Rebuilding global multi-asset portfolios between the innovation dividend and inflation risk

Overview

The report focuses on how the innovation cycle and inflation cycle jointly affect multi-asset returns. Goldman Sachs believes the recent macro environment has shifted toward reflation because of the Middle East war and rate pressures, but market-based inflation indicators have begun to normalize, and the growth and inflation backdrop may subsequently improve. At the same time, the long-running leadership of technology stocks has pushed equity weights, valuations, and index concentration to historical highs; investors cannot ignore the earnings support from innovation, but also need to prepare for valuation drawdowns, sticky inflation, and simultaneous declines in stocks and bonds.

Core views

First, the current environment is not simply a replay of the tech bubble: global earnings expectations have been unusually strong this year, and earnings have contributed more to equity returns than during the bubble period, but high valuations, market concentration, and pressure from AI capital expenditure are eroding the margin of safety. Second, risk appetite remains high, while cross-asset positioning and sentiment have returned to more neutral levels, and market narratives may switch rapidly among growth optimism, reflation, and rate pressure. Third, falling inflation is expected to ease stagflation shocks and downside risks for 60/40 portfolios, but positive stock-bond correlations and rapidly rising bond yields still limit traditional diversification benefits. Fourth, investors should maintain equity participation while improving diversification and tail protection through regions, styles, real assets, CTAs, equity long/short strategies, and long-dated options.

Analysis framework

The report combines analysis of macro growth and inflation forecasts, financial conditions indexes, business-cycle states, cross-asset risk appetite and positioning indicators, historical valuation and profitability, stock-bond correlations, historical technology-sector drawdowns, and long-term portfolio optimization. The risk assessment uses dynamic asset allocation models, random forests, cross-variable logistic regression, and Shapley value decomposition, and tests the performance of 60/40 and multi-asset portfolios through historical structural cycles and scenario analysis.

Methodology notes

• macro state identificationbusiness-cycle and reflation state framework

  Jointly observing standardized changes in growth, inflation, and policy variables.

  Used to identify the process by which markets shift from a moderate-growth, low-inflation state toward reflation or stagflation, and to explain rotation among different assets.

• market sentimentGoldman Sachs Risk Appetite Indicator

  Measures risk appetite by combining cross-asset prices, fund flows, futures positioning, options activity, and survey indicators.

  The report notes that the indicator rose above 1.2 in early June and remained elevated, while broader positioning and sentiment indicators have returned to relatively neutral levels.

• portfolio downside riskdynamic asset allocation random forest model

  Estimates the conditional probability of a drawdown in a 60/40 portfolio.

  The model uses growth, inflation, and cycle states as inputs and compares current probabilities with unconditional probabilities; rising inflation had pushed up portfolio drawdown risk, but that risk has now moderated.

• equity tail riskmultivariate logistic regression and Shapley attribution

  Estimates the probability of significant drawdowns or rallies in the S&P 500 over different horizons and decomposes the contributions of various inputs.

  Current tail-risk asymmetry has improved versus earlier, but valuation and business-cycle signals remain in conflict, so a large drawdown cannot be ruled out on this basis.

• long-term asset allocationstructural cycles and optimal portfolio framework

  Builds long-term return scenarios based on innovation, productivity, inflation, valuations, and stock-bond correlations.

  History shows that U.S. 60/40 portfolio performance varies significantly with structural cycles in innovation and inflation, and simple market-value weighting may not be the best starting point for active or passive multi-asset portfolios.

Asset mapping & comparison

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

• global equities

  neutral over 3 months, overweight over 12 months

  Strengths

  Earnings revisions are strong, macro conditions and corporate profitability still provide support, and equities late in the cycle can usually continue rising until the bull market peaks.

  Weaknesses

  Valuations are elevated, risk appetite is high, and index concentration is rising, leaving a limited margin of safety in the return distribution.

  Comparison

  Preferred over credit on a 12-month horizon; within regions, preference is for Asia Pacific ex-Japan, TOPIX, and S&P 500, with an underweight in European equities.

  Risks

  Sticky inflation, rising real yields, negative growth surprises, and reversal of crowded technology stock trades.

• U.S. large-cap technology stocks

  retain participation while controlling concentration risk

  Strengths

  This rally is supported by earnings growth, and the technology sector has improved the overall profitability of the S&P 500.

  Weaknesses

  Sector weight is already above tech-bubble levels, concentration and the Shiller P/E ratio are near extreme levels, and AI capital expenditure is depressing free cash flow.

  Comparison

  Low-volatility, high-dividend, and value styles have typically performed better during historical technology stock drawdowns, and non-U.S. equities have also often outperformed relatively.

  Risks

  Declining shareholder returns at hyperscalers, productivity gains falling short of expectations, momentum deleveraging, and valuation compression.

• government bonds

  neutral over both 3 months and 12 months

  Strengths

  Bond valuations are close to long-term average levels, and inflation normalization can provide relief for yields.

  Weaknesses

  Stock-bond correlations are more positive than before the COVID-19 pandemic, reducing the diversification capability of traditional 60/40 portfolios.

  Comparison

  U.S. and U.K. bonds are relatively preferred, while Japanese bonds are relatively less preferred.

  Risks

  A rapid rise in the U.S. 10-year yield, yields exceeding 5%, fiscal concerns, and renewed increases in inflation expectations.

• credit

  neutral over 3 months, underweight over 12 months

  Strengths

  Higher risk-free rates keep credit total yields superficially attractive.

  Weaknesses

  Credit spreads and term premia are low, compensation is insufficient, and the carry opportunity indicator is near lows since the global financial crisis.

  Comparison

  Relative preference for USD IG and USD HY, underweight EUR IG and EUR HY.

  Risks

  Growth shocks, rate shocks, spread widening, and negative convexity amid low risk premia.

• commodities and real assets

  commodities overall neutral, recommended as a portfolio diversification tool

  Strengths

  Historically, they have delivered better defensive and diversification performance during technology-led drawdowns and some inflationary phases.

  Weaknesses

  Sub-assets differ significantly in their sensitivity to growth, the U.S. dollar, and supply shocks, and short-term direction is not consistent.

  Comparison

  Relative preference for precious metals over 12 months, with energy and industrial metals maintained at neutral.

  Risks

  Slowing global growth, U.S. dollar appreciation, changes in commodity supply, and rapid decline in inflation.

• CTAs and equity long/short hedge funds

  supplementary allocation in inflationary and highly dispersed environments

  Strengths

  CTAs have historically tended to deliver stronger returns in higher-inflation environments; equity long/short strategies can benefit from rising single-stock dispersion.

  Weaknesses

  Performance depends on trend persistence, stock-selection skill, and risk management, and is not a stable substitute for traditional assets.

  Comparison

  When both stocks and bonds are under pressure, potential diversification value may be higher than simply adding traditional bond exposure.

  Risks

  Trend reversals, crowded trades, leverage, and manager selection risk.

• long-dated equity call options and collar strategies

  used to maintain equity exposure and improve tail asymmetry

  Strengths

  Long-dated call options can preserve upside convexity while limiting capital losses; selective put-spread collars are attractive in the current environment.

  Weaknesses

  Option prices are indicative levels and exclude transaction costs; protection depends on tenor, strike, and volatility.

  Comparison

  Compared with directly holding concentrated equity positions, they allow upside participation with a more explicit risk budget.

  Risks

  Premium decay, changes in volatility pricing, liquidity, execution, and transaction costs.

Key data

• 2026 global real GDP growth forecast2.5%Goldman Sachs forecast; the listed consensus forecast is 2.9%.
• 2027 global real GDP growth forecast2.8%Goldman Sachs forecast; the listed consensus forecast is 3.1%.
• 2026 global headline inflation forecast3.7%Goldman Sachs forecast, expected to fall to 3.2% in 2027.
• 2026 U.S. headline inflation forecast3.2%Expected to decline to 2.3% in 2027.
• global asset portfolio coverage sizeUS$289 trillionUsed to analyze the relative weights of global assets.
• historical drawdown of the technology, media, and telecom sectoraverage -26.7%Significant drawdown samples for the technology, media, and telecom sector listed in the report; the average drawdown of the S&P 500 over the same periods was -15.5%.
• performance of real asset index during technology, media, and telecom drawdownsaverage +2.0%In the same historical sample, real asset equities averaged -5.4%, indicating relatively better diversification benefits from the real asset index.
• risk appetite indicatorabove 1.2 in early JuneThe Goldman Sachs Risk Appetite Indicator was at a relatively high level at that time.
• market data observation date2026-08-04 to 2026-08-06Cross-asset volatility data as of August 4, relevant pricing as of August 6.

Impact & implications

The core portfolio implication is to avoid fully exiting equities late in the cycle because of valuation concerns, while reducing dependence on a small number of large U.S. technology stocks and on a single assumption of negative stock-bond correlation. Long-term investors can retain upside participation from the innovation theme, but should increase regional and style diversification and allocate to real assets, CTAs, and equity long/short strategies that can cope with inflation or technology stock drawdowns. Bond valuations are closer to long-term averages than equity valuations, but a rapid rise in yields could still hit both stocks and bonds at the same time; low risk premia also mean credit and carry positions should not be expanded excessively.

Risks

• The Middle East war further disrupts energy supply and creates a stagflation shock.
• Inflation data fail to normalize in line with market indicators, forcing major central banks to maintain more hawkish policies.
• The U.S. 10-year yield rises rapidly or breaks above 5%, triggering equity valuation compression and simultaneous declines in stocks and bonds.
• Deleveraging of technology stock momentum trades amplifies drawdowns in highly concentrated indexes.
• AI capital expenditure remains elevated for a prolonged period, but productivity and earnings returns fail to materialize sufficiently.
• Global growth optimism is priced ahead of macro data, followed by negative growth surprises.
• Credit spreads and term premia are too low, making carry and credit strategies more vulnerable to growth and rate shocks.
• Foreign exchange volatility weakens the diversification benefits of non-domestic equity allocations.
• Historical models and structural scenarios cannot fully capture changes in war, policy, and market regimes.

What to watch

• Whether U.S. and global headline inflation momentum can turn negative in the second half of 2026 and normalize further after year-end.
• The absolute level and speed of increases in the U.S. 10-year Treasury yield, and changes in long-term real yields.
• The Goldman Sachs Risk Appetite Indicator, cross-asset positioning, and the process of technology stock momentum deleveraging.
• Whether global earnings revisions can continue, especially profitability, free cash flow, and shareholder returns for the S&P 500 technology sector and AI hyperscalers.
• Technology sector weight, top-10 stock concentration, and the Shiller P/E ratio for U.S. equities.
• Whether the correlation between stocks and bonds remains positive, and whether the probability of a 60/40 portfolio drawdown rises again.
• Whether credit spreads, term premia, and carry opportunity indicators recover to offer more attractive risk compensation.
• Realized single-stock dispersion, implied correlation, and the cost of selective options protection.
• The Middle East situation, energy prices, and cross-asset performance under growth shocks and stagflation shocks.