ISSUE 2007/02

MARCH 2007
bruegelpolicybrief

by Alan Ahearne1 SUMMARY Policymakers in the US, Asia and Europe should not wait until
Bruegel
alan.ahearne@bruegel.org
William R. Cline
Peterson Institute for
International Economics
Kyung Tae Lee
Korea Institute for International
Economic Policy
Yung Chul Park
Korea Institute for International
Economic Policy
Jean Pisani-Ferry
Bruegel
jean.pisani-ferry@bruegel.org
John Williamson
Peterson Institute for
International Economics
1 0. 55
The Summary and
Policy Challenge sec-
tions on page 1 of this
policy brief were written
by Alan Ahearne and
Jean Pisani-Ferry. The
main text of this policy
brief, from page 2, was
written by all the
authors listed above. The
GLOBAL IMBALANCES:
TIME FOR ACTION
financial markets force adjustment in the large imbalances in global cur-
rent account positions. Although multilateral consultations organised by
the IMF began in Summer 2006, they have yet to be followed by policy
actions. The current stalemate is dangerous, as market participants are
likely to change their minds at some stage about the sustainability of
imbalances. Indications that the main players are able to agree on the
direction of desirable domestic policy changes and are willing to accept
the exchange rate implications of global current account adjustment
would help make this adjustment
Global Adjustment: Euro trade-weighted orderly. The time for action is now.
real exchange rate need not to change on
average, but bilateral rates need to move POLICY CHALLENGE
1. 59
Global adjustment requires substantial
Euro appreciation
US Dollar per Euro 1. 33 effective depreciation of the dollar and
1. 07
appreciation of the Asian currencies. The
trade-weighted exchange rates of the euro
0. 81
and sterling vis-à-vis their main partners
do not need to change, but this does not
173 mean that bilateral exchange rates
Yen per Euro 151 against the dollar should not move.
130 Although adjustment would see the euro
Euro depreciation
108
and sterling depreciate against currencies
87
in Asia, European currencies would need to
strengthen further against the dollar, to at
65

views presented in this 12. 7 least $1.45 per euro and to well over $2
policy brief do not
necessarily represent Renminbi per Euro
11. 6
per pound. Policymakers in Europe should
the opinions of the other
10.5

9. 4
not resist appreciation of their currencies
individuals who partici-
pated in the workshop Euro depreciation 8. 3
versus the dollar so long as it is matched
organised by Bruegel, 7. 2 by depreciation against the yen and the
KIEP and Peterson
Institute, or of their
6. 1
renminbi and happens in the context of a
5. 0
institutions. global currency adjustment.
Source: Bruegel
 GLOBAL IMBALANCES: TIME FOR ACTION

ONE of the principal dangers cur- of the simulation papers and sum- (IMF) China’s surplus swelled to
02 rently facing the world economy ari-
ses from the large and unsustaina-
marises the main policy conclu- an estimated $184 billion (7.2
sions that we draw from the analy- percent of GDP) in 20062, while
bruegelpolicybrief

ble imbalances in current account ses presented at the workshop. On Japan recorded an estimated sur-
positions. Some observers argue the basis of the discussions, we out- plus of $167 billion (3.7 percent of
that these imbalances will unwind line in Section 1 reasons why the GDP) last year. High oil prices pro-
gradually and non-disruptively, current situation is unsustainable. pelled the surplus for countries in
while others emphasise the risks of Adjustment must the Middle East to
a sudden change of sentiment in take place and will $282 billion last year.
financial markets that could result require significant ‘What effect financial
in an abrupt and damaging adjust- movements in globalisation and the There was broad
ment. No one knows which scenario exchange rates. agreement among
will materialise, but a priority for Section 2 argues that proliferation of deriva- the workshop parti-
policymakers should be to reduce adjustment induced tive instruments has cipants on a number
the risks of a crisis that could pro- by policy actions is of points. First, as a
duce a world recession and disrup- more likely to be had on the probability result of the
tions to the global trading system. orderly than one ini- of a smooth unwinding increase in global
For that, the global economy requi- tiated by financial financial integration
res official sponsorship of a crediblemarkets. We view the of global imbalances is over the last decade
and comprehensive adjustment current stalemate an open question.’ or so, larger and
programme. This policy brief outli- regarding policy more persistent cur-
2
nes what such a programme could actions as dange- rent account imba-
This estimate appears look like. rous, as financial market partici- lances are possible for many coun-
conservative. China’s
trade surplus in goods pants are likely to change their tries today than they were in the
was $178 billion in Bruegel, the Korea Institute for minds at some stage about the sus- past. Global capital markets are
2006, with imports International Economic Policy, and tainability of imbalances unless larger and more liquid, and new
reported on a cost, the Peterson Institute for they see that the main players are financial instruments have develo-
insurance, freight (CIF)
basis. When the
International Economics held a joint able to agree on the direction of des- ped that make it easier for inves-
imports data are adjus- workshop including about 30 of the irable policy changes. Section 3 pre- tors to manage risk. What effect
ted to free on board world’s leading experts on how to sents estimates of the exchange financial globalisation and the pro-
(f.o.b.), the trade in achieve such an rate implications of liferation of derivative instruments
goods surplus will likely orderly reduction in global current has had on the probability of a
come in at about $215
billion. Based on trends global imbalances in ‘The current situation account adjustment smooth unwinding of global imba-
in the other items in the Washington DC on 8 from a variety of lances is an open question.
first-half balance of and 9 February 2007. is unsustainable. models. Section 4
payments, Nicholas The purpose of the Adjustment must take describes the policy Second, the US is deriving signifi-
Lardy (2006) estima-
tes that China’s surplus
workshop was to place and will require implications that the cant benefits from the situation.
last year was $240 bil- compare analyses authors of this policy Financial inflows from abroad
lion (see Nicholas and evaluations of significant movements brief drew from these have boosted US asset prices and
Lardy, Toward a the requirements for in exchange rates.’ results and from the helped to keep US long-term inte-
Consumption-Driven an adjustment of this workshop discussions. rest rates low, thereby spurring
Growth Path, Policy
Briefs in International type. The discussions and financing domestic spending
Economics PB06-6, centred on two sets of contribu- 1. WHY THE CURRENT in the US. In addition, it is well
(Washington: Peterson tions: (1) country papers that provi- SITUATION IS known that the return on US gross
Institute for ded a perspective on the underlying foreign assets exceeds that on US
International
UNSUSTAINABLE
Economics), October
factors behind surpluses and defi- gross foreign liabilities. The effect
2006.) cits and the scope for adjustment in There has been a great deal of dis- is that, although US net foreign lia-
the current account, and (2) multi- cussion recently of global current bilities exceed 20 percent of GDP,
3
In fact, despite conti- country simulation papers that pro- account imbalances. Much of the net income payments on these lia-
nuously rising net duced estimates of the changes in attention has focused on the his- bilities are small3. Moreover,
foreign liabilities,
income receipts on US- policy variables and the correspon- torically large US current account because foreign claims on the US
owned assets abroad ding exchange rate adjustments deficit, which reached $857 billion are almost entirely priced or deno-
were greater than that are consistent with scenarios (6.5 percent of GDP) in 2006. The minated in dollars, while US direct
income payments on for a reduction in current account counterpart to this deficit can be and portfolio equity assets abroad
foreign-owned assets
in the United States
imbalances. found mainly in Asia and in the oil- as well as a portion of credit claims
until the fourth quarter exporting countries. According to on foreigners are priced or denomi-
of 2005. This policy brief reports the results the International Monetary Fund nated in foreign currency, the
 GLOBAL IMBALANCES: TIME FOR ACTION

significant depreciation in the

Chart 1: Current Account Balance
(% of world GDP)
dollar and corresponding apprecia-
tions in the currencies of other 03

bruegelpolicybrief
countries, as well as a rebalancing
United States Euro area % of demand and saving across the
Japan Emerging Asia globe5.
Oil exporters 1.5

1
2. WHAT ADJUSTMENT?
0.5
A key question is whether financial
0 markets or policy actions will ini-
-0.5 tiate the necessary and inevitable
adjustment. Market sentiment can
-1 change abruptly and the risk of a
-1.5
market-led adjustment is that it
might involve global recession,
-2 abrupt and excessive changes in
97

98

99

00

01

02

03

04

05

06

07

08

09

10

11
key exchange rates and asset pri-
19

19

19

20

20

20

20

20

20

20

20

20

20

20

20

ces and, as a consequence, aggra-

Source: IMF WEO September 2006, page 17 vated trade frictions. To reduce the
risk of such an outcome, policyma-
decline in the foreign exchange considering that gross foreign hol- kers need to initiate a policy-indu-
value of the dollar over recent dings of US assets would be far lar- ced adjustment in the near future.
years has boosted the dollar equi- ger. At some stage, foreign inves- The recent volatility in global finan-
valent of foreign assets, thereby tors will begin to demand ever cial markets is a reminder of the
reducing US net foreign liabilities higher returns on the US assets dangers of failing to act promptly.
as measured in dollars. As a result, that they buy, though where that
the increase in US net foreign liabi- limit might be is impossible to tell Agreement on the substance of a
lities over the past few years has at this point. policy-induced adjustment is the
been considerably smaller than purpose of the multilateral consul-
the cumulative current account The clear implication is that global tations at the IMF initiated in
deficits4. current account adjustment must 2006. However, they have not yet
take place. The most elementary achieved significant results.
Nevertheless, the current pattern theory tells us that this adjust- Meanwhile, the US is focusing on
of global imbalances is not sustai- ment will require movements in its bilateral relationship with
4
nable. Medium-term projections by exchange rates, including a China, and the Europeans have For comprehensive
data on the valuation
the IMF indicate that at unchanged effects, see Philip Lane
real effective exchange rates, large and Gian Maria Milesi-
current account imbalances will Chart 2: Net Foreign Assets Ferretti The External
persist (see Chart 1). Persistent (% of world GDP) Wealth of Nations Mark
external deficits and surpluses of II: Revised and
% Extended Estimates of
this scale imply an implausible United States Euro area
10 Foreign Assets and
Japan Emerging Asia
accumulation of foreign liabilities Oil exporters Liabilities, 1970-2004,
on the US side and an implausible IMF Working Papers
accumulation of assets on the 5 06/69, (Washington:
International Monetary
Chinese and Japanese sides. The Fund, 2006).
implied steep increase in US net 0
foreign liabilities (Chart 2) from 5
See, for example,
about 8 percent of world GDP (26 Maurice Obstfeld and
percent of U.S. GDP) in 2006 to -5 Kenneth Rogoff, Global
Current Account
roughly 15 percent of world GDP Imbalances and
(over 51 percent of U.S. GDP) by -10 Exchange Rate
2011 raises serious questions Adjustments, Brookings
about the willingness of foreign Papers on Economic
-15 Activity, 1, pp. 67–146
investors to continue accumula-
97

98

99

00

01

02

03

04

05

06

07

08

09

10

11

(Washington: Brookings
19

19

19

20

20

20

20

20

20

20

20

20

20

20

20

ting net claims on the US, especially Institution, 2005).

Source: IMF WEO September 2006, page 17
 GLOBAL IMBALANCES: TIME FOR ACTION

been getting vocal about the yen.
04 Yet the issue of adjustment has a
multilateral character. Thus, a mul-
bruegelpolicybrief
of the euro and sterling against
the dollar. Otherwise, there
would be an effective deprecia-
of policies that clearly signal their
intention to tackle the problem of
global imbalances. This stalemate

tilateral institution or forum, such
as the one convened by the IMF or
possibly an informal Group of Four
(US, euro area or the European
Union, Japan, and China), would
seem to be the appropriate venue
to deal with it.
There is a large degree ‘Effective depreciation
of convergence in the
economic interest of of the dollar and Asian Korea’s current
the key players: currencies implies a
• The US needs to further bilateral appre- as projected by
bring its current ciation of the euro and the IMF, then its
account deficit
down to an accep- sterling against the dol- need to appre-
table level and this lar.’
will require a signi-
ficant effective depreciation of
the dollar and higher US natio-
nal saving.
• China needs to curb its accu-
mulation of foreign exchange
reserves, rebalance growth
towards domestic demand,
and continue removing distor-
tions that favour exporting
industries.
• Although Japan’s weak
exchange rate and ultra-low
interest rates have been ins-
trumental in countering defla-
tion, economic recovery now
permits the return of monetary
policy and the exchange rate to
6
See, Economic a more neutral stance.
Forecasting for 2007,
(Korea Development • Europe’s currencies have
Institute, December already appreciated substan-
2006), and 2007
tially both against the dollar
Economic Forecasting ,
and in effective terms. For
(Samsung Economic
Research Institute, Europeans, the priority is to
November, 2006). avoid an overshooting of their
7
currencies that might result
Paul Krugman, ‘Will
There Be A Dollar from a disorderly adjustment.
Crisis?’ Paper presen- However, it is important to
ted at the Economic recognise that an effective
Policy Panel at the depreciation of the dollar and
Federal Reserve Bank of
New York, 12 February
an effective appreciation of
2007, available at the Asian currencies imply a
www.cepr.org. further bilateral appreciation
tion of the euro and sterling, is dangerous, and it is imprudent
eroding the extent of potential to delay a change in stance until
US external adjustment. financial markets conclude that
the present situation is unsustai-
• The Korean won, like the nable7. Of particular concern
European currencies, has already are:
already appreciated sharply
both against the • The trade frictions between
dollar and in effec- China and the US arising from
tive terms. If China’s exchange rate policy.
account remains • The weakness of the yen.
in small surplus Although it has strengthened
moderately recently, it
remains very weak on a histori-
currency would cal basis. This weakness not
only contributes to the US
ciate further trade deficit but also hurts
against the dollar other economies in Asia that
in the context of global adjust- have suffered a loss of compe-
ment. If instead its current titiveness against Japan. Yen-
account swings toward signifi- funded carry trades may have
cant deficit in 2007-08, as is begun to unwind, but any
being forecast by some Korean rebound in the amount of such
institutions, then more limited trades may weaken the yen
appreciation against the dollar again. A weaker yen would not
and corresponding partial be consistent with the ongoing
reversal of the trade-weighted recovery in Japan’s economy
appreciation experienced to and accompanying prospec-
date could be appropriate6. tive tightening of Japanese
monetary policy.
• Several key oil exporting coun-
tries have adopted a more pru- 3. ADJUSTMENT SCENARIOS
dent and forward-looking
approach than they did in the To examine what a return to sustai-
1970s and 1980s nability might
and are likely to mean for exchange
build their stocks ‘The main participants rates, participants
of foreign assets have not embarked on in the workshop
further. In other were asked to pre-
words, these a set of policies to sent estimates of
countries’ margi- tackle the problem of the exchange rate
implications of cur-
nal propensity to
spend out of oil
global imbalances. rent account
revenues is less This stalemate is adjustment scena-
than one. dangerous’ rios in which the US
However, their current account
surpluses need to deficit narrowed to
decline as domestic absorp- 3 percent of GDP in the medium
tion gradually expands. term. The scenarios differ in how
the burden of adjustment is sha-
In spite of this potential conver- red among individual countries in
gence of interest, the main partici- the rest of the world, but all scena-
pants have not embarked on a set rios assume that most of the
 GLOBAL IMBALANCES: TIME FOR ACTION

adjustment would be borne by Hurst are in this tradition, as changes in bilateral real exchange
China, Japan, other Asian econo-
mies, a few high-surplus European
are the Federal Reserve estimates
with a dynamic general equili-
rates against the US dollar.
Markets tend to focus on this 05

bruegelpolicybrief
economies not in the euro area, brium model referred to by figure, but it is the wrong figure in
and the oil-exporting countries. Christopher Erceg10. determining the extent of the eco-
The external balance of the euro nomic effects of exchange rate
area, which is projected to be in • Reduced-form estimates of changes. The average (i.e. effec-
slight deficit in 2007, is assumed equilibrium exchange rates: tive) exchange rate movements
unchanged. An important goal The Bénassy-Quéré, Lahrèche- shown in Table 1 are of far greater
behind all scenarios is that the Révil and Mignon estimates, importance, because they are
adjustment should take place and those by Stolper and what determine the change in
without depressing the rate of Fuentes using the Goldman trade outcomes for each country.
growth of world GDP. Sachs dynamic equilibrium They also tend to be much smaller,
exchange rate (GSDEER) model for the fundamental economic rea-
Three types of approaches were and by MacDonald and Dias son that many countries are pos-
used to assess those implications: using the behavioural equili- tulated to be appreciating their
brium exchange rate (BEER) exchange rates against the dollar
• Partial equilibrium ‘trade elas- model, are in this family11. So is simultaneously.
ticities’ models: the Baily the equilibrium real exchange 8
model of US trade rate approach • In principle, the extent of See papers presented
at the workshop by
performance, the described in the exchange rate adjustment Martin Baily, ‘Dollar
Cline model of opti- ‘The effective exchange IMF (2006) depends on the underlying fac- Adjustment To Reduce
mal exchange rate rate movements are of paper. tors behind surpluses or defi- The US Imbalance’;
realignment, and cits and on what policy actions William R. Cline
the Stolper and far greater impor- Box 1 describes are taken. Also, as the U.S. defi- ‘Estimating Reference
Exchange Rates’; and
Fuentes elasticity tance: they determine in more detail the cit shrinks, the assumed distri- Thomas Stolper and
model are in this tra- models and bution of the adjustment Monica Fuentes,
dition8. So are the the change in trade approaches used across the rest of the world ‘GSDEER and Trade
m a c r o e c o n o m i c outcomes.’ in the workshop matters. Specifically, the grea- Elasticities’.
balance and exter- papers. Table 1 ter the share of the adjustment 9
International
nal sustainability contains estima- that a country undertakes Monetary Fund (2006),
approaches outlined in the IMF tes presented to the workshop of through a decline in its current ‘Methodologies for CGER
(2006) review of methodolo- the changes in the real effective account balance, the larger the Exchange Rate
gies for equilibrium exchange exchange rates of the main curren- required appreciation of that Assessments’
(November 2006).
rate assessment9. cies required to meet the objecti- country’s real effective
ves for a reduced US current exchange rate. 10
See paper presented
• Macroeconomic models: The account deficit12. at the workshop by Ray
NiGEM model estimates prepa- • The models generally find that Barrell, Dawn Holland
and Ian Hurst
red by Barrell, Holland and Table 2 presents the equivalent a real effective depreciation of ‘Correcting US
Imbalances’. The
Table 1 results presented by
Christopher Erceg are
Real effective exchange rate change required to reduce U.S. current account deficit to 3 per cent of GDP in the based on the Federal
Reserve Board’s SIGMA
medium term (percent change; + implies appreciation, - implies depreciation) model. For more details
US dollar Japanese yen Chinese RMB Euro on SIGMA, see
www.ijcb.org.
Martin Baily -15 to -20 n.e. n.e. n.e. 11
See papers presented
by Agnès Bénassy-
Ray Barrell, Dawn Holland and Ian Hurst -11 to -19 +10 to +14 +3 to +7 -3 to +6 Quéré, Amina Lahrèche-
Révil and Valérie
Bill Cline (a) -18 +11 to +13 +11 to +18 0 Mignon, ‘World
Consistent Equilibrium
Thomas Stolper and Monica Fuentes (b) -16 +18 +5 +2 Exchange Rates’; and
Ronald MacDonald and
Preèthike Dias, ‘BEER
Ronald MacDonald and Preèthike Dias -11* +6 +27 0 Estimates and Target
Current Account
Chris Erceg (c) -8 to -25 n.e. n.e. n.e. Imbalances’.
n.e. (not estimated). (*) Using preferred coefficient estimate. (a) From Jan-Aug 2006 average. Range refers to two model variants applied to the three scenarios
considered by the workshop. (b) Only results from the elasticity model reported. (c) Range refers to the different shocks that are being unwound.
 GLOBAL IMBALANCES: TIME FOR ACTION

06 Table 2
Bilateral real exchange rate change against the U.S. dollar consistent with the REER movements in Table 1
bruegelpolicybrief

(percent change; + implies appreciation)

Japanese yen Chinese RMB Euro

Martin Baily n.e. n.e. n.e.

Ray Barrell, Dawn Holland and Ian Hurst +24 +18 +16
Bill Cline (a) +28 to +39 +31 to +44 +20

Thomas Stolper and Monica Fuentes (b) +25 +10 +15

Ronald MacDonald and Preèthike Dias n.e. n.e. n.e.

Chris Erceg n.e. n.e. n.e.
12 n.e. (not estimated). (a) See table 1, note a. (b) Nominal exchange rates. Only results from the elasticity model reported.
We exclude from the
comparisons estimates
that do not meet the the dollar of between 10 and ween roughly 3.5 and 6.5 per- effective depreciation of the dollar
specifications of the
scenario. Estimates
20 percent from the current centage points of GDP (with needed to bring about the targeted
from Bénassy-Quéré, level is needed to shrink the the low end of the range being adjustment at around 15 percent.
Lahrèche-Révil and U.S. current account deficit to 3 accompanied by an expansion Effective appreciations of around
Mignon, as well as those percent of GDP over the next of domestic demand in China 10 percent for the yen and 15 per-
of the GSDEER approach few years. that more than compensated cent for the renminbi would pro-
of Stolper and Fuentes,
did not examine what for the loss of vide part of the
exchange rate changes • To reduce the Japanese current foreign demand). counterpart. To bring
would be required to account surplus to levels speci- As in the case of ‘The models generally about these effec-
meet the 3 percent of fied in the scenarios (that is, to the Japanese find that a real effec- tive exchange rate
GDP target for the US
current account deficit
between $36 billion and $54 currency, this movements, much
specified in the work- billion, depending on the sce- strengthening of tive depreciation of the larger bilateral
shop terms of reference. nario, from $167 billion in the renminbi in dollar of between 10 appreciations
Both found surprisingly 2006), the models typically effective terms against the dollar
that the dollar was
undervalued, implying
find that a real effective appre- implies a subs- percent and 20 per- would be required,
that those models find ciation in the yen of between tantially larger cent is needed.’ of maybe 25 to 30
financial markets to be 10 percent and 15 percent is bilateral appre- percent for the yen
comfortable with a per- needed. This movement requi- ciation against and 30 percent for
sistent US current res a 25 percent to 30 percent the dollar. the renminbi. But there will also be
account deficit much
higher than this target,
real appreciation of the yen vis- a need for substantial bilateral
at least for an extremely à-vis the dollar, moving the • Since the scenarios assume a appreciations against the dollar by
long period. Both exchange rate to around 90 roughly unchanged current currencies whose effective
models are subject to yen/dollar compared with account deficit in the euro exchange rates do not need to
the possible problems roughly 118 yen/dollar today13. area, little or no change in the change. In particular, the euro
noted in box 1 (third
approach). Moreover, effective value of the euro is would need to strengthen to at
the alternative elastici- • The workshop produced a fairly needed. As the models find least $1.45 per euro, while sterling
ties model estimated by wide range of estimates for the that the euro depreciates would rise to well over $2 per
Stolper and Fuentes pro- required movement in the against the Asian currencies, a pound.
duced results more in
line with those of the
Chinese renminbi. This uncer- stable effective euro implies a
other papers. tainty in part reflects the diffi- marked bilateral appreciation Finally, although the primary focus
13
culty of estimating precisely of the euro vis-à-vis the dollar of the workshop was on the cur-
The implied yen/dollar
figure assumes that
the sensitivity of Chinese to between $1.45 and $1.50 rencies of countries taking part in
most of the required exports and imports to per euro compared with about the IMF multilateral talks, curren-
movement in the real exchange rate movements14. $1.32 today. cies of other economies running
bilateral exchange rate Effective appreciation of bet- large external surpluses would
comes about through a ween 5 and 25 percent was To summarise, the model estima- also need to appreciate on an
change in the nominal
exchange rate. calculated to be required to tes presented at the workshop pla- effective basis in order to meet the
reduce China's surplus by bet- ced the order of magnitude of targets for correcting global imba-
 GLOBAL IMBALANCES: TIME FOR ACTION

lances specified in the workshop

BOX 1
MODELS AND APPROACHES
Three broad modelling approaches were applied by the workshop paper
scenarios. Otherwise US external
adjustment would fall below target. 07

bruegelpolicybrief
The combined role of the smaller
authors to investigate equilibrium exchange rates. The most traditional surplus economies in Asia and
approach was the partial equilibrium “trade elasticities” method in Europe (outside of the euro area)
conjunction with judgmental current account targets. In this approach, in the adjustment process will be
exports and imports depend on the price incentive provided by the real more important than either China
effective exchange rate, and on the impact of foreign income on demand or Japan15. This consideration illus-
for exports and of domestic income on demand for imports. The more trates once again the multilateral
complete models in this genre include detailed treatment of foreign asset nature of the adjustment problem,
and liability changes and rates of return. These models examine the mag- which to date has arguably been
nitude of the exchange rate change needed to shift the current account addressed with an excessive
from baseline to target levels. A strength of this approach is its transpa- emphasis on just one facet: the US-
rency. It does, of course, require a judgment about the size of the current China relationship.
account deficit that is sustainable. There is no explicit modelling of how
an exchange rate change is to be achieved (under floating rates). A limita-
tion is that it does not attempt specific modelling of how the correspon- 4. POLICY CHOICES
ding change in absorption is composed (higher private saving, lower
public dissaving, and/or less investment). The authors of this brief draw the
following policy implications:
The second approach is to shock a macroeconometric model in such a
way that it generates a targeted change in the current account over a cer- • With the US economy currently
tain horizon. In principle, a strength of this approach is that it takes into operating close to full employ-
account feedback effects and explicitly incorporates the monetary and ment, adjustment requires a
fiscal policies needed to generate a desired shock. An important limita- rate of growth in US domestic
tion is the underlying return-to-equilibrium structure of the model, which demand below that of output 14
The time series of
over coming years to prevent data on Chinese exports
typically assumes the economy begins in equilibrium and therefore after and imports is relati-
a shock returns toward the same starting point over time from feedback inflationary excess demand. A vely short. In addition,
effects. However, the normal policy-feedback rules (such as a “Taylor prime candidate to facilitate the enormous structu-
rule” for monetary policy) that characterize such models are inappro- this adjustment is fiscal ral changes that the
priate when the economy starts in a disequilibrium requiring a policy contraction to offset the increa- Chinese economy has
sing contribution to growth undergone over the
change. This class of models also provides considerably less transpa- past decade complicate
rency than the partial equilibrium models in attributing the calculated from rising U.S. net exports. A econometric estimates
changes in outcomes to specific changes in the model inputs. rebalancing of world demand of China’s trade elastici-
between the United States and ties. Generally, the
East Asia is indispensable. more sensitive to
The third approach is econometric modelling of the influences that are exchange rate move-
found internationally to be associated with strong or weak real exchange ments that China’s
rates, including such variables as net foreign assets and productivity • Japan and China hold the key trade is estimated to
growth. In this approach, the coefficients estimated from international to the adjustment in Asia. be, the less apprecia-
experience are applied to the country in question to determine whether Unless both Japanese and tion of the renminbi is
Chinese policymakers accept needed.
its exchange rate is overvalued or undervalued, and by how much. A
major limitation of this approach is that it must assume not only that the the appreciation of their cur- 15
For four East Asian
period observed is one in which on average the countries in the sample rencies, it is difficult to see economies (Hong Kong,
are at equilibrium exchange rates, but also that the overall coefficients how the adjustment process Malaysia, Singapore,
can start in Asia as other Asian and Taiwan) and four
estimated for the panel of countries apply to the country of direct interest European economies
(e.g., the United States). A related limitation is that there is no explicit economies would in turn resist (Norway, Sweden,
attention to erosion or improvement of a country’s relative position over the appreciation of their cur- Switzerland, and
time from factors not directly in the model, such as the secular adverse rencies. Moreover, Japan has Russia) with large cur-
shift for the United States implied by the findings by Martin Baily. There been at the forefront of promo- rent account surpluses,
ting monetary integration in the combined weight in
are difficulties in measurement, such as the use of consumer to producer the Federal Reserve’s
price ratios to proxy productivity - thereby potentially placing what Asia. A Japan that is commit- broad real exchange
amounts to the real exchange rate on both sides of the equation. There ted to cooperation on rate index for the dollar
are also difficulties of policy interpretation, such as the result in the IMF exchange rate policy in Asia amounts to 13.2 per-
should take the lead in the cent, higher than that
model indicating that a larger fiscal deficit is associated with a stronger of either China (11.3
real exchange rate, even though running a larger fiscal deficit is not a sus- region on exchange rate percent) or Japan
tainable means of achieving equilibrium. adjustment against the dollar. (10.5 percent).
 GLOBAL IMBALANCES: TIME FOR ACTION

• Now that the yen has begun to by further appreciation, with maintain the expansion of
08 strengthen, it is important that
Japan should not intervene to
the aim, over a horizon of per-
haps three to four years, of
domestic spending.
bruegelpolicybrief

bail out speculators that enga- In conclusion, policymakers

ged in the yen carry trade when
the yen was very weak.
Moreover, it is reasonable to
expect that the yen will appre- • In Europe, policymakers should nal financial markets recently
ciate further in the near term
as the Bank of Japan continues
its moves towards normalising
monetary policy. If not, there
would be a case for interven-
tion in foreign exchange mar-
kets to push the yen higher. In
this regard, we note that on a
real effective basis the yen is
currently at its lowest level
since 1986 and that it stands
about 20 percent below its ave-
rage over the 20-year period
since then. By comparison,
when there was coordinated
intervention to boost the euro
in late 2000, the real effective
euro was 19 percent below its
20-year average16.
• In China, priority should be
given to further eli-
minating distor-
tions favouring ‘Japan and China hold tries, evidence pre- of the IMF will provide a crucial
exporting sectors
and gearing the key to adjustment
macro econom i c in Asia.’
policy towards pro-
moting domestic
demand. Given the substantial
real appreciation of the ren-
minbi that is required, a step
revaluation of the renminbi of,
say, 10 percent in the near
term would seem appropriate.
This move should be followed
fully eliminating intervention
should not wait until financial mar-
designed to prevent apprecia-
kets force global adjustment. The
tion of the renminbi.
heightened volatility in internatio-
not resist appre- underscores the
ciation of the euro risks of an abrupt
vis-à-vis the dollar ‘In Europe, policyma- market-led adjust-
so long as it hap- ment if they fail to
pens in the context kers should not resist act. It is unlikely
of global adjust- appreciation of the that the policyma-
ment and does not kers of each coun-
imply effective euro against the dollar try will resolve
euro appreciation. so long as it happens independently to
Otherwise if the take actions that
Asian currencies in the context of global add up to a cohe-
were to appreciate adjustment.’ rent package.
against the dollar, There needs to be
then the real effec- an international
tive exchange rate of the euro effort to persuade each country to
would depreciate. If Europe is contribute its fair share to a whole
not to run a current account capable of bringing about adjust-
surplus, then the euro will have ment without interrupting world
to strengthen vis-à-vis the dol- growth. In principle, the IMF’s mul-
lar. tilateral surveillance exercise pro-
vides an ideal context for organi-
• For the oil- sing such an international effort.
exporting coun- The forthcoming spring meetings
sented at the opportunity to assess progress
workshop by Brad made so far and move in earnest
Setser suggested towards reaching agreement on an
that significant adjustment package along the
adjustment is in lines sketched in this policy brief.
the pipeline, with domestic
absorption rising at a gradual
pace. Therefore, it is not clear
that further actions on their
part are called for (assuming
that the oil price does not
increase again), other than to

Bruegel is a European think tank devoted to international economics, which started operations in Brussels in 2005.
It is supported by European governments and international corporations. Bruegel’s aim is to contribute to the quality
of economic policymaking in Europe through open, fact-based and policy-relevant research, analysis and discussion.
© Bruegel 2007. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted in the original
language without explicit permission provided that the source is acknowledged. The Bruegel Policy Brief Series is publi-
shed under the editorial responsibility of Jean Pisani-Ferry, Director. Opinions expressed in this publication are those of
16
Pre-1999 calculation the author(s) alone.
uses the value of a syn-
thetic euro based on
the value of its legacy Visit www.bruegel.org for information on Bruegel's activities and publications.
currencies. Bruegel - Rue de la Charité 33, B-1210 Brussels - phone (+32) 2 227 4210 info@bruegel.org