Journal Issue

External Sector Report: Tackling Global Imbalances amid Rising Trade Tensions

International Monetary Fund
Published Date:
July 2018
  • ShareShare
Show Summary Details

I. Overview

1. The 2018 External Sector Report (ESR) discusses the evolution of and outlook for global external balances and provides an overview of the external assessments of a globally representative set of economies for 2017. This overview complements the assessments of external balances of 30 systemic economies detailed in the accompanying Individual Economy Assessments, providing a global view, identifying global patterns and discussing potential policies to address excess imbalances from a multilateral perspective (see also Box 1). The report is organized as follows: Section II documents recent trends in external flows (that is, current account balances), stock positions (that is, international investment positions) and exchange rates. Section III presents the normative assessment of external balances (see key definitions in Box 1) and Section IV discusses the outlook and policy recommendations. In light of the material developments since the end of 2017, the report pays particular attention to the outlook and risks stemming from excess external imbalances and related policies. Finally, Section V discusses the potential impact of trade costs (frictions and policy barriers) on external balances, and argues in favor of reviving liberalization efforts, especially in services and investment, where barriers remain relatively large.

Box 1.External Assessments: Objective and Concepts

Current account deficits and surpluses can be beneficial from both an individual-country and global perspective. Countries’ abilities to run current account deficits and surpluses at different times are essential for absorbing country-specific shocks and facilitating a globally efficient allocation of capital. Some countries may need to save through current account surpluses (for example, due to aging populations); their saving finances other countries that may need to borrow via current account deficits (for example, to import capital and foster growth). Similarly, countries facing temporary positive (negative) terms-of-trade changes may benefit from saving (borrowing) to smooth out those income shocks. Thus, nonzero external balances can be desirable from both the individual-country and global standpoints.

Current account balances are deemed excessive if they depart from the levels consistent with fundamentals and desired policies.

  • The current account gap, or excess imbalance, is the difference between the actual current account (stripped of cyclical and temporary factors) and the level assessed by IMF staff to be consistent with fundamentals and desirable medium-term policies (or “norm”). This staff-assessed gap reflects policy distortions vis-à-vis other economies identified in the External Balance Assessment (EBA) models as well as other policy and structural distortions not captured by the model. A current account balance that is higher (lower) than that implied by fundamentals and desired medium-term policies corresponds to a positive (negative) gap. Eventual elimination of such gaps is desirable, although there may be good reasons for a temporary gap, and/or for adjusting gradually. Calling a current account gap high (low) is another way to say it is positive (negative).
  • Assessments also include a view on the real effective exchange rate (REER)—normally consistent with the assessed current account gap, except when marked REER movements are not yet fully reflected in the current account, due to lagged effects. A positive (negative) REER gap implies an overvalued (undervalued) exchange rate. REER gaps do not necessarily predict future exchange rates and may occur in any economy, including those with floating exchange rates.

Although the overall assessment of a country’s external position hinges on the current account and the REER, it takes other indicators into consideration. These include the financial account balance, the international investment position, reserve adequacy, and other competitiveness measures, such as the unit-labor-cost REER. The overall external position is judged to be weaker (stronger) than warranted by fundamentals and desired policies when the current account gap is negative (positive) and/or the REER is deemed overvalued (undervalued). Assessments strive to be multilaterally consistent, meaning that negative staff-assessed current account/REER gaps in some economies are matched by positive staff-assessed gaps in others.

II. Evolution of External Positions

This section documents the recent evolution of global current account balances, exchange rates, and stock positions, with emphasis on developments during 2017. It highlights how progress in reducing global balances has stalled since 2013, and discusses the underlying drivers of the recent reconfiguration and concentration of surpluses and deficits in advanced economies (AEs). The discussion in this section is descriptive and does not yet take a view on whether, or where, adjustments are necessary. Normative assessments, identifying excess imbalances, are provided in Section III.

2017 Developments

2. Global current account balances were broadly unchanged in 2017, with minor shifts cementing the reconfiguration that has emerged since 2013 (see Table 1). Overall global balances—defined as the absolute sum of surpluses and deficits—remained at about 3¼ percent of world GDP in 2017 (Figure 1, panel 1). Last year’s small changes relative to 2016 continued the trend in recent years of greater concentration of surpluses and deficits in AEs. On the surplus side, China’s current account balance continued its gradual decline, largely offset by a rising surplus in Japan—against the backdrop of a depreciating yen—a further increase in current account balances of euro area debtor countries, and a resurgence of surpluses in oil-exporting countries on the back of recovering oil prices. On the deficit side, the United States continued to be the main global borrower, accompanied by growing current account deficits in some emerging market economies (Argentina, India, Turkey). These deficits were partially offset by smaller deficits in the United Kingdom—supported by further sterling depreciation—and in some large commodity-exporting economies (Australia, Brazil, Canada, Mexico, South Africa) thanks to strengthening commodity prices. Against the backdrop of a generalized pickup in global trade, trade tensions intensified during 2017, with generally minor concrete actions, but significant fears of escalation (some already in play in 2018).

Table 1.Selected Economies: Current Account Balance, 2014–17 1/
In billions of USDIn percent of World GDPIn percent of GDP
Top 15 Surplus Economies in 2017
Taiwan Province of China61.074.972.882.
United Arab Emirates54.517.613.
Top 15 Deficit Economies in 2017
United States–373.8–434.6–451.7–466.2–0.5–0.6–0.6–0.6–2.1–2.4–2.4–2.4
United Kingdom–161.4–150.0–153.9–106.7–0.2–0.2–0.2–0.1–5.3–5.2–5.8–4.1
India 2/–26.8–22.1–15.2––0.1–1.3–1.1–0.7–1.9
Memorandum item:
Euro Area332.3373.3397.8442.
Statistical discrepancy392.5228.1239.2405.
Surpluses (world)1,563.41,506.41,506.41485.
Surpluses (AEs)876.91004.41029.51110.
Deficits (world)–1,149.9–1,281.6–1151.3–1080.2–1.5–1.7–1.5–1.4–2.8–3.0–2.7–2.6
Deficits (AEs)–671.1–719.0–722.4–681.7–0.9–1.0–1.0–0.9–2.4–2.6–2.6–2.4
Source: World Economic Outlook and Fund Staff calculations.

Sorted by size (in USD) of surplus and deficit in 2017.

For India, data are presented on a fiscal year basis.

Source: World Economic Outlook and Fund Staff calculations.

Sorted by size (in USD) of surplus and deficit in 2017.

For India, data are presented on a fiscal year basis.

Figure 1.Evolution of Current Account Balances and Exchange Rates

Sources: World Economic Outlook, International Financial Statistics, Global Statistics Database and IMF staff calculations.

1/ Overall balance is the absolute sum of global surpluses and deficits. Surplus AEs: Hong Kong SAR, Korea, Singapore, Sweden, Switzerland, Taiwan POC; AE Commodity Exporters: Australia, Canada, New Zealand; Deficit EMs: Brazil, India, Indonesia, Mexico, South Africa, Turkey; Oil Exporters: WEO definition plus Norway.

2/ 2017 average relative to 2016 average.

3. Most exchange rates displayed relatively moderate movements over 2017 despite some intra-year volatility (Figure 1, panel 2). Uncertainty about monetary policy prospects led to intra-year volatility across major currencies, although year-average movements generally were small. Noticeable exceptions were the real depreciations of the British pound, reflecting continued Brexit-related uncertainty; the yen, reflecting interest rate differentials vis-à-vis the United States; and the renminbi, owing to capital outflow pressure in the earlier part of the year. Among emerging and developing economies (EMDEs), large real appreciations registered by Brazil, Russia, and South Africa, supported by stronger commodity prices and a better political environment in some cases, partly unwound the cumulative depreciations of previous years. Meanwhile, the Turkish lira depreciated sharply due to a vulnerable external position and uncertain political environment.

4. As discussed in earlier External Sector Reports, the broadly unchanged landscape of overall global balances since 2013 masks an important reconfiguration. Global surpluses and deficits have become increasingly concentrated in AEs, as China and oil exporters have seen their current account surpluses narrow and the deficits of some EMDEs (for example, Brazil, India, Indonesia, Mexico, South Africa) have shrunk. Key drivers of this reconfiguration were the sharp drop earlier this decade in oil prices, which have recovered somewhat after bottoming out in 2016, and the gradual tightening of global financing conditions reflecting prospects for monetary policy normalization in the United States. Also at work have been asymmetries in demand recovery and the associated policy responses in systemic economies (Figure 2). After 2013, higher or persistently large surpluses in key advanced economies (for example, Germany, Japan, the Netherlands) were underpinned by relatively weaker domestic demand, constrained by fiscal consolidation efforts—necessary in some cases, given compressed fiscal space. Meanwhile, higher or persistent current account deficits in other AEs (United Kingdom, United States) reflected a stronger recovery in domestic demand, supported by some recent fiscal easing. Meanwhile, the narrowing of China’s underlying current account surplus was supported by a marked relaxation of fiscal and credit policies, masking lingering structural problems and causing a buildup of domestic vulnerabilities. These asymmetries in demand strength have also led to differences in monetary policy (as seen by the evolution of longer term nominal bond yields) and currencies.

Figure 2.Systemic Economies: Fiscal Stance and Long-term Nominal Interest Rates, 2005–17 1/

Sources: WEO and IMF staff calculations.

1/ EA groups include ESR countries only. GDP-weighted averages are reported.

5. Real exchange rate movements have generally supported the reconfiguration of global balances (Figure 3). Since 2013, real currency appreciations have been associated with a pickup in domestic demand (relative to output), contributing to widening of the US current account deficit and narrowing of surpluses in China and Korea. In contrast, real currency depreciations have been associated with weaker domestic demand (relative to output) in some EMDEs (Brazil, Mexico, Russia, Turkey) and AEs (Australia, Canada), leading to higher current account balances, despite weaker terms of trade in some cases.

Figure 3.Relative Demand Growth vs. REER Changes, 2013–17

Sources: WEO and IMF staff calculations.

6. Capital flows to most EMDEs remained subdued, amid a resurgence of inflows to China (Figure 4).

  • Net non-reserve flows to EMDEs continued to be dominated by developments in China, which experienced a gradual reversal from sizable outflows and reserve sales in 2015–16 to renewed inflows and some reserve accumulation in late 2017, following improving domestic and external conditions as well as tight enforcement of capital flow management measures.
  • In most other systemic EMDEs, capital inflows and reserve accumulation remained subdued relative to earlier years, reflecting primarily slowly tightening global financial conditions; although some frontier market economies gained substantial reserves. Despite posting a current account surplus, Saudi Arabia experienced reserve losses as capital outflows continued.
  • In some advanced economies and financial centers (Hong Kong SAR, Singapore, Switzerland), the pace of reserve accumulation accelerated during 2017, supported by large current account surpluses.

Figure 4.Selected EMDEs: Capital Flows and Reserve Accumulation, 2012–17

(Percent of group GDP, four-quarter moving average)

Sources: International Financial Statistics and IMF staff calculations.

7. Global stock positions stabilized in 2017, mainly owing to valuation effects (Figure 5). At the global level, the growth of stock positions took a pause in 2017, despite a continuation of current account deficits in debtor countries (with the notable exception of euro area debtor countries) and of current account surpluses in creditor countries. A slight narrowing of overall debtor positions in 2017 was driven mostly by the United States, whose international investment position improved because of valuation changes linked to a weakening of the US dollar from the end of 2016 to the end of 2017 (which reduced the value of mostly dollar-denominated debt liabilities relative to mostly foreign-currency-denominated debt assets). 1 This was matched by smaller creditor positions in a few countries (most notably China) also reflecting mainly valuation changes. The above-mentioned narrowing of stock positions was offset by the expansion of debtor positions in some AEs (Canada, United Kingdom) and the creditor positions of others (Germany, Hong Kong SAR, the Netherlands, Singapore, Switzerland), the latter driven by sizable current account surpluses and favorable valuation effects.

Figure 5.External Stock and Flow Positions, 2002–17

Sources: World Economic Outlook and IMF staff calculations.

1/ Surplus AEs: Korea, Hong Kong SAR, Singapore, Sweden, Switzerland, Taiwan POC; AE Commodity Exporters: Australia, Canada, New Zealand; Deficit EMs: Brazil, India, Indonesia, Mexico, South Africa, Turkey; Oil Exporters: WEO definition plus Norway.

2/ Dot sizes proportional to US$ GDP.

III. Assessments of External Positions

This section presents an overview of how observed external balances align with the levels consistent with medium-term fundamentals and desired policies. It also briefly summarizes the process for arriving at the external assessments, including the use of the refined econometric model to underpin this year’s assessments. The configuration of excess external imbalances and the contributions from key policy distortions are also discussed.

8. External assessments compare actual external balances with those that are consistent with medium-term fundamentals and desired policies. This process, also summarized in an earlier IMF blog, involves combining numerical inputs from statistical cross-country models with country-specific judgment based on the IMF staff’s knowledge and insights regarding each economy gained during the Article IV consultation process. To arrive at estimates of external balances consistent with medium-term fundamentals and desired policies (that is, “norms”), IMF country teams rely on the numerical benchmarks of the various External Balance Assessment (EBA) models—although greater weight continues to be given to the current account model, because real exchange rates tend to be more volatile and difficult to explain econometrically.2 This year, as is done periodically, the EBA models were refined to reflect insights gained since the previous (2015) round of changes. Refinements entailed extending the estimation period and aimed at better capturing the role of certain fundamentals (demographics, institutions, and potential current account measurement biases), macroeconomic policies (foreign exchange intervention, credit excesses) and structural features could play in driving current account dynamics (for more details, see Box 2 and the 2018 External Sector Report—Refinements to the External Balance Assessment Methodology—Technical Supplement). While the refinements led to important improvements in the models, the models cannot capture every aspect of current account dynamics and, thus, should be treated only as numerical benchmarks for IMF staff assessments. Analytically grounded and transparently presented IMF staff judgment remains essential in arriving to a multilaterally consistent set of assessments. Considering estimation uncertainties, assessments are presented in ranges, which generally vary with the model standard errors and country-specific features.

9. IMD staff-assessed current account norms for 2017 vary considerably across countries, with relatively small changes over time, in most cases (see Tables 2 and 3). EBA model-based current account norms were generally positive in AEs. These positive norms reflected their higher income per capita and lower growth prospects, hence lower returns on capital (making them exporters of capital); their higher longevity and share of prime-aged savers (requiring them to save more than others); and need for tighter policies, especially fiscal (to address their generally higher public debt levels and future old-age commitments). Mirroring these features, current account norms are negative for most EMDEs—reflecting their higher growth potential, lower income per capita, and younger populations. Norms also varied within these groups depending on their institutional strength (which constrains their ability to borrow and invest), whether they issue reserve currencies (which allows them to finance larger external deficits), and the presence of non-renewable exports (where intergenerational equity considerations tend to increase optimal saving). In some cases, country-specific judgment was applied to arrive at staff-assessed (final) current account norms. External financing risk (Brazil, India, Spain, Turkey) or country-specific demographic features not fully captured in the model (for example, uncertainty about the effect of migration on national saving in Germany and high mortality risk in Indonesia and South Africa) were taken into consideration. In other cases, adjustments to the cyclically adjusted current account addressed measurement biases (Canada, South Africa, Switzerland, United Kingdom), temporary factors related to political uncertainty and not captured by the model (Russia, Thailand), and delays in investment plans financed by EU funds (Poland).

Table 2.ESR Countries: Summary of External Assessment Indicators, 2017
Current Account (% GDP)Staff-Assessed CA Gap (% GDP)Staff-Assessed REER Gap (Percent)Int’l Investment Position (% GDP) 1/CA NFA Stabilizing (% GDP) 2/CA/REER Elasticity 3/SE of CA Norm (%) 4/
CountryOverall AssessmentActualCycl Adj.MidpointLowHighMidpointLowHighNetLiabilitiesAssets
AustraliaBroadly Consistent–2.5–2.4–1.0–1.5––55185130–
BrazilBroadly Consistent–0.5–1.80.2–0.30.7–2.0–7.03.0–347542–
CanadaModerately Weaker–2.9–2.4–1.9–3.4–
ChinaModerately Stronger1.–3.0–13.07.01543581.60.231.6
Euro Area 5/Moderately Stronger3.–4.0–8.00.0–1222221–
FranceModerately Weaker–0.6–0.6–1.6–2.0––21324302–
GermanySubstantially Stronger8.–15.0–20.0–10.0601992591.80.230.9
Hong Kong SARBroadly Consistent4.33.30.0––5.05.040911971606............
IndiaBroadly Consistent–1.9–2.10.4–0.61.4–1.0–7.05.0–134128–
IndonesiaBroadly Consistent–1.7–1.60.1–1.41.6–1.1–9.47.2–346733–
ItalyBroadly Consistent2.82.1–0.3––7172165–
JapanBroadly Consistent4.03.60.5–0.81.8–3.5–13.06.0601241842.80.141.3
KoreaModerately Stronger5.–4.5–7.2–1.71678940.90.360.8
MexicoBroadly Consistent–1.7–1.40.5–0.51.5–4.0–12.04.0–4610155–
NetherlandsSubstantially Stronger10.–10.0–13.0–7.074117712512.60.740.9
PolandBroadly Consistent0.–2.5–5.00.0–6511852–3.20.430.6
RussiaModerately Weaker2.33.2–1.3–
Saudi ArabiaWeaker2.2....–2.0–3.0–
SingaporeSubstantially Stronger18.818.–10.0–16.0–4.02488691118............
South AfricaModerately Weaker–2.5–2.5–1.3–2.3––
SpainModerately Weaker1.91.5–1.5–2.5––85254169–
SwedenModerately Stronger3.–5.0–10.00.0102832930.40.251.2
SwitzerlandBroadly Consistent9.89.60.8–1.22.8–1.5–5.32.31275877148.90.531.3
ThailandSubstantially Stronger10.610.–10.5–14.0–7.0–7107100–0.50.641.7
United KingdomWeaker–4.1–4.0–3.0–5.0––13535523–
United StatesModerately Weaker–2.4–2.3–1.5–2.0––40183143–
Sources: IMF World Economic Outlook (WEO), International Financial Statistics (IFS), and Staff assessments.

The NIIP estimates come from WEO. Country team estimates (reported in ESR pages) could differ.

The current account balance that would stabilize the ratio of net foreign assets (NFA) to GDP at the benchmark NFA/GDP level.

Assumed elasticity linking a change in the current account (as percent of GDP) to a change in the REER (in percent).

The standard error of the 2017 estimated current account norms.

The staff-assessed euro area CA and REER gaps are calculated as the GDP-weighted averages of staff-assessed CA and REER gaps for the 11 largest Euro area economies.

Sources: IMF World Economic Outlook (WEO), International Financial Statistics (IFS), and Staff assessments.

The NIIP estimates come from WEO. Country team estimates (reported in ESR pages) could differ.

The current account balance that would stabilize the ratio of net foreign assets (NFA) to GDP at the benchmark NFA/GDP level.

Assumed elasticity linking a change in the current account (as percent of GDP) to a change in the REER (in percent).

The standard error of the 2017 estimated current account norms.

The staff-assessed euro area CA and REER gaps are calculated as the GDP-weighted averages of staff-assessed CA and REER gaps for the 11 largest Euro area economies.

Box 2.Implications of Refinements to the External Balance Assessment Model1

This box presents the implications of key methodological refinements on the estimated and IMF staff-assessed current account norms across the ESR sample.

External Balance Assessment (EBA) estimates: The distribution of EBA estimated norms was broadly unchanged following the refinements, with model-implied surpluses in most AEs and deficits in most EMDEs. That said, changes were nontrivial in some cases (see Box Figure 2.1), reflecting (1) refinements in the modeling of financial centers (Netherlands, Switzerland); (2) the new demographic specification aimed at disentangling compositional from longevity effects (Germany, Italy, Spain, and Sweden); and (3) changes in the institutional risk (Canada, Russia) and credit excess (Germany, Sweden) proxies. Demographic data updates played a role in some key cases (highlighted in red), including in Germany where recent migration flows suggest more favorable population dynamics. The median norm moved by -0.4 percent of GDP, although changes were somewhat smaller (-0.3 percent of GDP) for the key systemic economies. Changes in the EBA-estimated current account norms are comparable in size and distribution to the refinements undertaken in 2015, which introduced nonlinearities in the modeling of demographics.

Box Figure 2.1.ESR Economies: EBA Current Account Norms 2017

(in percent of GDP)

Sources: IMF Staff Estimates

Staff-assessed norms: Changes in EBA numerical estimates, however, did not necessarily translate into equivalent changes in staff-assessed norms (see Box Figure 2.2). In some cases, refinements reduced the need for staff adjustments because the new model addressed shortcomings with the previous specification (for example, demographics for Sweden). In other cases, necessary outside-the-model adjustments were identified and implemented (for example, Switzerland due to measurement; South Africa due to demographics). Changes in staff-assessed norms between 2016 and 2017 were largest in countries where there was some reassessment in the context of the refinements about: (1) the role of certain fundamentals (for example, Germany and Italy due to demographics); (2) factors affecting the underlying current account (Netherlands due to measurement, Thailand due to political uncertainty); and (3) past staff adjustments (for example, adjustments for structural distortions and offshoring were eliminated for Japan). Staff’s views in many of these areas (that is, demographics, measurement, structural policies), however, will continue to evolve with further analysis.

Box Figure 2.2.ESR Economies: Evolution of Staff-Assessed Current Account Norms

(in percent of GDP)

Sources: IMF Staff Estimates.

1 Prepared by Pau Rabanal and Zijiao Wang.

10. Despite some shifts in staff-assessed norms and cyclically adjusted current account balances, the configuration of excess external imbalances remained broadly unchanged (see Figure 6 and Tables 2 and 3):

  • Stronger positions: External positions were deemed “substantially stronger” than justified by medium-term fundamentals and desirable policies (current account gaps of more than 4 percentage points of GDP) in Germany, the Netherlands, Singapore and Thailand; “stronger” (2 to 4 percentage points of GDP) in Malaysia; and “moderately stronger (1 to 2 percentage points of GDP) in China, Korea, and Sweden. The euro area as a whole was assessed to be “moderately stronger,” compared with a broadly-in-line assessment last year. This overall assessment reflects wider positive current account gaps in some countries (Germany, Netherlands) and narrower negative gaps in others (France, Italy, Spain).
  • Weaker positions: Conversely, external positions were assessed to be “weaker (negative current account gaps in the range of 2 to 4 percent of GDP) in Argentina, Belgium, Saudi Arabia, Turkey and the United Kingdom; and “moderately weaker” (1 to 2 percent of GDP) in Canada, France, Russia, South Africa, Spain and the United States. A wide spectrum of excess external imbalances among euro area members, ranging from negative current account gaps (Belgium, France, Spain) to sizable positive gaps (Germany, Netherlands), underscores continued and marked asymmetries in competitiveness within the common currency area.
  • Changes since 2016: Despite relatively unchanged overall excess imbalances, there were some underlying shifts: narrower imbalances in some economies offset wider imbalances in others. Narrowing positive gaps in some economies, because of lower observed surpluses (Korea, Sweden) and higher staff-assessed norms (Japan), were generally offset by widening positive gaps in others, because of higher observed surpluses (Netherlands) and somewhat lower staff-assessed norms (China, Germany, the Netherlands). Meanwhile, the narrowing of negative gaps in some economies (Australia, France, Italy, Saudi Arabia, Spain), was offset by widening negative gaps in others (mainly United States).

Figure 6.Staff-assessed and EBA Estimated Current Account and REER Gaps1/

Source: IMF Staff assessments.

1/ Sorted by the mid-point of the staff-assessed gap. Hong Kong SAR, Saudi Arabia and Singapore do not have EBA estimates. For Saudi Arabia, the current account gap reflects a fiscal policy gap.

2/ EBA REER gap is defined as the average of index- and level-based regressions (see details in 2018 External Sector Report--Refinements to the External Balance Assessment Methodology--Supplemetary Information).

Table 3.ESR Countries: Summary of Staff-Assessed Current Account Gaps and Staff Adjustments, 2017(in percent of GDP)
Actual CA BalanceCycl Adj. CA BalanceEBACA NormEBACA Gap 1/Staff-Assessed CA Gap 2/ [E]Staff Adjustments 3/
Australia–2.5–2.4–0.6–1.9–1.0–0.9–0.9Large investment needs due to its size and low population density
Brazil–0.5–1.8– risks considerations
Canada–2.9–2.42.2–4.6–1.9–2.7–0.4–2.3Measurement biases and terms-of-trade; Demographics (updated data;
Euro Area 4/ individual country adjustments
Germany8. (uncertainty related to large/sudden immigration)
India–1.9–2.1– risks considerations
Indonesia–1.7–1.6–0.8–0.80.1–0.9–0.9Demographics (high mortality risk)
Japan4.–0.1–0.1Temporary boost in energy imports following the 2011 earthquake
Mexico–1.7–1.4– drop in investment (underlying CA)
Poland0.30.8– in disbursement of EU Funds that hold back public investment
Russia2.33.23.8–0.5– of economic sanctions
South Africa–2.5–2.50.7–3.2–1.3–1.9–1.1–0.8Demographics (high mortality risk); measurement biases
Spain1.– risks considerations
Switzerland9. biases
Thailand10.610. uncertainty, terms of trade, temporary tourism boom
Turkey–5.6–4.8–0.9–4.0–2.2–1.8–1.0–0.7NllP/financing risks considerations; Gold restocking
United Kingdom–4.1–4.01.0–5.0–3.0–2.0–2.0Measurement biases
United States–2.4–2.3–0.7–1.6–1.5–0.1
Hong Kong SAR4.33.30.0
Saudi Arabia2.2–2.0
Discrepancy 5/0.03
Source: Fund staff estimates.

Figures may not add up due to rounding effects.

Refers to the mid-point of the CA Gap.

Total staff adjustments include rounding in some cases. Breakdown between norm and other factors (which affect the underlying CA) are tentative.

The EBA euro area current account norm is calculated as the GDP-weighted average of norms for the 11 largest Euro area economies, adjusted for reporting discrepancies in intra-area transactions (which were equivalent to 0.6 percent of GDP in 2017). The staff-assessed CA gap is calculated as the GDP-weighted average of staff-assessed gaps for the 11 largest Euro area economies.

Weighted average sum of staff-assessed CA gaps.

Source: Fund staff estimates.

Figures may not add up due to rounding effects.

Refers to the mid-point of the CA Gap.

Total staff adjustments include rounding in some cases. Breakdown between norm and other factors (which affect the underlying CA) are tentative.

The EBA euro area current account norm is calculated as the GDP-weighted average of norms for the 11 largest Euro area economies, adjusted for reporting discrepancies in intra-area transactions (which were equivalent to 0.6 percent of GDP in 2017). The staff-assessed CA gap is calculated as the GDP-weighted average of staff-assessed gaps for the 11 largest Euro area economies.

Weighted average sum of staff-assessed CA gaps.

11. REER and current account assessments generally mapped closely to each other. Exchange rate assessments continued to be based, for the most part, on staff’s views of the current account gap, mapped into exchange rates by using trade elasticities estimated separately (Figures 6 and 7). In general, countries with current account balances that were higher (lower) than warranted by fundamentals and desirable policies were deemed to have undervalued (overvalued) exchange rates. In a few cases, discrepancies between the current account and exchange rate assessments reflected rapid exchange rate movements that were deemed temporary or not yet fully reflected in the current account (because of lags in the transmission of exchange rates to trade volumes and prices). China’s real exchange rate remained broadly in line amid a positive current account gap, which has narrowed and is projected to narrow further over time as the renminbi’s real appreciation of recent years continues to permeate to the current account. Other examples include Turkey, where the sharp lira real depreciation in 2017 is not yet reflected in a lower current account deficit, and Italy, where the current account balance rose to a level consistent with fundamentals and desirable policies as the current stance of the financial cycle (featuring still-high nonperforming loans and weak bank profitability) masks lingering competitiveness and structural concerns.

Figure 7.Staff-assessed Current Account and REER Gaps, 2017 1/

Source: IMF staff calculations.

1/ Gray bands depict broadly-in-line ranges for CA and REER gaps. REER gap based on 2017 average REER.

12. Factors driving excess external imbalances vary across countries, although some common patterns can be identified. Staff-assessed gaps can be decomposed into “identified policy gaps” and “other gaps” (or the residual). The former refers to differences between actual and desired policies in the medium term, when output gaps are expected to be closed (Table 4), and reflect domestic policy gaps relative to those of the rest of the world. Policy gaps for fiscal, public health spending, foreign exchange intervention, capital flow management, and credit policies are captured within the EBA model. Other staff-assessed gaps are interpreted as reflecting primarily distortions to saving and investment decisions that are not explicitly modeled in the EBA model. Overall, while positive (negative) identified policy gaps are associated with positive (negative) current account gaps, in some prominent cases, identified policy gaps fall significantly short of explaining excess external imbalances (Figure 8). In these circumstances, assessments need to rely on country-specific insights and complementary analysis to ascertain the role of other distortions, especially structural policies, in driving excess external imbalances. Complementary tools were developed in the context of this year’s refinements to shed light on the potential role of product and labor market policies (see Box 3 and the 2018 External Sector Report—Refinements to the External Balance Assessment Methodology—Technical Supplement), although country-specific insights remain necessary to properly tailor the structural policy advice.

Table 4.Selected ESR Countries: Current Account Regression Policy Gap Contributions, 2017(in percent of GDP)
Fiscal GapPublic Healh Exp GapPrivate Credit GapFXI GapOther (K-Controls)
EBA SapDomesticDomesticDomesticDomestic
Total 1/IdentifiedDom 2/ResidualTotal 1/Dom 3/CoeffPP*Total 1/Dom3/CoeffPP*Total 1/Dom 3/CoeffPP*Total 1/Dom 3/CoeffPP*Total 1/Dom 3/
Euro Area 4/–0.20.3–0.8–––0.1–7.8–3.5––0.10.0
South Africa––3.80.2–0.50.3–3.9–2.4–0.10.0––0.1–4.20.0–
United Kingdom–5.0–0.4–0.8–4.6–0.1–0.80.3–2.30.0–0.10.0––––0.10.0
United States–1.6–0.6–0.9–1.0–0.4–1.10.3–4.8–1.5–0.4–0.3––0.1–4.00.0––0.10.0
Source: IMF staff estimates.

Total contribution after adjusting for multilateral consistency.

Includes contribution of domestic policy gaps to the identified gap. The total foreign policy gap contribution is constant and equal to 0.4 percent for all countries.

Total domestic contribution is equivalent to coefficient*(P-P*)

The euro area EBA CA gap and policy gap contributions are calculated as the GDP-weighted averages of EBA CA gaps and policy gap contributions for the 11 largest Euro area economies.

Source: IMF staff estimates.

Total contribution after adjusting for multilateral consistency.

Includes contribution of domestic policy gaps to the identified gap. The total foreign policy gap contribution is constant and equal to 0.4 percent for all countries.

Total domestic contribution is equivalent to coefficient*(P-P*)

The euro area EBA CA gap and policy gap contributions are calculated as the GDP-weighted averages of EBA CA gaps and policy gap contributions for the 11 largest Euro area economies.

Box 3.Understanding Excess Imbalances: The Role of Structural Factors1

This box summarizes the complementary tools that have been developed (in the context of the refinements) to shed light on the relationship between excess imbalances and structural policies in labor and product markets.

Conceptual framework: The removal of distortions associated with product market and labor market policies are generally geared toward increasing the productive capacity of the economy, but they also affect the current account in the short to medium term (Obstfeld and Rogoff, 2006; Cacciatore and others 2016a, 2016b), through the

  • Productivity channel: Changes in structural policies increase investment opportunities, resource availability and productivity. These reforms improve the current account if output increases more than domestic demand, and productivity gains are mainly in the tradable sector.
  • Price-competitiveness channel: More wage flexibility may increase the current account through competitiveness gains. More goods market flexibility reduces the price-setting power of firms, but it could have an inflationary general equilibrium effect-stemming from the entry of new firms and increased labor demand-that hurts competitiveness and reduces the current account.
  • Uncertainty channel: Reforms that reduce uncertainty should increase firm investment, but their effect on precautionary saving and, thus, the current account is ambiguous (Gosh and Ostry, 1997).

Empirical approach: Lack of proper time and country coverage prevents including structural indicators directly into the External Balance Assessment (EBA) models. As an alternative, staff used available structural indicators from the Organisation for Economic Co-operation and Development (OECD) and World Economic Forum (WEF) for a subset of countries or years to examine their relationship with the estimated unexplained residual from the EBA current account Model. The focus is on understanding the extent to which the unexplained gaps are affected by product market and labor market regulatory deviations from best practice, while identifying policies that help reduce both domestic structural gaps and excess current account imbalances.

Findings and application (Box table): In line with the literature, both empirical (Jaumotte and others, 2010; Cheung and others 2013; IMF 2017; and Kerdrain and others 2010) and theoretical (Cacciatore and others 2016a,b), the results based on OECD data suggest that reducing burdens in the “licenses and permits system”—a type of product market regulation—can help reduce a country’s current account balance as investment by new firms rises, and their additional demand for labor puts upward pressure on wages and reduces competitiveness. Meanwhile, addressing certain labor market rigidities by easing employment protection laws can improve the current account through competitiveness gains as firms are more able to adjust labor inputs and costs, including as a result of changes in the bargaining power of the employed. Comparable results hold for WEF de facto measures of product and labor market rigidities, which indicate that the current account balance falls with a reduction in procedures to start a business, yet improves with better cooperation in labor-employer relations. Rigidities in product market regulations can help explain the positive residuals in key economies (Germany, Japan, Korea), while distortions in labor markets can explain negative residuals in others (Italy, South Africa). These complementary tools provide multilaterally consistent results that serve to guide policy recommendations, which will continue to rely on country-specific insights to properly tailor the advice. These tools will continue to be refined as experience is gathered and data availability constraints are eased.

Empirical Results

Dependent variable: EBA-CA residual

PMRs: LPS (+ = more burdens)0.0049**
LMRs: EPL (+ = stricter regulations)–0.0048**
PMRs: SBP (+ = more procedures)0.0242**
LMRs: CLER (+ = more cooperation)0.0508***
Number of Observations374533
Number of countries2449
* significant at 10%; ** significant at 5%; *** significant at 1%
* significant at 10%; ** significant at 5%; *** significant at 1%
1 Prepared by Carolina Osorio-Buitron. For a more detailed discussion, see 2018 External Sector Report—Refinements to the External Balance Assessment Methodology—Technical Supplement.
  • In many countries with higher-than-desirable current account balances (such as Germany, Korea, Netherlands, Sweden, Thailand), a fiscal stance that is tighter than desirable contributed, while other macroeconomic policies played a role elsewhere (for example, insufficient health spending in Korea; foreign exchange purchases in Thailand).3 Meanwhile, in China, positive contributions from insufficient health spending and renewed reserve accumulation were largely offset by continued fiscal and credit policies that were undesirably loose from a medium-term perspective. Meanwhile, product market regulations that inhibit firms’ entry because of hurdles to starting a business appear to have held back investment in a number of these economies (Germany, Korea, Malaysia).
  • At the other end, looser-than-desirable fiscal policy contributed significantly to the lower than warranted current account balances of large AEs (France, Spain, United Kingdom, United States) and some EMDEs (Argentina, Russia, South Africa, Turkey), while easy credit contributed to negative current account gaps in others (Canada, France, Turkey). Similarly, labor market regulations that increase labor costs through strict employment protection appear to have contributed to weak competitiveness in some of these cases.
  • In some countries without excess external imbalances, certain policies offset each other, masking underlying structural issues. In Japan, for example, while easier-than-desired fiscal policy has likely helped to contain the current account surplus, it seems to mask the effect of underlying product market distortions that hold back investment. In other economies (Brazil, Italy), lower-than-desirable credit, amid weak investment, pushed up current account balances, masking underlying competitiveness problems that pushed the current account in the opposite direction.

Figure 8.Policy Gap Contributions to Current Account Gaps, 2017

(Percent of GDP)

Sources: IMF staff calculations.

1/ Dot size proportional to external imbalances in percent of world GDP. Contribution of (domestic and external componets of) identified policy gaps to current account gap, based on estimated EBA coefficient and staff-assessed desirable policies.

2/ Domestic component of identified policy gap only.

13. Foreign exchange intervention remained muted during 2017, except in a few cases.4 Continuing the trend observed in previous years (see IMF 2017), foreign exchange intervention played a limited role in driving excess external imbalances during 2017, with some exceptions. Among ESR economies, Thailand stands out as having purchased a significant amount of reserves (and forward contracts) despite having more-than-adequate reserves and a higher-than-desirable current account balance (Figure 9). Reserve accumulation was also sizeable in India despite adequate levels of reserves, although consistent with preserving current account balances that were broadly-in-line. Meanwhile, after two years of marked decumulation, and accompanying the unwinding of short forward positions, China recorded a minor net positive accumulation of reserves in 2017. Foreign exchange purchases by Argentina and South Africa in 2017 reflected their need to build up buffers and contain further appreciation of already overvalued currencies. In the case of Turkey, the fall in FX reserves was offset by increased gold holdings.

Figure 9.Estimated FXI and Reserve Adequacy, 2016–17 1/

Sources: IMF staff calculations.

1/ Blue, gray and red dots denote higher-than-desirable, broadly-in-line and lower-than-desirable 2017 current account balances, respectively.

2/ Metric based on gross reserves. Adjusted for capital flow management measures (China, Malaysia, South Africa and Thailand) and commodity exporter buffer (Russia).

3/ Estimates of FXI, which are subject to uncertainty, are based on BOP reserve flows plus changes in off-balance sheet FX positions, as reported in BOP statistics and the International Reserves and Foreign Currency Liquidity Template. Excludes estimated interests on reserves, which are based on a 1 percent yield and are consistent with the median reported yields by EBA countries. FXI estimates for Malaysia may reflect valuation changes.

A Global View of Excess Imbalances

14. Overall current account excess imbalances remained unchanged, and concentrated in a few large economies, in 2017 (Figure 10). Global current account gaps remained relatively unchanged at about 1½ percent of global GDP in 2017, indicating that about 40–50 percent of global surpluses and deficits cannot be traced to fundamentals and desirable policies. Lower-than-desirable current account balances were concentrated in AEs: namely the United Kingdom, the United States and a few vulnerable EMDEs (Argentina, Turkey). Meanwhile, higher-than-desirable current account balances were concentrated in the euro area, other AEs (Korea, Singapore, Sweden) and China. Japan’s small positive current account gap—while within the broadly-in-line range—contributed to global current account gaps. Excess external imbalances of euro area member countries continue to explain a significant share of excess global imbalances.

Figure 10.Global Current Account Gaps, 2017 1/

(Percent of world GDP)

Source: IMF staff calculations.

1/ Only ESR countries are included, and systemic 5 economies reported individually. EA economies with positive (negative) CA gaps include Germany and the Netherlands (Belgium, France, Italy and Spain).

15. Persistence of excess external imbalances—especially on the surplus side—continues to be a feature of the global landscape (Figure 11). The same set of economies, especially on the side of positive excess imbalances, has displayed sizable excess external imbalances for much of the post-global financial crisis period. Higher-than-desirable current account balances have been recorded for a sustained period in northern European (Germany, Netherlands, Sweden) and Asian (China, Korea, Malaysia, Singapore) countries, although positive gaps narrowed somewhat in some (China, Korea, Malaysia, Sweden). Meanwhile, on the side of negative excess imbalances, more reconfiguration has occurred in the past and remains under way. Narrowing negative gaps in some EMDEs (Brazil, India, Indonesia, Mexico), debtor euro area countries (France, Italy, Spain) and key oil exporters (Saudi Arabia) were more than offset by higher negative gaps in key AEs (mainly United States). The constellation and evolution of excess external imbalances suggest that price adjustment mechanisms remain weak, notably for surplus countries, and that policy actions have been generally inappropriate or insufficient.

Figure 11.ESR Assessments, 2012–17 1/

1/ Grouping and ranking based on countries’ average excess imbalance during 2012–17. Coverage of Argentina in the ESR started in the 2018 ESR.

IV. Developments Since 2017, Outlook and Policies

This section discusses external developments since 2017, as well as prospects for the evolution of external balances and risks emanating from them. Policy actions to address excess external imbalances are discussed, both from an individual and multilateral perspective.

Developments since 2017 and outlook

16. Until recently, most major currencies had moved in a direction consistent with reducing excess external imbalances (Figure 12). These shifts generally reflected changes in growth prospects and associated policy responses in key systemic economies, although currency movements since April 2018—dominated by the strengthening of the US dollar—could instead aggravate future excess global imbalances.

  • Systemic currencies: Against a backdrop of some narrowing in interest rate differentials among key systemic economies, and partly reflecting changing fundamentals, REER movements through May 2018 (relative to the 2017 average) appeared supportive of reducing some systemically important external gaps, with US dollar weakness, mirroring strengthening of the euro and renminbi.5 However, conditions remain very fluid, as weaker-than-anticipated growth in the euro area and Japan, and stronger prospects for US monetary policy tightening have led to sharp US dollar appreciation since April. Political uncertainties, especially in Italy, have recently weakened the euro further, exacerbating previous trends.
  • Other currencies: Several EMDEs have witnessed sizable currency movements since 2017 against the backdrop of a tightening in external financing conditions, escalating trade tensions, and domestic political uncertainties in some cases. Depreciation pressure has been greatest in economies with larger excess external imbalances, including in Argentina and Turkey, where currency shifts should gradually support the narrowing of their weak external positions (see Box 4 on the recent EMDE volatility). Meanwhile, moderate real currency appreciations in some Asian economies (Korea, Malaysia, Thailand) have been consistent with those countries’ needs to narrow their excess surpluses.

Figure 12.2017 REER Gap vs. REER Changes in 2018 1/


Sources: INS, and IMF staff calculations.

1/ Blue (red) dots correspond to currencies that have moved in a direction consistent (inconsistent) with closing 2017 REER gaps.

17. Under baseline policies, current account surpluses and deficits are projected to further widen and concentrate in AEs. A significant share of global imbalances, especially surpluses, is stationed in economies where limited exchange rate flexibility constrains relative price adjustment, arguably contributing to sustaining imbalances over time. Under current policies, the United States would become an even more important contributor to global current account deficits because of its projected fiscal easing (Figure 13; Box 5), with a likely resulting increase in surpluses of other countries. Meanwhile, current account surpluses are also projected to further rotate toward AEs, primarily as China’s contribution to global current account surpluses is expected to fall by nearly half over the medium term. Surpluses in northern Europe and advanced Asia are expected to persist and possibly widen, partly reflecting relatively tight fiscal policy in these economies and further loosening fiscal policy in key trading partners, namely the US. The strengthening of oil prices will likely play a more muted role than in the past, primarily because the US current account has become less sensitive to energy price movements (see Box 6). Uncertainties persist over the implications of tighter global financial conditions for EMDEs, although countries with vulnerable external positions—such as Argentina and Turkey, whose currencies have already weakened substantially—would likely observe demand contraction and a consequent reduction in their deficits.

Box 4.Recent Financial Market Volatility in Key EMDEs: Role of External Fundamentals1

Recent volatility: EMDEs have witnessed considerable financial market volatility since April 2018. Against the backdrop of rising US interest rates, an appreciating US dollar, and rising vulnerabilities in some countries, some EMDE currencies have come under pressure, with the Argentine peso and Turkish lira being among the hardest hit, despite foreign exchange intervention.1 Other asset classes (credit default swap spreads, equities) came under pressure, and exchange-traded funds and mutual funds experienced outflows (notably in Indonesia, Malaysia, and Poland). Relative to the 2013 taper tantrum episode, the recent sell-off took place in a more supportive global context, with falling global risk aversion and a smaller rise in emerging market bond spreads. Commodity prices, especially the price of oil, rose during the recent sell off, partly shielding commodity exporters (Russia), while hurting others (Turkey).

Box Figure 4.1.Selected EMDEs: Portfolio Debt Liabilities and Currency Movements April-June 2018 1/

Sources: Haver, Balance of Payments and International Investment Position Statistics.

1/ Dot size is proportional to portfolio debt liabilities as a share of GDP.

Role of fundamentals: EMDEs with weaker fundamentals and policy frameworks have been most affected. This has been particularly true for countries with weaker external positions, reflecting a combination of high external financing needs, low reserve adequacy, large nonresident participation in local bond markets, and unhedged foreign currency exposure. The composition of external liabilities has mattered too, as countries with a higher share of portfolio debt liabilities faced significantly more pressure (see Box Figure 4.1). Fundamentals appear to have been a more important driver of asset price movements during the recent volatility episode than during taper tantrum, where EMDEs were hit more indiscriminately (see Box Figure 4.2).

Policy response: Policy responses have varied, although near-term actions have focused on raising policy rates, most notably in Argentina and Turkey, with foreign exchange intervention in some cases. Between late April and late May, Argentina and Turkey raised policy rates by 1275 basis points and 300 basis points, respectively, and foreign exchange reserves declined, in both cases. In general, countries with excess current account deficits will need to adopt a credible policy package involving growth-friendly fiscal consolidation, including to reduce excess reliance on monetary policy. Improving the composition of the international investment position also remains of essential to limit currency and maturity mismatches. Exchange rate flexibility will help buffer the shocks, although where reserves are adequate foreign exchange intervention could be considered to deal with disorderly market conditions.

Box Figure 4.2.Selected EMDEs: External Position and Currency Movements, Recent Episode and 2013 Taper Tantrum

Sources: Haver and IMF Staff estimates.

1/ For Argentina, 2017 CA gap is used in both the charts.

1 Prepared by Swarnali Ahmed Hannan and Kyun Suk Chang. The recent volatility episode refers to the period April 23-June 6, 2018. That said, some countries (Argentina and Turkey) started facing some market pressure earlier. For comparability, a 45-day window (starting May 22, 2013) is used when looking at the taper tantrum episode.

Figure 13.Evolution of Current Account Balances, 1996–20231/

(Percent of GDP)

Sources: WEO and IMF staff estimates.

1/ Projections assume constant global discrepancy (prorated among surplus economies).

Box 5.Tax Reform and the US Current Account1

The Tax Cuts and Jobs Act (TCJA), passed in December 2017, represents a significant change in the US tax system, with repercussions that go well beyond US borders. Its potential impact on the US current account balance can be thought of as operating along three dimensions or channels:

International business provisions (compositional effects): The TCJA changes the way the US tax system interacts with other jurisdictions. Previously, the US taxed worldwide income of US multinationals on repatriation with a nonrefundable credit for foreign taxes paid and a tax liability that was deferred until dividends were paid from the foreign subsidiary to the US parent. The TCJA moves the system toward a territorial system, excluding from US taxation business income that is earned abroad, via three provisions:

  • Global intangible low-taxed income (GILTI): To lessen the tax advantage afforded to profits accruing in low-tax jurisdictions, a minimum 10.5 percent corporate tax rate on the aggregate income of offshore subsidiaries of US corporations is imposed.
  • Base erosion and anti-abuse tax (BEAT): To prevent base erosion through transfer pricing and the holding of intellectual property offshore, a minimum tax is imposed on certain multinational companies with annual gross receipts higher than US$500 million, which is equivalent to the larger of: (1) a fixed percentage of “modified” taxable income that treats as income deductions claimed for cross-border payments to affiliates that are not related to goods trade (essentially encompassing service payments, interest, rents and royalties) or (2) the regular net tax liability under the normal corporate income tax base (with exceptions for research and development and other specific credits).
  • Foreign derived intangible income (FDII). FDII reduces from 21 to 13.125 percent the corporate tax rate for income arising from the sale of goods or services to non-US parties in excess of an amount equivalent to 10 percent of tangible assets. While GILTI and BEAT are the “sticks” applied to multinationals for taking their operations outside the US, FDII is the “carrot” for keeping their operations in the US.

The TCJA provides significant disincentives to producing and booking income outside the US, thus helping improve the US trade balance. However, it is unlikely to materially affect the US current account balance. Reduced profit booking outside the US will likely translate into lower investment income recorded in the income balance—that is, dividends and retained earnings from foreign subsidiaries—of a magnitude likely commensurate with the reduced profits booked abroad. Retaliatory actions by key low-tax jurisdictions may, however, mitigate the impact of the US reforms on the composition of the current account balance.

Domestic business provisions (efficiency effects): Domestic provisions in the TCJA, which effectively lower business tax rates, could affect the US current account through changed incentives to invest domestically:

  • Lower statutory and pass-through business tax rates: The TCJA lowered the federal statutory rate on incorporated businesses from 35 to 21 percent-bringing the US rate closer to the median tax rate of other OECD countries-and introduced a deduction for pass-through entities that file taxes under personal income (effectively lowering the top marginal rate from 39.6 percent to 29.6 percent).
  • Accelerated investment expensing: The TCJA shifted from a system of bonus depreciation (that allowed for a front-loaded write-off of 50 percent for certain types of corporate capital spending) to a system of full expensing of various forms of new and used tangible property (with a recovery period of less than 20 years). This immediate write-off applies to investment through the end of 2022, with the extent of expensing gradually falling during 2023–27.

The permanent cuts in the business rates may have a positive effect on the current account in the long term through increased domestic production capacity and exports. However, these are likely to be small since the new system is overall not assessed to be notably more efficient than the previous one). The accelerated investment provision is temporary and leads to a short-term stimulus and current account deterioration, but would not have any long-term effects on output and the current account.

Overall implications (stimulus and compositional effects): The TCJA is a reform law in nature. However, its business and personal provisions combined also involve a sizable near-term tax cut of 1¼ percent of GDP. With the US economy operating near potential, this fiscal stimulus is projected to have limited impact on output. Nonetheless, provided there are no retaliatory actions by US trading partners, it will still lead to a measurable deterioration of the current account in the near term. Over the medium term, the impact of the TCJA on the current account is expected to be mostly compositional—an improvement in the trade balance that would broadly wash out a weakening in the income balance.

1 Prepared by Ali Alichi (WHD).

Box 6.The Energy Revolution and the US Current Account Balance1

Evolution of external balances: Global balances, after peaking in 2006–07, corrected sharply in the years following the global financial crisis and have remained generally unchanged since 2013. Similar trends apply to the United states whose current account deficit has fluctuated around 2½ percent of GDP in recent years, well below the precrisis 6 percent deficit. A major force behind these trends has been energy commodities—whose prices boomed in the mid-2000s to later collapse in 2015 as the United States entered a new era of energy abundance (see Box Figure 6.1).

Box Figure 6.1.US Oil and Nonoil Current Account Balance, 2004–17

(Percent of GDP)

Sources: US Energy Information Administration and World Economic Outlook.

US shale gas revolution: High natural gas prices in the mid-2000s spurred innovation in the gas upstream sector, which led to the adoption of hydrofracking and directional drilling techniques. These allowed tapping vast amounts of shale gas reserves that were unprofitable before. Since the early 2000s, U.S. shale gas production has seen a 20-fold increase—making the United States the largest natural gas producer in the world and a net exporter of natural gas. Cheap and abundant natural gas also displaced coal from power generation, reducing energy imports. These technological improvements spread to the oil upstream sector and have resulted in a sharp increase in U.S. shale oil production. This surge in U.S. production contributed to the excess supply of oil, which eventually led to the collapse of oil prices in 2014.

Decomposition of Changes in US energy balance, 2005–17 1/(percent of GDP)
Avg.2015–17vs.Avg. 2005–07
Oil price effect0.14
Oil quantity effect1.39
Natural gas price effect0.03
Natural gas quantity effect0.16
Coal price effect0.01
Coal quantity effect0.02
Source: US Energy Information Administration and World Economic Outlook.

Average energy balance for 2005–07 and 2015–17 are calculated using average prices and quantities during those periods. The changes are shown as rations of the average GDP between 2005 and 2007. Petroleum products and crude oil are included in oil category.

Source: US Energy Information Administration and World Economic Outlook.

Average energy balance for 2005–07 and 2015–17 are calculated using average prices and quantities during those periods. The changes are shown as rations of the average GDP between 2005 and 2007. Petroleum products and crude oil are included in oil category.

US current account implications: The US energy deficit narrowed from an average of 2.5 percent of GDP in 2005–07 to a balance of nearly 0.3 percent in 2015–17 (see Box table). About two-thirds of this improvement (1.4 percent of GDP) can be attributed to an increase in shale oil production and exports of refined products, with the fall in oil prices playing only a minor role. Given today’s production structure, a $20 increase in oil price, all else equal, would lead to at most a 0.1 percent of GDP increase in the US current account deficit, although the impact could be even smaller (or positive), assuming price changes lead to higher natural gas and shale oil production. In a much-changed landscape of energy independence, the US trade balance has become largely insulated from energy price changes, and going forward higher world energy prices may improve the US trade balance.

Global implications: The projected absolute real annual world oil price increase for 2018 is similar in magnitude to that observed between 2004 and 2005 ($16-$17). Assuming unchanged net import oil volumes (which is reasonable given the low price-elasticity of oil consumption), the projected oil price increase would generate a substantial wealth redistribution from net oil importers to oil exporters (see Box Figure 6.2). The impact on global imbalances, however, will likely be more modest than during earlier episodes. IMF staff analysis suggests that both lower global oil intensity use and the US energy revolution may imply that the oil price increase will have a milder wealth redistribution effect relative to the mid-2000s period. In particular, the redistribution of wealth from the United States and other oil-exporting advanced economies towards major oil exporters (Russia, Saudi Arabia) could be much smaller. Further research on the global implications of these recent trends is of the essence.

Box Figure 6.2.Price change effect on oil balance 1/

(Percent of World GDP)

Sources: International Energy Agency, Global Assumptions and World Economic Outlook.

1/ Price change effect is calculated using following formula: {(Crude Oil Price_t – Crude Oil Price_t-1) * volume of net exports_t-1}/World GDP_t-1. Country sample includes EBA countries, non-EBA ESR countries and other major crude oil exporters. Crude oil price is a simple average of three spot prices provided by Global Assumptions database. Due to a restriction in the coverage of IEA database, quantity of oil (crude oil + oil products) net exports from 2016 (or 2015) is used to calculate 2017–18 price change effect. Oil exporters include NOR, RUS, SAU, IRN, NGA, VEN, ARE, IRQ, KWT, LBY, KAZ, AGO, DZA, OMN and QAT.

1 Prepared by Akito Matsumoto and Andrea Pescatori, with research assistance from Kyun Suk Chang and Lama Kiyasseh.

18. Meanwhile, global stock positions will continue widening. Absent valuation changes, net international investment positions (NIIPs) are projected to continue expanding as sustained current account surpluses remain in the largest creditor economies, and deficits in debtor economies.

  • Creditors: Particularly marked are the projected expansions of the NIIPs of Germany, Japan, Korea, the Netherlands, and other northern European countries. China’s NIIP will continue to grow in relation to global GDP, yet is expected gradually to fall in relation to its own GDP as sustained current account surpluses are more than offset by high GDP growth. Meanwhile, the NIIPs of many oil exporters are projected to expand moderately.
  • Debtors: Widening external creditor positions in the above-mentioned economies will be mirrored primarily by increased external indebtedness of the United Kingdom and the United States, where negative NIIPs are expected to reach 30 and 50 percent of GDP, respectively, over the next five years. This widening is expected to be accompanied by a narrowing of debtor positions of euro area countries (for example, Italy, Spain) owing to sustained current account surpluses. Debtor positions of some large EMDEs (Brazil, Indonesia, Mexico, Poland) will likely remain stable over the medium term.

19. Persistent excess external imbalances—amid weak automatic adjustment mechanisms and key policy actions contrary to stabilizing external positions—pose risks to the global economy, with distinct implications over the short and medium terms.

  • Short term: Tighter global financing conditions and protectionism. With a closed output gap, the fiscal impulse from the US tax reform could lead to a faster monetary tightening than currently envisaged.6 Ensuing US dollar appreciation would combine with strong demand to further widen the US current account deficit and the corresponding surpluses in other economies, possibly intensifying the US administration’s approach to reducing its deficit through trade measures. New trade barriers and possible retaliatory actions could derail global growth, with likely only limited impact on excess global imbalances (see Section V),7 Adverse balance sheet effects in many EMDEs from an even more rapid pace of US dollar appreciation and tightening of financing conditions could lead to disruptive adjustment, especially in countries where external balances and balance sheets are weak.
  • Medium-term: Disorderly adjustment. Continued reliance on demand from debtor countries (most notably, the United States) could constrain global growth going forward as debtor positions grow and become a drag on spending. Weakening stock positions could also lead to sharp and disruptive currency and asset price movements over the medium term as debt limits are approached and spending therefore falls abruptly. High and rising public debt levels exacerbate these risks. Sustained competitiveness asymmetries within the euro area, if unaddressed, would lead to protracted subpar demand growth, or a resurgence of unsustainable deficits, posing serious risks to the currency union and the global economy. Meanwhile, if not properly tackled, unbalanced domestic demand in China, which has been overly reliant on credit and investment, could culminate in an abrupt growth slowdown and a resulting resurgence of large external surpluses. The latter, in turn, could be met with stiff protectionist responses.

20. With the global economy operating at or above potential, macroeconomic policies will need to be properly sequenced and calibrated to achieve domestic and external objectives (Figure 14; Table 5). Surplus and deficit countries alike will need to normalize policies and rebuild policy space in a manner consistent with addressing excess external imbalances. In particular:

  • Most economies with weaker-than-warranted external positions need fiscal consolidation over the medium term (Figure 15), so in general, strengthening fiscal positions should be a priority. In key advanced economies (Canada, United Kingdom, United States), policies should be calibrated with a bias toward rebuilding fiscal space while protecting the most vulnerable segments of the population. Monetary normalization should proceed gradually in line with inflation objectives. In euro area debtor economies (France, Italy, Spain) growth-friendly fiscal consolidation will remain necessary to support the process of internal devaluation and strengthen external stock positions. In some cases (Canada, Turkey), keeping domestic credit in check would also facilitate external rebalancing. In some EMDEs with weak external positions, more ambitious and credible medium-term fiscal consolidation programs will be necessary (Argentina, Turkey), not only to reduce excess external imbalances, but also to stem capital outflows and reduce reliance on restrictive monetary policies. Flexible exchange rates should help support the adjustment process, while countries with adequate foreign reserve buffers should limit intervention to address disorderly market conditions.
  • In economies with stronger-than-warranted external positions and fiscal space, but without independent monetary policy (Germany, Netherlands), a more expansionary fiscal stance and policies to foster domestic credit growth would facilitate external rebalancing, including through real exchange rate adjustment, while monetary conditions remain somewhat accommodative to support demand in other currency union member countries. In other economies with stronger-than-warranted external positions and fiscal space (Korea, Thailand) looser fiscal policy would help to close negative output gaps. In China, a gradual tightening of fiscal and credit policies, necessary to reduce growing domestic vulnerabilities, should be accompanied by reforms to reduce precautionary saving and overcapacity in certain sectors.
  • In some countries with broadly-in-line external positions, policies will need to be carefully calibrated to avoid the emergence of excess external imbalances. In Japan, the fiscal consolidation necessary over the medium term should be accompanied by structural reforms to boost investment and prevent a further rise in the current account surplus (see ¶21), while monetary policy remains accommodative to facilitate reflation. Similarly, fiscal consolidation in some high-debt EMDEs (Brazil, India) should be met with reforms to boost investment and further improve the composition of external financing.

Figure 14.Evolution of Output Gaps, 2003–23 1/

(Percent of potential GDP)

Sources: WEO and IMF staff calculations.

1/ GDP-weighted averages per group. Only ESR countries are reported.

Table 5.2017 Individual Country Assessments: Summary of Policy Recommendations
Country NameOverall 2017 AssessmentPolicy recommendations 1/
ArgentinaWeakerFiscal consolidation to reduce the current account deficitImplement structural reforms that would increase productivity and competitiveness, and attract FDI
AustraliaBroadly ConsistentGradual, medium-term consolidationFurther monetary accommodation warranted, if growth was on the weak side, or commodity prices fell again
BelgiumWeakerSteady consolidation with labor tax reductionProduct and labor market reforms (to address labor market fragmentation), and wage moderation
BrazilBroadly ConsistentConsolidation (social security reforms and new federal spending cap)Structural efforts to improve overall competitiveness
CanadaModerately WeakerMedium-term consolidation, while increasing public infrastructure investmentMaintaining tight macroprudential policies to ensure financial stabilityImprove labor productivity, including by investing in R&D and physical capital, promoting FDI. Diversify export markets, especially into services
ChinaModerately StrongerSupport deleveraging by gradually consolidating to bring the primary balance to the debt-stabilizing levelGradually move toward more transparent, market-based MP framework and ER flexibility, which may require use of FX reserves to smooth excessive volatilityImprove social safety nets; SOE reform and open markets to more competition; create a more market-based and robust financial system; take steps to attract more inward FDI, including by ensuring that foreign investors receive the same treatment as domestic investors
Euro AreaModerately StrongerImplement more growth-friendly composition of national fiscal policies. Expand investment in countries with fiscal space. Centralized investment schemes at regional level. Strengthen euro area fiscal capacity for macroeconomic stabilizationRemain accommodative until inflation durably converges to ECB’s medium-term price stability objectiveEnhance productivity, increase competitiveness; strengthen private sector balance sheets; make currency union more resilient with banking and capital markets union; facilitate relative price adjustments at the national level by enabling greater inflation differentials across euro area members
FranceModerately WeakerSteady consolidationEnhance productivity; increase competitiveness through labor market reforms and wage moderation
GermanySubstantially StrongerA more growth-oriented fiscal policyReforms to foster entrepreneurship, and address aging costs by prolonging working life
Hong Kong SARBroadly ConsistentContinue prudent fiscal managementRobust and proactive financial supervision; maintain flexible wages and prices.
IndiaBroadly ConsistentExchange rate flexibility should remain the main shock absorber, with intervention limited to addressing disorderly market concernsGradual liberalization of portfolio flows. Ease domestic supply bottlenecks and revamp business climate, improve competitiveness and investment prospects, to attract FDI and boost exports.
IndonesiaBroadly ConsistentStrengthen fiscal position by accelerating tax reforms to boost investor confidence; Keep fiscal deficit below the legal limitContinued ER flexibility and market-determined bond yields would continue to underpin external stabilityExpand social safety nets. Ease non-tariff trade barriers and FDI restrictions. Continue infrastructure investment and strengthen human capital. Deepen financial markets
ItalyBroadly ConsistentGradual consolidation to maintain investor confidenceImplement reforms to better align wages with productivity at the firm level and to strengthen banks balance sheet to unlock investment potential
JapanBroadly ConsistentGradual fiscal consolidation anchored by a credible medium-term fiscal frameworkContinued accommodative stance by the Bank of Japan to achieve inflation objectivesAdopt measures to boost wages and labor supply, reduce labor market duality, enhance risk capital provision, reduce barriers to entry in some industries, and accelerate agricultural and professional services sector deregulation
KoreaModerately StrongerMore expansionary fiscal policy to boost domestic demandER flexibility with limited intervention to address disorderly conditionsStrengthen the social safety net to lessen incentives for precautionary savings. Address bottlenecks to investment
MalaysiaStrongerGradual medium-term consolidationExchange rate flexibilityImprove social protection; increase public healthcare spending; address structural bottlenecks (labor market skills mismatch; low female participation; weak education quality; further improve physical infrastructure
MexicoBroadly ConsistentGradual consolidationFree-floating ER policy, and use foreign exchange intervention occasionally to prevent disorderly market conditionsStructural reforms to improve competitiveness and strengthen non-oil exports
NetherlandsSubstantially StrongerExpansionary fiscal policyStructural reforms to raise the productivity of small domestic firms, repair household balance sheets, and strengthen the banking system
PolandBroadly ConsistentGradual structural fiscal consolidationTimely monetary policy responses to prevent nascent overheating and other market pressuresContinued reforms are crucial to boost structurally-low private investment and potential growth.
RussiaModerately WeakerUse new fiscal rule to reduce impact of oil volatility on non-oil sector; re-allocate from current to capital spending, while leaving space for health spendingStructural reforms to invigorate private sector and improve its compettiveness
Saudi ArabiaWeakerFurther consolidation over the short- and medium termStructural reforms to diversify the economy and boost the non-oil tradeable sector over the medium term
SingaporeSubstantially StrongerHigher public investment in physical infrastructure and human capital. More public health spending to reduce precautionary savingGradual normalization of monetary policy to support gradual appreciation of the real exchange rateStructural reforms to improve productivity and domestic investment incentives
South AfricaModerately WeakerPreserve debt sustainability, while allowing for greater public investmentSeize opportunities to build-up reserves to deal with FX liquidity shocksStrengthen education/skills; increase financial inclusion; foster entry into key product markets; accelerate labor and product market reforms
SpainModerately WeakerReduce the still-sizable structural fiscal deficitPress with additional reforms of the labor market and faster implementation of certain product market reforms
SwedenModerately StrongerAdopt a mildly expansionary fiscal stance consistent with the medium-term surplus targetContinued monetary accommodation to bring inflation back to target and support domestic demandFacilitate migrant integration into the labor market reduce household uncertainties around the sustainability of Sweden’s strong social model
SwitzerlandBroadly ConsistentMove to -and maintain- a structurally-neutral fiscal stanceForeign currency intervention should be reserved for addressing large exchange market pressuresReform corporate income tax to encourage SME investment, thereby reducing net saving
ThailandSubstantially StrongerBoost public infrastructure within available fiscal spaceGreater exchange rate flexibility with limited interventionStrengthen social safety nets, and reduce barriers to investment
TurkeyWeakerTighten fiscal and quasi-fiscal policies.Tighter monetary policy should aim at reanchoring inflation expectations. Further deceleration of credit growth is necessary. Increase net international reserves.
United KingdomWeakerFiscal consolidation with investment in public infrastructureMaintain financial stability through macroprudential policiesBroaden skill base; improve public infrastructure
United StatesModerately WeakerConsolidate over the medoim term, while upgrading public infrastructureEnhance schooling and training of workers; strengthen measures to support the working poor; increase labor force growth (including through skill-based immigration reform)
Source: 2017 Individual External Assessments.

This non-exhaustive list focuses on key recommendations for closing external imbalances.

Source: 2017 Individual External Assessments.

This non-exhaustive list focuses on key recommendations for closing external imbalances.

Figure 15.ESR Economies. Current Account Gaps and Policy Stance, 2017 1/

Sources: IMF staff calculations.

1/ Excludes non-EBA economies (Hong Kong SAR, Saudi Arabia, Singapore). Real short-term rates based on 1-year ahead inflation expectations. Fiscal gap measured as domestic component of the policy gap. Dot size proportional to absolute value of excess current account balance (percent of world GDP). Green (red) dots correpond to positive (negative) gaps. Broadly-in-line current accounts depicted in light blue.

21. Structural reforms will need to play a greater role to address excess external imbalances. With both surplus and deficit countries needing to rebuild policy space, structural reforms will be central to tackling excess global imbalances. Addressing product- and labor-market distortions is desirable in all countries, but product-market reforms that remove unnecessary burdens from starting a business in high-current-account-surplus countries and labor-market reforms that reduce labor costs in high-current-account-deficit countries would support the reduction of excess global imbalances.8 Specifically,

  • Economies with stronger-than-warranted external positions should focus their efforts on reforms to reduce excess saving, through safety net and pension reforms (China, Germany, Malaysia, Korea, Netherlands, Thailand) and policies to facilitate household balance sheet repair (Netherlands). Reducing entry barriers, including to boost investment, especially in the service sector (Germany, Korea, Thailand), reducing obstacles to residential investment (Sweden) and boosting public investment (Germany, Thailand) would also help reduce excess surpluses. Policy actions to tackle high and rising corporate saving would be desirable, although further analysis is needed in identifying the precise underlying distortions (see Box 7).
  • Meanwhile, reducing market frictions that increase labor costs could help improve competitiveness in countries with weaker-than-warranted external positions. Policy interventions could include enhancing schooling and training (France, United States) and broadening the skill base and labor force, including through immigration policies (United Kingdom, United States). Reforming wage bargaining mechanisms to moderate wage growth and better align wages with productivity gains (France, Italy) and reducing labor market segmentation (Spain) could help in some cases. Policies that strengthen euro area integration on the banking, fiscal, labor and regulatory fronts are necessary to boost investment throughout the currency area and to reduce its external imbalance.
  • In some economies with broadly-in-line current account balances, addressing underlying structural distortions masked by other policy gaps would also be important to prevent a resurgence of external excess imbalances. For example, In Japan, efforts are needed to boost wages and reduce entry barriers in some industries, while in Italy improving wage bargaining mechanism and strengthening bank balance sheets are necessary to strengthen competitiveness.

22. Overall, with persistent excess external imbalances and the global economy operating at or above potential, a multilateral approach to sustaining growth and rebuilding policy space is needed. While a move toward regaining policy space ahead of the next slowdown is essential, rebuilding policy buffers must be consistent with sustaining globally balanced growth and tailored to country-specific conditions, including within currency areas. Addressing external imbalances should not come at the expense of growing domestic imbalances. Similarly, allowing currencies to float freely remains of essence to facilitate external adjustment. As discussed in the next section, surplus and deficit countries alike should avoid protectionist policies and instead work toward reducing trade barriers and ensuring a level playing field while maintaining reasonable certainty about the rules of international trade—a key element to fostering global trade, investment, and growth.

Box 7.Corporate Sector Saving in Current Account Surplus Economies1

The 2017 External Sector Report highlighted the role of nonfinancial corporate sector saving in explaining high and persistent current account surpluses in key AEs. This box provides additional details on the sources and uses of corporate saving in key surplus economies with the aim of shedding light on potential policy responses in countries where these surpluses appear excessive.

Sources of nonfinancial corporate sector gross saving: Nonfinancial corporate sector gross saving can be decomposed into corporate profits (gross operating surplus), property income (including income from rent, interest, dividends and net retained earnings from foreign direct investment) minus dividend, interest, and tax payments. Other items such as transfer payments and net social security adjustments are quantitatively small. For all five major AE surplus countries, rising nonfinancial corporate gross saving has been driven by higher profits from both domestic production (aided by lower labor income shares) and expanding foreign operations, as well as lower interest payments. These higher profits have not been matched by higher dividend or income tax payments (see Box Figure 7.1). This trend has continued, albeit at a slower rate, after the global financial crisis, when dividend payments stopped growing with profits, even falling in some cases (for example, Germany).

Box Figure 7.1.Surplus Economies: Changes in NFC Savings, 1995–2016

(Percentage of GDP)

Sources: OECD National Accounts Dataset, Country Authorities, IMF WEO and Staff calculations.

Notes:. Chart shows GDP-weighted average of 5 surplus countries: Germany, Netherlands, Japan, Korea and Denmark. FDI RE is Net Retained Earnings on FDI. Other PI (Property Income) consists of investment income on financial assets and net rent receivable.

Uses of nonfinancial corporate gross saving: Corporate saving in excess of investment (that is, corporate net lending) can be used to buy back shares, pay down debt, or acquire financial assets. While all uses have been relevant in different countries and years, simple regressions using flow of funds and sectoral financial accounts data for surplus countries show that the accumulation of cash has been the most salient use of corporate net lending. The correlation between net lending and cash holding has been particularly strong in Germany and the Netherlands (see Box table and Box Figure 7.2).

Box Figure 7.2.Correlation between Cash Ratio and Corporate Net Lending in Germany and Netherlands

Sources: OECD National Accounts Dataset, IMF WEO and Fund Staff calculations

1 Prepared by Mai Chi Dao and Deepali Gautam.

Policy implications: Understanding the underlying motives for rising net lending and liquidity demand will help formulate policy advice. In addition, the fact that rising nonfinancial corporate saving is reflected in higher current account surpluses suggests that households are not “piercing the corporate veil,” as economic theory would predict. While various country-specific characteristics and frictions may explain the corporate veil puzzle, a few policy-related factors have likely contributed to the build-up of high nonfinancial corporate net saving and current account surpluses, including: (1) declining corporate income tax rates and increased profit shifting activity of multinationals (see Zucman 2018), (2) shifts in corporate governance structures that favor retained earnings and share buybacks over dividend payouts or long-term investment (seem for example, Philippon and Gutierrez 2016), and (3) unequal wealth distribution concentrated among households with low propensity to consume (see IMF 2013). Further research is required to better understand the role policies could play in this area.

Top Surplus Countries: Use of Net Lending
Dependent Variable: NFC Net Lending/GDP
Equity assets–0.0140.133**
Loan/Debt assets0.262***0.146
Debt repayment0.222**0.136
Share buyback0.330*0.373
Number of Observations6363
* significant at 10%; ** significant at 5%; *** significant at 1%Notes: Column 1 is pooled, column 2 within-country regression. Cash, equity and loan/debt assets are in percent of total financial assets. Debt repayment and share buyback are net negative transactions in debt and equity liability, in percent of total financial assets.
* significant at 10%; ** significant at 5%; *** significant at 1%Notes: Column 1 is pooled, column 2 within-country regression. Cash, equity and loan/debt assets are in percent of total financial assets. Debt repayment and share buyback are net negative transactions in debt and equity liability, in percent of total financial assets.

V. Other Considerations: Trade Costs and Current Accounts

23. Conventional economic wisdom holds that the drivers of external current account balances are primarily macroeconomic in nature: the current account represents the excess of national saving over investment, whose drivers in turn are macroeconomic. The IMF’s approach to assessing imbalances reflects this conventional wisdom and the vast theoretical and empirical literature on this topic. Current accounts in the EBA model are pinned down by macroeconomic fundamentals, including demographics, income per capita, and growth prospects, as well as by domestic policies.

24. Trade policies (the topic of this section) can have macroeconomic effects although those effects do not center on the current account balance. Protectionist policies, to the degree they impact macro outcomes, would have deleterious effects on the quantum of trade between nations, and on economic efficiency and productivity over time (Krugman 1982, Sen and Turnovsky, 1989, Ostry and Rose, 1992, Obstfeld, 2016). There might also of course be considerable sectoral allocative effects from trade policies, since such policies by definition favor certain sectors over others. It is difficult, however, to find in the trade literature a well-documented channel through which trade policies of realistic magnitude have quantitatively significant effects on current account or trade balances.

25. The conventional wisdom of the weak linkage between trade policies and current account balances has been challenged recently. The challenge comes from a stream of new models (Barattieri, 2014, Joy and others 2018; Reyes-Heroles, 2016) that have highlighted particular channels (e.g. interest rate channel, wealth effect) through which commercial policies might have transitional effects on saving and investment decisions. The precise impact, however, depends on model specifics and calibration which often require perfect foresight of the evolution of trade policies or substantial rigidities that temper consumption or investment adjustments. The challenge has also come from a policy perspective (Carney 2017, Joy and others 2018), which argues that asymmetries in the reduction of trade policy barriers across sectors (i.e. faster in manufacturing than in services) can partially explain the evolution of global imbalances, as countries specializing in services tend to run lower current account balances due to relatively higher trade policy barriers. Properly addressing this question, however, requires looking at a broad measure of trade costs, which captures all the costs associated with the cross-border movement of goods and services, ranging from tariff and shipment costs to the nontariff barriers. This is necessary since both natural and policy barriers to trade have differed markedly across sectors and over time. For example, trade in services is more sensitive to natural inhibiting factors such as geographic distance, and cultural and language differences.

26. This section sheds further light on the empirical impact of broader trade costs (natural and policy-related) on current account outcomes. Whether trade costs in prime export sectors help explain current account balances merits a rigorous investigation that addresses the empirical challenges of measuring trade costs and comparative advantage. Using data on bilateral trade flows and a parsimonious structure from the trade literature (Eaton and Kortum 2002), the analysis infers trade costs and comparative advantage across multiple sectors for a globally representative sample of countries (Box 8).9 Unlike in existing studies, estimated trade costs can vary between imports and exports, and capture both tariff and nontariff barriers as well as other behind-the-border barriers that are difficult to quantify but potentially pervasive, especially for services. The analysis constructs country-specific trade costs weighted by comparative advantage—that is, aggregate effective trade costs—to capture the height of net barriers to countries’ natural exports and natural imports. Finally, it uses the recently refined EBA current account model to test whether effective trade costs, separately when exporting and importing, are significant drivers of current account balances.

27. Results suggest that effective export costs have a statistically significant—but moderate—impact on current account outcomes. Specifically, for the sample period 1986–2009, a 10-percentage point unilateral reduction in aggregate effective export costs for an average country is associated with a current account balance improvement equivalent to ½ percent of GDP. The estimated effects are smaller when looking at the more recent period (2001–14). 10 Effective costs to import have generally statistically insignificant effects. These findings are consistent with theoretical predictions that generally suggest limited effects of trade costs on current account balances, coming mainly through mechanisms that induce transitory fluctuations in income and intertemporal prices. These results are generally robust across several empirical specifications and extensions, including using alternative weights to aggregate bilateral trade costs across partners, excluding countries that are known to engage in state support, or including the unexplained component of the gravity model in the trade costs estimates.

Text Table 1.Trade Costs and the Current Account 1/
Effective Exporting Cost–0.049***


Effective Importing Cost0.001


Sources: CEPII Gravity Dataset; IMF, EBA data set; Johnson and Noguera (2017) for 1986–2009; and World Input-Output Database 2016 for 2001–14.

Uses refined EBA model to estimate impact of effective trade cost on current account. P values are in parentheses. *** p<0.01; ** p<0.05; * p<0.1.

Sources: CEPII Gravity Dataset; IMF, EBA data set; Johnson and Noguera (2017) for 1986–2009; and World Input-Output Database 2016 for 2001–14.

Uses refined EBA model to estimate impact of effective trade cost on current account. P values are in parentheses. *** p<0.01; ** p<0.05; * p<0.1.

28. However, limitations of the approach entail that the results need to be interpreted with caution. While the estimated trade costs are comprehensive, in that they capture both tariff and nontariff barriers, and the comparative advantage measure is more reliable than other commonly used measures, it also has limitations. In particular, data constraints prevent a separate estimation of import barriers and export subsidies, implying that import costs would be underestimated for countries with significant export subsidies whereas comparative advantage would be overestimated in sectors receiving state support. In addition, the approach does not necessarily capture all barriers to investment that apply to both domestic and foreign firms, which could have a significant impact on current account imbalances (Box 3). Going forward, tackling these limitations require better measures of costs, especially those that pertain to services trade, where trade often requires commercial presence and thus, investment restrictions are more relevant.

Box 8.The Methodological Approach to Examining Trade Costs and Imbalances

Challenge: Measuring trade costs and comparative advantage is empirically challenging. Trade costs include all costs associated with the cross-border movement of goods and services, ranging from relatively easy-to-measure expenses for tariffs and goods shipment to the difficult-to-measure nontariff barriers especially in the case of services (Cerdeiro and Nam 2018). Comparative advantage depends on the differentials between domestic and foreign goods and services prices that would have prevailed had trade been impossible.

Approach: The proposed methodology uses the well-established gravity model established by Eaton and Kortum (2002) to estimate from bilateral trade flows normalized to remove domestic demand effect: (i) a country’s export capability; (ii) the toughness of import competition; and (iii) bilateral trade costs, which include both natural barriers (distance, common border or language, colonial relationship) and man-made policy barriers. The approach estimates all three factors simultaneously to reconcile with bilateral trade shares observed in the data.

  • The estimated export capability is used to proxy productivity and to construct comparative advantage (Hanson, Lind and Muendler, 2016). Specifically, the measurement of comparative advantage involves computing absolute advantage in each sector using the estimated export-specific factors, and normalizing that with country-specific averages across sectors to remove the effects of aggregate growth.
  • To infer trade costs, it assumes that for a country without barriers export capability and import competitiveness in a sector would be equivalent, since both are driven by productivity and marginal costs. In the event the estimated export capability (relative to a numeraire country) is lower than the estimated relative import competitiveness, it is reconciled by the existence of a positive (relative) import barrier. Under the approach any difference between the export capability and import competitiveness is attributed to the nondiscriminatory import barriers. The estimated nondiscriminatory import barriers are then combined with bilateral trade costs to arrive at the final estimates of the importing costs and exporting costs with a trading partner.
  • These estimates are inferred from two alternative data sets on bilateral trade. Johnson and Noguera (2017) provides a balanced panel of bilateral trade in the 1970–2009 period for 37 countries and the 2016 World Input-Output Database covers the 2001–14 period for a similar set of countries. The analysis focuses on agriculture, manufacturing and services.

Arriving at the effective trade cost: Two sets of trade costs are computed for each country and sector--those faced when exporting and those faced when importing--aggregating across trading partners using their respective lagged export and import shares as weights. Country-specific trade costs are weighted by comparative to arrive at the effective trade cost measure, which is meant to reflect the height of barriers on the basis of a country’s underlying trade potential. The measure effectively assigns weights based on trade that would have prevailed had there not been trade costs to avoid the underestimation inherent in the standard practice.

Augmenting EBA model: Measures of effective costs to export and to import are then added to the most recent version of the current account model in the EBA as additional explanatory variables, after taking their differences relative to their respective world averages.

29. As to policy implications of course, aside from the potential impact on imbalances, lowering trade costs remains essential to reaping the benefits of trade. While trade costs do not appear to be a key driver of current account imbalances, lower trade costs can boost trade and foster a more efficient allocation of resources and boost productivity by spurring innovation, technology transfer, and competition. Though trade-induced structural change can be costly to those who are dislocated, complementary policies can ease the accompanying adjustment process (IMF/WB/WTO 2017). Resisting protectionist policies, reviving liberalization efforts, and strengthening the multilateral trading system remain essential pro-growth strategies—particularly to promote trade in services, where barriers remain relatively high and potential under-exploited trade gains therefore are likely to be high as well.

30. Looking beyond a narrow focus on tariffs, policies remain necessary to address behind-the-border barriers and distortions as well as reduce the trade-impeding effects of natural factors. Sustained government interventions can distort trade and potentially have protracted effects on resource allocation, and aggregate savings and investment. Further efforts are needed to eliminate these distorting practices, which include state-owned enterprise subsidies, technology transfer requirements, weak investment and intellectual property protection, within a revamped multilateral rules-based trading system. Moreover, natural barriers to trade could be lowered through improved infrastructure, enhanced connectivity to shipping networks, and investment in digital technologies. Such comprehensive efforts would not only promote trade along the intensive margin, but also extend the spectrum of goods and services that can cross borders.


    BarattieriAlessandro2014Comparative Advantage, Service Trade, and Global ImbalancesJournal of International Economics Vol. 92 pp. 113.

    • Search Google Scholar
    • Export Citation

    CacciatoreM.R.DuvalG.Fiori andF.Ghironi (2016a) “Market Reforms in the Time of ImbalanceJournal of Economic Dynamics and Control Vol. 72 pp. 6993.

    • Search Google Scholar
    • Export Citation

    CacciatoreM.R.DuvalG.Fiori andF.Ghironi (2016b) “Short-Term Gain for Long-Term Pain: Market Deregulation and Monetary Policy in Small Open EconomiesJournal of International Money and Finance.

    • Search Google Scholar
    • Export Citation

    CarneyMark2017A Fine Balance” speech at the Mansion House.

    CerdeiroDiego andRachelNam2018A Multidimensional Approach to Trade Policy IndicatorsIMF Working Paper 18/32 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    CheungC.D.Furceri andE.Rusticelli2013Structural and Cyclical Factors behind Current Account BalancesReview of International Economics Vol. 21(5): pp. 923—944.

    • Search Google Scholar
    • Export Citation

    EatonJonathan andSamuelKortum2002Technology, Geography and TradeEconometrica Vol. 70 pp. 174179.

    GhoshA. R. andJonathan D.Ostry1997Macroeconomic Uncertainty, Precautionary Saving, and the Current AccountJournal of Monetary Economics” Vol.40 pp. 12139.

    • Search Google Scholar
    • Export Citation

    GutiérrezGerman andThomasPhilippon2016Investment-less Growth: An Empirical InvestigationNBER Working Paper 22897.

    HansonGordonNelsonLind and Marc-Andreas Muendler2016The Dynamics of Comparative Advantage” (unpublished).

    International Monetary Fund2017External Sector ReportWashington D.C.

    International Monetary Fund2013Taxing Our Way Out of—or Into?—TroubleFiscal Monitors World Economic and Financial Surveys October Chapter 2.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund World Bank and World Trade Organization (IMF/WB/WTO)2017Making Trade an Engine of Growth for AllWashington D.C.

    • Search Google Scholar
    • Export Citation

    JaumotteF. andP.Sodsriwiboon2010Current Account Imbalances in the Southern Euro AreaIMF Working Paper 10/139International Monetary FundWashington.

    • Search Google Scholar
    • Export Citation

    JohnsonRobert andGuillermoNoguera2017A Portrait of Trade in Value-Added over Four DecadesReview of Economics and Statistics Vol. 99 pp. 896911.

    • Search Google Scholar
    • Export Citation

    JoyMarkNoemieLisackSimonLloydDennisReinhardtRanaSajedi andSimonWhitaker2018Mind the (Current Account) GapFinancial Stability Paper No. 43Bank of England.

    • Search Google Scholar
    • Export Citation

    KerdrainC.I.Koske andI.Wanner2010The Impact of Structural Policies on Saving, Investment and Current Accounts.” OECD Economics Department Working Papers No. 815OECD Publishing.

    • Search Google Scholar
    • Export Citation

    KrugmanPaul1982The macroeconomics of protection with a floating exchange ratein Carnegie-Rochester Conference Series on Public Policy Vol. 16 pp. 14182. North-Holland.

    • Search Google Scholar
    • Export Citation

    ObstfeldMaurice2016Tariffs Do More Harm Than Good at Home

    • Search Google Scholar
    • Export Citation

    ObstfeldMaurice andKennethRogoff2007The Unsustainable US CA Position Revisited” in: ClaridaR. ed. G7 Current Account Imbalances: Sustainability and AdjustmentThe University of Chicago Press.

    • Search Google Scholar
    • Export Citation

    OstryJonathanD. andAndrew K.Rose1992An Empirical Evaluation of the Macroeconomic Effects of TarrifsJournal of international Money and Finance Vol. 11 No. 1 pp. 6379.

    • Search Google Scholar
    • Export Citation

    Reyes-HerolesRicardo2016The Role of Trade Costs in the Surge of Trade ImbalancesPrinceton University (unpublished).

    SenPartha andStephen J.Turnovsky1989Tariffs, Capital Accumulation, and the Current Account in a Small Open EconomyInternational Economic Review Vol. 30 pp. 81183.

    • Search Google Scholar
    • Export Citation

    ZucmanGabriel2014Taxing across Borders: Tracking Personal Wealth and Corporate ProfitsJournal of Economic Perspectives Volume 28.

    • Search Google Scholar
    • Export Citation

Although the U.S. dollar’s year-average value was the same in 2017 as in 2016, the currency depreciated by about 6½ percent in real terms between the end of 2016 and the end of 2017. It is the latter change that matters for computing IIP valuation changes, as these are based on year-end positions and asset prices.


These models estimate the average historical relationship between current accounts or real exchange rates and a set of country fundamentals and policy variables from a panel of 49 countries for the period 1986–2016. See IMF Working Paper 13/272 and the 2015 EBA Refinements Technical Background Note for a full description of the EBA methodology and earlier refinements, respectively.


The overall positive-fiscal-gap contribution also reflects looser-than-desired policies in the rest of the world, including in excess deficit economies such as the United Kingdom and the United States.


The analysis is based on staff estimates and subject to a margin of error. Access to timely, accurate and comprehensive FXI data is key for Fund bilateral and multilateral surveillance. The publication of FXI data is also desirable as a matter of transparency and good policy, as it would allow increasing the understanding and credibility of macroeconomic policies and frameworks.


Meanwhile, the yen has been stable over the same period, while the British pound has strengthened somewhat, following its sharp depreciation during 2016–17, reflecting market perceptions of improved prospects of an orderly Brexit.


Excess global imbalances could also widen as the normalization of monetary conditions raises the costs of sustaining net external debtor positions (for example, those of the United Kingdom, United States, etc.) and the returns on net external creditor positions (China, Germany, Japan, Netherlands, etc.). Associated valuation changes could, however, offset some of the impact on stock positions.


See also analysis in Box 6 of the 2017 External Sector Report.


See also Box 7 of 2017 External Sector Report.


A forthcoming IMF Working Paper by Emine Boz, Nan Li and Hongrui Zhang will provide a more detailed exposition of the analysis.


The average effects, however, mask noticeable differences across countries, where trade costs can account for a nonnegligible portion of the negative current account gap in countries with higher-than-average effective costs to export (for example, United States). That said, trade costs appear to influence current account balances especially through natural, rather than policy-related, components.

Other Resources Citing This Publication