Informe Del FMI
Informe Del FMI
Informe Del FMI
EXTERNAL
SECTOR
REPORT
External Rebalancing in
Turbulent Times
2023
INTERNATIONAL MONETARY FUND
EXTERNAL
SECTOR
REPORT
External Rebalancing in
Turbulent Times
2023
©2023 International Monetary Fund
Cataloging-in-Publication Data
IMF Library
The External Sector Report (ESR) is a survey by the IMF staff published once a year, in the summer.
The ESR is prepared by the IMF staff and has benefited from comments and suggestions by Executive
Directors following their discussion of the report on July 13, 2023. The views expressed in this pub-
lication are those of the IMF staff and do not necessarily represent the views of the IMF’s Executive
Directors or their national authorities.
Recommended citation: International Monetary Fund. 2023. External Sector Report: External Rebalancing
in Turbulent Times. Washington, DC, July.
Preface viii
Executive Summary ix
Figures
Figure 1.1. Global Current Account Balances, REER, and Capital Flows, 1990–2022 2
Figure 1.2. The COVID-19 Crisis and the War in Ukraine 2
Figure 1.3. Movements in Oil Trade Balance and Current Account for Oil Exporters and Importers 3
Figure 1.4. Trade in Goods and Services Compared with Pre-pandemic Trends 3
Figure 1.5. Global Current Account Balances, with the Contributions from Cyclical and
COVID-19 Factors 4
Figure 1.6. Average Fiscal Policy Changes, 2021–22 4
Figure 1.7. Current Account Decomposition 5
Figure 1.8. Currency Movements 5
Figure 1.9. Exchange Market Pressure and Its Components 6
Figure 1.10. Exchange Market Pressure and Inflation, 2022 7
Figure 1.11. Cross-Currency Basis Swap against US Dollar 7
Figure 1.12. Capital Flows to Emerging Market and Developing Economies 8
Figure 1.13. Incidence of Extreme Capital Flows: Number of Surges, Stops, Flights, and
Retrenchments 8
Figure 1.14. Cumulative High-Frequency Portfolio Flows to EMDEs January 2022–April 2023 9
Figure 1.15. Net International Investment Positions, 1990–2022 10
Figure 1.16. Sum of Cross-Border Assets and Liabilities 10
Figure 1.17. International Investment Position Valuation Change and Net International
Investment Position, 2022 11
Figure 1.18. The Evolution of Global Financial Safety Net, 1995–2021 11
Figure 1.19. External Balance Assessment Current Account Norms, 2022 12
Figure 1.20. Evolution of External Sector Assessments, 2012–22 13
Figure 1.21. Evolution of Headline Current Account Balances and IMF Staff Gaps 13
Figure 1.22. IMF Staff Current Account and Real Effective Exchange Rate Gaps 14
Figure 1.23. Change in 2023 Current Account Balances 17
Figure 1.24. Number of Trade Restrictions, 2009–22 18
Figure 1.25 Exchange Market Pressure and REER Gaps 19
Figure 1.1.1. China: Current Account Surplus and Capital Flows, Four-Quarter Trailing Sums 21
Figure 1.1.2. Saudi Arabia: Current Account Surplus and Capital Flows, Four-Quarter
Trailing Sums 21
Figure 1.1.3. Russia: Current Account Surplus and Capital Flows, Four-Quarter Trailing Sums 22
Figure 1.1.4. Bank for International Settlements Reporting Banks’ Cross-Border Positions on
Residents of Russia 22
Figure 1.1.5. United States: Current Account Surplus and Capital Flows, Four-Quarter
Trailing Sums 22
Figure 1.1.6. Net Foreign Purchases of US Securities 23
Figure 1.1.7. US Incurrence of Portfolio Investment Liabilities 23
Figure 1.1.8. Net Foreign Purchases of US Securities 24
Figure 1.1.9. Total Foreign Holdings of US Treasury Securities 25
Figure 1.1.10. Currency Composition of Official Foreign Exchange Reserves 25
Figure 1.1.11. China: Incurrence of Portfolio Investment Assets 25
Figure 1.2.1. Cumulative Distribution of Foreign Currency Exposures 26
Figure 1.3.1. Trade Openness and Current Account Balances since 1870 27
Figure 1.3.2. Effect of a Labor Productivity Shock on a Representative EMDE:
Baseline and Higher Trade Cost 28
Figure 2.1. Nominal US Dollar Trade-Weighted Index against Major Advanced Economies 41
Figure 2.2. The US Dollar Index and the Global Dollar Cycle 42
Figure 2.3. Spillovers from a US Dollar Appreciation: Macro Aggregates 45
Figure 2.4. Spillovers from a US Dollar Appreciation: External Sector Variables 46
Figure 2.5. Spillovers from a US Dollar Appreciation: Financial Variables 47
Figure 2.6. Country Characteristics 49
Tables
Table 1.1. Selected Economies: Current Account Balance, 2020–23 15
Table 1.2. Selected Economies: Net International Investment Position, 2019–22 16
Annex Table 1.1.1. Selected Economies: Foreign Reserves, 2019–22 29
Annex Table 1.1.2. External Sector Report Economies: Summary of External Assessment
Indicators, 2022 30
Annex Table 1.1.3. External Sector Report Economies: Summary of IMF Staff–Assessed Current
Account Gaps and IMF Staff Adjustments, 2022 31
Annex Table 1.1.4. External Sector Report Economies: Summary of IMF Staff–Assessed Real
Effective Exchange Rate and External Balance Assessment Model Gaps, 2022 32
Annex Table 1.1.5. Selected External Sector Report Economies: External Balance Assessment
Current Account Regression Policy Gap Contributions, 2022 33
Annex Table 1.1.6. 2022 Individual Economy Assessments: Summary of Policy Recommendations 34
Table 2.1. Correlates of the Global Dollar Cycle 43
Table 2.2. Categorization of Countries by Policy Regimes and Structural Characteristics 48
Table 3.A. Description in External Sector Report Overall Assessment 62
Table 3.B. Economies Covered in the External Sector Report 62
Table 3.1. Argentina: Economy Assessment 64
Table 3.2. Australia: Economy Assessment 65
Table 3.3. Belgium: Economy Assessment 66
Table 3.4. Brazil: Economy Assessment 67
Table 3.5. Canada: Economy Assessment 68
Table 3.6. China: Economy Assessment 69
Table 3.7. Euro Area: Economy Assessment 70
Table 3.8. France: Economy Assessment 71
Table 3.9. Germany: Economy Assessment 72
Table 3.10. Hong Kong Special Administrative Region: Economy Assessment 73
Table 3.11. India: Economy Assessment 74
Table 3.12. Indonesia: Economy Assessment 75
Table 3.13. Italy: Economy Assessment 76
Table 3.14. Japan: Economy Assessment 77
Table 3.15. Korea: Economy Assessment 78
Table 3.16. Malaysia: Economy Assessment 79
Table 3.17. Mexico: Economy Assessment 80
Table 3.18. The Netherlands: Economy Assessment 81
Table 3.19. Poland: Economy Assessment 82
Table 3.20. Russia: Economy Assessment 83
Table 3.21. Saudi Arabia: Economy Assessment 84
Table 3.22. Singapore: Economy Assessment 85
Print
Print copies of this External Sector Report can be ordered from the IMF Bookstore at http://IMF.org/external/
terms.htm
Digital
Multiple digital editions of the External Sector Report, including ePub, enhanced PDF, Mobi, and HTML, are
available on the IMF eLibrary at www.elibrary.IMF.org/ESR23
Download a free PDF of the report and data sets for each of the charts therein from the IMF website at
imf.org/en/Publications/ESR or scan the QR code below to access the External Sector Report web page directly:
Produced since 2012, the IMF’s annual External Sector Report analyzes global external developments and pro-
vides multilaterally consistent assessments of external positions of the world’s largest economies representing more
than 90 percent of global GDP, which include current accounts, real exchange rates, external balance sheets, capital
flows, and international reserves. Together with the World Economic Outlook and Article IV consultations, this
report is part of a continuous effort to assess and address the possible effects of spillovers from members’ policies
on global stability and to monitor the stability of members’ external positions in a comprehensive manner.
Chapter 1, “External Positions and Policies,” discusses the evolution of global external positions in 2022, exter-
nal developments through the COVID-19 pandemic and Russia’s invasion of Ukraine, and policy priorities for
reducing excess imbalances over the medium term. Chapter 2, “External Sector Implications of the Global Dollar
Cycle,” analyzes cross-border spillovers from US dollar appreciations. It finds large negative spillovers on emerging
markets, accompanied by increased current account balances. More flexible exchange rates and better anchored
inflation expectations can mitigate the negative spillovers. Chapter 3, “2022 Individual Economy Assessments,”
provides details on various aspects of the overall external assessment and associated policy recommendations for
30 economies. This year’s report and associated external assessments are based on the latest version of the IMF’s
External Balance Assessment methodology, external sector data as of May 31, 2023, and IMF staff projections in
the April 2023 World Economic Outlook.
This report was prepared under the overall guidance of Pierre-Olivier Gourinchas, IMF Economic Counsellor
and Director of Research, and under the direction of the External Sector Coordinating Group, comprising staff
from the IMF’s area departments (African, Asia and Pacific, European, Middle East and Central Asia, and Western
Hemisphere) and several functional departments (Fiscal Affairs; Statistics; Strategy, Policy, and Review; Monetary
and Capital Markets; and Research): Ali Jawad Al-Eyd, Rudolfs Bems, Maria Borga, Emine Boz, Nigel Chalk,
Jiaqian Chen, Mariana Colacelli, Borys Cotto, Christopher Erceg, Kevin Fletcher, Kenneth Henry Kang, Purva
Khera, Nir Klein, Vitaliy Kramarenko, Jaewoo Lee (Chair), Amine Mati, Paulo Medas, Paolo Mauro, Papa M.
Bagnick N’Diaye, Marcos Poplawski Ribeiro, Lev Ratnovski, Umang Rawat, Christian Saborowski, Rani Salgado,
Mika Saito, Carlos Sánchez-Muñoz, Martin Sommer, and Anita Tuladhar.
Rudolfs Bems, Jiaqian Chen, and Giovanni Ganelli led the preparation of the report, which draws on contributions
from Cian Allen, Lukas Boer, Camila Casas, Allan Dizioli, Luciana Juvenal, Racha Moussa, Cyril Rebillard, and
Robert Zymek. Important input was provided by country teams as well as David Florián Hoyle, Adam Jakubik, Parisa
Kamali, Robin Koepke, Svitlana Maslova, Dimitre Milkov, Faezeh Raei, Marco Rodriguez Waldo, Silvia Sgherri, and
Hui Tong. Excellent research and editorial assistance were provided by Mariela Caycho Arce, Anna Konopatskaya,
Jane Haizel, Abreshmi Nowar, Jair Rodriguez, Xiaohan Shao, and Emelie Stewart.
Cheryl Toksoz from the Communications Department led the editorial team for the report, with production
and editorial support from Absolute Services and the Grauel Group.
The analysis benefited from comments and suggestions by staff members from other IMF departments, as well
as by the IMF’s Executive Directors following their discussion of the report on July 13, 2023. However, both pro-
jections and policy considerations are those of the IMF staff and should not be attributed to Executive Directors or
to their national authorities.
G
lobal current account balances (defined Over the medium term, global current account
as the sum of absolute values of current balances are expected to narrow as the impacts of the
account deficits and surpluses) increased pandemic and Russia’s war in Ukraine recede. However,
for the third consecutive year in 2022 and several risks surround this outlook, including a renewed
are projected to narrow in 2023. This widening over increase in commodity prices, a slower-than-expected
the three years reflects several factors, including the recovery in China, or a slower fiscal consolidation in
unequal impact of the COVID-19 crisis in 2020–21 economies with current account deficits. While the
and the increase in commodity prices fueled by the impact of geoeconomic fragmentation on global current
economic recovery in 2021 and later by supply con- account balances is unclear, it would unambiguously
cerns following Russia’s invasion of Ukraine in 2022. reduce global welfare.
The absence of widespread sudden stops during the The excess global current account balances (defined
pandemic has enabled deficit economies to avoid an as the sum of absolute values of current account
abrupt contraction of their current account deficits. surpluses and deficits in excess of their norms) have
Currency markets exhibited significant fluctuations remained unchanged since 2021, after being on a
in 2022, driven by changes in the terms of trade declining trend for several years. While the widening
and monetary tightening. The US dollar appreciated of global current account balances is not necessarily a
by about 8 percent in real effective terms, reaching negative development, excess global current account
its strongest level since 2002. Emerging market and balances can fuel trade tensions and protectionist
developing economies with preexisting vulnerabili- measures or increase the risk of disruptive currency
ties such as high inflation and misaligned external and capital flow movements. Narrowing excess global
positions experienced greater depreciation pressures, current account balances would reduce the risk of
while commodity-exporting economies benefited financial crisis and improve welfare.
from the increase in commodity prices. Historically, Policy efforts, in both excess surplus and deficit
US dollar appreciations have had large negative economies, are required to promote external rebalanc-
cross-border spillovers, disproportionately affect- ing. Where excess current account deficits in 2022
ing emerging markets, and have increased current partly reflected larger-than-desired fiscal deficits, fiscal
account balances, as the investment rate falls (see consolidation will help stabilize debt-to-GDP ratios and
Chapter 2). close current account gaps. In economies where excess
The uphill capital flows from emerging market current account surpluses persist, higher fiscal spend-
and developing economies to advanced economies ing in targeted areas will help them to meet their goals
reemerged in 2022. The net flows of capital from in climate, digital, and energy security, while reducing
emerging market and developing economies were their excess surpluses. Economies with lingering com-
mostly driven by China and commodity-exporting petitiveness challenges will need to address structural
economies, which have funded some large current bottlenecks. Multilateral cooperation will help counter
account deficits in advanced economies. In contrast risks of geoeconomic fragmentation, including efforts
to past episodes, however, an accumulation of official to strengthen the current rule-based trading system, and
foreign exchange reserves played a limited role in net facilitate the green transition. Successfully completing
capital outflows from emerging market and developing the 16th General Review of Quota would ensure that
economies. Meanwhile, net creditor and debtor posi- the IMF is adequately resourced to serve as an anchor of
tions remained at historically high levels. the global financial safety net.
The following remarks were made by the Acting Chair at the conclusion of the Executive Board’s discussion
of the External Sector Report on July 13, 2023.
E
xecutive Directors broadly agreed with the Directors observed that capital has moved from
findings of the 2023 External Sector Report emerging market and developing economies to
(ESR) and its policy recommendations. They advanced economies in 2022, in the context of
noted that global current account balances increased risk aversion triggered by the war in Ukraine
widened for the third consecutive year in 2022. Direc- and tighter monetary policy in advanced economies.
tors observed that the trend decline in excess current They noted that the net flows of capital from emerg-
account balances had stalled and that currency markets ing market and developing economies, mostly driven
exhibited significant fluctuations, driven by changes in by China and commodity-exporting economies, have
the terms of trade and monetary tightening. Concur- funded large current account deficits in some advanced
rently, stocks of foreign assets and liabilities remained economies.
at historically high levels in 2022. Directors noted that global current account bal-
Directors observed that elevated commodity prices, ances are expected to narrow over the medium term as
amid supply concerns following Russia’s invasion of the impact of pandemic and Russia’s war in Ukraine
Ukraine, significantly contributed to the widening of recede. They cautioned that significant risks surround
global balances in 2022. They noted that the pandemic this outlook, including tightening global financial
has continued to unevenly affect current account bal- conditions, renewed increase in commodity prices, or
ances, albeit to a lesser extent relative to prior years, slower fiscal consolidation in economies with current
as travel services remained subdued and high trans- account deficits.
portation costs persisted in some economies. Directors Directors reiterated that excess global current
observed that the US dollar appreciated substantially, account balances, which remained unchanged since
reflecting a rapid tightening of monetary policy and 2021, can fuel trade tensions and protectionist mea-
more favorable terms of trade. They noted that emerg- sures or increase the risk of disruptive currency and
ing market and developing economies with preexisting capital flow movements. They consequently encour-
vulnerabilities experienced greater depreciation pres- aged both excess surplus and deficit economies to take
sures, while commodity exporting economies benefited steps to promote external rebalancing.
from the higher commodity prices. Directors underscored that cooperation is key to
Directors generally welcomed the analysis of the addressing the complex challenges facing the global
external sector implications of the global dollar cycle. economy and preserving the benefits of global integra-
They highlighted that, in contrast to the historical tion and multilateralism. Noting that geoeconomic
trend, the recent strong dollar episode was accompa- fragmentation would unambiguously reduce global
nied by surging commodity prices. Directors noted welfare, Directors highlighted that coordinated policy
that US dollar appreciations have increased current efforts will be needed to counter the related risks,
account balances and have had large negative cross- including by strengthening the current rules-based
border spillovers, disproportionally affecting emerging trading system. Moreover, Directors noted that while
markets. They underscored that more flexible exchange industrial policy could be pursued to address well-
rates and more anchored inflation expectations can established market failures, they should not introduce
mitigate negative spillovers to emerging market distortions and should be consistent with interna-
economies. tional agreements and WTO rules. Directors stressed
that ensuring an adequate global financial safety net, and inclusive growth while boosting productivity. In
with the Fund at its core, remains critical at a time of economies where excess current account surpluses per-
heightened vulnerabilities in emerging markets with sist, prioritizing reforms that encourage investment and
high external liabilities. In this regard, Directors under- discourage excessive private saving, while also pursuing
scored the importance of successfully completing the domestic objectives, is warranted.
16th General Review of Quotas. Directors reiterated the need to ensure transpar-
Directors underscored that policies to promote ency, consistency, and evenhandedness of external
external rebalancing differ with positions and needs assessments across countries. They stressed the
of individual economies. They considered that in importance of continued caution in interpreting and
economies in which excess current account deficits communicating the assessment results. Directors
reflect excessive fiscal deficits, fiscal consolidation that encouraged further exploration of possible improve-
preserves space for critical infrastructure and well- ments to enhance the EBA methodologies, given
targeted social spending will be critical to supporting model limitations, and continued efforts to ensure
external rebalancing. Directors stressed that economies consistency across work streams. They called for
with lingering competitiveness challenges will need to greater analysis of vulnerabilities associated with large
address structural challenges to promote green, digital, external stock positions.
Recovery, War, and Policy Shocks balances—those beyond what can be explained by
Global current account balances widened further medium-term fundamentals and desirable policies—
in a third consecutive year in 2022 (Figure 1.1). One albeit gradually and over the medium term (see
prominent contributor to the widening in 2022 was 2022 External Sector Report, Box 1.2). However,
Russia’s invasion of Ukraine, which elevated com- there is a high degree of uncertainty surrounding
modity prices amid supply concerns. The uneven this outlook. Risks include a renewed increase in
recovery from the COVID-19 pandemic—across commodity prices and a slower-than-expected pace
countries and sectors—and the rapid tightening of of China’s recovery or of fiscal consolidation in
US monetary policy also contributed to the widen- economies with current account deficits. In addi-
ing of global balances, offsetting the impact from tion, a severe tightening of global financial con-
unwinding of pandemic-induced fiscal measures. ditions could trigger broad-based capital outflows
Concurrently, the US dollar appreciated substan- from vulnerable EMDEs, and further geoeconomic
tially, and the uphill capital flow—capital flowing fragmentation could potentially lead to large welfare
from faster-growing emerging market and developing losses, including through its effects on trade barriers
economies (EMDEs) to slower-growing advanced and foreign direct investment.
economies—reappeared.
China’s reopening and the US banking sector tur- Recent Developments in Current
moil were the new forces that could have important Account Balances
implications on global balances in early 2023. The
reopening of the Chinese economy led to a tempo- Elevated Commodity Prices and the War in Ukraine
rary rebound in exports in the first quarter of 2023 Commodity prices increased in 2022, enlarging
as supply chain conditions improved, contributing to the differences in current account balances between
a widening of global trade balances. The unexpected commodity importers and exporters (Figures 1.2 and
failures of two large regional banks in the United 1.3). In the aftermath of Russia’s invasion of Ukraine,
States and a systemically important global bank in commodity prices soared amid concerns about a
Europe have had limited impact on cross border shortfall in global supplies from Russia and Ukraine
capital flows and currency volatility so far, owing to and trade disruptions caused by the war itself. Oil
forceful policy actions undertaken to reassure markets prices then started falling from their peak in mid-
and shore up the banking sector. However, as bank- 2022, as demand growth from major economies,
ing sector turmoil has tightened credit conditions and such as China, slowed and trade diversion enabled
curtailed lending, market participants now expect a steady supply of Russian crude oil to the global
a shallower monetary policy path in the United market. European gas prices had risen to a strato-
States, which has provided some support to EMDE spheric level amid supply disruptions but declined,
currencies. owing to substitution efforts and an exceptionally
The widening of global current account balances mild winter that reduced demand. Food prices also
is expected to reverse in 2023, as the impacts of began to fall around the same period as supply and
the pandemic and Russia’s war in Ukraine recede. demand reacted to higher prices, including through
Policy actions will also help narrow excess global the reopening of the Black Sea corridor, increased
wheat production in Europe and India, and lower
The authors of this chapter are Cian Allen, Lukas Boer, Camila demand for price-elastic items. Despite the decline
Casas, Jiaqian Chen (co-lead), Giovanni Ganelli (co-lead), Luciana since mid-year, average commodity prices in 2022
Juvenal, and Cyril Rebillard, in collaboration with Robert Zymek,
under the guidance of Jaewoo Lee. Jair Rodriguez, Xiaohan Shao, were higher than those in 2021 and well above their
and Abreshmi Nowar provided excellent research support. pre-pandemic levels.
Figure 1.1. Global Current Account Balances, REER, and Figure 1.2. The COVID-19 Crisis and the War in Ukraine
Capital Flows, 1990–2022
1. International Cereal Prices 2. Oil Inventory Built and Price
1. Global Current Account Balance1 (Index, Jan. 2020 = 100) Oil price, US dollars
(Percent of world GDP) Corn Sorghum a barrel
8 Wheat Oil inventory built,
Deficits
260 140 millions of barrel 6
7 Surpluses
per day (right scale)
Global current account balance 120 5
6
210 100 4
5
3
4 80
160 2
3 60
1
2
110 40
0
1
20 –1
0
60 0
1990
94
98
2002
06
14
18
10
22
–2
Jan. Jan. Jan. Apr. Jan. Jan. Jan. Jan. Apr.
2020 21 22 23 2019 20 21 22 23
2. United States: Real Effective Exchange Rate
(Index, 1990 = 100) 3. Economic Activity and 4. Supply Chain
120 Commodity Prices (Index, Jan. 2017 = 100)
(Index, 2019 average = 100) Harper Petersen
115
All metals price index Charter Rates Index
110 Energy prices Baltic Dry Index
Industrial production Pressure Index
105 300 110 2,500 5
(right scale) (std. dev. pts.,
106 right scale) 4
100 254 2,000
95 101 3
208
97 1,500 2
90
163
85 93 1,000 1
117
1990
94
98
2002
06
14
18
10
22
89 0
71 500
84 –1
3. Emerging Market and Developing Economies: Net Financial Inflows
(Percent of world GDP) 25 80 0 –2
Jan. Jan. Jan. Jan. Apr. Jan. Jan. Jan. Apr.
1.5 2019 20 21 22 23 2017 19 21 23
1.0
0.5
Sources: CEIC Global Economic Data; Haver Analytics; IMF, Primary Commodity
0.0
Price System; Joint Organisations Data Initiative; and US Energy Information
–0.5 Administration.
–1.0 Note: In panel 2, oil inventory built is calculated as the six-month moving average
–1.5 of total world petroleum production minus total world petroleum consumption, and
–2.0 oil price refers to crude oil (petroleum), West Texas Intermediate 40 American
–2.5 Nonreserve flows Petroleum Institute (API), in US dollars a barrel.
–3.0 Reserve flows
–3.5 Total net inflows (including reserves)
–4.0
1990
94
98
2002
06
14
18
10
22
Figure 1.3. Movements in Oil Trade Balance and Current Figure 1.4. Trade in Goods and Services Compared with
Account for Oil Exporters and Importers Pre-Pandemic Trends
(Average of balances in percent of GDP)
1. Global Goods Imports 2. Global Services Imports
15 (Index, 2019 = 100) (Index, 2019 = 100)
Oil trade balance of importers
130 130
Oil trade balance of exporters Jan. 2020 WEO
Current account of importers 125 Jun. 2020 WEO 125
Current account of exporters Jan. 2022 WEO
120 Apr. 2023 WEO 120
10
115 115
110 110
5 105 105
100 100
95 95
0
90 90
Jan. 2020 WEO
85 Jun. 2020 WEO 85
Jan. 2022 WEO
80 80
–5 Apr. 2023 WEO
2011 12 13 14 15 16 17 18 19 20 21 22 75 75
2019 20 21 22 23 24 25 2019 20 21 22 23 24 25
Sources: IMF, April 2023 World Economic Outlook; and IMF staff calculations.
Note: Countries are defined as exporters or importers by their oil trade balance in Sources: IMF, April 2023 World Economic Outlook (WEO); and IMF staff
2021. Figure includes External Balance Assessment countries: Hong Kong SAR, calculations.
Saudi Arabia, and Singapore. Importer countries are Argentina, Australia, Austria, Note: Global imports are in volumes.
Belgium, Chile, China, Costa Rica, Czech Republic, Denmark, Egypt, Finland,
France, Germany, Greece, Guatemala, Hong Kong SAR, Hungary, India, Indonesia,
Ireland, Israel, Italy, Japan, Korea, Mexico, Morocco, The Netherlands, New
Zealand, Pakistan, Peru, Philippines, Poland, Portugal, Singapore, South Africa,
Spain, Sri Lanka, Sweden, Switzerland, Thailand, Tunisia, Türkiye, the United Contribution of Cyclical Factors
Kingdom, the United States, and Uruguay. Exporter countries are Brazil, Canada,
Colombia, Malaysia, Norway, Russia, and Saudi Arabia.
Cyclical factors played a more important role in the
widening of the global balances in 2022 compared
with previous years (Figure 1.5). The contribution
of cyclical factors to the global balances reflected the
growth rate, though remaining below its pre-pandemic
(temporary part of ) elevated commodity prices, which
level (Figure 1.4).1
pushed the terms of trade for commodity-exporting
Nonetheless, the emergence of especially contagious,
and -importing countries in opposite directions. It also
but less lethal, COVID-19 variants continued to mate-
reflects the impact from output gaps as economies were
rially affect some economies’ external balances in 2022
in different phases of recovery: weak domestic demand
(Figure 1.5). The resulting travel shock is estimated to
led to a stronger current account balance, via factors
have materially lowered the travel services and current
including lower investment, and vice versa for econo-
account balances of a few tourism-exporting countries
mies with stronger domestic demand.
such as Thailand. While shipping costs abated in the
second half of 2022, the yearly average remained high
compared with the historical average (Figure 1.2, panel Policy Actions
4). As a result, they continued to increase the current
Fiscal policies in 2022 likely moderated the increase
account balances of economies with large presences of
in global current account balances. On average,
shipping companies (for example, France).
economies with current account deficits consolidated
their fiscal policies in 2022 relative to 2021, while
economies with current account surpluses loosened
1Given those developments in pandemic-related factors, the
their stances (Figure 1.6). Among deficit countries,
medical and consumption shift adjustors have been discontinued for
2022, while the transportation and travel adjustors have continued Canada, Türkiye, the United Kingdom, and the
to be applied in the 2023 External Sector Report. United States reduced their (cyclically adjusted) fiscal
Figure 1.5. Global Current Account Balances, with the Figure 1.6. Average Fiscal Policy Changes, 2021–22
Contributions from Cyclical and COVID-19 Factors (Percentage points)
(Percent of world GDP)
Current account deficit economies
4.5 3.0 Current account surplus economies
4.0
2.5
3.5
2.0
3.0
2.5 1.5
0.0 –0.5
2015 16 17 18 19 20 21 22 23 24 25
Sources: IMF, April 2023 World Economic Outlook; and IMF staff calculations.
Sources: CEIC Data, Global Database; IMF, Primary Commodity Price System; Note: Figure shows the GDP-weighted average change in fiscal stance, measured
Refinitiv Datastream; UN, Comtrade; UN Conference on Trade and Development; as cyclically adjusted general government overall balance as percent of potential
and IMF staff calculations. GDP. An increase (decrease) denotes tighter (looser) fiscal policy relative to 2021.
Note: Global current account balance is the sum of absolute values of current Economies are grouped according to current account balances in 2021.
account; COVID-19 factors are the sum of absolute values of transportation and
travel COVID-19 adjustors for External Sector Report countries only; and cyclical
factors are the sum of absolute values of the contribution of cyclical factors to
current accounts of External Sector Report countries only. Data from 2023 onward
are projections, based on the April 2023 World Economic Outlook.
In the early months of 2023, trade data suggest that
global trade balances widened compared with their
levels at the end of 2022, driven by the reopening of
deficits; among surplus economies, China, Japan, China offsetting the impact from falling commodity
Korea, and The Netherlands increased theirs. How- prices. China’s exports temporarily improved in the
ever, the strengthening of the US dollar widened the first quarter of the year against the backdrop of relaxed
US current account deficit. testing and quarantine requirements and normalization
Government and household saving in advanced of supply chains; imports also increased from the pre-
economies moved in opposite directions, while vious quarter, but less than exports, reflecting subdued
corporate saving remained above pre-pandemic levels imports of intermediate goods amid growth led by
(Figure 1.7). Despite the budgetary support deployed private consumption that is less import intensive. The
(about 1.3 percent of GDP in the case of the Euro- improvement in China’s trade surplus has so far more
pean Union) to help households and firms weather than offset the narrowing of the surplus in commod-
the energy crisis, public sector saving improved in ity-exporting economies, but China’s trade surplus
2022 relative to 2021 in many economies, mostly is expected to shrink with a significant anticipated
reflecting the unwinding of temporary support pickup in tourism travel in the remainder of 2023.
measures deployed during the pandemic. Against
this background, household saving declined, nota-
Currencies, Financial Flows, and Balance Sheets
bly in the United States, where the saving rate fell
below pre-pandemic levels. On the other hand, since Exchange Rates
mid-2020, corporate saving has remained high in the In the past year and a half, the currency market
United States and several other advanced economies has experienced significant fluctuations (Figure 1.8,
compared with pre-pandemic levels. panel 1). The US dollar, in real effective terms, was
MEX
SAU
SGP
CHE
HKG
IDN
POL
MYS
AUS
CHN
GBR
IND
SWE
KOR
CAN
JPN
ZAF
TUR
RUS
ARG
BRA
USA
EA
THA
–8
–10
2019 20 21 22
2. US Dollar Real Effective Exchange Rate1
(Index, Jan. 2019 = 100)
2. Euro Area 120
8 With respect to advanced economies
With respect to emerging markets
6 115
4
2 110
0
105
–2
–4 100
–6
95
–8
2019:M01
19:M04
19:M07
19:M10
20:M01
20:M04
20:M07
20:M10
21:M01
21:M04
21:M07
21:M10
22:M01
22:M04
22:M07
22:M10
23:M01
23:M04
2019 20 21 22
–5
the US dollar appreciated more with respect to advanced negative link between the US dollar and commodity prices.
However, the 2021–22 US dollar appreciation coincided with a
economy currencies, on average, than with respect to significant upswing in commodity prices, linked to recovery from the
EMDE currencies (Figure 1.8, panel 2), in part due COVID-19 pandemic and Russia’s war in Ukraine.
6 percent on a real trade-weighted basis, reflecting a Figure 1.9. Exchange Market Pressure and Its Components
change in expectations of US monetary policy and
Exchange Market Pressure Index
improved risk sentiment. Despite this, the dollar (Percent change)
remains stronger than it has been since 2000. URY
•• By contrast, as of April 2023, other major curren- RUS FXIs
MEX Interest rate
cies have either remained broadly unchanged (such BRA Nominal exchange rate
as the euro and the pound sterling) or depreciated, HKG Total exchange market pressure
CHE Exchange market pressure in 2021
including the Japanese yen by 15.3 percent and SGP
the renminbi by 7.6 percent, in real effective terms MYS
compared with their 2021 averages. The deprecia- CAN
EA
tions were driven by interest rate differentials, high KOR
energy prices and different speeds of economic ZAF
AUS
recovery. CHN
•• In EMDEs, currency movements have been more THA
POL
heterogeneous. While currencies in some economies, PER
such as Brazil and Mexico, appreciated in nomi- ISR
CHL
nal effective terms in 2022 and early 2023, those NZL
in other economies—including Argentina, South GBR
DNK
Africa, and Türkiye—depreciated significantly. The NOR
monetary tightening in advanced economies has IND
SWE
put depreciation pressure on all EMDE currencies; JPN
however, country-specific factors such as earlier MAR
CZE
monetary tightening (than in advanced economies), TUN
preexisting vulnerabilities (such as lower perceived HUN
ROU
institutional quality), and commodity exposure have COL
led to these different currency movements. The –0.2 –0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Russian ruble appreciated significantly in the second
quarter of 2022 under restrictions on imports and Sources: Adler and others (2021); Goldberg and Krogstrup (2023); IMF,
International Financial Statistics; and IMF staff calculations.
capital outflows, but it has since depreciated against Note: Positive values correspond to exchange market pressure that would
the US dollar, largely owing to weaker terms of depreciate the nominal exchange rate. A country’s total exchange market pressure
in 2022 is the sum of scaled and weighted observed foreign exchange
trade and a sharp increase in parallel imports. interventions (FXIs), short-term interest rate changes, and nominal exchange rate
movements. Values of FXIs and interest rate changes are expressed in terms of
The widespread depreciation pressure of 2022 was counterfactual exchange rate adjustments that would have occurred if no FXI or
policy rate changes had been conducted. FXIs are spot interventions from an
evident in a more comprehensive measure of market updated data set of Adler and others (2021). EA = euro area. Data labels in the
pressure. The realized change in exchange rates may figure use International Organization for Standardization (ISO) country codes.
only be a partial measure of external pressure, as
economies can resort to foreign exchange interven-
tion or interest rate changes to cushion such pres- In 2022, many economies let their currencies adjust
sure. Figure 1.9 plots the Exchange Market Pressure fully (for example, Australia, Sweden), whereas many
Index and its components for 2022, incorporating others undertook foreign exchange intervention
both realized exchange rate movement and policy (for example, Czech Republic, Singapore) or raised
intervention (purchases and sales of foreign exchange the policy rate (for example, Colombia, Romania),
reserves and policy rate changes) by central banks.3 dampening depreciation pressures as a consequence.4
Compared with those in 2021, external pressures in
3The Exchange Market Pressure Index is based on Goldberg and
2022 were much larger, with many economies hiking
Krogstrup (2023). It is defined as the weighted and scaled sums
of exchange rate depreciation, foreign exchange intervention, and interest rates and offsetting depreciation pressures.
policy rate changes. It combines pressures observed in exchange
rate adjustments with model-based estimates of incipient pressures
that are absorbed by foreign exchange interventions and policy rate 4Singapore uses foreign exchange intervention as a monetary
adjustments. instrument.
Figure 1.10. Exchange Market Pressure and Inflation, 2022 Figure 1.11. Cross-Currency Basis Swap against US Dollar
(Percent) (Basis points)
0.8 0
Linear
trend line COL
–20
0.6
–40
Exchange market pressure index
ROU
–60
0.4
TUN
CZE
HUN –80
JPN MAR
0.2 NOR PER THA DNK
IND ISR GBR –100
SWE
CHN NZL
POL
KOR EA CHL –120 Japanese yen Mar. 10
0.0 MYS AUS Swiss franc Silicon Valley
ZAF Euro Bank Collapse
BRA MEX –140
URY CAN British pound sterling
CHE Korean won
–0.2 –160
–10 –5 0 5 10 15 20 Jan. 2022 Apr. 22 July 22 Oct. 22 Jan. 23 Apr. 23
Change in inflation
Source: Bloomberg Finance L.P.
Sources: Goldberg and Krogstrup (2023); and IMF staff calculations.
Note: Figure plots the cumulative Exchange Market Pressure Index for 2022 and
the change in inflation between 2021 and 2022. Russia is excluded. If policy rate
changes are excluded from the Exchange Market Pressure Index, the correlation
goes from 0.6 to 0.5. EA = euro area. Data labels in the figure use International
Organization for Standardization (ISO) country codes. External Sector Report Online Annex 1.2).6 However, in
2022, net capital outflows from EMDEs, and particu-
larly from China, took place not via an accumulation
of official foreign exchange reserves, but via other types
Over the year, countries with larger increases in of flows. Consistent with this pattern, private holdings
inflation tended to experience more external pressure of US assets increased (Box 1.1). This net flow of cap-
(Figure 1.10).5 ital from EMDEs, as a whole, is expected to diminish
The March 2023 turmoil in the banking sector in 2023.
had only limited impact on currency volatility, Turning to subcomponents of the financial account
thanks to the forceful policy responses. In particular, (Figure 1.12), a large share of overall net outflows
after a brief period of tightening, international dol- from EMDEs has been through net portfolio flows,
lar funding conditions eased, with the cross-currency which declined substantially in 2022. This decline
basis of advanced economy currencies with respect likely reflects monetary tightening in advanced
to the US dollar narrowing back to pre-March levels economies. Other investment inflows, and in partic-
(Figure 1.11). ular global cross border bank flows to EMDEs, have
also declined since 2021. The bulk of the decline was
inflows into China, which has experienced higher
Global Financial Flows funding costs amid dollar strength. Net foreign
In 2022, uphill capital flows from EMDEs to direct investment (FDI) inflows, which remained
advanced economies reemerged. This resembles a relatively stable in 2020 and 2021, also fell in 2022.
pattern of capital flowing from lower-income to high-
er-income economies that occurred in the lead-up to
6Standard economic models suggest that capital is expected to
the global financial crisis (Figure 1.1; see also the 2021
flow from slower-growing, capital-abundant richer economies to
faster-growing capital-scarce ones in search of higher returns (see
Boz, Cubeddu, and Obstfeld 2017). This is commonly referred to
5Nonetheless, the Exchange Market Pressure Index does not cap- as a downhill flow of capital, whereas the reverse is called uphill
ture the effect of capital flow management measures that were used (Gourinchas and Jeanne 2013; Lucas 1990; Prasad, Rajan, and
by some economies as part of the policy mix. Subramanian 2007).
Figure 1.12. Capital Flows to Emerging Market and Figure 1.13. Incidence of Extreme Capital Flows: Number of
Developing Economies Surges, Stops, Flights, and Retrenchments
(Percent of country group GDP)
Advanced economies Emerging markets
Emerging Europe Russia 1. Surges 2. Stops
China India 30 30
Saudi Arabia Emerging Asia excluding China and India
Latin America Total
1. Net Portfolio Inflows 2. Net FDI Inflows
4 4 20 20
3 3
2 2
10 10
1 1
0 0
0 0
–1 –1
1990:Q1
2000:Q1
10:Q1
20:Q1
1990:Q1
2000:Q1
10:Q1
20:Q1
–2 –2
–3 –3
2018:Q1
18:Q3
19:Q1
19:Q3
20:Q1
20:Q3
21:Q1
21:Q3
22:Q1
2018:Q1
18:Q3
19:Q1
19:Q3
20:Q1
20:Q3
21:Q1
21:Q3
22:Q1
22:Q3
22:Q3
3. Flights 4. Retrenchment
30 30
3 3
2 2 10 10
1 1
0 0
0 0
–1 –1
1990:Q1
2000:Q1
10:Q1
20:Q1
1990:Q1
2000:Q1
10:Q1
20:Q1
–2 –2
–3 –3
Sources: IMF, Balance of Payments; and IMF, International Financial Statistics.
2018:Q1
18:Q3
19:Q1
19:Q3
20:Q1
20:Q3
21:Q1
21:Q3
22:Q1
2018:Q1
18:Q3
19:Q1
19:Q3
20:Q1
20:Q3
21:Q1
21:Q3
22:Q1
22:Q3
22:Q3
Note: Capital flows are defined as gross inflows and outflows (excluding reserves).
Episodes are based on flows in billions of US dollars. Sample is External Balance
Assessment countries. Last observation is fourth quarter of 2022. A surge is a
Sources: IMF, International Financial Statistics; Institute of International Finance; sharp increase in gross capital inflows from foreign investors, a stop is a sharp
and IMF staff calculations. decrease in gross capital inflows from those investors, a flight is a sharp increase
Note: Group GDP is the total GDP of all economies considered in the figure, which in gross capital outflows from domestic investors, and a retrenchment is a sharp
include Brazil, Chile, China, Colombia, India, Indonesia, Malaysia, Mexico, Peru, decrease in gross capital outflows from domestic investors.
Philippines, Poland, Romania, Russia, Saudi Arabia, Thailand, and Türkiye. For
panels 1–3, positive numbers represent net inflows of capital. FDI = foreign direct
investment.
individual economies.7 The results suggest that after a Figure 1.14. Cumulative High-Frequency Portfolio Flows to
period of relative stability, characterized by “ripples” EMDEs January 2022–April 2023
(Billions of US dollars)
rather than “waves” (Forbes and Warnock 2021),
the frequency of extreme capital flow movements 250
China India
has increased since the onset of the pandemic, with Latin America Emerging Europe
200 Middle East and Africa Other
a notable rebound in gross flows from both foreign
Total
(surges) and domestic (flights) investors occurring 150
during the recovery from the pandemic in 2021,
likely fueled by mounting optimism in financial 100
Jan. 2022
Feb. 22
Mar. 22
Apr. 22
May 22
June 22
July 22
Aug. 22
Sep. 22
Oct. 22
Nov. 22
Dec. 22
Jan. 23
Feb. 23
Mar. 23
Apr. 23
the reopening of China and a shallower expected mon-
etary policy rate path in the United States. In particular,
there was a strong rebound in nonresident—and mostly
debt—flows to EMDEs (Figure 1.14). The banking sector Sources: Institute of International Finance; and IMF staff calculations.
turmoil in March 2023, while so far having had a limited
impact on short-term capital flows, calls for caution and
raises the risk of a potential risk-off episode, with decreas- declining asset prices (Figure 1.15). The largest debtor
ing inflows to EMDEs. economy remains the United States, though its net
international investment position improved from
–18.1 percent of world GDP in 2021 to –16.4 percent
International Balance Sheets and the Global Financial in 2022. Other large debtor economies include the
Safety Net euro area (excluding Germany and The Netherlands),
Creditor and debtor stock positions remained while the largest creditor economies remain, in
elevated in 2022, reflecting the offsetting effects descending order, Japan, Germany, and China.
of widening current account balances, the dollar’s Financial centers play an outsized role in global bal-
strength—which caused valuation gains in countries ance sheets, representing 36 percent of global holdings
with long positions in the dollar (Box 1.2)—and but only 7 percent of global GDP (see also Box 1.1).9
Stock positions remain even more elevated in gross
7Capital flow episodes are defined based on Forbes and Warnock terms (Figure 1.16).
(2012, 2021), a definition that is also used in David and Gonçalves Valuation changes, which induce wealth transfer
(2021). They are events in which the year-over-year changes in across countries, were more muted in 2022 compared
four-quarter flows are more than two standard deviations away from
the historical average (based on 20 quarters) during at least one with 2021 for all External Sector Report (ESR) econo-
quarter of the event. The event lasts for all consecutive quarters for mies. In 2022, creditor economies tended to have more
which the change in annual capital flows is more than one standard valuation losses, while debtors tended to experience
deviation away from the historical average. A surge is a sharp increase
in gross capital inflows by foreigners; a stop is a sharp decrease in
more valuation gains (Figure 1.17), dampening global
gross capital inflows by foreigners; a flight is a sharp increase in gross stock imbalances. For instance, in the United States,
capital outflows by domestic investors; and a retrenchment is a sharp declining asset prices led to (positive) valuation gains
decrease in gross capital outflows by domestic investors.
8Typically, global current account balances tend to widen when
in its external balance sheet, more than offsetting the
many economies recover from a sudden stop. This pattern was
not observed in the 2022 widening of global balances, reflecting
the absence of widespread sudden stops in EMDEs during the 9The list of financial centers is based on Lane and Milesi-Ferretti
Figure 1.15. Net International Investment Positions, Figure 1.16. Sum of Cross-Border Assets and Liabilities
1990–2022 (Percent of world GDP)
(Percent of world GDP)
560 USA CHN DEU and NLD
30 Surplus AEs Oil exporters AE commodity exporters 520 JPN EA (other) GBR
Deficit EMDEs Other deficit Other surplus Creditor AEs Oil exporters AE commodity exporters
480 Debtor EMs ROW Total
USA CHN DEU and NLD
20 JPN EA (other) GBR 440
Discrepancy 400
360
10
320
280
0 240
200
–10 160
120
80
–20
40
0
1990
92
94
96
98
2000
02
04
06
08
10
12
14
16
18
20
22
–30
1990
92
94
96
98
2000
02
04
06
08
10
12
14
16
18
20
22
Sources: External Wealth of Nations database; IMF, April 2023 World Economic
Sources: External Wealth of Nations database; IMF, April 2023 World Economic Outlook; and IMF staff calculations.
Outlook; and IMF staff calculations. Note: AEs = advanced economies; EA = euro area; EMs = emerging markets;
Note: Advanced economy (AE) commodity exporters comprise Australia, Canada, ROW = rest of the world. Data labels in the figure use International Organization
and New Zealand; creditor AEs comprise Hong Kong SAR, Korea, Singapore, for Standardization (ISO) country codes.
Sweden, Switzerland, and Taiwan Province of China; deficit emerging market and
developing economies (EMDEs) comprise Brazil, Chile, India, Indonesia, Mexico,
Peru, South Africa, and Türkiye; oil exporters comprise those classified as such in
the World Economic Outlook definition plus Norway. EA = euro area. Data labels in
the figure use International Organization for Standardization (ISO) country codes. Figure 1.17. International Investment Position Valuation
Change and Net International Investment Position, 2022
(Percent of GDP)
19 crisis, with the Federal Reserve’s bilateral swap MEX MYS KOR
5 ESP
lines playing a key role in stabilizing global financial CHN BEL
FRA
0
markets and capital flows to EMDEs. Its goal is to POL
BRA IND CAN NLD
provide countries with insurance against (financial) –5 AUS
THA RUS
shocks, as well as financing and incentives for sound –10
ARG
macro-economic policies (Aiyar and others 2023). The –15
ZAF
SAU
GFSN is composed of four main layers (Figure 1.18):
–20 CHE
gross international reserves, central banks’ bilateral swap
lines (BSLs, limited and unlimited), Regional Financing –25
–150 –100 –50 0 50 100 150
Arrangements (RFAs), and the IMF (borrowed and NIIP, 2022
quota resources). As of the end of 2021, it represented
Sources: IMF, April 2023 World Economic Outlook; and IMF staff calculations.
a combined firepower of about 19 percent of global Note: Valuation changes are calculated as the difference between changes in net
international investment position (NIIP) and current account. For some countries,
NIIPs are still projections. Bubble sizes are proportional to 2022 GDP in US dollars.
10Declining
domestic asset prices tend to improve the net interna- Hong Kong SAR and Singapore are excluded because of the size of their NIIPs.
tional investment position, as nonresidents hold some of these assets, Data labels in the figure use International Organization for Standardization (ISO)
leading to a decline in foreign liabilities. country codes.
Figure 1.18. The Evolution of Global Financial Safety Net, multilaterally consistent analysis of current accounts
1995–2021 and real exchange rates. Annex Tables 1.1.2 and 1.1.3
(Percent of world GDP)
summarize the IMF staff–assessed current account and
5 Gross international reserves (eop, right scale) 18 real effective exchange rate gaps and external sector
BSLs, advanced economies, unlimited1 assessments for these economies.
BSLs, limited2
RFAs3 15
4 IMF borrowed resources4
IMF quota resources5 Primer on Methodology
12
The primary numerical inputs for the IMF staff’s
3
assessments come from the models in the External
9 Balance Assessment (EBA) methodology.12 The models
2
produce medium-term current account and real
6 exchange rate benchmarks (or norms) that are consis-
tent with country fundamentals and desired policies
1 (Figure 1.19).13 The norms are compared with realized
3
current account and real exchange rate levels (after
adjusting for cyclical and other short-term factors) to
0 0 derive gaps, a measure of excess external balances. Posi-
1995
96
97
98
99
2000
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
This report presents individual assessments of affecting economies’ saving and investment decisions. Advanced
economies with higher incomes, older populations, and lower growth
external positions for 30 of the world’s largest econo-
prospects tend to have positive norms, while most EMDEs, which
mies, which represent 87.5 percent of global GDP.11 tend to be younger and are expected to import capital to invest and
The IMF staff’s assessment of external positions is a exploit their higher growth potential, have negative norms. Norms
also depend on desirable medium-term policies—that is, policies
deemed appropriate by IMF staff once cyclical factors are accounted
11Although the ESR presents assessments for 30 systemic econo- for. For instance, economies for which the staff recommends a
mies, the IMF staff conduct an assessment of the external sector of relatively loose fiscal policy will have lower norms than those that are
all members as part of bilateral surveillance. evaluated as needing fiscal consolidation.
Figure 1.19. External Balance Assessment Current Account economies with such positions were Germany,
Norms, 2022 Malaysia, Russia, Singapore, Sweden, and Thailand,
(Percent of GDP)
along with India, Mexico, and Saudi Arabia, which
7.5 entered the category in 2022.
EBA norm1 Net foreign assets
Demographics Oil
•• Moderately weaker, weaker, or substantially weaker
Other fundamentals2 Desirable policies2 than the level consistent with medium-term fundamen-
5.0 tals and desirable policies: The eight economies with
such positions were Argentina, Belgium, Canada,
South Africa, and the United States, along with
2.5
France, Italy, and Türkiye, which entered the cate-
gory in 2022, driven by decreases in their current
account balances that resulted in negative current
0.0
account gaps.
•• Broadly in line with the level consistent with medi-
–2.5 um-term fundamentals and desirable policies: The
13 economies with such positions were Brazil,
China, Hong Kong Special Administrative Region,
–5.0 Indonesia, Japan, Korea, Spain, Switzerland, and the
CHE
KOR
NLD
RUS
JPN
DEU
BEL
ZAF
CAN
SWE
CHN
ARG
ESP
EA4
MYS
TUR
AUS
GBR
IDN
MEX
IND
POL
ITA
THA
FRA
USA
BRA
Figure 1.20. Evolution of External Sector Assessments, Figure 1.21. Evolution of Headline Current Account Balances
2012–22 and IMF Staff Gaps
2022–21 change in
Indonesia 3 JPN MYS
ESP
Spain 2 CHE IND THA
Brazil 1 ZAF IDN
United Kingdom HKG RUS SWE SGP
0 USA
CAN BRA POL DEU
Weaker than Implied by Fundamentals –1
ITA GBR EA NLD
France –2 BEL
ARG KOR TUR
Türkiye –3 AUS
United States FRA
–4
South Africa –4 –3 –2 –1 0 1 2 3 4 5 6
Canada 2021 IMF staff–assesed CA gap
Italy
Argentina -- -- -- -- -- --
Source: IMF staff calculations.
Belgium
Note: CA = current account; EA = euro area; ESR = External Sector Report. Data
labels in the figure use International Organization for Standardization (ISO) country
Moderately Substantially codes.
Stronger 1
The headline CA for 2023 is a projection.
Broadly in line 2
Bubble sizes are proportional to 2022 GDP in US dollars.
Weaker
Figure 1.22. IMF Staff Current Account and Real Effective the surplus is projected to increase, with the change
Exchange Rate Gaps driven by lower liquefied natural gas prices and stronger
20
demand from Asia, while Japan’s current account surplus
ARG is also projected to increase mainly driven by lower com-
15 modity prices and inbound tourism.
ITA USA The narrowing of global current account balances
10 CAN
BEL
FRA
BRA is expected to continue over the medium term, as the
Staff REER gap (percent)
Memorandum item:
Euro Area –566 –433 62 283 –0.7 –0.5 0.1 0.3 –4.2 –3.3 0.4 2.0
Statistical Discrepancy –3,599 –4,706 –5,355 –5,197 –4.1 –5.6 –5.6 –5.2 ... ... ... ...
Overall Creditors1 17,367 19,634 21,125 20,004 20.0 23.2 22.0 20.0 ... ... ... ...
Of which: 13,532 15,602 17,184 15,392 15.5 18.4 17.9 15.4 ... ... ... ...
Advanced
Economies
Overall Debtors1 –20,966 –24,340 –26,481 –25,200 –24.1 –28.7 –27.6 –25.2 ... ... ... ...
Of which: –15,945 –19,696 –21,877 –20,413 –18.3 –23.3 –22.8 –20.4 ... ... ... ...
Advanced
Economies
Sources: IMF, April 2023 World Economic Outlook; US Bureau of Economic Analysis; and IMF staff calculations.
Note: “. . .” indicates that data are not available or not applicable; SAR = Special Administrative Region.
1Overall creditors and debtors (and the “of which” advanced economies) include non–External Sector Report economies.
Figure 1.23. Change in 2023 Current Account Balances material capital outflows and exchange rate adjust-
(Percent of baseline world GDP)
ments (see the April 2023 Global Financial Stability
0.12 Report). To the extent EMDE currencies—many of
which carry current account deficits—depreciate with
Deficit EMs
falling risk appetite, this would likely contribute to
0.08 narrowing global balances.
Rising commodity prices: Another surge in commod-
Change in CA balances
Figure 1.24. Number of Trade Restrictions, 2009–22 Further fragmentation would also weaken international
policy coordination on vital global public goods, such
3,000
Goods as climate change mitigation and pandemic resilience
Services
Investment (see Chapter 2 of the 2022 External Sector Report).
2,500
2,000
Policy Priorities for Promoting
External Rebalancing
1,500
While current account surpluses and deficits are not
necessarily an undesirable phenomenon to the extent
that they reflect differences in countries’ fundamentals
1,000
and desirable medium-term policies, excess current
account balances should be reduced. Excess bal-
500 ances reflect an inefficient allocation of resources and
frictions in domestic economies, leading to welfare
0 losses in societies. Economies with excessively large
2009 10 11 12 13 14 15 16 17 18 19 20 21 22 current account deficits and negative net international
Sources: Global Trade Alert; and IMF staff calculations.
investment positions are associated with larger real
Note: Data as of April 26, 2023. effective exchange rate gaps and subject to greater
exchange market pressures and risks of sudden stops
(Figure 1.25), the risk of which has likely risen—other
things equal—for ESR economies that have moved
Geoeconomic fragmentation further hampering farther away from the “broadly in line” category in
global trade and other international flows: The risk of 2022 while debtor stock position remained elevated.
geoeconomic fragmentation has been aggravated by Moreover, excess balances could have real or perceived
the US–China trade tensions and the war in Ukraine. distributional effects, raising discontent with globaliza-
Trade barriers have been rising (Figure 1.24), and in tion and fueling trade tensions. Therefore, correcting
the extreme, the world economy could splinter into excess balances can improve welfare and reduce the risk
geoeconomic blocs. Geoeconomic fragmentation could of disruptive capital flow reversals.
affect the currency composition of foreign exchange Promoting external rebalancing requires both excess
reserves, reduce capital flows, complicate provision of current account surplus and deficit economies to act
the global safety net, and lead to a reorganization of collectively. As the April 2023 World Economic Outlook
the international monetary system (Aiyar and others emphasizes, policymakers will need to tread a narrow
2023). The impact on global current account balances path toward restoring financial sector stability, normal-
would depend on the specific scenario: while further izing fiscal policy, and avoiding recession while also
increase in trade costs across country blocs would durably reducing inflation and achieving sustainable
likely contribute to reducing global balances (Box 1.3), and inclusive growth. In addition to being consistent
trade costs within each bloc could fall and contribute with these objectives, the policy priorities set out in the
to increasing global balances. Moreover, the risk of April 2023 World Economic Outlook, including efforts
extreme fragmentation could increase the incentive for to normalize fiscal policy and steadily increase policy
self-insurance and potentially increase global balances rates, would also help to facilitate trade, rebalance
if countries with current account surpluses increase excess external positions, and contain risks to external
savings more than those with deficits. In any case, fur- balances.
ther geoeconomic fragmentation would unambiguously In the event of global financial distress, EMDEs
lead to lower welfare, including through its effect on should let their currencies adjust to help their econo-
FDIs, the diffusion of technology, and flows of labor, mies absorb external shocks. However, in specific cases
goods, and capital (Aiyar and others 2023; Chapter 4 in which shocks are large and countries face vulnera-
of the April 2023 World Economic Outlook; Chapter bilities from shallow foreign exchange markets, sizable
4 of the April 2023 Global Financial Stability Report). balance sheet mismatches, or poorly anchored inflation
Figure 1.25. Exchange Market Pressure and REER Gaps further strengthen the rule-based system. Policies
(Percent)
to preserve global economic integration would also
0.8 mitigate the risks related to fragmentation of FDI and
Linear trend line
COL other capital flows along geoeconomic fault lines (see
0.7
Chapter 4 of the April 2023 World Economic Outlook).
0.6 Supporting availability of climate financing is also
0.5 important, given that green infrastructure investment
ROU in developing economies could mitigate the external
0.4
sector impact of climate change mitigation and adapta-
EMP, 2022
HUN TUN
0.3 tion efforts (see Chapter 2 of the 2022 External Sector
CZE MAR
CHN JPN Report). Industrial policy could be pursued to address
0.2 SWE DNK IND
ISR POL GBR well-established market failures and if other policies are
0.1 THA NZL CHL
NOR not available. However, industrial policy should not
AUS KOR ZAF PER CAN
0.0 introduce distortions and should be consistent with
CHE BRA
MYS
URY
international agreements and WTO rules, minimize
–0.1 RUS MEX
adverse spillovers, and avoid creating barriers to tech-
–0.2 nology transfer. They should also be, well-structured,
–20 –10 0 10 20 30 40
REER gap, 2021 cost-effective, transparent and accountable, while not
undermining competition (Cherif and others 2022).
Sources: Goldberg and Krogstrup (2023); and IMF staff calculations. Maintaining liquidity in the global financial
Note: CA = current account; EMP = exchange market pressure index. Real
effective exchange rate (REER) gaps are current-account-implied REER gaps system, via, among other things, the GFSN, will be
based on EBA CA gaps for 2021. The REER gaps shown in this figure can deviate essential to helping economies manage risks related
from the CA-implied REER gap in the 2021 external sector assessment due to data
updates and country-specific adjustors. Data labels in the figure use International to the tightening of global financial conditions and
Organization for Standardization (ISO) country codes. financial system fragmentation due to geopolitical
tensions. The GFSN has played a vital role in safe-
guarding the stability of the global economy. How-
expectations, temporary foreign exchange interventions ever, the coverage of the various layers of the GFSN
may be appropriate. Capital flow management mea- is uneven, and global liquidity provision is limited
sures on outflows may be used if disruptive outflows (IMF 2016). To this end, the IMF is the only layer
lead to (imminent) crisis circumstances, but these mea- that provides universal coverage, where its lending
sures should not substitute for needed macroeconomic programs help provide a safety net for countries hit
policy adjustment. by balance-of-payments shocks. To perform this
Coordinated policy efforts will help deal with a host function effectively, the IMF should remain repre-
of complex challenges facing the world. Over the last sentative of its global membership and adequately
three decades, the sharp growth in global trade has resourced to serve as an anchor of the GFSN, which
gone hand in hand with billions of people moving crucially depends on the successful completion of the
out of poverty. With the world at increasing risk of 16th General Review of Quotas.
geoeconomic fragmentation, it is therefore of para- Policies to promote external rebalancing differ
mount importance to preserve the benefits of global based on individual economies’ positions and needs,
integration and multilateralism. To achieve this, the as detailed in the Individual Economy Assessments in
current rule-based trading system must be strengthened Chapter 3 (and summarized in Annex Table 1.1.6).
to adapt to a changing world. Advancing multilat- •• Economies with weaker-than-warranted external
eral trade rules may require focusing on reforms with positions should focus on policies that boost sav-
high impact in which preferences of countries are ing and competitiveness. Where current account
broadly aligned. The package agreed upon at the 12th deficits in 2022 partly reflected fiscal deficits
Ministerial Conference of the World Trade Organiza- above desirable levels (as in Italy and the United
tion (WTO) in June 2022 is a step in this direction. States), medium-term fiscal consolidation would
Fully restoring the WTO dispute settlement system help stabilize debt-to-GDP ratios and close current
and implementing new WTO-based agreements would account gaps. However, fiscal consolidation should
be implemented in a growth-friendly way, while widespread informality should help reduce precau-
providing space for critical infrastructure investment tionary saving and support consumption, thus also
and well-targeted social spending to help reduce helping with external rebalancing.
poverty and inequality (for example, in Argentina •• Economies with external positions broadly in line with
and South Africa). Countries with competitiveness fundamentals should continue to address domestic
challenges also need to address structural bottlenecks imbalances to prevent excessive external imbal-
through labor, product market, and other structural ances. Some economies (such as China) should
reforms to promote green, digital, and inclusive address offsetting policy distortions. Relevant
growth while boosting productivity. policies include accelerating market-based struc-
•• Economies with stronger-than-warranted external posi- tural reforms—including state-owned enterprise
tions should prioritize policies aimed at promoting reform—to promote growth and shifting fiscal
investment and diminishing excess saving to support policy support toward strengthening social protec-
external rebalancing while also pursuing domestic tion to reduce high household saving and stimulate
objectives. For example, in Germany, higher fiscal private consumption. In countries with negative net
deficits than currently planned are likely to be international investment positions (such as Brazil
required over the medium term to achieve domes- and Spain), keeping current account balances in
tic climate, digital, and energy security goals. In line with their norms will require a combination of
Sweden, higher investment in the green transition fiscal consolidation efforts and higher private saving
and the health sector, needed to attain the country’s to provide room for investment in education and
ambitious medium-term climate goals and prepare other reforms to encourage innovation and improve
for demographic transition, would also lower the competitiveness. Reforms to boost productivity
external balance. In some emerging markets (such would also create space for investment needed to
as Malaysia and Thailand), efforts to reform and advance green transition and reduce dependence on
expand social safety nets and measures to address foreign energy.
2007
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
China and Saudi Arabia (Figures 1.1.1 and 1.1.2).
Accumulation of foreign exchange reserves has played
Sources: Refinitiv Datastream; and IMF staff calculations.
Note: Figure shows total cross-border total liabilities (amounts
outstanding) of reporting banks in all currencies. Last
Figure 1.1.1. China: Current Account Surplus observation: third quarter of 2022.
and Capital Flows, Four-Quarter Trailing Sums
(Billions of US dollars)
Current account Portfolio debt (–) a much smaller role than before the GFC. Instead,
Foreign exchange Net errors and net portfolio investment (debt in China, equity in
reserves (–) omissions (–) Saudi Arabia) and net other investment (bank loans
Other investment (–)
Foreign direct investment and portfolio equity (–) in China, currency and deposits in China and Saudi
800 Capital account and financial derivatives (–) Arabia) have become more important channels of
600
recycling (that is, investing) the recent surpluses in
these economies.1 In Russia, net other investment is
400 Outflows the main channel for financial outflows, with a notable
200 portion of those outflows headed toward the euro area,
0 with Belgium being a prime destination (Figure 1.1.3).
Outside the euro area, Switzerland has been a recipient
–200 Inflows of a substantial share of Russia’s investment since 2008
–400 (Figure 1.1.4).
–600
Who Funds the US Current Account Deficit?
2000
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
1In China, net errors and omissions account for part of the
This box was prepared by Cian Allen and Cyril Rebillard. recycling of the surplus.
–1,500
–2,000
Sources: Refinitiv Datastream; and IMF staff calculations.
2000
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
Note: Figure shows total cross-border total liabilities (amounts
outstanding) of reporting banks in all currencies. Last
observation: fourth quarter of 2022. Sources: Refinitiv Datastream; and IMF staff calculations.
Note: Figure shows total cross-border total liabilities (amounts
outstanding) of reporting banks in all currencies. Last
observation: fourth quarter of 2022.
2003:Q1
03:Q4
04:Q3
05:Q2
06:Q1
06:Q4
07:Q3
08:Q2
09:Q1
09:Q4
10:Q3
11:Q2
12:Q1
12:Q4
13:Q3
14:Q2
15:Q1
15:Q4
16:Q3
17:Q2
18:Q1
18:Q4
19:Q3
20:Q2
21:Q1
21:Q4
22:Q4
Sources: Federal Reserve; and IMF staff calculations.
Note: Estimated flows are essentially constructed using changes
in foreign holdings of US Treasury securities adjusted for
valuation effects as discussed in Bertaut and Judson (2014).
Tabova and Warnock (2021) assess the different sources Sources: US Bureau of Economic Analysis; and IMF staff
available for measuring foreign transactions in US Treasury calculations.
securities and support the use of holdings-based estimates of Note: “Other” includes Canada, Latin America and other Western
flows. “Corporate” includes bonds and stocks. Hemisphere, Africa, and international organizations.
corporate securities compiled by the Federal Reserve the share of gold in its reserves has increased since
(Figure 1.1.8)3: 2007, reaching 21 percent at the end of 2022.
•• In China, while the 2015–16 sale of US Treasur- •• On the other hand, two countries have signifi-
ies coincided with exchange rate depreciation, the cantly increased their holdings of US securities
2018–20 sale occurred with a modest net purchases and as a result accounted for the largest share
of US government agency bonds. Since late-2021, of the external portfolio debt financing of US
the purchase of agency bonds has increased, broadly current account deficits in the recent period. They
offsetting the decline in the purchase of Treasury were the United Kingdom, with a total of about
securities. US$600 billion, comparable to the pre-GFC peak,
•• Russia has been divesting away from US Trea- although the composition is now more tilted toward
sury bonds, especially since 2014, following the US Treasuries and away from corporate bonds;
annexation of Crimea and subsequent US and and the Cayman Islands, with a total of about
international sanctions. Its divestment of US US$500 billion, also tilted toward US Treasuries.
securities appears to have peaked about 2018, with In light of the United Kingdom’s current account
no significant transactions since mid-2019. Instead, deficit and the Cayman Islands’ small size, both
countries are likely to be only intermediaries provid-
ing financial and banking sector services.
3The patterns in these data can be distorted by “custodial
Echoing the increased role of financial centers in
bias,” where a foreign holder of the US liability chooses to use financing the US current account deficit, the share of
a custodian in a different country. This can be an issue in major
official holdings (among total holdings) of US Treasury
financial centers, such as Belgium, the Caribbean banking
centers, Luxembourg, Switzerland, and the United Kingdom (see securities has steadily decreased, from a peak of 76
Bertaut and Judson 2014). percent in mid-2009 to about 50 percent at the end
2013:M1
14:M1
15:M1
16:M1
17:M1
18:M1
19:M1
20:M1
21:M1
22:M1
23:M1
3. Cayman Islands 4. UK
200 200
150 150
100 100
50 50
0 0
–50 –50
–100 –100
–150 –150
–200 –200
2013:M1
14:M1
15:M1
16:M1
17:M1
18:M1
19:M1
20:M1
21:M1
22:M1
23:M1
2013:M1
14:M1
15:M1
16:M1
17:M1
18:M1
19:M1
20:M1
21:M1
22:M1
23:M1
of 2022, while the share of private holdings exceeded The interdependence between large surplus and defi-
40 percent at the end of 2022 (Figure 1.1.9). However, cit economies appears to be largely intact. At the same
the currency composition of official foreign exchange time, the role of financial centers has increased, as
reserves has remained largely stable in recent years, with their rising share in financing the US current account
the US dollar still accounting for about 60 percent of deficit (Figure 1.1.8) or in China’s overseas portfolio
the total of (allocated) global reserves (Figure 1.1.10). investment (Figure 1.1.11) shows.
1999
2000
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
Sources: Refinitiv Datastream; and US Department of the Source: IMF, Currency Composition of Official Foreign Exchange
Treasury, Treasury International Capital System. Reserve (COFER).
1,200
British Virgin Islands
United States
1,000
Hong Kong SAR
Cayman Islands
800
600
400
200
0
2015:H1
15:H2
16:H1
16:H2
17:H1
17:H2
18:H1
18:H2
19:H1
19:H2
20:H1
20:H2
21:H1
21:H2
22:H1
Cumulative probability
Long foreign currency positions. Aggregate foreign AEs, 2007
currency exposures, which measure net foreign assets 0.6 EMDEs, 2020
AEs, 2020
in foreign currency (as a percentage of total assets
and liabilities), have improved significantly since 0.4
1990, particularly in emerging market and devel-
0.2
oping economies (EMDEs). In fact, most EMDEs
have moved from a negative aggregate net position in 0.0
foreign currency (indicated by negative x-axis values in –0.8 –0.6 –0.4 –0.2 0.0 0.2 0.4 0.6 0.8
Figure 1.2.1, panel 1) to a positive one, as evidenced Foreign currency exposure
(percent of total assets and liabilities)
by the rightward shift of the corresponding curve. This
transition took place mainly before the global financial 2. EMDEs: Aggregate 3. EMDEs: Portfolio Debt
crisis and is largely attributable to the currency com- Net Foreign Net Foreign Currency
position of other investments (mainly bank related) Currency Exposure2 Exposure2
(Percent of GDP) (Percent of GDP)
and a greater reliance on portfolio equity financing. 0.4 0.00
Currency-induced valuation effects. Positive net
0.3
positions in foreign currency have reduced risks –0.02
0.2
associated with depreciations in domestic currency, –0.04
increasing the insurance role of national balance 0.1
sheets in response to negative shocks to economies. In 0.0 –0.06
1990, a 10 percent depreciation in domestic currency, –0.1
–0.08
all else equal, resulted in a median valuation loss of –0.2
1.6 percent of GDP for EMDEs. However, by 2020, –0.10
–0.3
this median effect had become positive, equivalent –0.4 –0.12
to 2.4 percent of GDP (as illustrated in Figure 1.2.1,
1990
95
2000
05
10
15
20
1990
95
2000
05
10
15
panel 2). Advanced economies also experienced a 20
similar trend, with a 10 percent depreciation leading Source: Allen, Gautam, and Juvenal (2023).
to a median valuation gain of 0.5 percent of GDP in Note: AEs = advanced economies; EMDEs = emerging market
1990 and a valuation gain of 9.2 percent of GDP in and developing economies.
1
Aggregate foreign currency exposure is defined as total net
2020. The proportion of EMDEs with net long posi- foreign assets denominated in foreign currency as a share of
tions in foreign currency increased significantly, from total assets and liabilities.
2
17 percent in 1990 to 75 percent in 2020. However, A 1 percent uniform shift in the value of the domestic currency
against all foreign currencies leads to a median valuation
92 percent of EMDEs were short on foreign currency change of x percent of GDP.
in portfolio debt in 2020, resulting in a median
valuation loss of 1 percent of GDP in portfolio debt
when there is a 10 percent depreciation in domestic
currency (Figure 1.2.1, panel 3). sectors, institutions, or more granular asset classes.
Risks. Aggregate positions may mask significant cur- For example, when debt and equity are examined
rency mismatches on the balance sheets of individual separately, currency-driven valuation effects in debt
and equity tend to offset each other for many econ-
omies. Nonetheless, the prevalence of short positions
This box was prepared by Cian Allen and Luciana Juvenal.
1These economies are included in either the External Balance in foreign currency for debt among EMDEs keeps
Assessment or the External Sector Report and taken together EMDEs with such positions vulnerable to depreciation
represent more than 85 percent of world GDP. pressures.
costs, which is one aspect of GEF. But GEF could easily bring
Figure 1.3.2 show the response to the same productivity
about asymmetric changes in trade costs: were the world to be shock in a representative EMDE when the country’s
divided into several blocs, trades costs across blocs would rise trade barriers with the rest of the world are raised in line
to very high levels, but trade costs within each bloc would fall with the baseline GEF scenario in Bolhuis, Chen, and
significantly. GEF could also increase and alter frictions in inter- Kett (2023). As the figure shows, relative to the baseline
national transactions of all stripes, including financial market
transactions (see Aiyar and others 2023).
with lower trade costs, the trade and current account
2For details on modeling assumptions and calibration, see imbalances following the shock are smaller, while the
Cuñat and Zymek (2023). initial decline in consumption is larger. Simulating the
model for the size, frequency, and persistence of pro- imbalances is a diminished capacity to smooth the
ductivity shocks experienced by the typical EMDE, the impact of shocks on consumption. In the model simula-
average absolute value of the trade balance is found to tions, the standard deviation of real aggregate consump-
be 10 percent lower after GEF, and the average absolute tion is 20 percent larger. Higher trade costs thus expose
value of the current account balance is found to be EMDEs to greater consumption volatility, even if the
8 percent lower—a small but noticeable decline. The frequency and magnitude of domestic economic shocks
flip side of the reduction in trade and current account remain unchanged.
is equivalent to the change in reserve assets in the financial account series from the World Economic Outlook (which excludes valuation effects but includes interest
income on official reserves) plus the change in off-balance-sheet holdings (short and long FX derivative positions and other memorandum items) from International
Reserves and Foreign Currency Liquidity minus net credit and loans from the IMF.
4 The ARA metric reflects potential balance-of-payments FX liquidity needs in adverse circumstances and is used to assess the adequacy of FX reserves against potential
FX liquidity drains (see IMF 2015). The ARA metric is estimated for selected EMDEs and includes adjustments for capital controls for China. For Argentina, the adjusted
measure uses a four-year average to smooth the temporary effect of the sharp reductions in short-term debt and exports, and a collapse in the valuation of debt portfolio
investments in the wake of the sovereign debt restructuring. Additional adjusted figures are available in the individual country pages in Chapter 3.
5 The aggregate is calculated as the sum of External Sector Report economies only. The percent of GDP is calculated relative to total world GDP.
Annex Table 1.1.2. External Sector Report Economies: Summary of External Assessment Indicators, 2022
Current
Account International Investment
(Percent of IMF Staff CA Gap IMF Staff REER Position
GDP) (Percent of GDP) Gap (Percent) (Percent of GDP) CA NFA
Stabilizing SE of CA
Cycl. (Percent Norm
Economy Overall Assessment Actual Adj. Midpoint Range Midpoint Range Net Liabilities Assets of GDP) (Percent)
Argentina Weaker –0.6 –0.8 –1.8 ±1 17.5 ±2.5 18 49 67 1.0 0.5
Australia Broadly in line 1.2 –2.1 –0.5 ±0.8 2.6 ±4 –34 183 149 –1.9 0.8
Belgium Substantially weaker –3.5 –1.7 –4.6 ±0.4 6.3 ±0.6 54 365 419 2.6 0.4
Brazil Broadly in line –3.0 –3.3 –0.8 ±0.5 6.0 ±3.9 –40 90 49 –2.1 0.5
Canada Moderately weaker –0.3 –1.3 –1.8 ±0.5 6.8 ±1.7 30 235 265 2.3 0.5
China Broadly in line 2.2 2.2 0.8 ±0.6 –5.7 ±4.7 14 37 51 0.8 0.6
Euro Area1 Broadly in line –1.0 0.1 –0.1 ±0.6 0.2 ±1.8 2 249 251 0.1 0.6
France Moderately weaker –2.1 –1.5 –2.0 ±0.5 7.1 ±1.6 –24 326 302 –1.5 0.5
Germany Stronger 4.2 5.3 2.8 ±0.5 –7.8 ±1.4 71 239 310 4.3 0.5
Hong Kong SAR Broadly in line 10.5 10.3 0.6 ±1.5 –1.4 ±3.9 486 1,192 1,678 ... ...
India Moderately stronger –2.0 –0.9 1.5 ±0.7 –7.8 ±3.6 –11 37 26 –1.2 0.7
Indonesia Broadly in line 1.0 –1.5 0.3 ±0.6 –2.0 ±3.6 –19 53 34 –1.6 0.6
Italy Weaker –1.2 0.6 –2.5 ±0.7 9.3 ±2.7 4 171 174 0.3 0.7
Japan Broadly in line 2.1 3.2 0.0 ±1.1 0.0 ±6.6 75 165 240 3.2 1.1
Korea Broadly in line 1.8 4.2 –1.0 ±0.9 2.9 ±2.7 46 84 130 2.4 0.9
Malaysia Stronger 3.1 2.4 4.0 ±0.5 –8.0 ±1 4 121 125 0.8 0.5
Mexico Moderately stronger –1.3 –0.4 1.7 ±0.5 –4.9 ±1.3 –42 94 52 –2.0 0.5
The Netherlands Broadly in line 4.4 5.5 0.0 ±0.6 0.1 ±0.9 75 968 1,043 4.3 0.6
Poland Broadly in line –3.0 –1.8 0.9 ±0.5 –2.0 ±1 –34 91 57 –2.6 0.5
Russia Stronger 10.4 6.7 2.3 ±1.1 –13.6 ±6.5 34 38 72 1.2 1.1
Saudi Arabia Substantially stronger 13.6 12.5 4.7 ±2.5 –21.6 ±12.5 62 58 119 ... ...
Singapore Substantially stronger 19.3 21.8 5.1 ±1.8 –10.2 ±3.6 176 949 1,126 ... ...
South Africa Moderately weaker –0.5 –1.4 –1.3 ±0.7 5.0 ±2.9 17 114 131 0.6 0.7
Spain Broadly in line 0.6 1.4 0.7 ±0.8 –2.2 ±2.6 –61 259 199 –3.1 0.8
Sweden Stronger 4.3 5.0 3.8 ±0.4 –9.7 ±5.7 40 285 325 2.1 0.4
Switzerland Broadly in line 10.1 10.6 0.0 ±0.8 0.1 ±1.4 93 588 681 4.4 0.8
Thailand Stronger –3.2 –2.3 2.9 ±0.7 –6.2 ±1.6 –3 121 118 0.0 0.7
Türkiye Moderately weaker –5.3 –2.5 –1.9 ±0.7 6.5 ±2.5 –31 65 34 –1.9 0.7
United Kingdom Broadly in line –3.8 –2.2 –0.8 ±1 2.9 ±3.6 –11 574 563 –0.8 0.3
United States Moderately weaker –3.7 –3.5 –1.1 ±0.7 9.0 ±5.6 –65 176 112 –3.5 0.7
Sources: IMF, International Financial Statistics; IMF, April 2023 World Economic Outlook; US Bureau of Economic Analysis; and IMF staff assessments.
Note: CA = current account; Cycl. Adj. = cyclically adjusted; NFA = net foreign assets; REER = real effective exchange rate; SAR = Special Administrative Region; SE = standard error.
1 The IMF staff–assessed euro area CA gap is calculated as the GDP-weighted average of IMF staff–assessed CA gaps for the 11 largest euro area economies.
Annex Table 1.1.3. External Sector Report Economies: Summary of IMF Staff–Assessed Current Account Gaps and IMF Staff
Adjustments, 2022
(Percent of GDP)
IMF Staff Adjustments3
IMF
Actual CA Cycl. Adj. EBA CA EBA CA Staff-Assessed Other
Balance CA Balance Norm Gap1 CA Gap2 Total COVID-19 CA Norm Comments on Non–COVID-19-related
Economy [A] [B] [C] [D=B–C] [E=D+F] [F=G+H-I] [G] [H] [I] Adjustments
Argentina –0.6 –0.8 0.3 –1.2 –1.8 –0.6 0.1 0.0 0.7 NIIP/financing risk considerations
Australia 1.2 –2.1 –1.0 –1.1 –0.5 0.6 0.6 0.0 0.0
Belgium –3.5 –1.7 2.8 –4.5 –4.6 0.0 0.0 0.0 0.0
Brazil –3.0 –3.3 –2.2 –1.1 –0.8 0.3 0.3 0.0 0.0
Canada –0.3 –1.3 2.2 –3.4 –1.8 1.6 0.0 1.6 0.0 Measurement biases
China 2.2 2.2 0.7 1.5 0.8 –0.7 –0.7 0.0 0.0
Euro Area4 –1.0 0.1 –0.3 0.5 –0.1 –0.5 0.1 –0.5 0.1 Country-specific adjustments
France –2.1 –1.5 –0.3 –1.1 –2.0 –0.9 –0.9 0.0 0.0
Germany 4.2 5.3 2.8 2.5 2.8 0.4 0.4 0.0 0.0
India –2.0 –0.9 –2.3 1.5 1.5 0.0 0.0 0.0 0.0
Indonesia 1.0 –1.5 –1.1 –0.4 0.3 0.8 0.4 0.0 –0.4 High mortality rate, norm
Italy –1.2 0.6 3.4 –2.9 –2.5 0.4 0.4 0.0 0.0
Japan 2.1 3.2 3.5 –0.3 0.0 0.3 0.3 0.0 0.0
Korea 1.8 4.2 4.8 –0.6 –1.0 –0.4 –0.4 0.0 0.0
Malaysia 3.1 2.4 –0.5 2.9 4.0 1.1 1.1 0.0 0.0
Mexico –1.3 –0.4 –1.6 1.2 1.7 0.4 0.4 0.0 0.0
The Netherlands 4.4 5.5 4.8 0.7 0.0 –0.7 –0.2 –0.5 0.0 Measurement biases
Poland –3.0 –1.8 –2.7 1.0 0.9 –0.1 –0.1 0.0 0.0
Russia 10.4 6.7 4.0 2.7 2.3 –0.4 –0.4 0.0 0.0
South Africa –0.5 –1.4 2.2 –3.6 –1.3 2.3 0.2 1.5 –0.6 SACU transfers and measurement
biases (CA), demographics (high
mortality risk, norm)
Spain 0.6 1.4 –0.1 1.5 0.7 –0.8 0.2 0.0 1.1 NIIP/financing risk considerations
Sweden 4.3 5.0 0.8 4.2 3.8 –0.3 –0.3 0.0 0.0
Switzerland 10.1 10.6 6.5 4.1 0.0 –4.1 –0.1 –4.0 0.0 Measurement biases
Thailand –3.2 –2.3 0.9 –3.2 2.9 6.1 6.1 0.0 0.0
Türkiye –5.3 –2.5 –0.8 –1.7 –1.9 –0.2 –0.2 0.0 0.0
United Kingdom –3.8 –2.2 –1.0 –1.2 –0.8 0.4 –0.3 0.7 0.0 Measurement biases
United States –3.7 –3.5 –2.2 –1.2 –1.1 0.2 0.2 0.0 0.0
Hong Kong SAR 10.5 10.3 ... ... 0.6 ... 0.9 ... ...
Singapore 19.3 21.8 ... ... 5.1 ... –3.1 ... ... Measurement biases, NFA
composition, health spending
Saudi Arabia 13.6 12.5 ... ... 4.7 ... 0.0 ... ...
Absolute sum ... ... ... 1.2 0.9 ... ... ... ...
of excess
surpluses and
deficits5
Discrepancy 6 ... ... ... ... –0.01 ... ... ... ...
Source: IMF staff estimates.
Note: “. . .” indicates that data are not available or not applicable; CA = current account; Cycl. Adj. = cyclically adjusted; EBA = External Balance Assessment; ESR = External Sector
Report; NIIP = net international investment position; SACU = Southern African Customs Union.
1 Minor discrepancies between constituent figures and totals are due to rounding.
2 Refers to the midpoint of the IMF staff–assessed CA gap.
3 Total IMF staff adjustments include rounding in some cases. See Online Annex 1.1 for a description of COVID-19 adjustors. The last column explains country-specific adjustments
The IMF staff–assessed CA gap is calculated as the GDP-weighted average of IMF staff–assessed gaps for the 11 largest euro area economies.
5 Sum of absolute value of IMF staff–assessed CA gaps in percent of aggregate GDP for economies included in the ESR exercise.
6 Sum of IMF staff–assessed CA gaps in percent of aggregate GDP for economies included in the EBA and/or ESR exercise.
Annex Table 1.1.4. External Sector Report Economies: Summary of IMF Staff–Assessed Real Effective
Exchange Rate and External Balance Assessment Model Gaps, 2022
REER Gap Implied REER
IMF by IMF EBA EBA (Percent Change)
Staff-Assessed Staff-Assessed REER-Level REER-Index CA/REER Average 2022/ April 2023/
Economy REER Gap1 CA Gap2 Gap Gap Elasticity3 Average 2021 Average 2022
Argentina 17.5 15.2 10.8 25.0 0.12 21.0 1.4
Australia 2.6 2.6 23.4 –20.1 0.20 0.2 –1.5
Belgium 6.3 6.3 31.3 16.9 0.72 –0.4 0.8
Brazil 6.0 6.0 –14.4 –29.1 0.13 12.1 2.3
Canada 6.8 6.8 –10.5 1.9 0.27 –0.7 –4.3
China –5.7 –5.7 12.7 16.1 0.14 –1.2 –6.5
Euro Area 0.2 0.2 8.0 7.6 0.35 –4.1 5.0
France 7.1 7.1 5.3 –4.8 0.28 –4.6 2.3
Germany –7.8 –7.8 –9.5 6.7 0.37 –3.6 3.2
India –7.8 –7.8 10.6 12.5 0.19 0.9 –2.8
Indonesia –2.0 –2.0 –16.3 –2.7 0.16 2.5 0.4
Italy 9.3 9.3 15.4 12.3 0.27 –2.0 2.8
Japan 0.0 0.0 –31.4 –31.7 0.17 –13.7 –1.3
Korea 2.9 2.9 3.4 –1.9 0.34 –5.4 –1.4
Malaysia –8.0 –8.0 –29.3 –25.2 0.50 –1.5 –1.2
Mexico –4.9 –4.9 14.9 –3.8 0.34 5.3 12.9
The Netherlands 0.1 0.1 15.0 27.8 0.66 0.1 0.8
Poland –2.0 –2.0 –19.0 2.7 0.43 1.4 8.9
Russia –13.6 –13.6 –4.7 5.7 0.17 36.8 –7.1
South Africa 5.0 5.0 12.8 –3.5 0.25 –2.2 –9.1
Spain –2.2 –2.2 29.2 10.6 0.31 –1.1 0.2
Sweden –9.7 –10.3 –17.0 –15.9 0.37 –6.1 –0.8
Switzerland 0.1 0.1 17.6 11.9 0.55 0.3 2.1
Thailand –6.2 –6.2 –2.6 6.7 0.47 –1.1 1.6
Türkiye 6.5 6.5 –56.7 –46.3 0.29 –8.5 6.9
United Kingdom 2.9 2.9 2.3 –8.4 0.28 –1.4 1.1
United States 9.0 9.0 22.8 10.7 0.12 9.5 –0.5
Hong Kong SAR –1.4 –1.4 ... ... 0.39 3.7 0.5
Singapore –10.2 –10.2 ... ... 0.50 6.0 6.1
Saudi Arabia –21.6 –21.6 ... ... ... 4.1 –0.2
33
34
Annex Table 1.1.6. 2022 Individual Economy Assessments: Summary of Policy Recommendations
Economy Overall 2022 Assessment Policy Recommendations
Argentina Weaker Implement growth-friendly fiscal consolidation, combined with tight monetary policy and a streamlined FX regime to strengthen the trade balance,
rebuild international reserves, regain market access, and ensure debt sustainability; introduce reforms to boost export capacity and encourage FDI.
Australia Broadly in line Withdraw fiscal and monetary stimulus at an appropriate pace. Boost investment by executing planned infrastructure spending, streamlining product
market regulation, and promoting R&D and innovation.
Belgium Substantially weaker Strengthen competitiveness by addressing structural challenges, including labor and product market reforms, to foster green, digital, and inclusive
growth. Rebuild fiscal buffers through expenditure-led consolidation.
Brazil Broadly in line
2023 EXTERNAL SECTOR REPORT
Raise national saving including by implementing medium-term fiscal consolidation. Reduce the cost of doing business by fostering a skilled labor force
and implementing structural reforms to increase competitiveness.
Canada Moderately weaker Develop a medium-term fiscal consolidation plan; boost nonfuel exports through improved labor productivity, removal of nontariff trade barriers,
(Continued)
Annex Table 1.1.6 (continued)
Economy Overall 2022 Assessment Policy Recommendations
Malaysia Stronger Strengthen the social safety net, including through a reorientation of fiscal spending that should target a gradual and growth friendly consolidation;
implement structural policies to encourage private investment and boost productivity growth.
Mexico Moderately stronger Implement structural reforms to address investment obstacles, including tackling economic informality and governance gaps. Continue using floating
ER as the main shock absorber, with FXI used only to prevent disorderly market conditions.
The Netherlands Broadly in line Support investment in physical and human capital to foster robust potential growth, including through structural investment and reform plans to
safeguard energy security, allay housing market shortages, facilitate access to finance for SMEs, reinforce the education system, and advance the
climate transition and digitalization.
Poland Broadly in line Reduce fiscal deficit while boosting public investment by deploying Next Generation EU grants to tackle infrastructure gaps, digitalization, and climate
change; use structural policies to encourage corporate investment and productivity and incentivize credit allocation to the private sector.
Russia Stronger ...
Saudi Arabia Substantially stronger Implement structural reforms with an accompanying investment program to help diversify the economy, lift productivity and align the external position
in the medium term; avoid procyclical fiscal policy amid high hydrocarbon windfalls; minimize risks associated with industrial policies.
Singapore Substantially stronger Increase public investment, including spending on health care, green and other physical infrastructures, and human capital, to help reduce external
imbalances over the medium term by lowering net public saving.
South Africa Moderately weaker Implement structural reforms and stronger fiscal consolidation under a credible medium-term framework, while providing space for critical
infrastructure and social spending; improve governance, efficiency of key product markets (to promote private sector participation), and functioning
of labor markets; seize opportunities to build up reserves.
Spain Broadly in line Implement fiscal consolidation. Improve productivity to increase private saving by enhancing education outcomes, encouraging innovation, and
improve energy efficiency. Spain’s recovery plan foresees investments and reforms in these areas.
Sweden Stronger Once inflation recedes, increase private and public investment in the green transition and the health sector.
Switzerland Broadly in line Fiscal policy should remain in line with debt-brake rule framework in the near term. As inflation pressures ease, small deficits should support
necessary expenditures; consider targeted FXI to mitigate disruptive volatility.
Thailand Stronger Focus public spending on targeted social transfers as well as infrastructure investment to support a green recovery and reorientation of affected
sectors. Continue the effort to reform and expand social safety nets; implement measures to address widespread informality.
Türkiye Moderately weaker Strengthen the policy framework to help underpin external sustainability. Implement a tighter monetary and fiscal policy stance, and rebuild policy
credibility.
United Kingdom Broadly in line Implement gradual fiscal consolidation, while preserving the quality of key public services and protecting the vulnerable. Implement structural reforms
to boost competitiveness, including via upgrading the labor skill base to support labor reallocation to fast-growing sectors, bolstering national
savings to help finance investment, including in support of the climate transition.
United States Moderately weaker Implement fiscal consolidation over the medium term. Implement structural policies to increase competitiveness, including upgrading infrastructure;
enhancing schooling, training, and mobility of workers; supporting the working poor; and policies to increase growth in the labor force (including
skill-based immigration reform). Roll back tariff barriers and resolve trade and investment disagreements supporting an open, stable, and
transparent global trading system.
Source: 2022 Individual External Balance Assessments.
35
2023 EXTERNAL SECTOR REPORT
References Cuñat, Alejandro, and Robert Zymek. 2023. “Trade Costs and
Adler, Gustavo, and Daniel Garcia-Macia. 2018. “The Stabilizing Imbalances in a Small Open Economy.” Unpublished, Inter-
Role of Net Foreign Asset Returns.” IMF Working Paper 18/79, national Monetary Fund, Washington, DC.
International Monetary Fund, Washington, DC. David, Antonio C., and Carlos Eduardo Gonçalves. 2021. “In
Adler, Gustavo, Kyun Suk Chang, Rui Mano, and Yuting Shao. Search of Lost Time: Examining the Duration of Sudden
2021. “Foreign Exchange Intervention: A Dataset of Public Stops in Capital Flows.” Journal of International Money and
Data and Proxies.” IMF Working Paper 2021/047, Interna- Finance 117 (C).
tional Monetary Fund, Washington, DC. Denbee, Edd, Carsten Jung, and Francesco Paternò. 2016.
Aiyar, Shekhar, Jiaqian Chen, Christian Ebeke, Roberto “Stitching Together the Global Financial Safety Net.” Finan-
Garcia-Saltos, Tryggvi Gudmundsson, Anna Ilyina, Alvar cial Stability Paper 36, Bank of England, London. https://
Kangur, and others. 2023. “Geoeconomic Fragmentation and www.bankofengland.co.uk/financial-stability-paper/2016/
the Future of Multilateralism.” IMF Staff Discussion Note stitching-together-the-global-financial-safety-net.
2023/001, International Monetary Fund, Washington, DC. Eaton, Jonathan, Samuel Kortum, and Brent Neiman. 2016.
Allen, Cian, Camila Casas, Giovanni Ganelli, Luciana Juvenal, “Obstfeld and Rogoff’s International Macro Puzzles: A Quan-
Daniel Leigh, Pau Rabanal, Cyril Rebillard, Jair Rodriguez, titative Assessment.” Journal of Economic Dynamics and Con-
and João Tovar Jalles. 2023. “2022 Update of the External trol 72: 5–23. https://doi.org/10.1016/j.jedc.2016.06.002.
Balance Assessment Methodology.” IMF Working Paper 23/47, Forbes, Kristin J., and Francis E. Warnock. 2012. “Capital Flow
International Monetary Fund, Washington, DC. Waves: Surges, Stops, Flight, and Retrenchment.” Journal
Allen, Cian, Deepali Gautam, and Luciana Juvenal. 2023. of International Economics 88 (2): 235–51. https://doi.
“Currencies of External Balance Sheets.” Unpublished, Inter- org/10.1016/j.jinteco.2012.03.006.
national Monetary Fund, Washington, DC. Forbes, Kristin J., and Francis E. Warnock. 2021. “Capital Flow
Antolín-Díaz, Juan, Ivan Petrella, and Juan F. Rubio-Ramírez. Waves—or Ripples? Extreme Capital Flow Movements since
2021. “Structural Scenario Analysis with SVARs.” Journal of the Crisis.” Journal of International Money and Finance 116:
Monetary Economics 117: 798–815. 102394. https://doi.org/10.1016/j.jimonfin.2021.102394.
Bank for International Settlements (BIS). 2023. “Statistical Goldberg, Linda S., and Signe Krogstrup. 2023. “International
Release: BIS International Banking Statistics and Global Capital Flow Pressures and Global Factors.” NBER Working
Liquidity Indicators at End-December 2022.” Bank for Inter- Paper 30887, National Bureau of Economic Research, Cam-
national Settlements, Basel, Switzerland. bridge, MA. https://www.nber.org/papers/w30887.
Bertaut, Carol, and Ruth A. Judson. 2014. “Estimating U.S. Gourinchas, Pierre-Olivier, and Olivier Jeanne. 2013. “Capital
Cross-Border Securities Positions: New Data and New Flows to Developing Countries: The Allocation Puzzle.”
Methods.” International Finance Discussion Papers 1113, The Review of Economic Studies 80 (4): 1484–1515.
Board of Governors of the Federal Reserve System, International Monetary Fund (IMF). 2015. “Assessing
Washington, DC. Reserve Adequacy–Specific Proposals.” IMF Policy Paper,
Blanchard, Olivier J., and Mitali Das. 2017. “A New Index of Washington, DC.
Debt Sustainability.” NBER Working Paper 24068, National International Monetary Fund (IMF). 2016. “Adequacy of the
Bureau of Economic Research, Cambridge, MA. Global Financial Safety Net.” IMF Policy Paper, Washington,
Bolhuis, Marijn A., Jiaqian Chen, and Benjamin Kett. 2023. DC. https://www.imf.org/en/Publications/Policy-Papers/
“Fragmentation in Global Trade: Accounting for Commod- Issues/2016/12/31/Adequacy-of-the-Global-Financial-Safety-
ities.” IMF Working Paper 23/73, International Monetary Net-PP5025.
Fund, Washington, DC. Joy, Mark, Noëmie Lisack, Simon Lloyd, Dennis Rein-
Boz, Emine, Luis Cubeddu, and Maurice Obstfeld. 2017. “Revisit- hardt, Rana Sajedi, and Simon Whitaker. 2018.
ing the Paradox of Capital.” Vox: CEPR Policy Portal, March 9. “Mind the (Current Account) Gap.” Financial Stabil-
Boz, Emine, Nan Li, and Hongrui Zhang. 2019. “Effective ity Paper 43, Bank of England, London. https://www.
Trade Costs and the Current Account: An Empirical Anal- bankofengland.co.uk/financial-stability-paper/2018/
ysis.” IMF Working Paper 19/08, International Monetary mind-the-current-account-gap.
Fund, Washington, DC. Jordà, Òscar, Moritz Schularick, and Alan M. Taylor. 2017.
Cherif Reda, Fuad Hasanov, and Nikola Spatafora and compris- “Macrofinancial History and the New Business Cycle
ing Rahul Giri, Dimitre Milkov, Saad Quayyum, Gonzalo Facts.” NBER Macroeconomics Annual 31: 213–63. https://
Salinas, and Andrew Warner. 2022. “Industrial Policy for doi.org/10.1086/690241.
Growth and Diversification: A Conceptual Framework.” IMF Lane, Philip R., and Gian Maria Milesi-Ferretti. 2018. “The
Departmental Paper DP/2022/017, International Monetary External Wealth of Nations Revisited: International Financial
Fund, Washington, DC. Integration in the Aftermath of the Global Financial Crisis.”
IMF Economic Review 66 (1): 189–222. https://doi.org/10.1057/ a Common Cause?” NBER Macroeconomics Annual 15:
s41308-017-0048-y. 339–90. https://doi.org/10.1086/654423.
Lane, Philip R., and Jay C. Shambaugh. 2010. “The Long or Perks, Michael, Yudong Rao, Jongsoon Shin, and Kiichi
Short of It: Determinants of Foreign Currency Exposure in Tokuoka. 2021. “Evolution of Bilateral Swap Lines.” IMF
External Balance Sheets.” Journal of International Economics Working Paper 21/210, International Monetary Fund,
80 (1): 33–44. https://doi.org/10.1016/j.jinteco.2009.09.007. Washington, DC. https://www.imf.org/en/Publications/WP/
Lucas, Robert. 1990. “Why Doesn’t Capital Flow from Rich Issues/2021/08/06/Evolution-of-Bilateral-Swap-Lines-463358.
to Poor Countries?” American Economic Review 80 (May): Prasad, Eswar, Raghuram G. Rajan and Arvind Subramanian. 2007.
92–96. “The Paradox of Capital.” Finance & Development 44 (1): 16.
Obstfeld, Maurice. 2017. “Assessing Global Imbal- Reyes-Heroles, Ricardo. 2017. “The Role of Trade Costs in the
ances: The Nuts and Bolts.” IMF Blog, June 26, 2017. Rise of Trade Imbalances.” Working Paper. 2017 Meeting
https://www.imf.org/en/Blogs/Articles/2017/06/26/ Papers 212, Society for Economic Dynamics.
assessing-global-imbalances-the-nuts-and-bolts. Tabova M. Alexandra, and Francis E. Warnock. 2021. “Foreign
Obstfeld, Maurice, and Kenneth Rogoff. 2000. “The Six Investors and US Treasuries.” NBER Working Paper 29313,
Major Puzzles in International Macroeconomics: Is There National Bureau of Economic Research, Cambridge, MA.
US dollar appreciations can inflict sizable negative spill- emerging markets. A large literature has highlighted
overs on emerging markets. Building on the methodology the impact of global financial cycles on economic
of Obstfeld and Zhou (2023), this chapter investigates activity and policy trade-offs and studied the channels
implications of the “global dollar cycle” for the current of transmission (see, for example, Rey 2013; Bruno
account balance and other external sector indicators. It and Shin 2015; and Kalemli-Özcan 2019). A more
finds that negative real sector spillovers from US dollar recent strand of this literature has put the US dollar at
appreciations fall disproportionately on emerging mar- the center of global financial market booms and busts
kets. In contrast, effects on advanced economies are small (see, for example, Druck, Magud, and Mariscal 2018;
and short-lived. Current account balances increase in Shin 2020; Shousha 2022; Akinci and others 2022;
both country groups, with larger and more persistent Obstfeld and Zhou 2023; and Fukui, Nakamura, and
effects on emerging markets, driven by a fall in invest- Steinsson 2023). In particular, Obstfeld and Zhou
ment. Emerging market commodity exporters historically (2023) find that the US dollar is closely related to
experienced larger negative spillovers than commodity global financial conditions even after established factors
importers, reflecting a strong negative link between such as US monetary policy and US domestic financial
the US dollar and commodity prices. More flexible conditions are controlled for, and they link the “global
exchange rates and more anchored inflation expectations dollar cycle” to large negative spillovers to economic
can mitigate negative spillovers to emerging markets. activity in emerging markets, through both financial
and trade channels. Given the US dollar’s dominant
role in global finance, the global dollar cycle is a
Introduction convenient barometer for studying the implications of
During the post–Bretton Woods era of flexible booms and busts in global financial markets, capturing
exchange rates, the US dollar has followed pronounced factors such as changes in investor risk appetite and
decade-long swings. The most recent sharp US dollar preference for liquidity.
appreciation in 2021–22 is part of these oscillations. Building on Obstfeld and Zhou (2023), this chapter
An extensive literature has studied determinants of zooms in on the external sector implications of the
US dollar fluctuations, including contributions from global dollar cycle for a sample of emerging markets
established macroeconomic factors and policies, albeit and small advanced economies. The chapter’s exter-
recognizing their limited explanatory power (see, for nal sector focus is motivated by the centrality of the
example, Frenkel 1976; Dornbusch 1976; Obstfeld and current account for exchange rate–induced macro-
Rogoff 1996; Engel and West 2005; and Gourinchas economic adjustment, capturing the propensity of
and Rey 2007). More recent research has focused on countries to buffer or magnify the impact of US dollar
the close association between the US dollar and global fluctuations. The chapter addresses three questions
financial conditions, with appreciations accompanied pertaining to the global dollar cycle:
by tightening financing constraints (see, for example, •• Are there systematic external sector spillovers from
Rey 2013; and Miranda-Agrippino and Rey 2022). the global dollar cycle?
Policymakers scrutinize strong US dollar episodes •• Do effects differ across countries, and if so, what
closely because of potential negative cross-border explains the heterogeneity?
spillovers and ensuing policy challenges, especially in •• What are the implications for global current account
balances?1
The authors of the chapter are Cian Allen, Rudolfs Bems (lead),
Lukas Boer, Allan Dizioli, and Racha Moussa, under the guidance of
Jaewoo Lee. Abreshmi Nowar and Xiaohan Shao provided research 1Global current account balances are defined as the sum of abso-
support and Jane Haizel editorial assistance. The chapter also ben- lute current account balances across all countries. It is a key metric
efited from comments by Şebnem Kalemli-Özcan, internal seminar in the External Sector Report that can signal increasing financial
participants, and reviewers. vulnerabilities and rising trade tensions (see Chapter 1).
To answer these questions, the chapter studies cur- economic terms: a 10 percent US dollar appre-
rent account determinants, including the behavior of ciation decreases output by 1.9 percent after one
investment and saving, components of trade and capi- year, and the negative effect dissipates only after
tal flows, and foreign asset and liability positions. The 10 quarters. In contrast, the negative effects in
chapter further examines the heterogeneous impact of advanced economies are considerably smaller in
US dollar fluctuations based on countries’ policies and size and short-lived, peaking at 0.6 percent after
structural characteristics, which can shed light on the one quarter.
trade and financial channels of transmission. Given the •• Current account balances, as a share of GDP, increase
lagged nature of the current account response, both in both country groups, but the effect is again
short- and longer-term responses for variables of inter- larger, peaking at 1 percent of GDP for a 10 percent
est are examined. To benchmark findings, the chapter appreciation in the US dollar, and more persistent
contrasts emerging markets with smaller advanced for emerging markets. A depressed investment rate
economies. accompanying the negative spillovers is the main
The chapter estimates cross-border spillovers with a contributor to the current account increase. Exchange
state-dependent local projections (LP) methodology, rate depreciation and accommodative monetary
following Obstfeld and Zhou (2023). To isolate the policy facilitate the external sector adjustment for
role of the global dollar cycle, the analysis simultane- advanced economies, while “fear of floating” and less
ously controls for other established factors influencing accommodation limit the shock-absorbing contribu-
US dollar fluctuations, including monetary policy tion of exchange rates in emerging markets, where
developments, broader US financial conditions, and income compression dominates.
economic activity trends in the rest of the world. •• Among structural characteristics, the chapter finds
Estimated impulse responses are allowed to vary by commodity exposure to be a key contributor to
different characteristics of interest, with commodity spillovers from US dollar appreciations. Com-
exporter or importer status as a key exogenous struc- modity exporters exhibit larger negative spillovers
tural feature. owing to a pronounced deterioration in their terms
Leveraging the close correlation between the global of trade, reflecting a strong negative link between
dollar cycle and uncovered interest parity (UIP) devia- commodity prices and the US dollar, in which
tions, the chapter employs model-based simulations to most commodities are invoiced. The opposite holds
shed light on its empirical findings. Analyzing global for commodity importers. The ensuing economic
risk premium shocks in the Flexible System of Global adjustment has contrasting implications for current
Models (FSGM; Andrle and others 2015) helps disen- account changes: sizable surpluses for commodity
tangle some of the mechanisms behind the chapter’s importers, in contrast to broad balance or even
external sector findings. The model employed in this deficits for commodity exporters.
chapter also provides an interpretation for the link •• Policies can mitigate negative spillovers to emerging
between the global dollar cycle and other key global markets from US dollar appreciations. In line with
variables studied, including commodity prices and Obstfeld and Zhou (2023), the chapter finds that
global trade openness. monetary policy credibility facilitates accommodative
The chapter’s main findings confirm that US dollar policy responses to a US dollar appreciation, includ-
appreciations inflict negative spillovers on emerg- ing through reduced policy rates and real effective
ing markets and expand on this result along several exchange rate (REER) depreciations. The result is a
dimensions: shallower initial negative spillover. Meanwhile, a more
•• Negative spillovers from US dollar appreciations flexible exchange rate regime systematically speeds up
fall disproportionately on emerging markets economic recovery. These mitigating policies moder-
when compared with smaller advanced econo- ate current account increases.
mies.2 Impacts on emerging markets are large in •• The chapter estimates that global current account
balances decline significantly in response to US
2The chapter uses terms such as “US dollar appreciation” and
dollar appreciations, with a 10 percent appreciation
“upswing in the global dollar cycle” interchangeably to refer to an
increase in the value of the US dollar against that of currencies in associated with a decline in global current account
other major advanced economies. balances by 0.4 percent of GDP after one year.
The chapter’s empirical strategy puts some limits on Figure 2.1. Nominal US Dollar Trade-Weighted Index against
the interpretation of the global dollar cycle. The latter Major Advanced Economies
(Index, 100 = January 2006)
is estimated as a residual, potentially containing many
endogenous factors that the chapter does not attempt The US dollar exhibits pronounced decade-long swings, with the recent
to further disentangle. Instead, following established sharp appreciation constituting the most recent strong dollar episode.
practices in macroeconomics, the focus is on the 170
unexplained residuals, that is, the portion of US dollar 161
fluctuations that cannot be attributed to established
152
factors.3 The chapter estimates these residuals and
shows, with the help of model simulations, that global 143
financial market shocks—distinct from other funda-
134
mentals such as interest rate differentials—could con-
tribute to the global dollar cycle. However, the analysis 125
does not preclude other interpretations, which could 116
be made possible by further advances in analyzing the
107
drivers of US dollar fluctuations.
98
89
Characterizing the Global Dollar Cycle
This section links fluctuations in a US dol- 80
1973:Q1 81:Q1 89:Q1 97:Q1 2005:Q1 13:Q1 22:Q4
lar index4 to contributing factors, distinguishing
between established exchange rate determinants Sources: Federal Reserve Bank of St. Louis, Federal Reserve Economic Data
(FRED); and Haver Analytics.
and a residual contribution from global financial Note: Series retrieved from Haver Analytics, based on the Nominal Advanced
factors. The latter contribution—the global dollar Foreign Economies US Dollar Index from FRED, using goods and services trade
weights. Values before 2006 are constructed with services trade data estimates
cycle—is then related to other financially motivated from the Federal Reserve Board. Index constructed as the trade-weighted average
indicators, including UIP deviations and the global against the currencies of seven major advanced economies: Australia, Canada,
financial cycle. euro area, Japan, Sweden, Switzerland, and the United Kingdom.
effective, and real effective, in terms of foreign currency per US 6To boost the sample of advanced economies, those with weights
dollar, so that an increase represents an appreciation of the US dollar in the US dollar index smaller than 4 percent in 2020 are included
and a depreciation of the foreign currency (or a basket of currencies, in the sample for spillover estimates. Results are robust to excluding
in the case of an effective exchange rate). such countries from the sample (Online Annex 2.3).
would also be accounted for. The specific explanatory Figure 2.2. The US Dollar Index and the Global Dollar Cycle
(Index, 0 = 1999:Q4)
variables used to capture interest rate developments
are (1) policy rates, including shadow rates7; (2) dif- The global dollar cycle closely tracks movements in the US dollar
ferences between US policy rates and those of major trade-weighted index against the currencies of advanced economies.
advanced economies; and (3) an index for US finan- 20
US dollar index
cial conditions to capture longer-term interest rate Global dollar cycle
developments. The estimation further controls for 10
(4) a common component of economic activity in the
rest of the world and (5) the lagged change in the US
0
dollar index.8 Established factors are found to exhibit
expected relationships with the US dollar index. For
example, a tightening of measured financial condi- –10
tions is associated with a US dollar appreciation, as
is an increase in the policy rate differential in the –20
United States with respect to that in other advanced
economies. With the US financial conditions index
–30
and the lagged change in the US dollar index making
the largest contributions, established factors alto-
gether explain about one-fifth of US dollar fluctua- –40
2000:Q1 05:Q1 10:Q1 15:Q1 20:Q1
tions (see Online Annex Table 2.2.1).
However, a significant unexplained residual in Sources: Federal Reserve Bank of St. Louis, Federal Reserve Economic Data;
Haver Analytics; and IMF staff calculations.
the estimation—labeled “global dollar cycle” in this Note: Global dollar cycle constructed as cumulated residuals after established
chapter—remains. This unexplained residual accounts factors are controlled for (1) monetary policy, (2) policy rate differences with major
for the bulk of US dollar fluctuations over the last two advanced economies, (3) US financial conditions, and (4) an economic activity
factor.
decades (Figure 2.2). Its correlation with the US dollar
index is 84 percent. Zooming in on recent years, the
exchange rate movement attributable to established interest rates, and the addition of commodity market
factors, represented in Figure 2.2 by the difference developments.9
between the US dollar index and the residual global Recent literature views this residual as reflecting
dollar cycle, closely traces exchange rate fluctuations global financial market forces. With the rise of finan-
during the 2020–21 pandemic-related downturn and cial globalization, the literature has focused on the role
recovery, but the global dollar cycle accounts for a of global financial markets in driving and magnifying
sizable portion of the sharp US dollar appreciation in exchange rate fluctuations, as captured by, for example,
2022. Extensive robustness tests, results of which are
reported in Online Annex 2.4, show that the estimated 9However, the analysis refrains from directly including commodity
role of the global dollar cycle is broadly unchanged prices or the terms of trade as explanatory variables, as the global
under a wide variety of alternative specifications of dollar cycle (as proxied by risk premium shocks) can be an import-
ant driver of commodity prices and, hence, the terms of trade. This
explanatory variables, including alternative series channel is confirmed in the FSGM simulations. Furthermore, the
for monetary policy shocks, alternative horizons for focus on the US dollar index against currencies of major advanced
economies weakens the applicability of the commonly held assump-
tion that the terms of trade in a small open economy can be treated
as exogenous. To account for commodity market developments,
robustness tests instead consider global commodity supply shocks,
7Shadow rates used are the Wu-Xia shadow federal funds rate proxied with oil supply shocks (Baumeister and Hamilton 2019),
(Wu and Xia 2016) for the United States; the LJK Limited shadow as an additional explanatory variable. Results in Online Annex 2.4
rates for Australia, Canada, euro area, Japan, Switzerland, and show that in historical data this variable has only a marginal explan-
the United Kingdom (Krippner 2015); and the shadow rate from atory power. However, this finding does not preclude the possibility
De Rezende and Ristiniemi (2023) for Sweden. that commodity price surges in 2021–22, linked to recovery from
8Online Annex 2.2 reports details on the empirical specification the COVID-19 pandemic and Russia’s war in Ukraine, have contrib-
and regression results of this exercise. uted to the strong US dollar.
the portfolio-balance approach to capital flows and Table 2.1. Correlates of the Global Dollar Cycle
exchange rates (see, for example, Gabaix and Maggiori Comparison of the global dollar cycle with other global financial
indicators reveals the strongest correlation with uncovered interest
2015) and renewed interest in the exchange rate dis- parity deviations and the global financial cycle.
connect puzzle (Itskhoki and Mukhin 2021). The liter- Indicator Correlation
ature also emphasizes the unique role of the US dollar
Uncovered interest parity deviations from 0.69*
in global financial markets, linked to safe-haven and major advanced economy currencies
liquidity considerations. Financial markets can be a Global financial cycle −0.53*
key transmission channel through which conventional Chicago Board Options Exchange Volatility 0.04
macroeconomic shocks and policies (such as monetary Index (VIX)
tightening) affect the exchange rate (see, for example, Global uncertainty index 0.09
Miranda-Agrippino and Rey 2022; and Kalemli-Özcan Sources: Consensus Economics; Davis (2016); Federal Reserve Board;
2019). Perhaps more important, financial markets Haver Analytics; Miranda-Agrippino, Nenova, and Rey (2020); Refinitiv
Datastream; and IMF staff calculations.
can also be a source of financial shocks that trigger
Note: Quarterly correlations over 2000:Q1–22:Q4 depending on data availability
short- and longer-term exchange rate fluctuations. An (global financial cycle variable ends in 2019:Q2).
example would be a decrease in investor risk appetite “ * ” indicates the correlation is significant at the 1 percent level.
and resulting appreciation of a safe-haven currency,
such as the US dollar.10 A notable empirical challenge expected rate of exchange rate depreciation12; that is,
for studying the role of financial markets is that the US dollar appreciations coincide with expected dollar
underlying financial shocks that have an impact on depreciations, while cross-border interest rate dif-
the US dollar are not directly observable. The chap- ferentials vary relatively less.13 Zooming in on UIP
ter addresses this issue by resorting to identification deviations of individual advanced economy curren-
by exclusion, linking financial market forces to the cies reveals comparable positive correlations for all
residual not explained by established exchange rate currencies except the Japanese yen and Swiss franc,
determinants. which could reflect safe-haven considerations.
The global dollar cycle can be related to other finan- •• The global dollar cycle shows a strong negative cor-
cial indicators. The chapter examines several measures: relation with the global financial cycle, emphasized
•• An index of UIP deviations is found to be strongly by Bruno and Shin (2015) and Miranda-Agrippino
positively correlated (69 percent) with the global and Rey (2022) (Table 2.1). The global finan-
dollar cycle.11 During episodes of US dollar appre- cial cycle is the global common factor estimated
ciations, investments in currencies of other major from a worldwide cross-section of risky asset
advanced economies carry excess returns relative to prices, covering equity, bonds, and commodities
US dollar investments, stemming from decreased (Miranda-Agrippino, Nenova, and Rey 2020).
risk appetite for other advanced economies, and the Tightening of financial conditions, as captured by a
opposite is true when the US dollar depreciates. A downswing in the global financial cycle, accompa-
statistical decomposition reveals that most move- nies an upswing in the global dollar cycle.
ments in UIP deviations are associated with the
12UIP deviations, λi , for a foreign currency i against the US dollar,
t
10Examples of recent studies that examine financial market shocks capturing excess returns on the foreign currency, can be statistically
include Itskhoki and Mukhin (2021); Devereux, Engel, and Wu decomposed into changes in the interest rate differential between the
(2023); and Lilley and others (2022). yields on comparable assets (term in bold) and an expected exchange
11UIP is an arbitrage condition in international financial markets rate adjustment (bracketed term), expressed as
stating that the difference in interest rates between two countries will λti = iti − itUS − (ln(E(St+k
LC/$
)) − ln(StLC/$)),
equal the expected relative change in the exchange rates over the cor-
responding horizon. Deviations from UIP indicate excess returns in where iti is the interest rate in country i, ln denotes the natural
one market, which in the case of US dollar fluctuations could stem logarithm, StLC/$ the nominal exchange rate expressed in terms of local
from frictions in global financial markets. Online Annex 2.3 reports currency per US dollar, and E(St+k )
LC/$ the expectation of the exchange
UIP deviations for individual currencies, along with index construc- rate k periods out (the same horizon as the interest rate maturities).
tion details. Bilateral deviations of advanced economies included in 13See Online Annex Figure 2.3.2. This is in contrast to UIP devia-
the US dollar index are aggregated using trade weights to arrive at a tions in emerging markets, which are predominantly associated with
measure that can be directly compared with the global dollar cycle. changes in interest rates (Kalemli-Özcan 2019).
•• The Chicago Board Options Exchange Volatility of UIP deviations—in a general equilibrium model
(VIX) Index—a measure of US stock price volatility can help provide deeper understanding of the channels
and one of the components of the global finan- through which spillovers from the global dollar cycle
cial cycle—has been explored by the literature (di to emerging markets operate. Second, constructed UIP
Giovanni and others 2022; Obstfeld, Ostry, and deviations offer an alternative source of global financial
Qureshi 2019) as an indicator of global risk senti- market shocks whose spillovers to emerging markets
ment, but does not correlate significantly with the can be estimated (see Online Annex Figure 2.4.5 for
global dollar cycle for the period of investigation, details). The chapter explores both avenues.
although a somewhat stronger correlation is present
for subperiods. This is due to an already low cor-
relation of 0.2 between the VIX and the US dollar Empirical Analysis: Spillovers from the
index during our sample period and the fact that Global Dollar Cycle
the VIX is one among a large set of factors of the This section examines the differential impact of US
US financial conditions index for which the global dollar appreciations on emerging market and advanced
dollar cycle controls. economies, explores the contribution of policies and
•• Finally, the global uncertainty index from Davis structural features to negative spillovers to identify
(2016), which is another news-based indicator of potential channels of transmission, and examines the
global financial distress, shows only a weak positive impact of fluctuations in the US dollar index on global
correlation with the global dollar cycle. current account balances.
to these global controls, the specification includes a Figure 2.3. Spillovers from a US Dollar Appreciation:
set of lagged country-specific controls—GDP growth, Macro Aggregates
(Percent change)
the policy rate, and the bilateral exchange rate against
the US dollar—as well as lags of the global control A US dollar appreciation affects emerging markets more adversely than
variables, the change in the US dollar index, and the advanced economies.
dependent variable. Lastly, as the sample of countries Advanced economies Emerging markets
includes potentially heterogeneous smaller advanced 1. Real GDP 2. Real Investment
10 10
and emerging market economies, the empirical frame-
work estimates state-dependent LP, following Ramey 5 5
and Zubairy (2018), allowing for differential responses
for sets of countries split by policy and structural char- 0 0
acteristics.16 Overall, this empirical specification makes
–5 –5
it possible to interpret the estimated impulse responses
as spillovers from the global dollar cycle discussed in –10 –10
the previous section.
–15 –15
0 4 8 12 0 4 8 12
Quarter Quarter
Spillovers to Advanced and Emerging Market Economies
Negative spillovers from a US dollar appreciation are 3. Real Imports 4. Real Exports
10 10
concentrated in emerging markets. Emerging markets
experience a deeper and longer-lasting contraction 5 5
than advanced economies (Figure 2.3, panel 1). An
appreciation of the US dollar index by 10 percentage 0 0
points is associated with a decline in real output by
–5 –5
1.9 percent in emerging markets and 0.5 percent in
advanced economies 2 quarters after the initial appre- –10 –10
ciation. Output in advanced economies recovers 3
quarters after the appreciation, while emerging market –15 –15
0 4 8 12 0 4 8 12
output remains depressed 10 quarters out. An outsized Quarter Quarter
decline in real investment in emerging markets drives
the differential impact on output (Figure 2.3, panel 2). Source: IMF staff calculations.
Note: Impulse responses show a 10 percent appreciation in the nominal US dollar
Trade volumes decline disproportionately more than index with 90 percent confidence intervals. Macro aggregates are measured in
economic activity for both country groups, with the national currencies at constant prices. Advanced economies exclude countries
with weights in the US dollar index that are larger than 4 percent in 2020: Canada,
magnitude of the decline in imports roughly double France, Germany, Ireland, Italy, Japan, Switzerland, and the United Kingdom.
the decline in exports (Figure 2.3, panels 3 and 4).
The chapter’s estimated large negative real spillovers
for emerging markets confirm the findings in Obstfeld economic terms: a 10 percent appreciation in the
and Zhou (2023) and are consistent with results of US dollar increases the current account after five
several other recent studies, including Druck, Magud, quarters by about 1 percent of GDP in emerg-
and Mariscal (2018), Shousha (2022), and Fukui, ing markets and 0.7 percent of GDP in advanced
Nakamura, and Steinsson (2023). economies. Further analysis from the saving-invest-
In response to US dollar appreciations, the cur- ment perspective, linking the current account to
rent account, as a share of GDP, increases in both changes in investment and saving rates, all expressed
emerging markets and smaller advanced econo- in percent of GDP, reveals that a decline in invest-
mies. Mimicking output responses, the impact is ment drives the current account increases around
larger and more persistent for emerging markets one year out in both country groups (Figure 2.4,
(Figure 2.4, panel 3). The impact is sizable in panels 1 and 2). Investment is also the main driver
of the divergent longer-term current account
16Online Annex 2.2 reports details of the regression specification. response, recovering strongly in advanced economies
Figure 2.4. Spillovers from a US Dollar Appreciation: group, the REER depreciates persistently on impact,
External Sector Variables allowing the expenditure switching channel to con-
When the US dollar appreciates, the current account increases in both
tribute to the external sector adjustment (Figure 2.4,
emerging market and advanced economies, but through distinct panel 4). Subsequent analysis of the role of exchange
channels, as investment is persistently depressed in emerging markets rate flexibility (see Online Annex Figure 2.4.2) further
but recovers quickly in advanced economies. Because of “fear of highlights the benefits stemming from the shock-ab-
floating,” in emerging markets income compression drives the fall in
imports and the external adjustment, while in advanced economies sorbing role that the exchange rate plays in response to
depreciation in the real effective exchange rate (REER) and the resultant US dollar appreciations. By contrast, in emerging mar-
expenditure switching facilitates the adjustment. kets the REER does not respond to a US dollar appreci-
Advanced economies Emerging markets ation on impact, consistent with well-documented fear
1. Saving 2. Investment of floating for this country group and depreciates only
(Percent of GDP) (Percent of GDP)
1.5 1.5 gradually over subsequent quarters.17 In the absence of
an exchange rate adjustment, income compression plays
1.0 1.0
an outsized role, driving a large fall in imports.18
0.5 0.5 Net trade in goods and services contributes dif-
0.0 0.0 ferently to external sector adjustment in advanced
economies and emerging markets. Detailed gross
–0.5 –0.5
and net trade flow responses reveal that in advanced
–1.0 –1.0 economies, where (as noted) the REER depreciates on
impact, the current account increase is driven mainly
–1.5 –1.5
0 4 8 12 0 4 8 12 by an increase in the services trade balance and, in
Quarter Quarter particular, a boost to service exports, as a share of
3. Current Account 4. REER
GDP (see Online Annex Figure 2.4.1). In emerging
(Percent of GDP) (Percent change) markets, where (again, as noted) the REER does not
2 10 adjust on impact, the current account increase is
driven mainly by a fall in imports of goods, as a share
of GDP, consistent with the income compression
1 5
channel.19
Financial transmission channels magnify the adverse
0 0
spillovers in emerging markets. Contemporaneously
with the US dollar appreciation, capital inflows to
emerging markets, both private and public, decline (see
–1 –5 Figure 2.5, panels 1 and 2).20 There is also evidence
0 4 8 12 0 4 8 12
of systematic negative valuation effects impacting the
Quarter Quarter
net international investment position (NIIP) over the
Source: IMF staff calculations. examined horizon, as NIIP does not increase despite
Note: Impulse responses show a 10 percent appreciation in the nominal US dollar
index with 90 percent confidence intervals. An increase in the REER is a
17Fear of floating here, as well as in subsequent estimation results,
depreciation. Advanced economies exclude countries with weights in the US dollar
index that are larger than 4 percent in 2020: Canada, France, Germany, Ireland, is applied in a more expansive manner to refer to all non-floating
Italy, Japan, Switzerland, and the United Kingdom. exchange rate regimes. However, this emerging market REER
response is not driven by the sample’s limited number of pegged
exchange rate observations.
18The expenditure switching channel is further hindered by the
but remaining depressed for emerging markets. US dollar invoicing in trade, which is more prevalent in emerging
markets (see Online Annex Table 2.4.1 and Gopinath and others
Meanwhile, saving does not reveal a clear systematic
2020).
response or differences between the two country 19The fall in imports of goods is observed in all broad economic
groups, except for a contemporaneous significant categories, including capital goods, intermediate consumption goods,
but short-lived drop in emerging markets. and final consumption goods.
20Private and public inflows are normalized by lagged foreign
Exchange rate depreciation facilitates external sector liabilities to account for the differences in financial integration across
adjustment in advanced economies. For this country countries.
Advanced economies continue to borrow externally after US dollar appreciations and implement countercyclical monetary policy to mitigate
negative spillovers. Neither of these channels operates in emerging markets.
Advanced economies Emerging markets
1. Private Inflows 2. Public Inflows 3. Net International Investment Position
(Percent of lagged foreign liabilities) (Percent of lagged foreign liabilities) (Percent of GDP)
4 4 8
6
2 2 4
2
0 0
0
–2 –2 –2
–4
–4 –4 –6
0 4 8 12 0 4 8 12 0 4 8 12
Quarter Quarter Quarter
persistent current account surpluses (see Figure 2.5, reveal no systematic pattern and are even procyclical
panel 3). These findings contrast with advanced econo- on impact.21 Domestic credit declines persistently,
mies, where the NIIP increases, driven by both current extending beyond the 12-quarter horizon. Stock prices
account surpluses as well as an initial positive valuation decline by more in emerging markets than in advanced
effect stemming from the US dollar appreciation. Fur- economies (see Figure 2.5, panel 6). These findings are
thermore, public capital inflows to advanced econ- broadly consistent with an extensive literature that has
omies increase, smoothing the impact of the global focused on financial transmission channels of global
dollar cycle. In terms of domestic financial conditions financial shocks to emerging markets (see, for example,
and policies, in advanced economies US dollar appre- Gourinchas 2018; di Giovanni and others 2022; and
ciations are systematically associated with accommo- Kearns and Patel 2016).
dative monetary policy, mitigating negative spillovers.
Accordingly, the decline in domestic credit is shallow
and short lived (see Figure 2.5, panels 4 and 5). In 21Using short-term interest rates instead of policy rates yields
contrast, policy rate responses in emerging markets similar findings (De Leo, Gopinath, and Kalemli-Özcan 2023).
The Role of Policy Regimes and Structural one another, making it difficult to identify individual
Characteristics impacts on spillovers. An instructive example is the
To investigate why emerging markets experience relation between exchange rate regimes and the extent
larger negative spillovers than advanced economies, this of US dollar invoicing of exports; countries with float-
section analyzes how US dollar appreciation differ- ing exchange rate regimes disproportionately exhibit
entially affects economies based on their policies and low shares of US dollar invoicing, while countries with
structural characteristics. For each factor considered, less flexible exchange rate regimes exhibit high shares
the analysis estimates state-dependent responses based of US dollar invoicing (Figure 2.6, panel 2). Another
on a sample split into two corresponding subgroups, important example in this regard relates to commod-
mirroring the estimation procedure for the whole ity-exporting status. Categorization results reveal that
sample. The set of examined factors is motivated by commodity-exporting countries are disproportionately
the commonly studied policies at countries’ disposal associated with less flexible exchange rate regimes and
and structural characteristics impacted by US dollar lower trade openness, as well as higher shares of US
fluctuations, including commodity prices and financial dollar invoicing in exports and US dollar liabilities.22
and trade exposures to the US dollar (see Table 2.2). The chapter uses commodity exporter or importer
Identifying contributions to spillovers from individ- status as a key exogenous structural feature. Using
ual country characteristics presents several challenges. commodity exporter status avoids problems arising
First, the examined characteristics are closely correlated from the fact that most characteristics are endogenous,
with the split of the sample between emerging market collinear, or both, which complicates identification.
and advanced economies. The issue is most striking for Moreover, this status is slow moving over the study’s
the US dollar liability exposure and the extent of mon- time frame and should arguably not respond to poli-
etary policy anchoring, where, based on categorization cies and other structural features. The contribution of
in Table 2.2, all of the more exposed and less anchored other characteristics is then estimated, after the role
countries are found among emerging markets. Hence, of commodity exporter status is controlled for. Where
any identification of these characteristics’ contribution overlap with the split in the sample between advanced
to spillovers requires limiting the sample to emerging and emerging market economies is severe, estimation
markets. This issue is a concern for the other examined is limited to the emerging market sample. Monetary
characteristics as well, except commodity exporter policy credibility is found to be the least correlated
status, which is more evenly distributed within the
two country groups (Figure 2.6, panel 1). Second, 22Online Annex Table 2.4.1 details the country composition of
many of the characteristics are closely correlated with each examined policy and structural feature.
Country characteristics are closely correlated with the split in the sample between advanced economies and emerging markets, complicating the
identification of contributions to differential spillovers from a particular characteristic. Some country characteristics are closely correlated with each
other, further complicating the identification of the role of an individual characteristic.
1. Share of a Category in a Characteristic by Country Group 2. Sample Split Between Categories of Exchange Rate Regime and
(Percent) US Dollar Export Invoicing
80
Within emerging markets Within advanced economies US dollar export US dollar export
invoicing below invoicing above
70 75 percent of exports 75 percent of exports
60
Other ARG; CHL; COL;
50 exchange ISR; IDN; IND;
rate KOR; MYS; PAK;
40 regime PHL; RUS; THA
30
20
AUT; BEL; CZE;
Freely
DNK; ESP; FIN;
10 floating
GRC; HUN; NLD;
exchange AUS; BRA
NOR; NZL; POL;
rate
0 PRT; ROU; SWE;
regime
Other Less open Commodity Less- High High TUR; ZAF
exchange (trade) exporter anchored US dollar US dollar
rate regime inflation export liability
(not freely expectations invoicing exposure
floating)
Sources: Bems and others (2021); Bénétrix and others (2019); Boz and others (2022); Ilzetzki, Reinhart, and Rogoff (2019); IMF, Annual Report on Exchange
Arrangements and Exchange Restrictions; IMF, Balance of Payments Statistics; IMF, Global Data Source; UN, Comtrade; and IMF staff calculations.
Note: Advanced economies exclude countries with weights in the US dollar index that are larger than 4 percent in 2020: Canada, France, Germany, Ireland, Italy,
Japan, Switzerland, and the United Kingdom. Countries that are not freely floating that are anchored to a currency other than the US dollar, that is freely floating
against the US dollar, are classified as freely floating. Data labels in the figure use International Organization for Standardization (ISO) country codes. In panel 2, a
country is in the freely floating exchange rate regime category if it spent any part of the sample period in that category.
with commodity-exporting or -importing status and is (Figure 2.7, panel 2). In contrast, policy rates increase
thus studied separately.23 in emerging markets with less anchored monetary pol-
Monetary policy anchoring mitigates negative icy, though with only marginal statistical significance,
spillovers from US dollar appreciations by facilitating and the REER appreciates, rather than depreciating
accommodative policy responses. Emerging markets on impact, thereby contributing to larger negative
with more anchored inflation expectations exhibit spillovers.
a shallower initial decline in output. The difference In response to US dollar appreciation, commodity
between emerging markets with more and those with exporters exhibit larger negative spillovers owing to
less anchored inflation expectations is statistically sig- concurrent deterioration in their terms of trade.24 The
nificant (Figure 2.7, panel 1). When inflation expec- magnitude of the terms-of-trade deterioration is sizable
tations are anchored, the REER depreciates, and the and persistent, with a 10 percent US dollar apprecia-
policy rate becomes more accommodative (Figure 2.7, tion decreasing the terms of trade by 10 percent after
panels 3 and 4). Credibility of monetary policy limits five quarters (Figure 2.8, panel 2). On the flip side,
imported inflation (not shown) and thus creates room the terms of trade improve for commodity import-
for these policy adjustments, which support investment ers. These contrasting terms-of-trade responses drive
rate in the aftermath of the US dollar appreciation the difference in spillovers between the two country
groups. Commodity exporters smooth the temporary
23Controlling for commodity-importing or -exporting status does
not change the chapter’s findings with respect to the role of mone- 24A country’s terms of trade are defined as the ratio of its export
tary policy credibility for spillovers from US dollar appreciations. prices to its import prices.
Figure 2.7. Spillovers from a US Dollar Appreciation by among commodity exporters (Figure 2.8, panel 7).25
Degree of Anchoring of Inflation Expectations Overall, the strong negative link between the US dollar
In the aftermath of a US dollar appreciation, investment remains stable in
and commodity prices is an important cross-border
countries with more anchored monetary policy, contributing to a transmission channel for the negative spillovers. The
shallower decline in output. More accommodative exchange rate and importance of this channel is further highlighted by
interest rate responses contribute to more muted negative spillovers. the 2021–22 strong US dollar episode, which was
Above-median anchoring Below-median anchoring
accompanied by a commodity price surge, rather than
1. Real GDP 2. Investment
(Percent change) (Percent of GDP)
a decrease, driven by the unique nature of the pan-
4 1 demic recovery and commodity supply disruptions
stemming from Russia’s war in Ukraine. An event
2
0 study of this episode, presented in Box 2.1, reveals that
0 the commodity price surge significantly muted, or even
–2 –1 reversed, the negative spillovers from the US dollar
appreciation for commodity-exporting countries.
–4
–2 For commodity importers, the improvement in the
–6 terms of trade partially offsets the negative spillovers
–8 –3
from a US dollar appreciation. The decline in out-
0 4 8 12 0 4 8 12 put is shallower and the REER and monetary policy
Quarter Quarter
further buffer the impact of the negative shock. The
3. Real Effective Exchange Rate 4. Policy Rate current account increase is magnified, as the initial
(Percent change) (Percentage point change) fall in investment (not shown) is accompanied by a
15 2 significant increase in saving from the fifth quarter
1
onward, leading to a gradual improvement in the NIIP
10 (Figure 2.8, panel 5).
0 Among other examined country characteristics,
5 exchange rate flexibility is found to significantly
–1 impact output spillovers, after the influence of com-
0
–2
modity trade is accounted for. In support of the
shock-absorbing properties of flexible exchange rates,
–5 –3 emerging markets with freely floating exchange rate
0 4 8 12 0 4 8 12
Quarter Quarter
regimes exhibit systematically faster recoveries in output
than emerging markets with less flexible exchange rates
Source: IMF staff calculations. (see Online Annex 2.4 and Online Annex Figure 2.4.2).
Note: Emerging markets sample only. Inflation expectations are anchored when
the country average of the measure in Bems and others (2021) is above the
Current account balances in the latter country group
sample median. Impulse responses show a 10 percent appreciation in the nominal show a larger increase, as both saving increases and
US dollar index with 90 percent confidence intervals. An increase in the real investment falls. However, a floating exchange rate
effective exchange rate is a depreciation.
Figure 2.8. Spillovers from a US Dollar Appreciation by Net Commodity Exporter Status
Commodity exporters are hard hit by a US dollar appreciation as a result of a concurrent deterioration in their terms of trade. On the flip side, the
terms of trade improve in commodity importers, which helps counter the effect of the appreciation.
Commodity importers Commodity exporters
1. Real GDP 2. Terms of Trade 3. Current Account 4. Trade Balance: Goods
(Percent change) (Percent change) (Percent of GDP) (Percent of GDP)
4 5 2 2
2
0 1 1
0
–5 0 0
–2
–10 –1 –1
–4
–6 –15 –2 –2
0 4 8 12 0 4 8 12 0 4 8 12 0 4 8 12
Quarter Quarter Quarter Quarter
5. Net International Investment 6. Real Effective Exchange Rate 7. Policy Rate 8. Saving
Position (Percent change) (Percentage point change) (Percent of GDP)
(Percent of GDP)
10 12 1.5 2
10 1.0 1
5 8
0.5
6 0
0 0
4 –1
–0.5
–5 2
–1.0 –2
0
–10 –2 –1.5 –3
0 4 8 12 0 4 8 12 0 4 8 12 0 4 8 12
Quarter Quarter Quarter Quarter
regime might not be an option readily available to all credibility is motivated by concerns about identifi-
countries. Emerging market economies with severe cation of conditional impulse responses to US dollar
financial frictions and balance sheet vulnerabilities fluctuations among sample countries of this chapter,
should resort to complementary policy tools, such as which examines aggregate data. It should not be
macroprudential measures and capital flow management interpreted as evidence that other policies or structural
measures, which can play a useful role in mitigating features do not affect spillovers from US dollar fluctua-
negative cross-border spillovers under limited exchange tions to emerging markets.
rate flexibility (see IMF 2020). For such emerging mar-
ket economies, adopting flexible exchange rate regimes
and benefiting from their shock-absorbing properties Implications for Global Balances
would have to wait until preexisting structural vulner- Beyond negative cross-border spillovers, US dol-
abilities are overcome, including by strengthening the lar appreciations are associated with a compression
domestic financial market and policy framework. of global balances. To estimate the impact on global
Finally, the focus of this section on commodity balances, a time-series LP exercise is applied, similar
exporter or importer status and monetary policy to the panel approach used in this section to estimate
Figure 2.9. Impact of a US Dollar Appreciation on of global balances resulting from US dollar appreci-
Global Balances ations is also consistent with Gopinath and others
(Percent of GDP)
(2020), who link a stronger US dollar with lower trade
An increase in the US dollar index leads to a sustained decrease in flows in the presence of dominant currency pricing.
global balances. This effect can be further amplified when US dollar
0.5 appreciation tightens collateral constraints for importers
that borrow in US dollars (Casas, Meleshchuk, and
Timmer 2022).
0.0
•• UIP: Deviations from UIP in the model are based on UIP deviations as the primitive exogenous shock in the
risk premiums.27 Different borrowers (households, FSGM and studies their implications for cross-border
firms, government) in the model face varying inter- spillovers and key global variables, drawing parallels
est rates depending on their time horizons and risk with empirical findings of the previous section. There
profiles. The UIP condition holds in the short term are different ways to introduce UIP deviations into the
only for the sovereign, and only if the sovereign risk model. The one that most closely links to the chapter’s
premium is set to zero. However, the calibrated model empirical findings is a global (excluding the United
has a nonzero exogenous sovereign risk premium and States) disturbance to sovereign spreads, so that the
a term premium on long-term bonds. More generally, direct effect of the disturbance is an increase in financ-
a UIP equation holds when all risk premiums are ing costs for firms and households.28
accounted for. The model includes an endogenous cor- Figure 2.10 plots impulse responses for key vari-
porate risk premium, which depends on the business ables of interest to this global persistent 1 percentage
cycle and on commodity prices. The sovereign risk point shock to the sovereign premium, reported in
premium affects all interest rates in the model, while the figure’s panel 1. To facilitate comparison with the
the corporate risk premium affects only those for the empirical findings, results for the G20 economies
private sector. Risk premiums vary across private sector distinguish between an aggregated region of advanced
borrowers because shocks affect the cost of financing economies, excluding the United States, and an
differently or can apply to different borrowers. aggregated region of emerging markets, with some
•• Commodity exposure: Data-driven calibration makes results further distinguishing between emerging market
the FSGM particularly well suited to examining the commodity exporters and importers.
differential impacts of economic disturbances on One of the direct effects of the sovereign premium
commodity exporters and importers. The FSGM shock is a US dollar appreciation. The shock increases
incorporates three types of commodities: oil, food, the demand for US dollars by reducing risk-free returns
and metals and their associated prices. The model is on foreign bonds (short-term interest rates do not
calibrated using countries’ commodities production, immediately change, and the risk premium increases)
consumption, and trade. Commodities are priced in and creating an incentive to invest in US bonds absent
the dominant currency: the US dollar. changes in the policy rate (Figure 2.10, panel 2).29
•• External sector: Foreign and domestic economic Another direct effect is an increase in financing costs,
activity and the exchange rate determine exports and which leads to a reduction in domestic consumption,
imports, with producer pricing assumed. Investment through the channel of intertemporal substitution, as it
decisions of firms, saving decisions of households, becomes more costly to borrow to smooth out con-
and fiscal policy determine the current account and sumption. The increase in financial costs also lowers
implied net-foreign-asset positions. investment, and the combined result is a fall in output
in the rest of the world (see Figure 2.10, panel 3).30
Thus, the modeled global risk premium shock generates
Simulation Setup and Model Results
The chapter’s analysis of the global dollar cycle 28Consistent with empirical literature (Kalemli-Özcan 2019)
documents its strong association with UIP deviations, and findings in Online Annex 2.3, FSGM simulations show that
suggesting that economic disturbances driving UIP exchange rate adjustment contributes more to UIP deviations in
advanced economies than in emerging markets; as in the latter coun-
deviations contribute to the cycle. This section takes the try group, the examined global risk premium shock endogenously
triggers other mechanisms that increase the cost of capital, including
27At the normative level, there are two distinct approaches for mod- through lower commodity prices tightening financing conditions.
eling UIP deviations, with differing implications for policy, one based 29To facilitate comparison with the empirical findings, the
on risk premiums and the other on intermediary frictions. The former figure reports US dollar index against currencies of other advanced
approach builds on nondiversifiable risk or reduced appetite for risk economies, but the US dollar appreciation is broad based. Central
but does not feature price distortions. By contrast, the latter approach banks in advanced economies react to the increase in financing cost
is based on market distortions, as intermediaries require rents to absorb by easing policy rates, which contributes to a further US dollar
risk (see, for example, Gabaix and Maggiori 2015), with a potential appreciation.
role for policy. The semistructural FSGM does not feature financial 30Fiscal automatic stabilizers are allowed to operate and partially
intermediaries, so that UIP deviations are a proxy for risk premiums. cushion the negative effects on activity.
Figure 2.10. Impulse Responses to a Global Risk Premium Shock in the Flexible System of Global Models
The Flexible System of Global Models’ response to a global sovereign risk premium shock reveals that a US dollar appreciation is accompanied by
(1) a fall in output in the rest of the world, with a more negative impact on emerging markets; (2) a fall in commodity prices; and (3) a contraction
in trade openness, while (4) the current account increases in commodity-importing countries. These model results are consistent with empirical
findings for spillovers from US dollar appreciations.
1. Sovereign Risk Premium in the Rest of 2. US Nominal Effective Exchange Rate against 3. Real GDP in the Rest of the World
the World Advanced Economies (Percent change)
(Percentage point difference) (Percent change)
1.2 1.6 0.0
1.4 Advanced economies
1.0
1.2 Emerging markets –0.2
0.8
1.0
0.6 0.8 –0.4
0.6
0.4
0.4 –0.6
0.2
0.2
0.0 0.0 –0.8
–1 0 1 2 –1 0 1 2 –1 0 1 2
Year Year Year
4. Global Real Oil Price 5. World Trade Openness 6. Current Account Balance in the Rest of
(Percent change) (Percentage point of GDP change) the World
(Percentage point of GDP change)
0.0 0.0 Advanced economies 0.2
–0.5 Emerging market commodity
importers
–1.0
–0.2 0.1
–1.5
–2.0
–2.5
–0.4 0.0
–3.0 Emerging market commodity
–3.5 exporters
–4.0 –0.6 –0.1
–1 0 1 2 –1 0 1 2 –1 0 1 2
Year Year Year
the empirically observed negative real spillover, linking proportional fall in the commodity price is magnified by
US dollar appreciations with falling foreign economic the higher commodity intensity in the rest of the world,
activity. The fall is larger in emerging markets mainly compared to the United States, and the pricing of com-
because of their more limited exchange rate flexibility modities in terms of the appreciating US dollar.31
(see Figure 2.4, panel 4). As countries invest less, there is a large worldwide
Model simulations also generate a strong negative drop in imports due to the high import propensity of
link between US dollar and commodity prices through investment goods. The combined effect of less trade in
the demand channel. As global demand declines, the both commodities and investment goods lowers global
demand for commodities is depressed and the real trade openness (Figure 2.10, panel 5).
price of commodities falls (Figure 2.10, panel 4). For
the simulated shock, a 1 percent appreciation in the 31The model decomposition of the quantitative results shows that
US dollar is associated with a 2.3 percent decline in the US dollar pricing channel accounts for about 10 percent of the
commodity prices at a one-year horizon. The more than overall fall in the commodity price after one year.
The commodity-induced terms-of-trade adjustment as a share of GDP increases in both emerging market
benefits commodity importers. As their import values and advanced economies, with weak investment driving
temporarily fall, real income increases and households the increase, but the dynamics differ, with investment
increase saving to smooth out consumption, repre- rebounding in advanced economies but remaining per-
senting an income effect. A substitution effect also sistently negative in emerging markets. A depreciation
operates, whereby the temporary fall in commodity in the REER facilitates adjustment in advanced econo-
prices, by lowering the consumption-based real interest mies. Consistent with fear of floating, the REER does
rate households face, increases contemporaneous not adjust on impact in emerging markets and depre-
consumption, reducing saving. In the model calibra- ciates only gradually. Financial channels contribute
tion these two effects broadly offset one another, and to the adverse spillovers in emerging markets through
the fall in investment is the main driver of the current reduced capital inflows, both public and private, and a
account increase (Figure 2.10, panel 6). For commod- decline in domestic credit. More broadly, global current
ity exporters, two opposing forces are at work. On the account balances decline in response to a US dollar
one hand, the rise in the cost of capital and resultant appreciation, reflecting a broad-based contraction in
fall in investment increase the current account. On the trade, facilitated by a fall in commodity prices.
other hand, falling commodity prices make commodity Commodity exporter status magnifies spillovers
exporters temporarily worse off, as their export values from a US dollar appreciation. Given the histori-
decrease. This effect is buffered by reduced saving, cally negative relationship between commodity prices
which decreases the current account. In the model and the US dollar index, a US dollar appreciation
simulation, the investment response and the saving is accompanied by deteriorating terms of trade for
response broadly offset one another, leaving the current commodity exporters. In the absence of a real exchange
account unchanged. Overall, consistent with the rate depreciation that could buffer both shocks,
empirical findings of the previous section, the current emerging market commodity exporters smooth the
account increases only in commodity-importing coun- temporary drop in income through reduced saving
tries, more so in emerging market commodity import- and decreased current account balances. In contrast,
ers because of the larger fall in investment. commodity importers experience improved terms of
It is worth stressing that the model omits several trade, which partly offsets the negative spillovers from
potentially important factors. One relates to additional the US dollar appreciation. In 2021–22, in contrast to
financial vulnerabilities stemming from balance sheet the historical evidence, the simultaneous strengthening
mismatches and a more nuanced modeling of the of commodity prices and the US dollar mitigated the
degree of central bank credibility, both of which are impact to the US dollar appreciation on the vulnerable
not captured by FSGM, and could potentially mag- emerging market commodity exporters.
nify the negative spillovers. Another important caveat Policies can mitigate negative spillovers from
relates to the modeling of spillovers. In some models US dollar appreciation to emerging markets. More
(Georgiadis, Müller, and Schumann 2021), emerging anchored inflation expectations mitigate the negative
market economies are directly exposed to a fraction of effect on real output through accommodative policy
the shock imposed to the sovereign risk premium in an responses, as the real exchange rate depreciates and pol-
advanced economy. In the FSGM, this spillover is cap- icy rates decrease. A more flexible exchange rate regime
tured by an exogenous shock to financial conditions, systematically speeds up economic recovery. Imple-
representing a shortcut for incorporating financial spill- mentation of such policies should be supported by
overs not directly modeled but believed to be present complementary factors. Flexible exchange regimes can
in global risk-off episodes. be supported and facilitated by domestic financial mar-
ket development that helps deepen foreign exchange
markets and expand foreign exchange hedging options
Conclusion (IMF 2020). The anchoring of inflation expectations
Negative spillovers from US dollar appreciations are can be strengthened by a sustained longer-term com-
more pronounced in emerging market economies, with mitment to improving fiscal and monetary frame-
larger declines in output that are longer lived compared works, including through ensuring a well-balanced
with those in advanced economies. The current account mix of fiscal and monetary policies, consolidating and
enhancing central bank independence, and continuing could affect the global dollar cycle would require a
to strengthen the transparency and effectiveness of deeper understanding of UIP deviations, which this
communications (see Chapter 3 of the October 2018 chapter has uncovered as a key driver of the global
World Economic Outlook). More broadly, findings of dollar cycle. UIP deviations can be attributed to the
this chapter highlight the importance of precaution- market-wide risk appetite and variations in the risk
ary policy tools, such as global safety nets as well premia demanded by global financial intermediaries,
as Integrated Policy Framework-linked policy tools which in turn reflect intermediary frictions, including
(IMF 2020), in addressing global financial market spillover from financial regulation in other segments
cycles and their spillovers. In emerging markets with of the financial system. One indirect contribution of
severe financial frictions and balance sheet vulnerabil- the chapter is to bring attention to these issues that
ities, macroprudential and capital flow management warrant further research and would enrich policy anal-
measures could help mitigate negative cross-border ysis. Concrete avenues for such research would include
spillovers under the global dollar cycle. understanding the spillover of national and global reg-
Beyond these policy recommendations for emerg- ulation of financial intermediaries as well as examining
ing markets to manage the spillovers from the global sources of intrinsic fluctuations in the market-wide risk
dollar cycle, an analysis of multilateral policy that appetite.
Box 2.1. The 2021–22 Strong-Dollar Episode and Spillovers to Commodity Exporters
Historically, US dollar appreciations have been Figure 2.1.1. US Dollar Index and
accompanied by significant declines in commod- Commodity Prices
ity prices, as captured by the negative comovement (Percent change, year-over-year)
between the two variables.1 The recent 2021–22 40 US dollar index Commodity prices
strong-dollar episode stands out in this context
because of the marked surge in commodity prices, 30
linked to recovery from the COVID-19 pandemic and
to Russia’s war in Ukraine. 20
This box presents results of an event study contrast-
ing the most recent US dollar appreciation with the 10
only comparable year-over-year US dollar appreciation
0
in the post-2000 period, which took place during
2014–15 (Figure 2.1.1):
–10
•• 2014–15 episode: The US dollar index appreciated
by 16 percent, while commodity prices fell by
–20
32 percent, in line with the historical relationship
between the two variables (see Figure 2.1.1).2 –30
•• 2021–22 episode: The US dollar index appreciated
by 10 percent, while commodity prices increased by –40
34 percent. A comparable simultaneous large and 2015 2022
persistent positive comovement in the two variables
Sources: Federal Reserve Bank of St. Louis, Federal
has not been observed in recent decades. Reserve Economic Data (FRED); Haver Analytics; IMF,
How did cross-border real output spillovers from Global Data Source; and IMF staff calculations.
Note: Percent change is calculated using the year
the US dollar appreciation differ for these two epi- average for monthly data between 2015 (2022) and 2014
sodes? The study proxies output spillovers with real (2021).
GDP forecast errors for each episode, constructed as
actual GDP for 2015 and 2022 minus the GDP fore-
In 2022, by contrast, the real GDP of emerging
cast prior to the US dollar appreciation (Figure 2.1.2).
market commodity exporters was systematically
Results reveal reversed spillovers to emerging market
revised upward following the US dollar appreciation,
commodity exporters for the recent strong-dollar
with the notable exception of Russia. Meanwhile,
episode. In 2015, the US dollar appreciation was
small downward revisions were observed for advanced
associated with systematic negative revisions to output
commodity-exporting economies.
for commodity exporters, more so for exporters with
Findings of this event study suggest that emerging
larger commodity trade surpluses (see Figure 2.1.2).
market vulnerabilities from the most recent US dollar
Notably, the negative spillovers were driven entirely
appreciation episode require a nuanced interpretation.
by emerging market commodity exporters, while there
The accompanying surge in commodity prices—
were no systematic negative GDP forecast errors for
uncharacteristic by historical standards and triggered by
advanced commodity-exporting economies. These
unique circumstances—mitigated the impact of the US
findings are broadly consistent with the outsized
dollar appreciation on the more vulnerable commodity-
negative spillovers for emerging market commodity
exporting emerging markets during 2022. Instead, the
exporters, compounded by less flexible exchange rate
negative spillovers fell disproportionately on emerging
regimes (see Figure 2.8).
market commodity importers. However, the vulnerabil-
ity of commodity importers was muted by their more
The authors of this box are Cian Allen, Rudolfs Bems, Lukas limited exposure to commodities, when compared to
Boer, and Racha Moussa. commodity exporters (see x-axis range in Figure 2.1.2),
1The correlation between the US dollar index and commodity
and their more flexible exchange rate regimes. A return
prices for the sample period is −0.38.
2Obstfeld (2022) reports a coefficient of −2.45 (standard error to the historically observed relationship between the US
of 0.42, R 2 = 0.15) for a simple ordinary least squares regression dollar and commodity prices could reverse the mitigating
of the oil-price change on dollar appreciation. role that commodity prices played in 2022.
–4 –4
RUS
RUS
BRA
–6 –6
Linear trend line: Linear trend line:
y = –0.16x + 0.17 y = 0.05x + 0.54
–8 –8
–15 –10 –5 0 5 10 15 20 25 –10 –5 0 5 10 15 20
Commodity trade balance, percent of GDP Commodity trade balance, percent of GDP
Sources: IMF, World Economic Outlook database; and IMF staff calculations.
Note: The forecast error for real GDP growth in 2015 is calculated as actual minus the IMF World Economic Outlook data
for April 2014. The forecast error for real GDP growth in 2022 is calculated as the IMF World Economic Outlook data for
April 2023 minus that for January 2022. Commodity trade balance is defined as the ratio of commodity exports to GDP
minus the ratio of commodity imports to GDP. Trend line includes only emerging market economies. For 2015, the trend
line excludes Brazil, and the coefficient is statistically significant at the 5 percent level. For 2022, the trend line excludes
Russia. Data labels in the figure use International Organization for Standardization (ISO) country codes.
Lilley, Andrew, Matteo Maggiori, Brent Neiman, and Jesse Obstfeld, Maurice, and Haonan Zhou. 2023. “The Global Dol-
Schreger. 2022. “Exchange Rate Reconnect.” Review of Eco- lar Cycle.” NBER Working Paper 31004, National Bureau of
nomics and Statistics 104: 845–55. Economic Research, Cambridge, MA.
Miranda-Agrippino, Silvia, and Hélène Rey. 2022. “The Global Ramey, Valerie A., and Sarah Zubairy. 2018. “Government
Financial Cycle.” In Handbook of International Econom- Spending Multipliers in Good Times and in Bad: Evidence
ics: International Macroeconomics, vol. 6, edited by Gita from US Historical Data.” Journal of Political Economy 126
Gopinath, Elhanan Helpman, and Kenneth Rogoff, 1–43. (2): 850–901.
Amsterdam: Elsevier. Rey, Hélène. 2013. “Dilemma not Trilemma: Global Cycle and
Miranda-Agrippino, Silvia, Tsvetelina Nenova, and Hélène Rey. Monetary Policy Independence.” Paper presented at Jackson
2020. “Global Footprints of Monetary Policies.” CFM Dis- Hole Economic Policy Symposium “Global Dimensions of
cussion Paper 2020-04, Centre for Macroeconomics, London. Unconventional Monetary Policy,” Jackson Hole, Wyoming,
Obstfeld, Maurice. 2022. “The International Financial System August 23.
after COVID-19.” Working Paper 22-2, Peterson Institute for Shin, Hyun Song. 2020. “Global Liquidity and Procyclicality.”
International Economics, Washington, DC. In The State of Economics, the State of the World. Cambridge,
Obstfeld, Maurice, Jonathan D. Ostry, and Mahvash S. Qureshi. MA: MIT Press.
2019. “A Tie That Binds: Revisiting the Trilemma in Emerg- Shousha, Samer F. 2022. “The Dollar and Emerging Markets:
ing Market Economies.” Review of Economics and Statistics Channels and Impacts.” Unpublished.
101 (2): 279–93. Wu, Jing Cynthia, and Fan Dora Xia. 2016. “Measuring the
Obstfeld, Maurice, and Kenneth Rogoff. 1996. Foundations of Macroeconomic Impact of Monetary Policy at the Zero Lower
International Macroeconomics. Cambridge, MA: MIT Press. Bound.” Journal of Money, Credit and Banking, 48: 253–91.
Technical Endnotes by Economy of GDP to 10.6 percent (midpoint of the IMF staff–assessed
norm range). First, a deduction of 5.7 percentage points of GDP
Argentina
(midpoint of an estimated 4.7–6.7 percentage point range) is
1A band of ±1 percent of GDP (two standard errors of the CA made to the EBA model’s implied contribution of the NIIP. This
norm) is applied to account for elevated country-specific uncer- deduction is made because the positive NIIP contribution in EBA
tainty in the context of external vulnerabilities. captures average income effects that are less relevant for Hong
Kong Special Administrative Region, since the income balance
relative to its NIIP is systematically lower than that in other peer
Belgium economies, due to a persistently higher share of debt instruments
1Methodological and source data changes in September 2019 on the asset side than on the liability side. Second, a deduction
led to major revisions of balance-of-payments data from 2015 of 4¼ percentage points of GDP (midpoint of an estimated
onward, causing a break in the data series. 4–4½ percentage point range) is made to account for a decline
in the gold trade balance that reflects not changes in wealth but
rather the increased physical settlement of gold futures contracts
Canada resulting from the opening of a precious metals depository. Third,
1The statistical treatments of retained earnings on portfolio a deduction of 1¼ percentage points of GDP (midpoint of an
equity and of net interest outflows (which are recorded in estimated 1–1½ percentage point range) is made to account for
nominal terms and thus artificially boosted by currently high China’s increased onshoring, which led to a decline in logistics
inflation) are estimated to have generated a downward bias in and trading activities in Hong Kong Special Administrative
the income balance of 0.6 and 1 percent of GDP, respectively, Region but did not result in lower consumption because it is
totaling 1.6 percent of GDP. viewed as temporary and to be replaced with increased provision
2The semielasticity of the CA with respect to the REER is set of high-value-added services as Hong Kong SAR’s own economy
to 0.27. rebalances in response to demand in China. See “People’s Republic
of China—Hong Kong Special Administrative Region: Selected
Issues” (Country Report No. 17/12) for more details.
2The range is calculated by applying the average semielastici-
China
ties of Hong Kong Special Administrative Region and similar
1See the IMF’s 2021 Taxonomy of Capital Flow Management
economies.
Measures for a list of China’s existing CFM measures and related 3The financial linkages with the Mainland have deepened in
policy advice.
recent years with the increase in cross-border bank lending, cap-
ital market financing, and the internationalization of the RMB.
As of end-2022, banking system claims on Mainland non-bank
Euro Area
entities amounted to HK$6.4 trillion, or about 225 percent of
1The export and import elasticities are obtained as the average GDP, down by about 9 percentage points from end-2021.
of estimates from Consultative Group on Exchange Rate Issues–
inspired export and import equations using REERs relevant for the
euro area with an autoregressive distributive lag (2,2,2) model on India
quarterly data 2000–19. The trade balance elasticity is calculated 1The cyclical adjustment and COVID-19 adjustors have been
using the share of exports and imports in extra-EU trade in GDP.
computed based on the fiscal year (as opposed to the calendar
year) to take into account the quarterly dynamics of commod-
ity prices and travel and transport services between the second
Hong Kong Special Administrative Region
quarter of 2022 and the first quarter of 2023.
1Hong Kong Special Administrative Region is not in the EBA
sample, as it is an outlier along many dimensions of EBA analysis,
thus one possibility—though with obvious drawbacks—is to use Indonesia
EBA-estimated coefficients and apply them to Hong Kong Special 1Indonesia is among the few countries with low life expec-
Administrative Region. Following this approach, the CA norm
tancy at prime age, and demographic indicators are adjusted to
in 2022 is estimated to have been about 21.8 percent of GDP,
account for this. As a result, the model-estimated CA norm is
implying a CA gap of –11.3 percent, which the model residuals
adjusted by subtracting 0.4 percentage point.
explain almost entirely. The EBA CA gap is overstated, as it does 2The standard error of the EBA norm is 0.6 percent of GDP.
not properly reflect the measurement issues that are relevant for 3The width of the range for the REER gap takes the standard
Hong Kong Special Administrative Region, so three adjustments
±3.6 percent interval applied to the midpoint of –2.0 percent,
are made that reduce the CA norm by 11.2 percentage points
leading to a range of –5.6 to 1.6.
and rhodium), which are not included in the IMF EBA model Switzerland
(terms-of-trade adjustment). 1Because of large revisions to historical balance-of-payments and
2Net current transfers related to the Southern African Customs
IIP data, particular caution is needed when the ESA results for
Union (SACU) in 2022 are assessed to have had a net nega-
different periods are compared. For example, after the initial
tive impact on the CA, are not accounted for in the regression
release in March 2022, the 2021 financial account net balance
model, and warrant an adjustment to the cyclically adjusted
was subsequently revised upward, from Sw F 27.5 billion to
CA by 0.7 percent of GDP. In addition, measurement issues
Sw F 79.2 billion (a revision of 7.1 percent of 2021 GDP),
pertaining to the income balance are likely to have contributed
driven by sizable revisions to both net acquisition of financial
to an underestimation of the CA by 0.8 percent of GDP in
assets and net incurrence of liabilities.
2022 overall. 2Valuation changes reflect fluctuations of exchange rates and prices
3Because South Africa is among the few countries with relatively
of securities and precious metals that interact with differences
high adult mortality rates, the demographic indicators have
among assets and liabilities in terms of currencies and instruments.
been adjusted to account for the younger average prime age
As a result, an appreciation (depreciation) of the Swiss franc has a
and exit age from the workforce. This results in an adjustor of
negative (positive) effect on the NIIP. Other stock-flow adjust-
–0.6 percent of GDP to the model-based CA norm for 2022.
ments include changes in statistical sources, such as changes in the
Overall, important positive contributors for the norm include
number of entities surveyed and items covered.
demographics (even after downward adjustment to account for 3Part of the positive EBA CA gap may reflect institutional
the aforementioned lower life expectancy), net foreign assets, and
pension features, such as replacement and coverage rates, in
a desirable policy stance, especially in regard to reserves.
Switzerland rather than other economic policy gaps.
4These significant flows in 2021 can largely be attributed to
4The underlying CA is adjusted for Switzerland-specific factors
Prosus N.V. acquiring about 45 percent of Naspers Ltd N
in the income account: (1) retained earnings on portfolio equity
ordinary shares from existing Naspers Ltd shareholders (direct
investment that are not recorded in the income balance of the
investment inflows) and to both resident and nonresident inves-
CA (or, the PE RE bias) under the sixth edition of the IMF
tors exchanging Naspers Ltd N ordinary shares for Prosus N.V.
Balance of Payments and International Investment Position Manual
ordinary shares (portfolio investment outflows).
(BPM6) and (2) recording of nominal interest on fixed-income
securities under the BPM framework, which compensates for
expected valuation losses (due to inflation and/or nominal
Spain
exchange rate movements), even though this stream compen-
1The EBA model suggests a cyclically adjusted CA norm of sates for the (anticipated) erosion in the real value of debt assets
–0.1 percent of GDP, with a standard error of 0.8 percent of and liabilities. The PE RE bias was estimated using the “stock
GDP. However, given external risks from a large and negative method” and “flow method” as explained in Adler and others
NIIP, the IMF staff’s assessment puts more weight on external (2019), and it is similar in size to estimates based on the Swiss
sustainability and is guided by the objective of raising the NIIP National Bank’s pilot BPM7 data.
to at least –50 percent over the medium term. Under current In addition, the CA balance is also adjusted for transitory
policies, the NIIP is projected to reach this target, though with impacts of the COVID-19 pandemic on trade of goods and ser-
high uncertainty, as valuation effects are assumed to be zero over vices, including adjustors for tourism (0.0 percentage point) and
the projection horizon. Allowing for a safety margin, the staff transport (–0.1 percentage point). Adjusted for these COVID-19
therefore considers the CA norm to be 1.0 percent of GDP, with related effects, the underlying CA would need to be reduced by
a range of 0.2 to 1.8 percent of GDP. about 0.1 percent of GDP.
2The range of the REER gap is ±2.6 percent, which is computed 5Prices of energy products, especially gas prices, were a main
based on Spain’s estimated standard error of the EBA CA norm driver underlying the PPI inflation differentials between
(0.8 percent of GDP) and a semielasticity of the CA to the Switzerland and other advanced economies such as the euro area
REER of 0.31. and the US. If core PPIs excluding energy products were used,
the depreciation of the PPI-based franc REER in 2021 and 2022
would be smaller.
Sweden
1The upper and lower bounds are derived by adding/subtracting
the standard deviation (5.7) from the average outcome (midpoint) Thailand
to reflect uncertainty surrounding the EBA estimated norm. 1For Thailand, the change in the transport services balance
2A $60 billion swap facility was agreed with the Federal
between 2019 and 2022 was –2.1 percent of GDP. In the IMF
Reserve to address dollar funding risks related to the pandemic. staff’s view, this change is too large relative to Thailand’s net
Although it was not utilized, it provided an important back- imports of global transportation services. Using an average of
stop function. percentage change in transport balances of comparable countries,
the staff estimates the impact of high freight costs on Thailand’s References
transport service balance and CA to be a worsening of about
Adler, Gustavo, Daniel Garcia-Macia, and Signe Krogstrup.
60 percent (1.3 percent of GDP). Therefore, the staff proposes a
2019. “The Measurement of External Accounts.” IMF Work-
transportation adjustor of 1.3 percent.
ing Paper 19/132, International Monetary Fund, Washington,
DC.
Allen, Cian, Camila Casas, Giovanni Ganelli, Luciana Juvenal,
United Kingdom
Daniel Leigh, Pau Rabanal, Cyril Rebillard, Jair
1For example, long-term access of EU clearing members (such as Rodriguez, and João Tovar Jalles. 2023. “2022 Update of the
banks and asset managers) to UK central counterparties (CCPs) External Balance Assessment Methodology.” IMF Working
remains uncertain. The EU has extended the equivalence for UK Paper 23/47, International Monetary Fund, Washington, DC.
CCPs only until the end of June 2025. Bank of England. 2022. “Financial Stability Report,” Box A
2The official NIIP data may understate the true position: estimates
(December), London.
of FDI stocks at market values imply a much higher NIIP, close to Bénétrix, Agustin S., Deepali Gautam, Luciana Juvenal, and
100 percent of GDP, as reported in Bank of England (2022). Martin Schmitz. 2019. “Cross-Border Currency Exposures.”
3Estimates in Bénétrix and others (2019) suggest that in 2017,
IMF Working Paper 19/299, International Monetary Fund,
about 94 percent of external assets were denominated in foreign Washington, DC.
currency, compared with 56 percent for external liabilities. International Monetary Fund (IMF). 2017. “People’s Republic of
China—Hong Kong Special Administrative Region: Selected
Issues.” Country Report No.17/12, Washington, DC.
United States
1While the fiscal policy gap is estimated to be rather small, at
–0.1 percent of GDP, the domestic fiscal policy gap is estimated
to amount to about –1.3 percent of GDP.
Download the
IMF PUBLICATIONS
Fall–Winter 2023 latest catalog to
learn more about
new titles on the
global economy
IMF.org/pubs