.80 –0.66 –0.14
0.00 –0.01 –0.04 0.00 –3.09 0.00 0.00 –0.04 –0.01 –3.30 –3.09 –0.21
0.00 –0.01 0.00 0.00 –0.02 1.05 0.00 –0.04 –0.02 0.90 1.05 –0.15
0.00 –0.01 0.00 0.00 –0.03 0.00 –0.15 –0.03 –0.03 –0.38 –0.15 –0.23
0.00 –0.01 0.00 0.00 –0.02 0.00 0.00 –1.95 –0.02 –2.07 –1.95 –0.12
0.00 0.00 0.00 0.00 0.00 0.00 0.00 –0.01 –0.59 –0.62 –0.59 –0.03
–0.03 –0.02 –0.01 –0.03 –0.09 0.01 0.00 –0.10 –0.19 –0.86 –0.73 –0.13
Source: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
TABLE 6A.4
PPP –0.01 –0.01 –0.01 –0.16 –0.08 –0.07 0.00 –0.05 –0.01 –0.07 –0.09 –0.04
weighted
average
–0.01 –0.01 –0.02 –0.04 0.00 –0.01 –0.05 –0.06 –1.4 –0.8 –0.59
–0.06 –0.01 –0.01 –0.08 –0.01 0.00 –0.11 –0.06 –1.7 –0.9 –0.79
0.00 0.00 0.00 –0.01 0.00 0.00 –0.01 –0.02 –0.3 –0.2 –0.10
–0.01 0.00 –0.01 –0.01 0.00 0.00 –0.02 –0.11 –1.0 –0.7 –0.29
–0.01 –0.01 –0.01 –0.07 0.00 0.00 –0.05 –0.04 –2.0 –1.6 –0.37
–0.02 0.00 –0.01 –0.04 0.00 0.00 –0.05 –0.04 –1.2 –0.9 –0.33
0.00 0.00 0.00 –0.01 0.00 0.00 –0.01 –0.01 –2.8 –2.7 –0.08
0.00 0.00 0.00 –0.01 0.00 0.00 –0.01 –0.03 –0.7 –0.6 –0.11
–0.02 –0.01 –0.01 –0.09 0.00 0.00 –0.21 –0.22 –2.2 –1.3 –0.88
–0.01 0.00 –0.01 –0.05 0.00 0.00 –0.03 –0.04 –2.0 –1.7 –0.28
0.00 0.00 0.00 0.00 0.00 0.00 –0.01 –0.03 –1.0 –0.9 –0.10
–0.01 0.00 –0.01 –0.01 0.00 0.00 –0.02 –0.10 –1.5 –1.1 –0.40
–1.08 –0.01 –0.01 –0.06 –0.01 0.00 –0.09 –0.04 –1.6 –1.1 –0.57
–0.01 –2.02 0.00 –0.17 0.00 0.00 –0.03 –0.02 –2.4 –2.0 –0.33
–0.01 0.00 –1.06 –0.02 0.00 0.00 –0.02 –0.03 –1.3 –1.1 –0.21
–0.01 –0.03 –0.01 –3.17 0.00 0.00 –0.04 –0.02 –3.4 –3.2 –0.26
–0.01 0.00 –0.01 –0.02 –0.38 0.00 –0.04 –0.03 –0.6 –0.4 –0.21
–0.01 0.00 –0.01 –0.03 0.00 –0.24 –0.03 –0.05 –0.6 –0.2 –0.35
–0.01 0.00 0.00 –0.02 0.00 0.00 –2.10 –0.04 –2.3 –2.1 –0.17
0.00 0.00 0.00 0.00 0.00 0.00 –0.01 –0.91 –1.0 –0.9 –0.04
–0.02 –0.01 –0.05 –0.10 0.00 0.00 –0.11 –0.30 –1.19 –1.0 –0.19
Source: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
TABLE 6A.5
From: Austria Belgium Brazil China France Germany Greece India Ireland Italy Japan
To:
Austria 0.12 0.00 0.01 0.04 –0.01 –0.13 –0.02 0.02 0.00 –0.06 –0.01
Belgium 0.00 –0.43 0.01 0.03 –0.04 –0.10 –0.02 0.04 0.00 –0.04 –0.01
Brazil 0.00 0.00 1.28 0.03 0.00 0.00 0.00 0.01 0.00 0.00 –0.01
China 0.00 0.00 0.01 1.28 0.00 –0.01 0.00 0.02 0.00 –0.01 –0.03
France 0.00 –0.01 0.01 0.02 –0.35 –0.04 –0.01 0.01 0.00 –0.03 –0.01
Germany 0.00 0.00 0.01 0.03 –0.01 –0.65 –0.01 0.01 0.00 –0.03 –0.01
Greece 0.00 0.00 0.00 0.00 0.00 –0.01 –3.81 0.00 0.00 –0.01 0.00
India 0.00 0.00 0.01 0.02 0.00 0.00 0.00 1.57 0.00 0.00 –0.01
Ireland 0.00 –0.02 0.01 0.03 –0.02 –0.04 –0.01 0.01 –0.75 –0.03 –0.03
Italy 0.00 0.00 0.01 0.02 –0.01 –0.03 –0.02 0.01 0.00 –1.13 –0.01
Japan 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.01 0.00 0.00 –0.64
Korea, Republic of 0.00 0.00 0.02 0.22 0.00 –0.01 –0.02 0.04 0.00 –0.01 –0.04
Netherlands 0.00 –0.02 0.00 0.02 –0.02 –0.09 –0.02 0.01 –0.01 –0.03 –0.01
Portugal 0.00 0.00 0.01 0.01 –0.01 –0.02 0.00 0.00 0.00 –0.01 0.00
Russian Federation 0.00 0.00 0.00 0.03 0.00 –0.02 –0.01 0.01 0.00 –0.02 –0.01
Spain 0.00 0.00 0.01 0.01 –0.01 –0.02 –0.01 0.01 0.00 –0.02 0.00
Sweden 0.00 0.00 0.01 0.02 0.00 –0.02 0.00 0.01 0.00 –0.01 –0.01
Switzerland 0.00 0.00 0.01 0.02 –0.01 –0.04 –0.01 0.06 0.00 –0.03 –0.01
United Kingdom 0.00 0.00 0.00 0.01 0.00 –0.01 0.00 0.01 –0.01 –0.01 0.00
United States 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 –0.01
PPP weighted 0.00 0.00 0.06 0.28 –0.02 –0.05 0.00 0.14 0.00 –0.05 –0.06
average
0.00 –0.01 –0.01 0.03 –0.04 0.01 0.00 –0.04 –0.01 –0.1 –0.23
0.00 –0.02 –0.01 0.02 –0.08 0.01 0.00 –0.08 –0.01 –0.7 –0.32
0.00 0.00 0.00 0.01 –0.01 0.00 0.00 –0.01 0.00 1.3 0.01
–0.01 0.00 0.00 0.02 –0.01 0.00 0.00 –0.02 –0.02 1.2 –0.07
0.00 0.00 –0.01 0.01 –0.07 0.00 0.00 –0.04 –0.01 –0.5 –0.18
0.00 –0.01 –0.01 0.02 –0.04 0.01 0.00 –0.03 –0.01 –0.7 –0.08
0.00 0.00 0.00 0.00 –0.01 0.00 0.00 –0.01 0.00 –3.8 –0.03
0.00 0.00 0.00 0.00 –0.01 0.00 0.00 –0.01 –0.01 1.6 –0.01
0.00 –0.01 –0.01 0.01 –0.09 0.01 0.00 –0.16 –0.05 –1.1 –0.40
0.00 0.00 –0.01 0.02 –0.05 0.00 0.00 –0.03 –0.01 –1.2 –0.10
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 –0.01 –0.6 0.03
–0.35 0.00 0.00 0.02 –0.01 0.00 0.00 –0.01 –0.02 –0.2 0.17
0.00 –0.38 –0.01 0.01 –0.06 0.01 0.00 –0.07 –0.01 –0.7 –0.28
0.00 0.00 –2.50 0.00 –0.17 0.00 0.00 –0.02 0.00 –2.7 –0.22
0.00 0.00 0.00 1.73 –0.02 0.00 0.00 –0.01 0.00 1.7 –0.05
0.00 0.00 –0.04 0.01 –3.16 0.00 0.00 –0.03 0.00 –3.3 –0.12
0.00 0.00 0.00 0.01 –0.02 0.73 0.00 –0.03 0.00 0.7 –0.06
0.00 0.00 0.00 0.01 –0.03 0.00 0.13 –0.02 –0.01 0.1 –0.07
0.00 0.00 0.00 0.01 –0.02 0.00 0.00 –1.57 –0.01 –1.6 –0.06
0.00 0.00 0.00 0.00 0.00 0.00 0.00 –0.01 –0.17 –0.2 –0.00
–0.01 –0.01 –0.01 0.08 –0.10 0.01 0.00 –0.08 –0.05 0.1 –0.04
Source: IMF, World Economic Outlook database, and Direction of Trade Statistics; and IMF staff estimates.
Note: PPP: purchasing power parity.
The discussion of global and regional imbalances has put the spotlight on a possible
link between current accounts and structural policies. Drawing on standard empirical
current account models, the paper finds that the commonly recommended structural
factors cannot explain the widening of imbalances prior to the 2008–09 crisis. That
said, structural factors do help explain some part of long-standing cross-country differ-
ences in the current account levels. In particular, countries with stricter credit market
regulation, higher taxes on businesses, lower minimum wage (in particular, in slow-
growing economies), and generous unemployment benefits tend to have higher current
account balances than others.
INTRODUCTION
Although the relationship between global current account imbalances and the
financial crisis of 2008–09 is far from obvious, concern remains that such imbal-
ances are a continuing source of global instability and a threat to a sustainable
recovery (Blanchard and Milesi-Ferretti 2009). The seriousness with which the
imbalances are viewed is reflected in the far-reaching actions that have been pro-
posed to limit them, including a suggestion for imposing quantitative targets on
the current account balances.1 Underlying these proposals is the premise that
some sizeable fraction of both the surpluses and the deficits represents
The author is grateful for invaluable support and guidance from Ashoka Mody. She also thanks Fabian
Bornhorst for allowing her to use some of his analysis. The chapter has benefited from the insightful
comments by participants at several workshops at the Germany in an Interconnected World Economy
conference organized by the Federal Ministry of Finance of Germany, with particular thanks to the
discussant, Carsten-Patrick Meier; “Preventing and Correcting Macro Economic Imbalances in the
Euro Area” co-organized by the Central Bank of the Netherlands and the IMF; and the IMF internal
group on macroeconomic imbalances in Europe. Special thanks also go to Akito Matsumoto for help-
ful discussions. Susan Becker provided valuable research assistance.
1
See, for example, the proposal by U.S. Treasury Secretary Timothy Geithner to the meeting of G-20
ministers in South Korea in 2010 (http://graphics8.nytimes.com/packages/pdf/10222010geithnerletter.
pdf ).
199
“distortions.” In other words, where the current account balance is the outcome
of an “optimal” allocation of resources (“good imbalance”), it is not a problem;
but the imbalances that result from policy distortions or externalities are “bad.”
Since distortions are undesirable even from the country perspective, their mitiga-
tion by policy action is twice blessed, since this also scales back the threat from
global instability.
In this paper, I empirically examine the contribution of structural factors—the
presumed locus of the distortions—to current account imbalances. While the
analysis covers an extended period, 1975–2009, I use the results to interpret
developments during the last phase of the sample period. It was in those years that
the unprecedented global expansion and exuberance were accompanied by widen-
ing imbalances. I conclude that a significant fraction of the imbalances in the
run-up to the crisis reflected the global cycle. Yet, since much policy attention has
been focused on possible structural causes and remedies, the bulk of the paper is
devoted to assessing the link between structural policies and the current account.
I apply these findings in particular to Germany, where the current account sur-
plus surged to 7.5 percent of GDP in 2007.
In practice, the specific distortions at the root of imbalances remain a matter
of some speculation, with competing explanations for the observed behavior of
the current account. For example, high current account surpluses due to low
investment may reflect a variety of factors, including lack of competition in the
financial sector, high corporate taxation, or expectations of low potential growth.
More seriously, the same package of structural policies is at times prescribed to
both surplus and deficit countries. That package often includes deregulation of
product, services, and credit markets, reduction in employment protections,
removal of rigidities in the labor market, and taxation. While these policies may
be good for many reasons, their impact on the current account is not clear a
priori. Structural policies, which may influence productivity growth and/or access
to credit, could impact both savings and investment decisions. The variety of
channels and the complex interactions between them make the issue an empirical
one, a perspective that I adopt.
For a panel of 106 advanced, emerging, and developing countries, I estimate
an equation to determine the correlates of the current account balance, using five-
year non-overlapping averages. As is standard practice, to represent the intertem-
poral consumption and investment decisions underlying the current account, I
include such control variables as income growth and level, population age struc-
ture, fiscal balance, initial net foreign asset position, and the degree of financial
integration. In line with other recent studies, I find that these standard determi-
nants of the current account did not evolve significantly during the final years of
the global exuberance and so cannot be used to explain the emergence of global
imbalances. I then add a number of variables representing structural factors.
Even more so than the standard variables, structural factors changed little over
time or changed in the same direction in both surplus and deficit countries.
Therefore, these factors can explain very little of the emergence of imbalances
prior to the crisis.
I infer from these findings that the emergence of imbalances was likely linked
to cyclical factors. Germany, in particular, was able to benefit from the global
increase in demand for technology-intensive goods, in the production of which
Germany has a comparative advantage. However, the “windfall” profits of
German firms due to their export success did not immediately translate into an
increase in domestic investment, since German firms apparently viewed the boom
as temporary, and the German growth potential remained low.
As a further consideration, I ask if structural policies, while not directly influ-
ential, may have helped shape the response of the current account to the standard
variables. The evidence presented in this paper suggests that even in their role as
absorbers or amplifiers of changes in fundamentals, structural factors account for
only a small fraction of the imbalances.
To be clear, even if they are not candidates for explaining the rise in imbal-
ances, some structural factors do have a meaningful correlation with the current
account balance and so can explain longstanding differences in the current account
balances across countries. Even these findings need to be qualified, however, as
they are often not robust across country samples and time periods, with some com-
monly recommended policies increasing and some reducing the current account
balance. With these caveats, the empirical results suggest that lower business taxa-
tion, credit market regulation,2 and unemployment benefits can reduce the current
account surplus. Consistent with earlier studies, I find that a lower minimum wage
and less strict employment protection, often recommended for making the labor
market “more flexible,” are associated with larger current account surpluses. In the
application to Germany, this would imply that the minimum wage would have to
be raised and employment protection strengthened to reduce the current account
surplus, although this may not be desirable since a higher minimum wage and
stricter employment protection might also raise unemployment. These findings’
relevance to Germany is therefore unclear. However, in some of Europe’s periph-
eral economies, reducing minimum wage and lowering employment protection
could contribute to reducing their current account deficits.
The empirical evidence therefore points to select structural measures that
would need to be tailored to particular countries, rather than a package of broad
structural policies for addressing imbalances. For Germany, these results suggest
that lower taxes on businesses, further reduction in the gross unemployment
replacement rate, and a smaller public share in the banking system3 could reduce
the surplus. Altogether, however, the impact on the German current account
surplus will likely be modest.
2
The measure of credit market regulation employed in this paper includes four components measur-
ing the degree of public ownership of the banking system, control of interest rates, percentage of
credit extended to private sector, and competition from foreign banks.
3
Germany scores well on all of the subcomponents of the index of credit market regulation except the
degree of public ownership of the banking system due to the large presence of publicly owned banks
(Landesbanken and Sparkassen).
LITERATURE REVIEW
The relationship between structural policies and the current account remains an
open one. The literature agrees that fundamentals such as income per capita,
demographics, fiscal policy, and other traditional factors are important determi-
nants of the current account. But beyond that, while several recent studies point
to imbalances in the run-up to the 2008–09 crisis as “excessive” compared to the
fundamentals, the role of structural factors in the emergence of these imbalances
remains an open question. The overall impact of the commonly recommended
package of structural policies—such as liberalization of product, services and
credit markets, reduction in employment protection, removal of other labor mar-
ket rigidities, and reduction in business taxation—remains unclear.
Chinn and Prasad (2003), Abiad, Leigh and Mody (2009), Jaumotte and
Sodsriwiboon (2010), and Lane and Milessi-Ferretti (2011) find that current
account balances are largely driven by such fundamentals as relative per-capita
income, fiscal stance, demographics, oil prices, the initial net foreign assets posi-
tion, and the degree of financial integration conditional on income level. The
studies find a positive and significant relationship between relative income per
capita and the current account, possibly capturing the fact that capital flows from
rich countries to poor countries, where there are higher growth “catching up”
opportunities. The current account balances are also found to be relatively large
where the fiscal balances are relatively large, suggesting that private sector savings
provide only a partial Ricardian offset to changes in public savings (the coefficient
is often found to be less than one).
Higher old and young dependency ratios are associated with lower current
account balances, since relatively higher dependency ratios are associated with the
lower aggregate savings. However, the expected change in the old dependency
ratio has a positive association with the current account, since countries that age
rapidly are saving more. For oil producers, the current account is positively
related to the oil balance, which captures fluctuations in the oil price. The litera-
ture also finds that the current account is positively associated with the initial net
foreign assets position. While it is somewhat counterintuitive, this finding likely
reflects the fact that the net foreign assets position is generating net investment
income, which is part of the current account. Financial integration is also found
to facilitate access to capital for poor countries; hence, poorer countries tend to
have lower current account balances at a given state of financial integration. Some
studies also find that among developing countries, the degree of trade openness is
negatively associated with the current account balance. Chinn, Eichengreen, and
Ito (2011) also find weak evidence that countries with more developed financial
markets have weaker current accounts, but their results are not robust.
While a substantial body of literature exists on the link between current
accounts and macroeconomic fundamentals, the literature on the link between
structural policies and the current account is scarce and inconclusive. Following
is a summary of the recent studies, which should allow one to view this chapter
in proper perspective.
Kennedy and Sløk (2005) conclude that current account imbalances are struc-
tural in nature because they deviate from the current accounts projected under
unchanged fiscal policies, unchanged real exchange rates, and monetary policy
aimed at closing the output gap in the medium term. They also find that cycli-
cally adjusted current accounts are correlated with the potential growth, although
this correlation is largely driven by cross-country differences. On the other hand,
they do not find a robust link between specific structural policies and the current
account in their reduced-form pooled time series and cross-country regressions,
which they conducted on a sample of 14 OECD countries. However, there is
some evidence that more open product and financial markets are associated with
weaker current accounts. The other variables under investigation included indica-
tors of labor market regulation, foreign direct investment (FDI) restrictiveness,
financial market development (stock market capitalization), and labor market
performance (trend participation rate and non-accelerating inflation rate of
unemployment, or NAIRU). Nevertheless, they encourage policymakers to
undertake structural policies because a faster growing economy will improve wel-
fare, though it may or may not reduce imbalances.
Kerdrain, Koske and Wanner (2010) estimate reduced-form regressions in a
large panel of 117 advanced, emerging, and developing countries to assess the
impact of structural policies on savings, investment, and the current account.
They conclude that structural policies may influence savings, investment, and the
current account, not only through their impact on macroeconomic conditions
such as productivity growth or public revenues and expenditures but also directly.
In particular, social spending, notably spending on health care, is associated with
lower savings rates, possibly due to lower precautionary savings, and with a lower
current account. Stricter employment protection is associated with lower savings
rates, if unemployment benefits are low, as well as higher investment rates, per-
haps due to a greater substitution of capital for labor, leading to lower current
account balances. Product market liberalization is found to temporarily boost
investment, though direct impact on the current account could not be detected.
Financial market deregulation may lower the savings rate, although only in less
developed countries, and again the direct impact on the current account could
not be detected.
While the Kerdrain, Koske and Wanner (2010) study is rather comprehensive,
their regression includes country-specific fixed effects, which may absorb some of the
cross-country variation in the current account, possibly related to the structural
variables, which do not change significantly over time. Also, some of their other
variables, such as user cost of capital and productivity growth, might reflect struc-
tural conditions. As a result, their study does not allow one to fully answer the ques-
tion of the individual impact of various structural policies on the current account.
Kerdrain, Koske, and Wanner (2010) find little evidence that structural poli-
cies affect the speed of adjustment of the current account to the equilibrium. In
contrast, Ju and Wei (2007) provide evidence that rigid labor markets reduce the
speed of adjustment of the current account to the long-run equilibrium. The lat-
ter authors use a two-step approach: first, they estimate a speed of convergence of
the current account ratio to the steady state for each country separately, using a
vector-error correction model, and second, they relate the speed of convergence
to the degree of labor market rigidity in a cross-section of countries. However,
large economies, such as the United States, Japan, and Germany, are excluded
from this analysis, because the authors suggest that the current accounts in large
economies could behave systematically differently due to the importance of not
only their domestic labor market flexibility but also foreign labor market
flexibility.
Jaumotte and Sodsriwiboon (2010) estimate pooled current account regres-
sions with traditional determinants as controls in a smaller sample of 49 advanced
and emerging economies to test for the importance of the European Monetary
Union and the potential impact of policies. They find that financial liberalization
and higher minimum wage lower the current account, while no direct link could
be detected between the level of employment protection or the level of unemploy-
ment benefits and the current account. In an econometric study covering 100
advanced, emerging, and developing countries for the period 2001–09 (annual
data), Bayoumi, Vamvakidis, and Vitek (2010) find that countries with more
(less) credit market regulation have higher (lower) current account balances while
controlling for traditional fundamentals.4
Berger and Nitsch (2010) investigate the link between employment protection
and product market regulation and the bilateral trade balances as a fraction of
total bilateral trade in a sample of 18 European countries over a long time horizon
(1948 through 2008). They find that countries with less flexible labor and product
markets exhibit systematically lower bilateral trade surpluses than others.
A recent body of literature also identifies imbalances in the period preceding
the crisis as “excessive” compared to fundamentals. These studies including
Barnes, Lawson, and Radziwill (2010), who estimate current account regression
with traditional factors in a sample of 25 OECD countries; Lane and Milessi-
Ferretti (2011), in a sample of 65 advanced and emerging economies; and Chinn,
Eichengreen, and Ito (2011), in a sample of 109 industrial and developing coun-
tries. Barnes, Lawson, and Radziwill (2010) and Chinn, Eichengreen, and Ito
(2011) find some evidence that such excesses could partly be explained by hous-
ing investment, real housing appreciation, and stock market performance.
However, large residuals remain, in particular, for the United States and China.
Lane and Milessi-Ferretti (2011) conclude that the countries with the largest
excesses before the 2008–09 crisis have experienced the largest corrections there-
after, and also find that the adjustment in deficit countries has been achieved
primarily through demand compression rather than expenditure switching.
They further conclude that the high output costs that have been associated
with the rapid current-account corrections provide support for research that
assesses whether current account deficits during good times might partly reflect
4
Bayoumi, Vamvakidis, and Vitek (2010) employ an index of credit market regulation constructed by
the Fraser Institute, which is also utilized in this paper (see Appendix for details).
distortions that fail to internalize the risk of a subsequent sudden stop. It is not
clear, however, what exactly these distortions are.
Finally, theoretical literature (see, for example, Vogel 2011) suggests that while
structural policies that mainly target supply-side weaknesses may help regain
competitiveness in economies with competitiveness problems in the short run, in
the longer run this effect is offset by the income effect as imports rise.
Consequently, the lasting long-term rebalancing of external accounts also requires
the correction of demand imbalances.
The current paper contributes to the existing empirical literature in the follow-
ing three dimensions. First, it attempts to shed some light of the role of struc-
tural policies in the emergence of imbalances in the run-up to the 2008–09
financial crisis. Second, it assesses the direct impact of a commonly recommended
package of structural policies on the current account in a large sample of
advanced, emerging, and developing countries while controlling for traditional
macroeconomic fundamentals. Third, it assesses the potential size of the current
account reduction due to these policies in Germany, which has come under a
spotlight due to its large current account surplus.
While the results in this paper support some of the earlier findings, they point
to the lack of robustness of many results in determining the level of the current
account. Moreover, the paper emphasizes the muted role of structural factors in
causing the growth of imbalances just prior to the recent crisis. For Germany, the
paper offers some policy directions for change but cautions that the quantitative
effects may be small.
BASELINE MODEL
This section introduces the results of the baseline econometric model. The base-
line model is estimated using a random effects model in a sample of 106
advanced, emerging, and developing countries. It includes traditional fundamen-
tals, which were found to be important current account determinants in the
earlier literature. As a robustness check, an Ordinary Least Squares OLS model
with cluster robust standard errors (not including fixed effects) is also estimated
and yields similar results. The current account is averaged over five-year non-
overlapping periods spanning the period of 1975–2009, since the goal is to iden-
tify the determinants of the medium term or so to speak “structural current
account.” Many of the explanatory variables enter as deviations from the PPP-
weighted sample average in a given period, which captures the fact that current
accounts are determined by the countries’ positions relative to their trading part-
ners. Data sources are described in the Appendix.
The baseline model (Table 7A.1)5 largely confirms the findings in the litera-
ture. Higher relative income per capita, fiscal balance, and initial net foreign
assets position as well as higher oil prices for oil producers are associated with the
5
Tables 7A.1 through 7A.6 may be found at the end of this chapter.
higher current account balances.6 Countries with relatively high current depen-
dency ratios have lower current account balances, as the elderly tend to draw on
savings more. However, countries with the higher expected increases in the
dependency ratio, capturing the speed of aging, are found to have higher current
account surpluses.
The regression also includes the degree of financial integration, measured by
the sum of foreign assets and liabilities in percent of GDP and the interaction of
the financial integration with the GDP per capita growth in the previous period
(column 2). The link between financial integration and the current account
works more robustly through growth than through the income level. In particu-
lar, it reduces the current account balance in the countries with higher previous-
period growth. However, high-growth countries also tend to be poorer countries,
so this finding is consistent with that in Abiad, Leigh and Mody (2009).
The model presented in this chapter does not include any crisis dummies,
unlike some of the earlier studies. The reason is that the goal is to explain the
developments in the current account with the known set of factors, including
structural policies, while the dummies could capture some of the effects without
identifying the policies and factors behind the crisis.
The relationship between the fundamentals and the current account balance
in the sample of OECD countries is somewhat different (Tables 7A.1, 7A.2,
7A.4, and 7A.5, columns 1 and 2). The relationship between the current account
and income per capita, fiscal balance, the ratio of net foreign assets to GDP, old
dependency ratio, an increase in the old dependency ratio, and the interaction of
financial integration and past growth remains broadly unchanged in the OECD
sample, although in some cases the coefficients become insignificant. In contrast,
the coefficient on the young dependency ratio becomes positive and significant.
While this result appears counterintuitive, it is consistent with the findings of
Kerdrain, Koske, and Wanner (2010) as well as Barnes, Lawson and Radziwill
(2010). It could perhaps be explained by the fact that richer OECD countries can
afford to save more for future generations, for example for education purposes.
The degree of trade openness also appears to matter more in a sample of OECD
countries; in particular, the higher the trade openness the higher is the current
account surplus, perhaps reflecting the fact that richer countries that are also more
open tend to export capital to the poorer countries.
The baseline model generates a fairly good fit, especially for advanced coun-
tries, explaining about 35 percent of the variation in the current account bal-
ances in the sample. The model explains cross-country variation better than
time-series variation, with the between R-square of 0.5 (Figures 7.1a and 7.1b).
6
Although some of the variables, e.g. financial integration, openness, and oil price, may be nonsta-
tionary, the residuals from the baseline regression estimated on annual data are found to be stationary
using an augmented Dickey-Fuller test, though they exhibit serial correlation. Therefore, the results
of the random effects model were estimated using standard errors robust for heteroscedasticity and
serial correlation.
0.3
0.2
Actual current account to GDP
0.1
– 0.1
– 0.2
– 0.3
– 0.4
– 0.5
– 0.1 – 0.05 0 0.05 0.1 0.15
Fitted Current Account to GDP
Figure 7.1a Actual and Fitted Current Account to GDP, Advanced Economies
0.3
0.2
Actual current account to GDP
0.1
– 0.1
– 0.2
– 0.3
– 0.4
– 0.5
– 0.15 – 0.1 –0.05 0 0.05 0.1
Figure 7.1b Actual and Fitted Current Account to GDP, Other Countries
Sources: Annual Macroeconomic Database of the European Commission (AMECO); World Economic Outlook data-
base; and IMF staff estimates.
Nonetheless, the residuals from the current account regression largely mirror
the imbalances that emerged in mid-2000. The “fundamentals” therefore did not
evolve to generate the imbalances. This is the case even when accounting for the
potential impact of the financial integration and trade openness. The fact that
imbalances widened across the globe suggests that some global forces were at
work, although country-specific factors probably determined the direction of
change in the current accounts (Figures 7.2a and 7.2b).
10
China Germany
8
Japan Spain
6
United States
4
2
0
–2
–4
–6
–8
–10
1975 1980 1985 1990 1995 2000 2005
Year indicates the beginning of the 5-year period
Figure 7.2a Current Account Balance, Germany and Four Large Economies, 1975–2005
(Five-year averages, percent of GDP)
10
China Germany
8
Japan Spain
6
United States
4
2
0
–2
–4
–6
–8
– 10
1975 1980 1985 1990 1995 2000 2005
Year indicates the beginning of the 5-year period
Figure 7.2b Residuals from the Baseline Model, Germany and Four Large Economies,
1975–2005 (Five-year averages, percent of GDP)
Sources: Annual Macroeconomic Database of the European Commission (AMECO); IMF, World Economic Outlook
database; and IMF staff estimates.
4.5
Japan
4.0 United States
Germany
3.5
Spain
3.0
2.5
2.0
1.5
1.0
0.5
0.0
1989 1992 1995 1998 2001 2004 2007
Figure 7.3a Employment Protection, Germany and Three Selected Countries, 1989–2007
0.6
0.5
0.4
0.3
China Japan
United States Germany
0.2
Spain
0.1
0.0
1980 1983 1986 1989 1992 1995 1998 2001 2004
Figure 7.3b Gross Unemployment Replacement Rate, Germany and Four Selected
Countries, 1980–2004
12
10
4
China Japan
2 United States Germany
Spain
0
1985 1988 1991 1994 1997 2000 2003 2006
Figure 7.3c Credit Market Regulation, Germany and Four Selected Countries, 1985–2006
2
United States Japan
1 Spain Germany
0
1980 1983 1986 1989 1992 1995 1998 2001 2004 2007
7
In Germany, the 2004 Hartz IV reform reduced unemployment benefits and social transfers and
increased the flexibility of temporary employment. The subcomponent of the employment protection
indicator, which measures protection of temporary employment, did decline, but the overall index
increased.
8
Some of the variables to replace the missing values were interpolated, as some of these variables are
not available on an annual basis. I also extrapolated the values of some structural variables to 2009,
since for this year many of the structural variables were not available. The index of employment pro-
tection is available only for OECD countries, but there are subcomponents of this indicator, such as
advance notice period and severance pay after nine months, available for a broader set of countries in
(Aleksynska and Schindler, 2011). I constructed an employment protection index for a broader set
using an out-of-sample forecast from the regression of the employment protection index on advance
notice period and severance pay after nine months.
sample). Generally, the results do not indicate a robust relationship between the
current account and structural policies, although in some specifications in the full
sample the coefficient on the unemployment gross replacement ratio is positive
and significant, while that on the ratio of the minimum-to-mean wage and
employment protection indicator is negative and significant. No significant asso-
ciation is found for OECD countries, although the sample there is rather small.
The positive association between the current account and the gross unemploy-
ment replacement rate could reflect the fact that generous unemployment systems
might contribute to higher unemployment rates by reducing incentives to seek
new jobs (Bassani and Duval 2006, Vandenberg 2010). In such an environment,
the unemployment rate and the probability of becoming unemployed are higher,
which could lead to higher precautionary savings by households. However, there
might be a counteracting impact as high unemployment benefits provide higher
income in the event of job loss. However, to have a negative impact on the current
account-to-GDP-ratio, this higher income would have to lead to an increase in
the marginal propensity to consume. The results suggest that the latter effect has
not been important historically.9
The negative association between the ratio of the minimum wage to the mean
wage and the current account is consistent with earlier findings and may reflect
the fact that higher minimum wage may lead to higher labor costs and, therefore,
hurt competitiveness. This, in principle, could work through both savings and
investment channels. Higher labor costs may reduce corporate profitability and
savings. However, higher labor costs may also encourage companies to substitute
capital for labor, when the latter is expensive.
Finally, higher employment protection is associated with a lower current
account, which is consistent with the findings in the literature that higher
employment protection reduces savings and increases investment. Higher
employment protection raises implicit and explicit labor costs, so the impact can
be similar to that of the minimum wage.
Not surprisingly, the residuals from the regression where three of the struc-
tural variables are included (unemployment gross replacement rate, ratio of the
minimum-to-mean wage, and employment protection index) continue to mirror
the imbalances (Figure 7A.2). Hence, structural factors on their own did not
evolve to generate the imbalances either.
While structural policies may not have contributed directly to the emergence
of imbalances, they may have helped shape the response of the current account to
macroeconomic shocks and changes in the fundamentals. In other words, struc-
tural factors might have played a role as macroeconomic shock absorbers or
amplifiers. This hypothesis is tested by analyzing the interaction of structural
factors with the more dynamic fundamentals.
9
At the time of the financial crisis, however, the impact of the reduction in unemployment benefits
may have been different from that observed historically, since the level of unemployment may be
largely a reflection of lower demand for labor rather than lower labor supply. Hence, the finding on
unemployment benefits should be interpreted in the medium-term context.
10
Japan United States
8
Germany Spain
6
4
2
0
−2
−4
−6
−8
−10
1990 1995 2000 2005
Figure 7.4 Residuals from the Model with Structural Variables, Germany and Three
Selected Countries, 1990–2005 (Five-year averages, percent of GDP)
Source: IMF staff estimates.
Impact of Selected Structural Reforms on Current Account, All Countries and OECD Countries
Advanced, Emerging, and Developing Economies OECD Sample
Impact on the current account Impact on the current account
Statistically Statistically
Direction significant Possibly strengthened by Direction significant
Structural reforms that could REDUCE the current account
balance
Deregulation of the credit market p Yes p No
Reducing taxes (profit, labor and other business taxes) and p Yes p No
simplifying procedures for tax payments
Reducing Unemployment gross replacement rate p Yes the higher initial value of the net foreign p Largely Yes
assets and lower previous period growth
Product market deregulation NA NA p No
Deregulation in retail trade NA NA p No
Ivanova
213
©International Monetary Fund. Not for Redistribution
214 Current Account Imbalances: Can Structural Policies Make a Difference?
protection, often recommended for making the labor market “more flexible,” are
associated with larger current account surpluses.
The two new indicators that become significant in the overall sample when
structural variables enter as averages over time are corporate income tax rate/
indicator of doing business paying taxes10 and credit market regulation (the
higher value of this index means less regulation). Countries with a long-standing
tradition of relatively high business taxes are found to have, on average, higher
current account balances. This could reflect the fact that higher corporate taxa-
tion reduces investment incentives and so may raise the current account bal-
ance.11
The credit market regulation index, which is constructed by the Fraser
Institute, includes several components, namely: the degree of public ownership of
the banking system, control of interest rates, percentage of credit extended to the
private sector, and competition from foreign banks. For example, in the case of
Germany this index indicates strict regulation largely on account of the high
public ownership of the banking system. The results suggest that stricter credit
market “regulation” raises the current account. Stricter credit market regulation
can work through both savings and investment channels. In particular, the lack
of access to credit may constrain investment. However, lack of access to credit
may also encourage household and corporate savings. Given that the index cap-
tures a broader set of components than just credit extended to the private sector,
the results could indicate that it is the broader effectiveness and efficiency of the
banking sector that affects the current account.
To be clear, though, these relationships are not evident in the OECD sample.
The indicators of the degree of regulation in product and services markets, which
are available only for OECD countries, generally are not significantly associated
with the current account. The results for the OECD sample, however, should be
interpreted with caution due to a relatively small number of observations.
Following Chinn and Ito (2007) and Abiad, Leigh, and Mody (2009), as a
robustness check two additional financial measures were included, namely the
degree of financial development measured by the ratio of private credit to GDP
and the measure of capital account openness constructed in Chinn and Ito
(2008). Unlike Chinn and Ito (2007), however, I included a measure of financial
development at the start of the period rather than the five-year period average to
mitigate the potential endogeneity problem, since financial development is mea-
sured by the ratio of private credit to GDP. Both financial development and
capital account openness were not significant when included on their own.
However, similarly to Abiad, Leigh and Mody (2009), I find that fast growing
countries (typically, these are poorer countries), which a have higher degree of
capital account openness, also have lower current account balances, which could
10
This variable captures the amount and administrative burden of paying taxes and contributions for
a medium-size company; it is a rank of a country among all countries.
11
There is evidence from firm-level data that lower corporate tax rates or higher depreciation allow-
ances are associated with higher investment (e.g., Vartia, 2008; Schwellnus and Arnold, 2008).
12
Tables 7A.1 through 7A.6 are located in the appendix to this chapter.
13
The analysis included various interaction terms, but the table reports only a subset of the results. In
particular, no robust link between the interaction of credit market regulation/demographics and the
current account could be established, although some theoretical research (Coeurdacier, Guibaud, and
Jin, 2012) suggests that such interaction may be important.
The impact of the gross unemployment replacement rate depends on the ini-
tial net foreign assets position and the previous period’s per capita income growth.
In particular, the positive impact of the unemployment benefits on the current
account may be reduced in countries that experience rapid income growth. This
finding would be consistent with the explanation that high unemployment ben-
efits increase the rate of unemployment and the probability of becoming unem-
ployed, which in turn lead to higher precautionary savings, since such a probability
would be reduced in an environment of rapid income growth. The finding that
the positive impact of unemployment benefits on the current account is strength-
ened in countries with a high initial net-foreign-assets position is difficult to
interpret; it could be related to the fact that the net-foreign-asset position might
capture the persistence of the current account beyond the factor-income contri-
bution.
Nonetheless, the residuals from the regression with interaction terms (Table
7A.5, column 2; and Figure 7.5) track the imbalances, though they are closer to
zero than in the baseline model for all countries except Japan. So even as absorb-
ers or amplifiers of changes in the fundamentals, the commonly evoked struc-
tural policies cannot account for the emergence of imbalances. There might be
other important structural differences in the economies of the surplus and deficit
countries, not necessarily representing policy distortions, which translated global
shocks into the differing responses of the current accounts.
In addition, the emergence of imbalances coincided with the global cyclical
upswing and a rapid expansion of world trade; cyclical factors have therefore
likely played a role. The correlation of the “excess imbalances” with the housing
investment/housing real price as well as with the performance of the stock market
found in the literature provide further support to this proposition. A further
investigation into the role of structural policies and broader structural factors in
the impact of cyclical shocks on the current account may therefore be warranted.
10
8 China Japan
6 United States Germany
Spain
4
2
0
–2
–4
–6
–8
1975 1980 1985 1990 1995 2000 2005
Year indicates the beginning of the 5-year period
Figure 7.5 Residuals from the Model with Structural Variables Interacted with
Fundamentals, Germany and Four Selected Countries, 1975–2005 (Five-year averages,
percent of GDP)
Source: IMF staff estimates.
12
World market share in specialized
CHN
10 DEU
USA
product varieties
6
ITA
4 JPN FRA
NLD
KOR ESP
2 MYS IND
POL
IRL
0 GRC PRT
0 100 200 300 400 500
Number of specialized product varieties
14
The charts on competitiveness and imports were provided by Fabian Bornhorst as part of the joint
column on VOXEU, which can be found at http://www.voxeu.org/index.php?q=node/6873.
15
In view of East Asia’s deep and extensive industrial division of labor, China’s exports to Germany
include export value added from other countries.
Germany’s export growth is mainly due to growth in world trade, not increas-
ing market share.
500
Increased market share
World trade effect
400
300
200
100
–100
ina
d
ia
rea
d
s
y
ce
ain
ly
l
e
sia
dS n
es
ga
nd
an
lan
nc
lan
Ita
Ind
ee
tat
p
lay
Ch
Sp
rtu
Ko
rla
rm
Fra
Ja
Po
Ire
Gr
Po
Ma
the
Ge
ite
Ne
Un
Figure 7.7 Contribution of World Trade Growth and Changes in Market
Share to Export Increases, 15 Selected Countries
Sources: UN Comtrade Database; and IMF staff estimates.
Note: Increase in exports 2001–08, as percent of exports in 2001, decomposed into the effect of
world trade growth and that of increased market share. Computed with Standard International
Trade Classification (SITC) 4 level trade data.
The share of imports from China has grown rapidly from a low base.
Rest of the
World
28%
European
Union
United States 60%
9%
China
3%
suggests that while all sectors contributed to increased current account surplus,
the largest contributor was German corporate sector (Figure 7.9), which did not
match a substantial increase in profits with increased investment despite the latter
being consistently low. Germany’s corporate investment remained low compared
Rest of the
World
27%
European
United States Union
3% 56%
China
14%
German investment has been low compared to EU peers even after accounting
for outward foreign direct investment.
25 EU
Germany
23
21
19
17
15
13
11
9
1999 2001 2003 2005 2007 2009
to European peers, even accounting for foreign direct investment (FDI) outflows.
The reluctance to invest domestically reflects long-standing low returns to invest-
ment in Germany, but pinning down particular policy distortions that could hold
back investment is difficult. One possible explanation, consistent with the find-
ings for the German labor market in the years preceding the 2008–09 crisis
(Burda and Hunt, 2011), is manufacturing employers’ lack of confidence that the
boom would last. The estimated potential growth in Germany remained
low (close to 1 percent) during those years, and the companies chose to save a
TABLE 7.2
CONCLUSION
This chapter reported on my econometric investigation into the possible links
between the current account balance and the commonly recommended package
of structural policies, including financial regulation, tax policy, and labor market
flexibility. I find little evidence that this set of policies contributed substantially
to the emergence of global imbalances. The large imbalances likely reflected
mainly a booming world economy. Moreover, while the structural factors might
have helped shape the response of the current account to macroeconomic shocks
and fundamentals, even in their role as shock absorbers/amplifiers those factors
only partially account for the emergence of imbalances.
16
In addition, overall low private investment in the 2000s reflected a prolonged period of normaliza-
tion in housing construction following the reunification boom and restructuring in the commercial
real estate construction.
REFERENCES
Abiad, A., D. Leigh, and A. Mody, 2009, “Finance and Convergence,” Economic Policy (April).
Aleksynska, M., and M. Schindler, 2011, “Labor Market Regulations in Low-, Middle-, and
High-Income Countries: A New Panel Database,” IMF Working Paper No. 11/154
(Washington: International Monetary Fund).
Barnes, S., J. Lawson, and A. Radziwill, 2010, “Current Account Imbalances in the Euro Area:
A Comparative Perspective,” OECD Working Paper No. 826 (Paris: Organisation for
Economic Co-operation and Development).
Bassani, A., and R. Duval, 2006, “Employment Patterns in OECD Countries: Reassessing the
Role of Policies and Institutions,” OECD Social Employment and Migration Working Paper
No.35 (Paris: Organisation for Economic Co-operation and Development).
Bayoumi, T., T. Oni, A. Vamvakidis, and F. Vitek, 2010, “How Far Do Differences in Financial
Regulation Drive Global Imbalances?,” Mimeo (Washington, DC: International Monetary
Fund).
Berger H., and V. Nitsch, 2010, “The Euro’s Effect on Trade Imbalances,” IMF Working Paper
No. 10/226 (Washington: International Monetary Fund).
APPENDIX
Data Description
The analysis included a sample of 106 advanced, emerging, and developing coun-
tries with populations exceeding one million. The OECD sample included 27
countries. The new EU member states are included starting from the year 1994
to avoid structural breaks. Most of the traditional variables determining the cur-
rent account were computed following Abiad, Leigh, and Mody (2009).
The current account as a ratio to GDP was taken from the Annual
Macroeconomic Database (AMECO) of the European Commission’s Directorate
General for Economic and Financial Affairs (http://ec.europa.eu/economy_
finance/indicators_en.htm) where available, and from the IMF’s World Economic
Outlook (WEO) database in other cases. Income per capita is real PPP GDP per
capita in 2005 constant prices with 1996 reference year from Penn World Tables
7.3 up to year 2007 (http://pwt.econ.upenn.edu). The rest of the years were
extrapolated using per capita real GDP growth from the WEO database. Fiscal
balance as a share of GDP was computed as general government net lending/
borrowing from the WEO database where available, otherwise general govern-
ment overall fiscal balance was used from the same database. Net foreign assets as
a ratio to GDP were computed as foreign assets minus foreign liabilities divided
by GDP. All the variables are from the External Wealth of Nations (1970–2007)
database, which can be downloaded from http://www.philiplane.org/EWN.html.
Financial integration was computed as the sum of foreign assets and foreign lia-
bilities divided by GDP from the same data source.
Old (young) dependency ratios were computed using the data from the World
Development Indicators (WDI) database. The old (young) dependency ratio was
defined as the ratio of the population aged above 64 (below 15) relative to the
population aged 15–64. The increase in the old dependency ratio was computed
over the five-year period (see below) to capture the underlying demographic
trend. Trade openness is calculated as the sum of exports and imports divided by
GDP; it is obtained from the Penn World Tables 7.3 database (‘openc’/100). Oil
price is taken from IMF’s WEO database.
Several macroeconomic variables (current account to GDP ratio, GDP per capita
growth, fiscal balance, oil price) were averaged over the 5-year non-overlapping
periods, namely, 1975–79, 1980–84, 1985–89, 1990–94, 1995–99, 2000–04, and
2005–09. Other variables were included as of the year preceding the beginning of
the five-year period, e.g. 2004 for the period 2005–09. Many of the variables were
also included as the deviations from the PPP-weighted sample average (growth, fiscal
balance, young and old dependency ratios) while real GDP per capita was computed
as the ratio to the U.S. real GDP per capita in a given year.
Credit market regulation is obtained from the Fraser Institute (http://www.
freetheworld.com/) and comprises an index consisting of four components, mea-
suring the degree of public ownership of the banking system, control of interest
rates, percentage of credit extended to the private sector, and competition from
foreign banks. The index ranges between zero and 10 with the higher values
implying less regulation.
Current Account and Structural Policies: Random Effects Model with Robust Standard Errors, Structural Variables are Averages over Five Year
Periods, Total Sample
Dependent variable=current account to GDP (1) (2) (3) (4) (5) (6) (7) (8) (9)
(5-year average) 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009
Log of GDP per capitaa,b 0.0247*** 0.0236*** 0.0265*** 0.0211*** 0.0176** 0.0107 0.0047 0.0107 0.0083
[4.49] [4.36] [4.33] [3.01] [2.55] [1.18] [0.47] [1.18] [1.26]
Previous period growthc,d –0.0002 0.0010 0.0012 0.0003 0.0011 0.0016 0.0020 0.0016 0.0022
[–0.25] [1.03] [1.18] [0.34] [1.09] [1.09] [1.25] [1.09] [1.41]
Fiscal balance to GDPc,d 0.3853*** 0.3782*** 0.3790*** 0.1882** 0.2325*** 0.1328 0.0219 0.1328 0.1450
[4.56] [4.42] [3.73] [2.42] [2.84] [1.13] [0.21] [1.13] [1.45]
Net foreign assets to GDPb 0.0146** 0.0197*** 0.0194*** 0.0253*** 0.0248** 0.0229** 0.0282** 0.0229** 0.0196
[2.26] [3.28] [2.62] [2.61] [2.05] [2.21] [2.15] [2.21] [1.45]
Old dependency ratiob,d –0.3226*** –0.3400*** –0.3744*** –0.3810*** –0.1273 0.0191 0.0373 0.0191 –0.2631**
[–3.73] [–3.93] [–4.15] [–3.64] [–1.29] [0.20] [0.29] [0.20] [–2.09]
Young dependency ratiob,d –0.0138 –0.0177 –0.0253 –0.0116 0.0354 0.0699** 0.0899** 0.0699** 0.0110
[–0.54] [–0.67] [–0.88] [–0.43] [1.25] [2.33] [2.08] [2.33] [0.31]
Trade opennessb –0.0080 –0.0117 –0.0079 –0.0033 0.0022 0.0046 0.0146 0.0046 0.0152
[–0.82] [–1.12] [–0.64] [–0.23] [0.25] [0.41] [1.17] [0.41] [1.19]
Increase in the old dependency ratio over 5 years 0.7330*** 0.6511** 0.6117** 0.5964** 0.2593 0.6761*** 0.8173** 0.6761*** 1.0855***
[2.91] [2.54] [2.19] [2.02] [0.96] [3.37] [2.28] [3.37] [3.35]
Contemporaneous oil price*Oil producerc 0.0005** 0.0004** 0.0005** 0.0006*** 0.0003 0.0001 0.0003* 0.0001 0.0004**
[2.29] [2.07] [2.09] [2.73] [1.42] [0.89] [1.88] [0.89] [2.10]
Financial integrationb 0.0034 0.0035* 0.0031 0.0032* 0.0004 –0.0010 0.0004 0.0009
[1.61] [1.73] [1.44] [1.71] [0.28] [–0.69] [0.28] [0.38]
Financial integration*Previous period growthe –0.0010** –0.0013*** –0.0011*** –0.0011*** –0.0003 –0.0005 –0.0003 –0.0010**
[–2.53] [–3.59] [–3.44] [–3.46] [–0.77] [–1.49] [–0.77] [–2.04]
Credit market regulationc,d –0.0025 –0.0010 –0.0018 –0.0020 –0.0017 –0.0020
[–1.45] [–0.63] [–1.04] [–1.16] [–0.88] [–1.16]
Ivanova
(continued)
225
©International Monetary Fund. Not for Redistribution
226
TABLE 7A.1 (continued)
Observations 548 548 501 371 242 153 124 153 172
Number of countries 106 106 101 77 65 48 48 48 59
Source: IMF staff estimates, see Data Description for data sources.
a
Deviation from US level in a given year.
b
At the beginning of the period, for example for a 5-year period covering 2005–2009, 2004 value was used.
c
5-year period average.
d
Deviation from a PPP GDP-weighted sample average.
e
Financial Integration is one year before the beginning of a given 5-year period; growth is the average over the previous 5-year period.
f
Gross replacement rate is the average over 2 years of unemployment.
g
For OECD countries OECD employment protection index was used. For a broader sample an index was constructed as an out-of-sample forecast from the regression of the employment protection index on
advance notice period and severance pay after 9 months. The latter two indicators are available for a large sample of advanced, emerging and developing countries (Aleksynska & Schindler, 2010).
Current Account and Structural Policies: Random Effects Model with Robust Standard Errors, Structural Variables are Averages over the
Whole Period, Total Sample
Dependent variable=current account to GDP (1) (2) (3) (4) (5) (6) (7) (8) (9)
(5-year average) 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–1994 1995–2009
Log of GDP per capitaa,b 0.0264*** 0.0205*** 0.0206*** 0.0157*** 0.0158*** 0.0187*** 0.0191*** 0.0130 0.0263***
[4.83] [3.58] [3.42] [3.01] [3.00] [3.49] [3.65] [1.50] [2.99]
Previous period growthc,d 0.0012 0.0014 0.0015 0.0016 0.0016 0.0017 0.0018 0.0009 –0.0009
[1.17] [1.31] [1.35] [1.31] [1.31] [1.45] [1.51] [0.26] [–0.35]
Fiscal balance to GDPc,d 0.3941*** 0.3123*** 0.3369*** 0.3202** 0.3201** 0.2954** 0.2925** 0.1979** 0.2683
[4.58] [2.90] [3.02] [2.39] [2.38] [2.31] [2.32] [2.23] [1.59]
Net foreign assets to GDPb 0.0211*** 0.0194** 0.0193** 0.0228** 0.0226** 0.0277*** 0.0283*** 0.0209 0.0611***
[3.34] [2.47] [2.29] [2.49] [2.40] [3.11] [3.18] [1.53] [3.33]
Old dependency ratiob,d –0.3481*** –0.3966*** –0.4024*** –0.4652*** –0.4657*** –0.4644*** –0.4671*** –0.0167 –0.4961***
[–4.07] [–4.46] [–4.39] [–4.93] [–4.92] [–4.94] [–5.00] [–0.10] [–3.39]
Young dependency ratiob,d –0.0211 –0.0193 –0.0170 –0.0236 –0.0232 –0.0191 –0.0203 0.0647* 0.1191
[–0.80] [–0.79] [–0.66] [–0.88] [–0.86] [–0.76] [–0.81] [1.69] [1.49]
Trade opennessb –0.0056 0.0032 0.0100 0.0114 0.0117 0.0080 0.0064 –0.0169 0.0155
[–0.48] [0.25] [0.81] [1.00] [1.01] [0.63] [0.52] [–1.17] [1.06]
Increase in the old dependency ratio over 5 years 0.6131** 0.6208** 0.6741** 1.0845*** 1.0905*** 0.9556*** 0.9197*** 0.6809* 1.6603***
[2.37] [2.13] [2.22] [3.13] [3.14] [2.92] [2.82] [1.90] [2.95]
Contemporaneous oil price*Oil producerc 0.0004** 0.0005** 0.0005** 0.0005** 0.0005** 0.0005** 0.0005** 0.0000 0.0004*
[2.06] [2.48] [2.38] [2.16] [2.15] [2.11] [2.08] [0.08] [1.83]
Financial integrationb 0.0034 0.0032 0.0030 0.0016 0.0015 0.0023 0.0027 0.0098 –0.0002
[1.60] [1.64] [1.54] [0.72] [0.67] [0.98] [1.14] [1.23] [–0.09]
Financial integration*Previous period growthe –0.0011*** –0.0013*** –0.0014*** –0.0013** –0.0013** –0.0013*** –0.0014*** 0.0012 –0.0006
[–2.81] [–4.10] [–4.00] [–2.37] [–2.36] [–2.63] [–2.69] [0.33] [–1.35]
Credit market regulationd,f –0.0047** –0.0081*** –0.0080*** –0.0059** –0.0059*** –0.0070*** –0.0067*** –0.0001 –0.0088
[–2.16] [–3.11] [–2.70] [–2.29] [–2.58] [–3.07] [–2.81] [–0.04] [–1.44]
Ivanova
Gross replacement rated,f,g 0.0971*** 0.0998*** 0.0929** 0.0945** 0.1130*** 0.1007** 0.0511 0.1128*
[2.73] [2.65] [2.16] [2.50] [2.80] [2.12] [1.15] [1.79]
(continued )
227
©International Monetary Fund. Not for Redistribution
228
TABLE 7A.2 (continued)
Observations 532 426 400 323 323 349 349 114 118
Number of countries 101 78 73 60 60 65 65 43 59
Source: IMF staff estimates. See Data Description for data sources.
a
Deviation from US level in a given year.
b
At the beginning of the period, for example for a 5-year period covering 2005–2009, 2004 value was used.
c
5-year period average.
d
Deviation from a PPP GDP-weighted sample average.
e
Financial Integration is one year before the beginning of a given 5-year period; growth is the average over the previous 5-year period.
f
Structural variable are country averages over all available years in a given period.
g
Gross replacement rate is the average over 2 years of unemployment.
h
For OECD countries OECD employment protection index was used. For a broader sample an index was constructed as an out-of-sample forecast from the regression of the employment protection index on
advance notice period and severance pay after 9 months. The latter two indicators are available for a large sample of advanced, emerging and developing countries (Aleksynska & Schindler, 2010).
Current Account and Structural Policies: OLS with Cluster Robust Standard Errors, Structural Variables are Averages over the Whole Period,
Total Sample
Dependent variable=current account to GDP (1) (2) (3) (4) (5) (6) (7) (8) (9)
(5-year average) 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–1994 1995–2009
Log of GDP per capitaa,b 0.0215*** 0.0149*** 0.0143** 0.0113** 0.0111** 0.0146*** 0.0153*** 0.0113 0.0202**
[4.51] [2.90] [2.58] [2.16] [2.09] [2.87] [3.07] [1.35] [2.31]
Previous period growthc,d 0.0017* 0.0016 0.0015 0.0018 0.0018 0.0020* 0.0021* 0.0009 –0.0003
[1.72] [1.48] [1.43] [1.45] [1.44] [1.72] [1.79] [0.26] [–0.12]
Fiscal balance to GDPc,d 0.3643*** 0.3084*** 0.3418*** 0.3219** 0.3222** 0.2965** 0.2910** 0.1589* 0.2666
[4.31] [3.34] [3.44] [2.62] [2.61] [2.55] [2.55] [1.80] [1.59]
Net foreign assets to GDPb 0.0339*** 0.0347*** 0.0354*** 0.0383*** 0.0386*** 0.0431*** 0.0429*** 0.0338*** 0.0661***
[5.72] [4.47] [4.26] [4.16] [4.03] [4.78] [4.74] [2.70] [3.47]
Old dependency ratiob,d –0.3061*** –0.3373*** –0.3240*** –0.4374*** –0.4378*** –0.4487*** –0.4504*** –0.0533 –0.6214***
[–3.35] [–4.19] [–3.91] [–5.12] [–5.13] [–5.22] [–5.24] [–0.31] [–4.48]
Young dependency ratiob,d –0.0204 –0.0238 –0.0216 –0.0318 –0.0313 –0.0281 –0.0294 0.0502 0.0442
[–0.76] [–1.03] [–0.88] [–1.19] [–1.17] [–1.16] [–1.21] [1.33] [0.57]
Trade opennessb –0.0061 –0.0002 0.0042 0.0057 0.0060 0.0051 0.0034 –0.0184 0.0207
[–0.54] [–0.01] [0.35] [0.48] [0.49] [0.43] [0.29] [–1.44] [1.67]
Increase in the old dependency ratio over 5 years 0.3301 0.4243 0.4938 0.8303** 0.8340** 0.7158** 0.6772* 0.4677 1.5275***
[1.16] [1.27] [1.50] [2.41] [2.41] [2.15] [1.98] [1.25] [2.71]
Contemporaneous oil price*Oil producerc 0.0005** 0.0006*** 0.0006*** 0.0005** 0.0005** 0.0005** 0.0005** 0.0002 0.0005**
[2.10] [3.29] [3.30] [2.57] [2.59] [2.41] [2.33] [0.43] [2.01]
Financial integrationb 0.0061*** 0.0047*** 0.0048*** 0.0044* 0.0043* 0.0050* 0.0054** 0.0122 0.0027
[2.95] [2.69] [2.71] [1.74] [1.68] [1.91] [2.03] [1.50] [0.78]
Financial integration*Previous period growthe –0.0016*** –0.0015*** –0.0016*** –0.0017*** –0.0017*** –0.0018*** –0.0018*** 0.0009 –0.0014**
[–4.21] [–5.29] [–5.40] [–3.05] [–3.05] [–3.57] [–3.58] [0.26] [–2.15]
Credit market regulationd,f –0.0039** –0.0069*** –0.0066** –0.0050** –0.0052** –0.0064*** –0.0061*** –0.0006 –0.0082
[–2.04] [–3.04] [–2.57] [–2.21] [–2.51] [–3.05] [–2.76] [–0.32] [–1.39]
Ivanova
Gross replacement rated,f,g 0.0789** 0.0760** 0.0768** 0.0798** 0.0955*** 0.0837* 0.0492 0.1009*
[2.46] [2.30] [2.01] [2.46] [2.81] [2.00] [1.18] [1.70]
(continued )
229
©International Monetary Fund. Not for Redistribution
230
TABLE 7A.3 (continued)
Observations 532 426 400 323 323 349 349 114 118
R-squared 0.371 0.399 0.401 0.368 0.368 0.363 0.361 0.311 0.549
Source: IMF staff estimates. See Data Description for data sources.
Note: OLS: ordinary least squares standard linear regression procedure.
a
Deviation from US level in a given year.
b
At the beginning of the period, for example for a 5-year period covering 2005-2009, 2004 value was used.
c
5-year period average.
d
Deviation from a PPP GDP-weighted sample average.
e
Financial Integration is one year before the beginning of a given 5-year period; growth is the average over the previous 5-year period.
f
Structural variable are country averages over all available years in a given period.
g
Gross replacement rate is the average over 2 years of unemployment.
h
For OECD countries OECD employment protection index was used. For a broader sample an index was constructed as an out-of-sample forecast from the regression of the employment protection index on
advance notice period and severance pay after 9 months. The latter two indicators are available for a large sample of advanced, emerging and developing countries (Aleksynska & Schindler, 2010).
Current Account and Structural Policies: Random Effects Model with Robust Standard Errors, Structural Variables are Averages over Five Year
Periods, OECD Sample
(1) (2) (3) (4) (5) (6) (7) (8)
Dependent variable=current account to GDP (5-year average) 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009
Log of GDP per capitaa,b 0.0372** 0.0401** 0.0402** 0.0384* 0.0444** 0.0127 0.0140 0.0207
[2.05] [2.28] [2.12] [1.94] [2.29] [0.89] [0.80] [0.92]
Previous period growthc,d –0.0021 –0.0005 –0.0005 –0.0007 –0.0022 0.0006 0.0012 0.0015
[–1.09] [–0.21] [–0.27] [–0.31] [–0.97] [0.37] [0.56] [0.70]
Fiscal balance to GDPc,d 0.2813** 0.2814** 0.2796** 0.2928** 0.3608*** 0.3052* 0.1279 0.1275
[2.21] [2.16] [2.16] [2.24] [2.70] [1.66] [0.69] [0.67]
Net foreign assets to GDPb 0.0291 0.0307 0.0304 0.0348 0.0310 0.0311** 0.0530*** 0.0692***
[1.41] [1.35] [1.33] [1.50] [1.31] [2.32] [3.96] [5.90]
Old dependency ratiob,d 0.0055 –0.0562 –0.0588 –0.0376 –0.0430 0.0862 0.0525 –0.1104
[0.05] [–0.47] [–0.46] [–0.27] [–0.28] [0.66] [0.40] [–0.55]
Young dependency ratiob,d 0.1513** 0.1422** 0.1447* 0.1843*** 0.1483** 0.0994* 0.1431 0.1367
[2.14] [1.98] [1.88] [2.76] [2.33] [1.94] [1.61] [1.21]
Trade opennessb 0.0261* 0.0295** 0.0295** 0.0346*** 0.0420*** 0.0380*** 0.0527*** 0.0534***
[1.95] [2.18] [2.00] [2.71] [3.44] [2.96] [3.61] [3.36]
Increase in the old dependency ratio over 5 years 0.5264** 0.4982* 0.4958* 0.5353* 0.5232* 0.9795*** 1.5232*** 1.6074***
[1.97] [1.76] [1.76] [1.89] [1.85] [4.05] [4.00] [4.19]
Contemporaneous oil price*Oil producerc 0.0002 0.0002 0.0002 0.0001 0.0001 0.0001 0.0002 0.0001
[0.94] [0.81] [0.80] [0.34] [0.45] [0.64] [1.61] [0.53]
Financial integrationb 0.0002 0.0001 –0.0004 –0.0004 –0.0017 –0.0029** –0.0018
[0.08] [0.06] [–0.22] [–0.23] [–1.49] [–2.28] [–1.10]
Financial integration*Previous period growthe –0.0008** –0.0008** –0.0007** –0.0006 –0.0004 –0.0007** –0.0010***
[–2.07] [–1.98] [–1.97] [–1.59] [–1.35] [–2.04] [–3.81]
Credit market regulationc,d 0.0002 0.0004 –0.0027 –0.0020 0.0016 0.0014
[0.09] [0.15] [–0.73] [–0.69] [0.27] [0.20]
Ivanova
Gross replacement ratec,d,f 0.0090 –0.0043 –0.0128 0.0282 0.0384
[0.32] [–0.16] [–0.46] [0.46] [0.83]
(continued )
231
©International Monetary Fund. Not for Redistribution
232
TABLE 7A.4 (continued)
Current Account and Structural Policies: Random Effects Model with Robust Standard Errors, Structural Variables are Averages over the Whole
Period, OECD Sample
Dependent variable=current (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
account to GDP (5-year average) 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009 1975–2009
Log of GDP per capitaa,b 0.0407* 0.0302 0.0314 0.0168 0.0089 0.0185 0.0227 0.0153 0.0185 0.0133 0.0113
[1.67] [1.30] [1.33] [0.62] [0.30] [0.71] [1.16] [0.76] [0.71] [0.67] [0.65]
c,d
Previous period growth –0.0005 –0.0010 –0.0012 0.0002 0.0004 0.0002 0.0005 0.0007 0.0002 0.0007 0.0007
[–0.23] [–0.42] [–0.46] [0.08] [0.16] [0.10] [0.19] [0.31] [0.10] [0.33] [0.30]
Fiscal balance to GDPc,d 0.2807** 0.2643** 0.2612* 0.0893 0.0746 0.0795 0.1092 0.0971 0.0795 0.0717 0.0831
[2.13] [1.98] [1.94] [0.59] [0.48] [0.51] [0.72] [0.65] [0.51] [0.55] [0.59]
Net foreign assets to GDPb 0.0302 0.0330 0.0310 0.0612*** 0.0641*** 0.0666*** 0.0665*** 0.0712*** 0.0666*** 0.0728*** 0.0720***
[1.34] [1.44] [1.30] [4.42] [4.82] [5.29] [4.49] [4.57] [5.29] [5.74] [5.44]
Old dependency ratiob,d –0.0606 –0.0893 –0.0911 –0.2026* –0.2107* –0.2047* –0.3575*** –0.4064*** –0.2047* –0.3607*** –0.3429***
[–0.47] [–0.72] [–0.70] [–1.77] [–1.73] [–1.69] [–2.64] [–3.07] [–1.69] [–2.83] [–3.62]
Young dependency ratiob,d 0.1438** 0.1384** 0.1435** 0.0557 0.0612* 0.0719** 0.0503* 0.0389 0.0719** 0.0465* 0.0491*
[1.96] [2.08] [2.08] [1.41] [1.96] [2.14] [1.87] [1.45] [2.14] [1.80] [1.69]
Trade opennessb 0.0299** 0.0321** 0.0321** 0.0373*** 0.0377*** 0.0378*** 0.0385*** 0.0323*** 0.0378*** 0.0360*** 0.0388***
[2.10] [2.38] [2.29] [2.88] [3.11] [2.89] [3.55] [3.36] [2.89] [3.61] [3.41]
Expected increase in the old 0.4963* 0.4769* 0.4774 0.7237** 0.7191* 0.7086* 1.0304** 0.9610** 0.7086* 0.9749** 0.9351**
dependency ratio
[1.75] [1.65] [1.64] [2.00] [1.89] [1.94] [2.22] [2.14] [1.94] [2.09] [2.07]
Contemporaneous oil price*Oil 0.0002 0.0002 0.0001 0.0001 0.0001 0.0000 0.0002 0.0002 0.0000 0.0002 0.0001
producerc
[0.81] [0.78] [0.78] [0.66] [0.53] [0.25] [0.94] [1.08] [0.25] [0.99] [0.65]
Financial integrationb 0.0001 –0.0001 –0.0002 –0.0014 –0.0017 –0.0018 –0.0015 –0.0015 –0.0018 –0.0015 –0.0014
[0.06] [–0.03] [–0.11] [–0.99] [–1.41] [–1.42] [–1.03] [–1.07] [–1.42] [–1.21] [–1.06]
Financial integration*Previous –0.0008* –0.0007* –0.0006 –0.0009*** –0.0009*** –0.0010*** –0.0012*** –0.0013*** –0.0010*** –0.0012*** –0.0011***
period growthe
[–1.88] [–1.84] [–1.55] [–3.05] [–3.28] [–3.57] [–2.94] [–2.97] [–3.57] [–3.21] [–2.72]
Credit market regulationd,f –0.0001 0.0006 0.0013 –0.0003 –0.0039 –0.0032 –0.0011 –0.0012 –0.0032 –0.0044 –0.0031
Ivanova
[–0.02] [0.10] [0.20] [–0.06] [–0.67] [–0.44] [–0.16] [–0.22] [–0.44] [–1.03] [–0.72]
Gross replacement rated,f,7 0.0367 0.0360 0.0226 0.0804* 0.0821** 0.0785* 0.0808** 0.0821** 0.0823** 0.0777*
[0.81] [0.79] [0.58] [1.95] [2.00] [1.87] [2.01] [2.00] [2.27] [1.82]
Corporate income tax rated,f –0.0004 0.0011 0.0012 0.0006 0.0008 0.0011 0.0011
[–0.63] [0.99] [1.18] [0.81] [1.24] [1.41] [1.38]
233
©International Monetary Fund. Not for Redistribution (continued )
234
TABLE 7A.5 (continued)
Source: IMF staff estimates. See Data Description for data sources.
a
Deviation from US level in a given year.
b
At the beginning of the period, for example for a 5-year period covering 2005–2009, 2004 value was used.
c
5-year period average.
d
Deviation from a PPP GDP-weighted sample average.
e
Financial Integration is one year before the beginning of a given 5-year period; growth is the average over the previous 5-year period.
f
Structural variable are country averages over all available years in a given period.
g
Gross replacement rate is the average over 2 years of unemployment.
h
For OECD countries OECD employment protection index was used. For a broader sample an index was constructed as an out-of-sample forecast from the regression of the employment protection index on advance
notice period and severance pay after 9 months. The latter two indicators are available for a large sample of advanced, emerging and developing countries (Aleksynska & Schindler, 2010).
Current Account and Structural Policies (interaction with fundamentals): Random Effects Model with Robust Standard Errors, Structural
Variables are Averages over the Whole Period, Total Sample
(1) (2) (3) (4) (5)
Dependent variable=current account to GDP (5-year average) 1975–2009 1975–2010 1975–2009 1975–1994 1995–2009
Log of GDP per capitaa,b 0.0130** 0.0169*** 0.0134*** 0.0111 0.0200**
[2.53] [3.17] [2.74] [1.45] [2.07]
Previous period growthc,d 0.0011 0.0011 0.0016 0.0012 –0.0037
[0.60] [0.96] [1.24] [0.49] [–1.60]
Fiscal balance to GDPc,d 0.3104** 0.3107** 0.2846** 0.1732* 0.2735*
[2.43] [2.17] [2.55] [1.81] [1.69]
Net foreign assets to GDPb 0.0237** 0.0293** 0.0424*** 0.0268** 0.0769***
[2.41] [2.26] [5.44] [2.19] [5.03]
Old dependency ratiob,d –0.4416*** –0.4878*** –0.4353*** –0.0229 –0.4896***
[–4.86] [–5.33] [–5.21] [–0.14] [–3.48]
Young dependency ratiob,d –0.0238 –0.0133 –0.0265 0.0463 0.0566
[–0.89] [–0.50] [–1.04] [1.22] [0.80]
Trade opennessb 0.0083 0.0078 0.0050 –0.0171 0.0102
[0.73] [0.75] [0.45] [–1.00] [0.51]
Increase in the old dependency ratio over 5 years 1.0614*** 1.0954*** 1.0508*** 0.8079** 1.5315**
[3.16] [2.98] [3.29] [2.37] [2.48]
Contemporaneous oil price*Oil producerc 0.0005** 0.0004* 0.0005** –0.0002 0.0004*
[2.26] [1.84] [2.39] [–0.63] [1.69]
Financial integrationb 0.0020 0.0048 0.0018 0.0099 –0.0020
[0.90] [1.12] [0.96] [1.16] [–1.14]
Financial integration*Previous period growthe –0.0013** –0.0007 –0.0014*** 0.0001 –0.0001
[–2.16] [–1.37] [–3.35] [0.03] [–0.21]
Credit market regulationd,f –0.0053** –0.0110*** –0.0059*** –0.0008 –0.0105***
[–2.34] [–2.58] [–3.11] [–0.34] [–2.79]
Ivanova
Gross replacement rated,f,g 0.0843** 0.1608*** 0.1275*** 0.0385 0.1962***
[2.38] [3.12] [4.41] [0.81] [3.91]
Corporate income tax rated,f 0.0015*** 0.0011* 0.0011** 0.0003 –0.0004
[2.81] [1.72] [2.10] [0.39] [–0.53]
235
©International Monetary Fund. Not for Redistribution
(continued )
236
TABLE 7A.6 (continued)
Current Account and Structural Policies (interaction with fundamentals): Random Effects Model with Robust Standard Errors, Structural
Variables are Averages over the Whole Period, Total Sample
Ivanova
237
©International Monetary Fund. Not for Redistribution
238 Current Account Imbalances: Can Structural Policies Make a Difference?
3
3
2
2
1
1
0
−1 0
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
EU and Germany: Household Savings EU and Germany: Household Investment
(Net, percent of GDP) (Net, percent of GDP)
8.0 4
European Union European Union
7.5
7.0 Germany Germany
3
6.5
6.0
5.5 2
5.0
4.5
1
4.0
3.5
3.0 0
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
0.8 Germany
0
−1
0.4
−2
−3
0.0
−4
−5 −0.4
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Figure 7A.1 Germany and EU: Savings and Investment by Sector, 1999–2009
Source: Eurostat.
Discussion
Comment on Chapters 2 and 3
MALTE HÜBNER
239
3.0
IMF estimate GCEE estimate
2.0
1.0
0.0
2004 2006 2008 2010 2012 2014 2016
−1.0
3.0
Hours worked Capital stock Total factor productivity
2.0
1.0
0.0
2003 2005 2007 2009 2011 2013 2015
−1.0
to ageing and shrinking of the population already evident in prior years. This
estimate already relies on fairly optimistic assumptions, for instance that
unemployment can decrease further without causing an acceleration in infla-
tion and that net migration inflows will add 200,000 persons per year to the
workforce. The slightly higher potential growth rates in Schindler’s estimate
are thus most likely caused by other determinants of potential output, such as
faster capital accumulation or a rise in total factor productivity growth. Both
developments are plausible. For instance, it is conceivable that Germans are
Comment on Chapter 4
WERNER EICHHORST
MAIN ARGUMENTS
Martin Schindler provides a consistent analysis of the German employment
response to the 2008–09 global economic crisis. He demonstrates that the moder-
ate impact of the crisis on the German labor market can be explained by the
propitious nature of existing institutions, discretionary action, and the type of
external shock experienced. He refers to five main explanatory elements:
• First, prior labor market reforms adopted in the early to mid-2000s, in
particular the Hartz package, contributed to reducing unemployment in
Germany both with respect to measured open unemployment and struc-
tural unemployment. One can therefore argue that the institutional envi-
ronment of the German labor market has become more employment-
friendly during the past decade. Schindler infers that the Hartz IV reform
has been the most crucial factor, since it merged unemployment assistance
with social assistance and reinforced activation policies for the long-term
unemployed so that the likelihood of the unemployed (re)entering the labor
market increased as incentives to take up paid work were boosted.
• Second, automatic stabilizers helped stabilize employment at the core of the
German labor market, namely in export-oriented manufacturing (Eichhorst
et al. 2010; Möller, 2010). German manufacturers were most heavily hit by
the crisis, which affected them mainly through a trade shock, with foreign
orders declining steeply in late 2008 and early 2009. However, work-time
adjustments meant that they were able to protect their skilled workforce
without having to resort to major redundancies. On the one hand, work-
time accounts, which had run into surplus due to the full utilization of
production capacities during the boom phase between 2007 and 2008,
facilitated a phase of shorter actual weekly working hours. On the other
hand, employers were able to rely on the long-established short-time work
allowance, which provides workers with a partial replacement of pay losses
for hours not worked amounting to at least 60 percent of net hourly pay.
However, without further policy changes, employers would have encoun-
tered major non-wage labor costs in social security contributions to be paid
for hours not worked.
• Third, as a response to this and in order to encourage the take-up of short-
time work instead of dismissing workers, the short-time work allowance was
made more attractive in late 2008 and early 2009. On the one hand, the
maximum duration was increased from six months to 24 months on a tem-
porary basis. On the other hand, and most importantly, employers’ social
security contributions were paid by the unemployment insurance fund if
training was provided to short-time workers or if short-time work lasted for
more than six months. This considerably lowered employers’ costs of using
short-time work.
• Fourth, German manufacturing employers acted very cautiously during the
crisis. Past experience had taught them that dismissing skilled workers dur-
ing a temporary downturn can lead to severe skill shortages when demand
recovers. This is particularly true in situations of imminent demographic
change, which result in smaller cohorts of young workers entering the labor
market. In fact, there is evidence that those sectors where firms had experi-
enced difficulties in recruiting before the crisis were most affected by the
crisis, and employers were most reluctant to dismiss workers at short notice
(Möller, 2010).
• Fifth, and finally, the existing institutional repertoire tallied well with the
character of the shock that hit Germany. It was possible to cope with a
temporary external shock, which predominantly affected manufacturing,
through strong reliance on work-time flexibility in the form of working-
time accounts and subsidized short-time work. Had the shock not been so
temporary—and had it then spread into other sectors—unemployment
would probably have risen more dramatically.
TABLE 8.1
TABLE 8.2
1,600,000
Stock, East
Stock, West
1,400,000 Notifications
600,000
400,000
200,000
0
Jan. 08
Feb. 08
March 08
Apr. 08
May 08
June 08
July 08
Aug. 08
Sept. 08
Oct. 08
Nov. 08
Dec. 08
Jan. 09
Feb. 09
March 09
Apr. 09
May 09
June 09
July 09
Aug. 09
Sept. 09
Oct. 09
Nov. 09
Dec. 09
Jan. 10
Feb. 10
March 10
Apr. 10
May 10
June 10
July 10
Aug. 10
Sep. 10
Oct. 10
Nov.10
Dec. 10
Figure 8.2 Short Time Work in Germany
Source: Bundesagentur fuer Arbeit.
Note: FTE: full time equivalents; PP: percentage point
500,000
400,000
300,000
200,000
100,000
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
from fairly stable employment in manufacturing but also from robust employment
growth in the private and public service sectors. This ongoing sectoral shift implies
a growing variety of job types and a wider dispersion of working conditions and
wages across occupations. Policymakers face a new set of challenges, which are
probably harder to tackle than mere fine-tuning of existing policy schemes.
From a policy perspective, the German experience demonstrates two themes.
On the one hand, automatic stabilizers are important for promoting internal flex-
ibility as a (limited) alternative to external flexibility, that is, dismissals. Flexible
work-time agreements and publicly subsidized short-time work schemes can help
avoid major job losses in the skilled core workforce if, and only if, economic
shocks are temporary. On the other hand, the German case is also characterized
by a specific function of the second tier of employment, consisting of fixed-term
contracts and temporary agency work, which helps buffer the core against major
short-term adjustment of the number of jobs. Nevertheless, the more volatile
forms of employment at the margin of the labor market create new challenges
regarding employment stability and promotion as well as social protection. This
calls for the recalibration of employment protection and unemployment benefits,
in particular better coverage by unemployment insurance.
Comment on Chapter 5
FELIX HÜFNER
Hélène Poirson and Sebastian Weber’s chapter is important in the German con-
text from an analytical as well as policy perspective. Analytically, it challenges the
conventional wisdom that German growth is an important driver of other coun-
tries’ (especially European) economies, by showing that growth spillovers from
Germany, despite the size of the German economy, are actually smaller than those
originating in other large economies. This obviously bears an important policy
message, namely that increased fiscal spending in Germany is less effective in
stimulating growth abroad than is often assumed. This point was underlined in
Chapter 6. At a time of increasing divergences in Europe, these findings are an
important contribution to the debate.
The study presented in Chapter 5 estimates a VAR for 14 European countries
plus Canada, Japan and the United States. One of its key findings is that the
German economy is more sensitive to external shocks than are other countries:
the sensitivity is particularly notable to shocks originating in the United States
and Asia. In contrast, outward spillovers from Germany to other countries are
estimated to be less than half the size of spillovers of other large economies.
However, the spillovers from Germany to some smaller euro area countries that
have tight trade links, such as Austria and the Netherlands, are large, not least
because these smaller countries are integrated into the supply chains of German
companies. Among the larger economies, spillovers are significant only to Italy.
Importantly, however, spillovers to the peripheral countries like Greece, Ireland,
and Portugal are much smaller than, for example, spillovers to these countries
emanating from France. Overall, the finding is that Germany acts as a conduit of
international shocks to other countries, rather than being an independent source
of spillovers. In line with these results is the fact that identified third-country
effects matter less for Germany than for other countries in the sample, showing
that Germany is more directly affected by foreign shocks.
Felix Hüfner was senior economist on the Germany desk in the OECD Economics Department at
the time of writing. The views expressed are those of the author and do not necessarily reflect those
of the OECD or its member countries.
1
In addition to including a crisis dummy, an interesting exploration would be to test whether the
importance of the trade channel has changed over time by identifying its importance for the split
sample. This would probably help to better understand the result that the trade channel is less impor-
tant for the transmission of shocks originating in the euro area relative to those coming from non-euro
area countries. At first glance, this finding is surprising, given the tight trade integration within the area.
TABLE 8.3
0.40
0.35
Import propensity of domestic demand
0.30
0.25
0.20
0.15 Germany
0.10
0.05
0.0
0 0.2 0.4 0.6 0.8
Import propensity of exports
Germany materializes, increased imports are unlikely to exert large effects on the
other major euro area countries. This hypothesis is supported by the fact that
import propensities out of domestic demand are estimated to be relatively low in
Germany, again limiting the outward growth effects of any domestic stimulus
(Figure 8.4). However, import propensities for exports are much higher, again
supporting the chapter’s finding that Germany acts as a conduit to other countries
of changes in external demand for its goods.
10 25
9 Germany on Spain
8 20
Spain on Germany
7
6 15
5
4 10
3
2 5
1
0 0
1999:Q4
2000:Q2
2000:Q4
2001:Q2
2001:Q4
2002:Q2
2002:Q4
2003:Q2
2003:Q4
2004:Q2
2004:Q4
2005:Q2
2005:Q4
2006:Q2
2006:Q4
2007:Q2
2007:Q4
2008:Q2
2008:Q4
2009:Q2
2009:Q4
Figure 8.5 Consolidated Banking Claims between Germany and Spain (Percent
of GDP)
Source: Bank for International Settlements, Consolidated Banking Statistics database and Organisation for
Economic Co-operation and Development, Economic Outlook database.
claims by German banks on Spanish banks have increased from around 5 per-
cent of Spanish GDP in 2002 to close to 20 percent by 2008, while the claims
of Spanish banks on German banks have remained fairly stable in terms of
German GDP over the same period (Figure 8.5). German surplus savings,
originating not least through the lack of domestic investment, were thus partly
channelled to Spain, supporting domestic demand there, not least in the resi-
dential investment sector. It is worth noting that this channel works in both
directions, with negative spillovers from Spain affecting German growth during
and after the crisis; this may be one explanation behind the significant spillovers
from Spain to Germany (the second largest after the United States in the more
recent period) identified by the authors, notably in the more recent period and
the crisis.
The mechanics of monetary policy in the monetary union are another avenue
of growth transmission. According to this channel, slow growth in Germany dur-
ing most parts of the last decade kept monetary conditions looser than they
otherwise would have been (as monetary policy is set on the basis of data for the
euro area average). Given Germany’s weight in the euro area average, slow growth
in Germany translated via lower interest rates to higher growth elsewhere. Thus,
this channel provides an indirect growth spillover from Germany to other coun-
tries in the euro area, however with lower growth in Germany leading to higher
growth elsewhere (Figure 8.6).
The framework applied by the authors might not be able to directly capture
both the financial and the monetary policy channel, since it focuses solely on
GDP growth rates. One possible, though likely imperfect, identification would
be to include (residential) investment growth as an additional variable into the
VAR. This would allow for the identification of an ‘investment channel’
1 Germany
0
−1
Interest rate gap
−2
−3
−4
−5
−6
−7
−8
0 1 2 3 4 5 6 7
Annual real GDP growth
Figure 8.6 Monetary Conditions and Real GDP Growth in the Euro Area (Average
1999–2007)
Sources: Organisation for Economic Co-operation and Development; and author’s calculations.
(in analogy to the inclusion of export growth, which the authors apply to iden-
tify the trade channel), which is the most likely way in which differences in
financial conditions are transmitted across countries.
Comment on Chapter 7
CARSTEN-PATRICK MEIER
Dr. Carsten Patrick Meier is founder and Managing Director of Kiel Economics.
and proposes policy measures that could help to reduce it. The basis of its policy
prescriptions is an econometric study of the current account movements of a
total of 106 advanced, emerging, and developing countries over the years from
1970 to 2009 and their likely determinants. The analysis is conducted within a
random-effects panel framework that uses five-year averages of variables in an
attempt to filter out cyclical forces. The conclusion is that while structural factors
such as taxes, labor market, and product market regulations do not play a sig-
nificant role in the emergence of large current account positions over the last
decade, they may in some cases explain part of the differences in the current
account positions of the countries in the sample. In particular, for some subsets
of countries it is found that the current account is likely to be more positive the
higher the corporate income tax rate, the stricter the credit market regulation,
and the higher the gross replacement rate. Consequently, the tentative policy
prescription of the paper for Germany is to lower the replacement rate, the cor-
porate income tax rate, and regulation of the banking sector (which in the discus-
sion is identified with reduced public ownership of banks) in order to reduce its
current account surplus.
SOME QUARRELS
The paper’s empirical findings and the policy prescriptions derived for Germany
raise questions. With respect to the empirical analysis, what strikes me most is
the lack of robustness in the findings across different country samples. To be
sure, the paper emphasizes this lack of robustness several times and concludes
that “the impact of these structural reforms on the surplus will likely be mod-
est….” Importantly, none of the three structural determinants that the policy
prescriptions for Germany are to address turn out to be significant when the
sample is restricted to the OECD countries. Evidently, there is a structural dif-
ference in the sample; the parameters of the panel model for the non-OECD
countries are significantly different from the parameters for the OECD coun-
tries. Given this difference, it is difficult to draw conclusions from the full-
sample analysis, especially for the OECD countries, to which Germany
belongs.
But even if one accepts the econometric results despite their somewhat shaky
statistical grounding, they look rather puzzling. This is especially the case for the
gross replacement rate, which turns out to have a positive sign in the regression
results. According to standard labor market theory (e. g. Layard, Nickell, and
Jackmann, 2005), a higher replacement rate—in a manner similar to a higher
minimum wage and higher unemployment protection—should increase the res-
ervation wage, thus raising overall wages and unit labor costs and leading to a loss
of price competitiveness, which should ultimately dampen exports and stimulate
imports. The sign should, therefore, be negative. While the paper presents some
economic reasoning for this counterintuitive result, the most likely explanation
is that the statistical result is driven by the high amount of collinearity of the
three labor market indicators used: replacement rate, minimum wage, and
employment protection index. Indeed, countries that have high replacement
rates often also feature high employment protection and minimum wages. A
simple remedy for the empirical strategy would be to aggregate the three indica-
tors into one overall indicator of “worker protection.”
When the result with respect to the replacement rate is applied to Germany, it
becomes even more puzzling. According to the econometric finding of a positive
relationship between this indicator and the current account surplus, the signifi-
cant reduction in unemployment benefits and social assistance from 2004
onwards mentioned in the paper should have reduced Germany’s current account
surplus in the second half of the last decade. Instead, the German current account
surplus increased at an accelerated pace until the global financial crisis took hold.
This lack of dynamic correspondence in the case of Germany also applies to
the second determinant identified in the paper. While the estimated positive
association between the corporate income tax rate and the current account sur-
plus seems plausible—since, everything else being equal, higher corporation taxes
mean lower investment and thus higher capital exports—it is again tricky to find
the pattern for Germany that is suggested by the regression results. Indeed, cor-
porate income taxes were lowered substantially in 2000—two years before the
German current account surplus started to take off.
Finally, there is the conclusion that stricter banking regulation is a factor that
contributes to Germany’s current account surplus. Again, the economic rationale
that tighter credit resulting from a lack of competition in the banking sector low-
ers investment and increases the current account surplus seems plausible.
However, it is hard to see that Germany, with its 2,000 independent banks, suf-
fers from a lack of competition in the banking sector. While most of the savings
banks that make up one-third of the banks are not held by private entities but by
the local communities, they cannot simply be counted as belonging to the state
and therefore hampering competition. Indeed, in the past it has sometimes been
argued that the German banking sector is over-competitive, resulting in low mar-
gins and relatively low capitalized banks. Moreover, the banking sector has been
deregulated substantially over the past 15 years, notably with the elimination of
the public guarantee obligation for the state-owned Landesbanken enforced by
European Union law in 2005. In the case of Germany, neither the level of bank-
ing regulation nor its dynamics lend themselves readily to confirm the positive
relationship between strict banking regulation and the current account as sup-
posed by the regression results.
driving them may have a longer duration than the conventional four to five
years—a fact that casts some doubt on the validity of taking simple five-year
averages for cyclical adjustment—but it remains a cycle. Typically, this cycle
reflects the longer-term adjustment of the economies in question to external
shocks.
The German current account is a case in point. After unification, it turned
negative as government deficits widened, investment in housing soared, and price
competiveness was eroded by sharp wage hikes. It went back into positive terri-
tory in the second half of the 1990s when investment in housing slowed down
and fiscal deficits declined. However, until 2001, the trade account was never
more than 1.5 percent of GDP, thanks partly due to the hike in consumption and
in business investment spurred by the tech-bubble, which happened to be par-
ticularly pronounced in Germany. It was only after the effects of the European
Monetary Union began to take hold that trade and current account surpluses
widened significantly. While the common monetary policy was too strict for
Germany, it was too lax for the economies at the European periphery, which were
already stimulated heavily by the decline in long-term interest rates brought about
by monetary union. As a consequence, consumption and investment boomed and
wage growth accelerated in those countries while the German economy was
dampened, leading to widening current account imbalances between the two
regions of the euro area. The comprehensive tax, social security, and labor market
reforms implemented in Germany between 2000 and 2005 enhanced this pro-
cess. Intended to stop the rise of unemployment and fiscal deficits, the reforms
not only depressed private consumption temporarily, but also lowered workers’
reservation wage and thus improved the price competitiveness of German firms,
not least vis-à-vis its competitors from the rest of the Euro area.
Today, the macroeconomic setting looks completely different. The German
economy has recovered quickly from the global financial crisis, thanks mainly to
the long-term effects of the reforms. Employment continues to rise and unem-
ployment is falling. Most other advanced countries have not come out of the
crisis as well as Germany has. In the rest of the euro area, unemployment has risen
sharply and capacity utilization is well below normal levels. However, adjustment
is already visible. Not only have European periphery countries, such as Ireland
and Spain, managed to improve their current accounts, the German trade surplus
has also declined from its peak of 8.5 percent at the end of 2007 to about 5 per-
cent at the end of 2011. On the domestic side, we now have interest rates that are
too low for the German economy while they are too high for most of the rest of
the euro area countries, a fact that is enhanced by the debt crisis. A small con-
struction boom is already building up in Germany as property prices have risen
in real terms in 2011—for the first time in a decade.
At the same time, with the unemployment rate heading for 6.5 percent in
2012, 5.5 percentage points lower than in 2005, labor shortages have become an
important concern for many firms, and wage pressure is rising. At Kiel Economics,
we expect unit labor costs to increase significantly in Germany over the coming
years while unit labor costs in the rest of the euro area are set to stagnate or even
REFERENCES
Arvanitis, S., 2005, “Computerization, Workplace Organization, Skilled Labour and Firm
Productivity: Evidence for the Swiss Business Sector,” Economics of Innovation and New
Technology, Vol. 14, No. 4, pp. 225–49.
Bresnahan, T. E. Brynjolfsson, and L.M. Hitt, 2002, “Information Technology, Workplace
Organization, and the Demand for Skilled Labour: Firm Level Evidence,” Quarterly Journal
of Economics, Vol. 117, No. 1, pp. 339–76.
Cheung, C., and S. Guichard, 2009, “Understanding the World Trade Collapse,” OECD
Economics Department Working Paper No. 729 (Paris: Organisation for Economic
Co-operation and Development).
Doms, M., Dunne, T., Koske, T., 1997, “Workers, Wages and Technology,” Quarterly Journal of
Economics, Vol. 112, No. 1, pp. 253–90.
Eichhorst, Werner, Mathias Dolis, Paul Marx, Andreas Peichl, Stefan Ederer and Thomas Leoni,
2010, The Role of Social Protection as an Economic Stabiliser: Lessons from the Current Crisis,
Report based on a study conducted for the European Parliament, IZA Research Report 31,
Bonn: IZA.
Eichhorst, Werner, and Paul Marx, 2011, “Reforming German Labour Market Institutions,
Journal of European Social Policy, Vol. 21, No. 1, pp. 73–87.
Fabiani, S., F. Schivardi, S. Trento, 2005, “ICT Adoption in Italian Manufacturing: Firm Level
Evidence,” Industrial and Corporate Change, Vol. 14, No. 2, pp. 225–49.
Fuchs, Johann, M. Hummel, S. Klinger, E. Spitznagel, S. Wanger, E. Weber, and G. Zika, 2011,
“Rekorde und Risiken: neue Arbeitsmarktprognose,” IAB Kurzbericht 7/2011.
GCEE, 2009, “Die Zukunft nicht aufs Spiel setzen,” Wiesbaden.
———, 2010, “Chancen für einen stabilen Aufschwung,” Wiesbaden.
———, 2011, “Herausforderungen des demografischen Wandels,” Wiesbaden.
Hohendanner, Christian, 2010, “Unsichere Zeiten, unsichere Verträge? : Befristete Arbeitsverträge
zwischen Auf- und Abschwung, IAB-Kurzbericht,” 14/2010.
Ivanova, A., and S. Weber, 2011, “Do Fiscal Spillovers Matter?”, IMF Working Paper No.
11/211 (Washington, DC: International Monetary Fund).
Layard, Richard, Stephen Nickell, and Richard Jackmann, 2005, “Unemployment:
Macroeconomic Performance and the Labour Market,” 2nd edition (Oxford: Oxford
University Press).
Möller, Joachim, 2010, “The German Labor Market Response in the World Recession:
De-Mystifying a Miracle,” Zeitschrift für Arbeitsmarktforschung, Vol. 42, No. 4, pp. 325–36.
OECD, 2010a, Economic Surveys: Germany (Paris: Organisation for Economic Co-operation
and Development).
———, 2010b, Economic Surveys: Euro Area (Paris: Organisation for Economic Co-operation
and Development).
Fabian Bornhorst
Fabian Bornhorst is an economist in the European Department of the
International Monetary Fund (IMF) and works on the surveillance teams for
Germany and the euro area. Previously he worked in the Fiscal Affairs and African
departments of the IMF and was an economist in the Ministry of Finance in
Namibia as a fellow of the Overseas Development Institute (ODI). He has pub-
lished papers on fiscal policy, natural resource related tax revenue, and public
investment and growth. He obtained his Ph.D. in economics from the European
University Institute (Italy) and holds master’s degrees in economics from
University College London (U.K.) and the Freie Universität Berlin (Germany).
Werner Eichhorst
Werner Eichhorst studied sociology, political science, psychology, and public
policy and administration at the universities of Tübingen and Konstanz
(Germany), where he graduated as Diplom-Verwaltungswissenschaftler in 1995.
He received his Ph.D. from the University of Konstanz in 1998. From 1999 to
2004 he was project director at the Bertelsmann Foundation, a private think tank
in Germany, where he was responsible for comparative analyses of the German
labor market and related policy areas. After working with the Institute for
Employment Research (IAB) from 2004 to 2005, he joined the Institute for the
Study of Labor (IZA) as research associate in July 2005, becoming a senior
research associate there in February 2006 and deputy director of labor policy in
April 2007. His main research area is the comparative analysis of labor market
institutions and performance as well as the political economy of labor market
reform strategies.
Malte Hübner
Malte Hübner is an economist at the German Council of Economic Experts,
where he is responsible for macroeconomic analysis and energy policy. Before
joining the staff of the Council, he conducted theoretical and empirical research
in the area of fiscal federalism. He obtained his Ph.D. in economics from the
University of Mannheim (Germany) in 2009 and holds a master’s degree in com-
puter science from the Universität des Saarlandes (Germany).
Felix Hüfner
Felix Hüfner is deputy director in the Global Macroeconomic Analysis Department
at the Institute of International Finance (IIF), where he is responsible for the IIF’s
257
Anna Ivanova
Anna Ivanova is a senior economist in the European Department of the IMF
and works on the surveillance team for Germany. Previously she worked as an
economist in the Fiscal Affairs, Middle East, and Central Asia departments
of the IMF and as a physicist at the Institute of Nuclear Problems in Belarus.
She has published a series of papers in the areas of fiscal policy, the role of
international financial institutions, and growth. She obtained her Ph.D. in
economics from the University of Wisconsin-Madison (USA), a master’s
degree in economic development from Vanderbilt University (USA), and a
master’s degree in nuclear physics from the Belarussian State University
(Belarus).
Christian Kastrop
Christian Kastrop is deputy director-general of the Economics Department and
director of the Public Finance, Macroeconomics, and Research Directorate of the
German Federal Ministry of Finance. He has been director of the European
Economic and Monetary Union Directorate and director of the International and
Financial and Monetary Policy Directorate. He is chairman of the OECD Senior
Budgetary Officials (SBO) Network on Performance and Results. From 2008 to
2009, he chaired the EU Economic Policy Committee (ECOFIN-EPC), of which
he was vice-chairman from 2005 to 2008. He is a lecturer at the Free University
of Berlin at the Hertie School of Governance, a policy fellow of the Cologne
Institute of Public Finance, and a State Member for Germany at BRUEGEL, a
think-tank located in Brussels. After starting his career as an assistant researcher
and lecturer in economics at the University of Cologne, he joined the Federal
Ministry of Finance in 1989, where he held various positions, including chief
press officer, director of the Press and Communication Division in the Minister’s
Office, and director of the Fiscal Policy Division. He holds a master’s degree and
a Ph.D. in economics from the University of Cologne, after studies in economics,
psychology, medieval philosophy, and methodology of science at Cologne and at
Harvard University (USA).
Carsten-Patrick Meier
Carsten Patrick Meier is founder and managing director of Kiel Economics, a
macroeconomic and risk management consultancy. He studied economics in
Göttingen and Kiel (Germany) and obtained his Ph.D. under Prof. Dr. Juergen
B. Donges. Between 1998 and 2008 he led the research groups on the German
business cycle and, later, on risks in the banking sector at the Kiel Institute for the
World Economy. He is the author of numerous articles on the business cycle,
capital markets, and banks, as well as on economic modeling and forecasting.
Ashoka Mody
Ashoka Mody was deputy director in the IMF’s European and then Research
Department when this book was written. He is now Charles and Marie Robertson
visiting professor in International Economic Policy at Princeton University. He
worked for many years at the World Bank and has also been a visiting professor
at the University of Pennsylvania’s Wharton School (USA). His academic
research, motivated by and drawing closely on his policy responsibilities, has
recently focused on international finance and domestic political economy.
Hélène Poirson
Hélène Poirson is senior economist in the European Department at the
International Monetary Fund (IMF), where she is responsible for financial sector
surveillance for France. Previously, she worked on the surveillance teams for
France and Germany, and held positions in the Research, Asia and Pacific, and
Finance Departments. Before joining the IMF in 1999, she was economist at the
Observatoire Francais des Conjonctures Economiques (OFCE), a private think-
tank. She has published a range of policy and research papers, especially in the
areas of economic growth, exchange rate and monetary policy issues, and capital
markets issues. Her most recent research focuses on bank-level spillovers within
and from the euro area. She holds a Ph.D. in economics from DELTA, École des
Hautes Études en Sciences Sociales (France) and a MSc in mathematics from
Ecole Polytechnique (France). She is a CFA charterholder.
Martin Schindler
Martin Schindler is a senior economist at the Joint Vienna Institute (JVI), on
leave from the IMF. He is also a policy fellow at the Institute for the Study of
Labor (IZA). At the JVI, he directs and contributes to capacity building and
training activities for public sector officials in Central, Eastern and Southeastern
Europe, mostly on policies for macroeconomic management and financial stabil-
ity. At the IMF, he most recently worked on the surveillance teams for Germany
and the euro area. He has published on a range of topics in macroeconomics and
international finance, and his work has appeared in the Journal of Monetary
Economics, the Journal of International Economics, and the Journal of International
Money and Finance, among other outlets. He holds a Diplom-Kaufmann degree
from the Universität des Saarlandes and a Ph.D. in economics from the University
of Pennsylvania (USA).
Sebastian Weber
Sebastian Weber currently works as an economist at the IMF, where he has held
positions in the European and African departments. He has written policy papers
and articles on labor markets, monetary policy, and international macro and
finance. His current research agenda focuses on the role of domestic institutions
for the transmission of shocks. He has been a consultant for the OECD and the
World Bank. He holds a Ph.D. and a master’s degree in international economics
from the Graduate Institute of International and Development Studies in Geneva
(Switzerland) and studied politics and economics at the University of Cape Town
(South Africa) and the University of Hamburg (Germany), where he obtained a
B.Sc. (Hons.) in economics.
261