Global imbalances are back in the policy debate. The persistence of large current account deficits and surpluses is again being discussed alongside trade tensions, industrial policy, and concerns about financial stability risks. Recent VoxEU columns have asked whether today’s imbalances differ from those before the global financial crisis, whether tariffs can reduce trade deficits, and if industrial policy can durably alter external balances (Weder di Mauro and Zettelmeyer 2026, Itskhoki and Mukhin 2025, Gourinchas et al. 2026).

A key question is often missing from this debate: how have large current account imbalances unwound in the past? Drawing on new OECD research (Frohm et al. 2026), we examine 70 large and durable current account adjustment episodes across 51 economies from 1980 to 2024, excluding crisis-driven adjustments. Three lessons stand out: adjustments are fairly common but slow; deficit and surplus narrowing work through different channels; and larger imbalances – especially deficits – are significantly more likely to correct.

Large and durable adjustments are common, but they typically unfold over several years

Current account positions are highly persistent. Some economies have run deficits or surpluses for much of the past three decades. This persistence reflects structural and policy factors, including demographics, corporate saving behaviour, fiscal positions and tax policies, trade specialisation, as well as financial development, and countries’ role in the global financial system.

But many economies have experienced sizeable and lasting current account adjustments. We identify 70 episodes in which cyclically adjusted current account balances move by at least 3 percentage points of GDP and are sustained for at least three years, with at least half the adjustment maintained afterwards. Episodes coinciding with systemic banking crises are excluded to focus on underlying adjustment rather than crisis-driven compression. The approach builds on Milesi-Ferretti and Razin (1998) and Freund and Warnock (2007) but applies the same criteria symmetrically to both deficit and surplus positions.

Over three-quarters of these episodes involve narrowing, meaning that an initial deficit or surplus becomes smaller over the course of the episode (Figure 1). The median narrowing episode delivers a current account shift of around 5 percentage points of GDP and lasts approximately eight years. That duration is critical: durable rebalancing is a multi-year process involving sustained changes in trade flows, saving-investment behaviour, and relative prices. Policy debates focused on quick bilateral fixes are likely to be disappointed.

Figure 1 Large current account adjustment episodes are sizeable and gradual

Notes: Summary statistics for cyclically adjusted and smoothed current account changes, by episode type. The sample covers 51 countries over 1980–2024. Current account changes are cumulative in percentage points of GDP; values are medians across episodes.
Source: OECD National Accounts Database, IMF Balance of Payments Database and authors’ calculations. 

Deficit adjustment is mostly about exports and private net saving

Deficit-narrowing episodes are typically associated with stronger exports (Figure 2, Panel A). Weaker imports also contribute, but the dominant pattern is an improvement in export performance. These episodes are accompanied by real exchange rate depreciation, lower inflation, and a temporary weakening of activity, including a reduction in r output gaps and higher unemployment.

The saving-investment perspective adds an important layer. In deficit-narrowing episodes, the current account improvement is concentrated in the private sector, especially non-financial corporations. Their net lending rises strongly, reflecting both higher saving and lower investment relative to the start of the episode. Household net lending also increases, although more modestly. The government sector initially weighs on the adjustment, but its net lending position increases later in the episode.

Taken together, these patterns suggest that durable deficit narrowing is not simply import compression. It is more often a combination of improved external competitiveness, stronger export performance, higher private net saving.

That said, the adjustment is not costless. Even though GDP growth remains positive on average, a temporary reduction in output gaps and labour markets shows why deficit correction can be politically difficult. A country may improve its external position while facing weaker domestic demand in the transition.

Surplus adjustment is mostly about stronger domestic absorption

Surplus-narrowing episodes are not mirror images of deficit-narrowing episodes (Figure 2, Panel B). They are associated with rising imports and stronger domestic absorption, while exports actually increase. Real exchange rates appreciate, inflation rises, output gaps improve, and unemployment falls.

The sectoral saving-investment adjustment is also broader. In surplus-narrowing episodes, the reduction in net lending is spread across households, firms and governments. This points to stronger domestic demand rather than a loss of export capacity. Households save less, firms invest more, and governments contribute through higher public investment.

The contrast between deficit and surplus adjustment is central for policy. It suggests that surplus adjustment should not be understood mainly as a competitiveness problem. In many cases, durable surplus narrowing occurs when domestic demand strengthens. Such a process requires policies that reduce precautionary saving, support household consumption, improve investment opportunities, or use fiscal space in ways that strengthen investment and demand.

Figure 2 Deficit and surplus narrowing operate through different channels

Notes: Values show average changes from the start of current account narrowing episodes, in percentage points of GDP. Imports are sign-reversed, so negative values indicate higher imports. 
Source: OECD National Accounts, IMF Balance of Payments Database and authors’ calculations. 

Larger deficits are more likely to narrow

We then ask what predicts the onset of narrowing episodes, by estimating a pooled probit model in which the probability of entering a narrowing episode depends on lagged initial conditions. The model distinguishes between deficit and surplus observations and allows the effects of key predictors to differ across the two regimes. Years following the start of an episode are excluded from the baseline sample, so that the model focuses on observations genuinely at risk of a new episode.

The strongest result is intuitive: larger initial imbalances are more likely to narrow.

For deficit countries, the relationship is particularly strong. A larger initial deficit is associated with a substantially higher probability of entering a deficit-narrowing episode. The estimated marginal effect is almost three times larger when the initial deficit is around 6% of GDP than when it is around 1% of GDP (Figure 3). This points to non-linear adjustment pressure: deficits may persist for some time, but once they become very large, the probability of correction rises sharply.

For surplus countries, larger initial surpluses are also associated with a higher probability of narrowing, but the effect is smaller and varies less across different surplus levels. This asymmetry is consistent with a simple economic intuition. Large deficits create external financing needs and can generate pressure from creditors, markets, policymakers, or domestic balance sheets. Large surpluses do not face the same immediate financing constraint. Their unwinding may depend more on domestic policy choices, household saving behaviour, investment opportunities, or changes in domestic demand.

Figure 3 Larger initial deficits predict a higher probability of narrowing

Effect of unit change in CA on the probability of episode onset, depending on initial CA balance (% pts)

Notes: Dots show the estimated change in the probability of entering a narrowing episode for a one-unit increase in the current account balance, evaluated at different initial current account positions. Whiskers show 90% confidence intervals.
Source: OECD National Accounts, IMF Balance of Payments Database and authors’ calculations. 

Policy implications

The evidence has several implications for today’s global imbalance debate.

First, large imbalances can contain information about future adjustment pressures. This is especially true for deficits. Larger deficits are more likely to narrow even outside crisis settings. Adjustment can involve weaker activity, tighter financing conditions and difficult policy trade-offs. But it does suggest that large deficits tend to generate stronger correction forces than large surpluses.

Second, rebalancing is asymmetric. Deficit countries typically adjust through stronger exports, real depreciation and higher private net saving. Surplus countries adjust through stronger domestic absorption and lower net lending across sectors. 

Third, trade policy is unlikely to be a sufficient tool for durable external adjustment. Tariffs or sectoral measures may alter bilateral trade flows or the composition of production, but the historical evidence points to broader macroeconomic channels. Durable current account adjustment involves saving, investment, domestic demand, competitiveness, and financing conditions.

Finally, the slow pace of adjustment matters. Most large narrowing episodes unfold over many years. This makes early policy action important. Waiting for adjustment pressures to become acute can raise the risk that rebalancing occurs through weaker demand or financial stress rather than through a smoother mix of competitiveness gains, investment, and domestic-demand rebalancing.

Global imbalances are inherently multilateral: one country’s surplus is another’s deficit. But domestic policy actions can help to shape the path of adjustment. The history of narrowing episodes shows that large imbalances do not last forever, but neither do they correct automatically or symmetrically. Understanding how they unwind is essential for designing policies that make global rebalancing less abrupt, less costly, and more durable.

References

Calvo, G (1998), “Capital flows and capital-market crises: The simple economics of sudden stops”, Journal of Applied Economics 1(1): 35–54.

Freund, C and F Warnock (2007), “Current Account Deficits in Industrial Countries: The Bigger They Are, The Harder They Fall?”,  G7 Current Account Imbalances: Sustainability and Adjustment, NBER. 

Frohm, E, J Hooley, F Ozturk, L Rawdanowicz and N Sivakumar (2026), “Current account imbalances: Facts, drivers and policy challenges”, OECD Economics Department Working Paper No. 1869.

Gourinchas, P-O, G Kindberg-Hanlon, M Patnam, L Rotunno, and M Ruta (2026), “Industrial policy, tariffs, and the return of global imbalances”, VoxEU.org, 27 May. 

Itskhoki, O and D Mukhin (2025), “Tariffs, global imbalances, and the dollar”, VoxEU.org, 7 November.

Milesi-Ferretti, G M and A Razin (1998), “Current account reversals and currency crises: Empirical regularities”, NBER Working Paper No. 6620.

Weder di Mauro, B and J Zettelmeyer (2026), “Why global imbalances matter again – and what to do about them,” VoxEU.org, 13 April.