U.S. Deficits, Yields, and the Dollar: Exploring Alternative Paths 

The U.S. dollar has come under increasing pressure in recent months, reflecting shifting global dynamics and deteriorating investor sentiment. A confluence of factors, including aggressive and unpredictable disruptions to free trade frameworks, concerns about the stability and openness of U.S. financial markets, threats to central bank independence, and unease over widening deficits and mounting public debt, has challenged the long-standing notion of U.S. exceptionalism that has traditionally underpinned the U.S. dollar’s safe-haven status. 

The U.S. dollar has posted one of its weakest first-half performances on record. While it has recently found some support from tariff reprieves amid newly signed trade deals and signs that the U.S. economy is proving more resilient to tariffs than initially feared, the trade-weighted broad dollar index remains 6% below its level at the start of the year (as of mid-August).  

We have seen this play out in bond markets as well. Long-end Treasury yields initially declined earlier in the year, partly reflecting expectations of a weakening economy. However, yields rose sharply following “Liberation Day” and again after the passage of the One Big Beautiful Bill Act (BBB), with the latter widely expected to add approximately US$3 trillion to U.S. national debt over the next decade. While yield movements reflect several factors, including central bank communications, issuance patterns, and shifting growth and inflation expectations, the broader trend has been one of elevated long-term yields even as shorter-term yields have declined and inflation break evens are little changed from the start of the year. 

Notably, the relationship between the dollar and U.S. 10-year Treasuries appeared to decouple after “Liberation Day” and more markedly following the passage of the BBB (chart 1). According to the Bank for International Settlements (BIS), the dollar’s sharp depreciation in April was not solely driven by selling but was amplified by ex-post FX hedging activity by non-U.S. investors, particularly from Asia. The BIS notes that the dollar’s trend appreciation since the pandemic, the dollar’s usual strength during risk-off episodes, and the flattening of the U.S. yield curve during the U.S. Federal Reserve’s hiking cycle, which increases the costs of hedging, had previously led to reduced FX hedge ratios among foreign investors. The April depreciation, coupled with a modest steepening in the yield curve, prompted a wave of delayed hedging, which tends to exert more pronounced pressure on the dollar than simultaneous hedging, contributing to its continued weakness through May. As hedge ratios adjust upward, the BIS suggests that the impact of ex-post hedging is likely to fade, and future dollar movements will increasingly reflect broader macroeconomic and U.S.-specific developments. 

Rather than offering a verdict on the future of U.S. exceptionalism, this section explores different dollar and yield outcomes under alternative deficit scenarios, given the central role these deficits are likely to play in the medium to long term, as suggested by emerging evidence. Recent research from the International Monetary Fund (IMF) has found that the relationship between fiscal balances and long-term rates is not stable over time and has in fact strengthened as fiscal positions have deteriorated. The anticipated spending increases under the BBB are expected to further erode the U.S. fiscal position, amplifying these pressures on term premia and the dollar.  

In our July base case, in which the federal deficit rises to and holds near 6.8% of nominal GDP through 2028 before gradually narrowing—in line with Congressional Budget Office (CBO) projections and BBB estimates—we project an additional 3.5% depreciation in the dollar by the end of 2026, building on the 7% decline already observed between early and mid-year 2025. The timing of this adjustment is uncertain; the depreciation may be limited in the near term, but structurally higher deficits are likely to weigh on the dollar more meaningfully as the horizon extends. By 2030, the cumulative decline reaches roughly 7.5%. This reflects an environment where policy, geopolitical, and fiscal concerns are undermining investor confidence, keeping long-term yields elevated in the near term, even as policy rates decline. Unlike the typical pattern where falling policy rates would lead to lower long-term yields and a weaker dollar, these concerns delay the drop in yields while further weakening the dollar. In this case, yields eventually resume a downward trend in line with fundamentals. 

Larger deficits, reaching 7.3% or 7.8% of GDP, accelerate this trend (charts 2–5). In these scenarios, the dollar weakens by around 4% by the end of 2026 and by as much as 8%–9% by the end of 2030. The erosion of confidence is more pronounced here, leading to stronger upward pressure on long-term rates and a slower normalization in yields despite expected policy easing. The result is a more pronounced and persistent weakening of the currency. 

Conversely, a more disciplined fiscal path—driven by stronger economic growth or higher-than-expected tariff revenues, for example—in which the deficit narrows to 6.3% of GDP in 2026 through 2028, would support a more stable outcome. With CBO’s August update showing that recent tariff changes would increase revenues, resulting in reduced deficit projections, the probability of this more favourable deficit scenario has somewhat increased. In this case, the dollar still depreciates, but more modestly: 3% by 2026 and 6.9% by 2030 relative to Q2/25. Improved investor sentiment reduces term premia, allowing long-term rates to adjust more quickly to falling policy rates. The dollar benefits from a more conventional easing environment and relationship with yields differential. 

Importantly, the range of outcomes is determined not only by the size of the fiscal deficit, but also by the sensitivity of yields and the dollar to those deficits. In a final extreme case, we consider what happens when we increase the coefficients on the deficit in both the 10-year yield and dollar equations by two standard deviations relative to the mid-point estimates (charts 6–8). This adjustment reflects the inherent uncertainty around coefficient estimates, the potential for nonlinear and time-varying market reactions to worsening fiscal positions, and broader uncertainty in the policy environment. Under this extreme case specification, the base case itself shifts materially: Long-term yields are persistently higher, by as much as 50–60 bp, and the dollar weakens more substantially, depreciating by roughly 5% by 2027 and around 10% by 2030. The opposite would be true if we were to decrease the coefficients on the deficit in both equations by two standard deviations to reflect the uncertainty in the other direction.