Carry remains supported by the macro environment, while adding fiscal and energy factors can improve the risk-adjusted performance of FX portfolios

Overview

The report examines the current macro environment through three FX factors: carry, fiscal conditions, and energy. It concludes that low global volatility, low recession risk, and elevated and differentiated policy rates remain supportive of carry, but long-term yields and fiscal risks are rising. Adding exposure to low-debt and energy currencies can produce a portfolio with better risk-adjusted performance and more diversified sources of macro risk.

Core views

Carry is the report's first major theme. In 2026, high-carry, energy-exporter, procyclical, and positive-momentum themes have generally performed strongly, with the carry factor delivering a year-to-date total return close to 15%. In the sample beginning in 2010, its January-August performance exceeded that of every other year. The report attributes this result to low volatility, low recession risk, and policy rates that remain high and substantially differentiated across countries. Although carry already has the highest average annual return among the seven factors examined, the strength seen over the corresponding period in 2026 is still considered significant. Recent events provided a stress test for this theme. Coordinated yen intervention triggered a sharp appreciation in the yen, which is the second-largest short in the carry factor, directly weighing on factor performance. However, spillovers to other high-yielding longs and fund unwinds were relatively limited. The report contrasts this with 2024, when intervention by Japan's Ministry of Finance coincided with deteriorating global growth sentiment and therefore triggered more contagious carry-trade unwinding. This time, there was no comparable degree of macro resonance. Similarly, the Swiss franc significantly outperformed following the August 19 US Treasury buyback announcement, pressuring the carry factor. However, broad US dollar weakness and relief for some high-yielding currencies limited the shock, and the carry factor subsequently reached another year-to-date high during the report week. The second theme is using a fiscal factor to hedge carry risk. Global long-term yields have risen recently, and the fiscal fundamentals of the United States, United Kingdom, and Japan have come under renewed scrutiny. The report believes this increases the risk of digestion pressure on high-yielding currency positions. Its "country debt factor" favors currencies with low debt/GDP ratios. Because FX carry and debt/GDP are typically slightly negatively correlated, combining this factor with carry can reduce exposure to fiscal vulnerability while retaining positive carry. The resulting model portfolio is long BRL, COP, IDR, PEN, and CHF and short JPY, USD, GBP, and EUR. Historical testing shows that the blended carry and debt factor outperforms the pure carry factor when fiscal pressure is elevated. The improvement is primarily related to yield levels and curve steepness rather than short-term changes in yields. The blended factor's excess performance is particularly pronounced when the US 10-year-to-30-year curve is in its steepest historical quintile. This threshold is approximately 1% or higher, compared with a current level of about 0.5% cited in the report. The report also acknowledges that gross debt/GDP cannot fully capture the details of fiscal risk, yield shocks, and market absorption capacity, making it a practical but incomplete proxy. BRL and COP are examples: both remain vulnerable to rising global yields, but their pure carry advantage outweighs their fiscal weaknesses in the model, particularly when held alongside defensive longs such as CHF. The third theme is the effect of energy terms of trade on exchange rates. The weights of the energy and carry factors are currently only slightly positively correlated, indicating that energy is not simply a substitute indicator for carry performance. The report finds that the FX impact of energy terms of trade develops slowly but persists longer than changes in energy prices themselves. This is evident both in energy-sensitive relative-value currency pairs such as NOK/SEK and in the energy factor's relatively steady upward trend throughout the year. The energy factor's correlation with risk volatility has also turned significantly negative in recent months. A negative correlation during an energy shock is not surprising, but this reversal is clearer than during other periods of elevated and volatile energy prices. Given restricted flows through the Strait of Hormuz and a net upside skew in energy-price risks, the report believes positive exposure to energy terms-of-trade winners relative to losers remains valuable. This negative correlation with general risk volatility can also improve diversification in FX portfolios. Combining the three themes, the carry, energy, and debt three-factor portfolio has delivered higher volatility-adjusted returns this year than any individual component factor, and the report believes the strong performance has room to continue. The model produces positive signals for COP, BRL, IDR, AUD, NOK, and CHF and clearly negative signals for JPY and THB. Among G10 currencies, NOK and AUD are energy net exporters with relatively sound fiscal positions and also offer some carry, although the report retains a relative-value view expressed through an AUD/NZD downside structure. CHF is expensive to hold but offers clear hedging value when fiscal risks rise. Among emerging-market currencies, COP remains one of the report's preferred longs due to its exceptionally strong carry and energy resilience. Goldman Sachs is more cautious on IDR and BRL because of domestic factors. Both COP and BRL are vulnerable to rising global yields, a risk that the gross debt/GDP factor does not fully capture. The report believes placing them alongside more defensive longs such as CHF, NOK, and AUD makes the associated vulnerabilities more manageable. JPY and THB can serve as funding currencies on fundamental grounds, although yen performance still depends primarily on whether Japanese officials intervene or encourage capital repatriation. The report concludes by listing 14 active cross-asset trade views: short SGD/MYR, initiated at 3.13 on January 24, 2026, with a target of 2.90, stop at 3.30, and current level of 3.18; equal-weight long TRY, NGN, and KZT versus USD, initiated at 0% on February 18, 2026, with a revised total-return target of 10%, stop at 4%, and current return of 7.5%; long 3-year SOFR swap spreads, initiated at -22.6bp on April 17, 2026, with a revised carry-inclusive target of -14bp, revised stop at -19bp, and current level of -17.4bp; short USD/EGP, initiated at 0% on April 24, 2026, with a revised total-return target of 12%, stop at 5%, and current return of 9.2%; hold a 1-year forward 2s10s GBP OIS curve steepener, initiated at 0.38 on May 29, 2026, with a target of 0.55, stop at 0.25, and current level of 0.35; long 30-year Indian rupee bonds, initiated at 7.34% on June 27, 2026, with a target of 6.90%, stop at 7.65%, and current level of 7.48%; hold a 2s10s NZD curve steepener, initiated at 72bp on July 3, 2026, with a revised target of 95bp, revised stop at 80bp, and current level of 82bp; hold a 2s10s CAD curve steepener, initiated at 53bp on July 10, 2026, with a target of 80bp, revised stop at 58bp, and current level of 62bp; short AUD/NZD via a 6-month 1.1650 put option expiring January 14, 2027, initiated at 1.2000 on July 14, 2026, with a current level of 1.1984; long the NIFTY Bank Index and short the NIFTY Pharma Index, initiated at 100 on July 15, 2026, with a target of 115, stop at 90, and current level of 97; hold a 2s5s SOFR curve steepener hedged against a European OIS curve flattener, initiated at -7bp on July 31, 2026, with a target of 10bp, revised stop at -7bp, and current level of -6bp; short PLN/HUF, initiated at 84.12 on August 5, 2026, with a target of 80.5, stop at 87, and current level of 84.83; receive the fixed rate on a 10-year AUD swap, recorded in the source as initiated at 5.10 on August 14, 2016, with a target of 4.80, stop at 5.25, and current level of 5.13; buy a 6m1y A/A-40/A-80 receiver butterfly structure, recorded in the source as initiated at 9bp running on August 14, 2016, with a target of 20bp, stop at 4bp, and current level of 8bp.

Analysis framework

The report first compares the year-to-date total returns of seven FX factors and their corresponding historical performance since 2010, then uses yen intervention and the US Treasury buyback announcement to test the event resilience of carry trades. It subsequently constructs single and blended factors based on currency carry, debt/GDP, energy terms of trade, and risk correlations, comparing volatility-adjusted performance using monthly data from 2010 to 2026, yield levels, and curve percentiles. Finally, it cross-checks model weights against Goldman Sachs' macro views and explains the economic meaning and limitations of each major currency signal.

Methodology notes

• Quantitative/factor/portfolio theoryMulti-factor model

  Carry, country debt, and energy three-factor portfolio

  The report forms long-short weights based on each currency's exposure to carry, debt/GDP, and energy terms of trade, then blends the three factors to diversify the fiscal, energy, and risk-sentiment shocks faced by a standalone carry strategy.

• Quantitative/factor/portfolio theory

  Volatility-adjusted return comparison

  The report compares not only total returns but also returns earned per unit of volatility. On this basis, the three-factor portfolio has outperformed each of the three individual component factors this year.

• Fixed income and credit analysisYield curve analysis

  US 10-year-to-30-year curve steepness percentile

  The report uses yield levels and 10s30s curve steepness to measure fiscal pressure, finding that when the curve reaches its steepest historical quintile, approximately 1% or higher, the blended carry and debt factor has a more pronounced advantage over the pure carry factor.

• Macroeconomic framework

  Energy terms-of-trade transmission

  The report analyzes how energy prices gradually transmit to exchange rates through the relative currency performance of energy-exporting winners and losers, emphasizing that this FX impact usually persists longer than energy-price fluctuations.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• COP (Colombian peso)

  The three-factor model produces a positive signal, and it is also one of the report's preferred emerging-market longs.

  Strengths

  Exceptionally strong carry and energy resilience.

  Weaknesses

  Sensitive to rising global yields.

  Comparison

  Ranks strongly in the fiscal-versus-carry trade-off, with its vulnerability becoming more manageable when paired with CHF, NOK, and AUD.

  Risks

  The gross debt/GDP measure does not fully capture its sensitivity to a high-yield environment.

• BRL (Brazilian real)

  Both the carry/debt portfolio and the three-factor portfolio select it positively.

  Strengths

  Its pure carry advantage is sufficient to outweigh fiscal vulnerability in the model.

  Weaknesses

  The report remains cautious because of domestic factors.

  Comparison

  Similar to COP, it performs strongly in the fiscal-versus-carry trade-off.

  Risks

  It is vulnerable to rising global yields, and gross debt/GDP statistics do not fully reflect the risks.

• IDR (Indonesian rupiah)

  The three-factor model produces a positive signal.

  Strengths

  Ranks highly in the combined weighting of carry, energy, and fiscal factors.

  Weaknesses

  The report remains cautious because of domestic factors.

  Comparison

  The model signal is positive, but the discretionary view is weaker than for COP.

  Risks

  Domestic factors may weaken the advantage indicated by the model.

• AUD (Australian dollar)

  The three-factor model produces a positive signal.

  Strengths

  It has relatively sound fiscal conditions, is an energy net exporter, and provides some carry.

  Weaknesses

  The report also retains a relative-value expression of AUD/NZD downside.

  Comparison

  Together with NOK, it is a fiscally sound energy-exporting currency favored by the model.

  Risks

  The absolute factor signal and relative-value view may point in different directions.

• NOK (Norwegian krone)

  The three-factor model produces a positive signal, while energy factors also support NOK/SEK relative-value performance.

  Strengths

  It is fiscally sound, is a net energy exporter, and provides some carry.

  Comparison

  Together with AUD, it is a fiscally sound energy-exporting currency favored by the report.

• CHF (Swiss franc)

  It ranks highly in both the carry/debt and three-factor portfolios, primarily serving as a hedge against fiscal risk.

  Strengths

  It offers clear defensive and hedging value when fiscal risks rise.

  Weaknesses

  The hedge is expensive to hold.

  Comparison

  It is more defensive than high-yielding emerging-market currencies and can balance the yield risks of COP and BRL.

  Risks

  A sudden strengthening of the Swiss franc would directly weigh on traditional carry portfolios.

• JPY (Japanese yen)

  The three-factor model produces a clearly negative signal, while it is also viewed as a potential funding currency.

  Strengths

  It has the fundamental characteristics of a funding currency.

  Weaknesses

  The carry and fiscal factor trade-off is distinctly unfavorable, making it a prominent outlier in the model.

  Comparison

  It is one of the clearest negative currencies among the report's positive and negative signals.

  Risks

  Performance is highly dependent on intervention by Japanese officials and capital-repatriation responses.

• THB (Thai baht)

  The three-factor model produces a clearly negative signal, and it is viewed as a potential funding currency.

  Strengths

  It has the fundamental characteristics of a funding currency.

  Weaknesses

  It ranks poorly across the combined carry, energy, and fiscal factors.

  Comparison

  Together with JPY, it constitutes the model's clearest negative signal.

• PEN (Peruvian sol)

  The blended carry and country debt factor includes it as a long.

  Strengths

  It helps construct a currency portfolio that retains positive carry while reducing fiscal vulnerability.

  Comparison

  It joins BRL, COP, IDR, and CHF in the fiscally hedged carry long basket.

• USD, GBP, and EUR

  The blended carry and country debt factor places them on the short side.

  Weaknesses

  They rank unfavorably in the report's trade-off between carry and low debt/GDP.

  Comparison

  Together with JPY, they form the main shorts in the fiscally hedged carry portfolio.

Key data

• Carry factor total return year to date in 2026Close to 15%The January-August 2026 performance was the highest for the same period since 2010.
• Carry factor historical sample2010—2026Monthly data are used for factor and curve-percentile comparisons.
• Correlation between energy and carry factor weightsOnly slightly positiveThis indicates that the energy factor is not simply a substitute for the carry factor.
• Significant stress range for the US 10s30s curveApproximately 1% or higherThis corresponds to the steepest historical quintile; the current level cited in the report is approximately 0.5%.
• Positive three-factor signalsCOP, BRL, IDR, AUD, NOK, CHFDerived from the blended carry, energy, and debt factor model.
• Negative three-factor signalsJPY, THBBoth are also viewed as potential funding currencies, with JPY constrained by policy-intervention risk.
• Short SGD/MYRInitiated at 3.13; target 2.90; stop 3.30; current 3.18Initiated on January 24, 2026.
• Equal-weight long TRY, NGN, and KZT versus USDInitiated at 0%; revised target 10%; revised stop 4%; current 7.5%Initiated on February 18, 2026; figures represent total return.
• Long 3-year SOFR swap spreadsInitiated at -22.6bp; revised target -14bp; revised stop -19bp; current -17.4bpInitiated on April 17, 2026; the target includes carry.
• Short USD/EGPInitiated at 0%; revised target 12%; revised stop 5%; current 9.2%Initiated on April 24, 2026; figures represent total return.
• 1-year forward 2s10s GBP OIS steepenerInitiated at 0.38; target 0.55; stop 0.25; current 0.35Initiated on May 29, 2026.
• Long 30-year INR bondsInitiated at 7.34%; target 6.90%; stop 7.65%; current 7.48%Initiated on June 27, 2026.
• 2s10s NZD curve steepenerInitiated at 72bp; revised target 95bp; revised stop 80bp; current 82bpInitiated on July 3, 2026.
• 2s10s CAD curve steepenerInitiated at 53bp; target 80bp; revised stop 58bp; current 62bpInitiated on July 10, 2026.
• Short AUD/NZD via a put optionInitiated at 1.2000; current 1.1984Initiated on July 14, 2026, using a 6-month 1.1650 put option expiring January 14, 2027.
• Long NIFTY Bank, short NIFTY PharmaInitiated at 100; target 115; stop 90; current 97Initiated on July 15, 2026, denominated in local currency INR.
• 2s5s SOFR steepener versus European OIS flattenerInitiated at -7bp; target 10bp; revised stop -7bp; current -6bpInitiated on July 31, 2026.
• Short PLN/HUFInitiated at 84.12; target 80.5; stop 87; current 84.83Initiated on August 5, 2026.
• Receive fixed on a 10-year AUD swapInitiated at 5.10; target 4.80; stop 5.25; current 5.13Recorded in the source as initiated on August 14, 2016.
• 6m1y A/A-40/A-80 receiver butterfly structureInitiated at 9bp running; target 20bp; stop 4bp; current 8bpRecorded in the source as initiated on August 14, 2016.

Impact & implications

The report argues that the current environment does not require abandoning carry, but rather changing how carry portfolios are constructed. Low-debt currencies can cushion fiscal and long-end yield pressures, while exposure to energy winners can address rising energy prices and provide risk diversification. The resulting three-factor portfolio is more resilient than simply chasing high yields, but the model cannot fully capture domestic political and economic factors, sensitivity to global yields, or official intervention. Quantitative signals must therefore still be interpreted alongside each currency's own fundamentals.

Risks

• A further rise in global long-term yields and renewed concerns about the fiscal fundamentals of the United States, United Kingdom, and Japan could trigger digestion and unwinding pressure on carry longs.
• Yen intervention or Japanese capital repatriation could cause JPY funding trades to reverse suddenly and directly weigh on the carry factor.
• BRL and COP are relatively sensitive to rising global yields, a risk the gross debt/GDP factor does not fully capture.
• Domestic factors in IDR and BRL may weaken the positive signals indicated by the three-factor model.
• Although CHF can hedge fiscal risk, it is expensive to hold.
• Gross debt/GDP is a relatively crude fiscal proxy and cannot fully describe the fiscal and market-absorption risks of each currency.

What to watch

• Track whether global volatility, recession risk, and policy-rate differentiation continue to support carry performance.
• Monitor whether global long-end yields and the US 10s30s curve move from approximately 0.5% toward the steepest historical quintile of approximately 1% or higher.
• Watch whether restrictions on flows through the Strait of Hormuz and net upside risks to energy prices persist.
• Monitor whether the negative correlation between the energy factor and general risk volatility is sustained.
• Track intervention by Japanese officials and capital-repatriation responses, which are key variables for the performance of JPY funding trades.
• Monitor domestic factors in BRL and IDR, as well as the sensitivity of BRL and COP to rising global yields.