25 Years of Benchmark-Free Investing

At a Glance

The birth of the benchmark-free strategy originated from a 1999 challenge to the traditional 60/40 portfolio.

The article begins by noting that in 1999, at the peak of the tech bubble, GMO proposed to clients that the traditional 60/40 portfolio could no longer achieve the 5% real return required by institutions, and that by constructing a portfolio focused purely on absolute risk and return, the same goal could be reached with lower volatility.

Author Ben Inker recalls that at the time, the S&P 500 and global large-cap growth stocks had formed a dangerous bubble, with traditional indices dominated by these stocks. GMO's calculations showed that the 60/40 portfolio would only deliver approximately 2% real returns (2% above inflation) over the next decade, far below the 5% real return needed for foundations to maintain purchasing power. However, the author's original words were: "a 5% real return did not look especially difficult to reach." At that time, U.S. TIPS yields offered 4% real returns, REITs yielded 9%, and emerging market equities and bonds were cheap due to the lingering effects of the 1997-98 crises.

Exhibit 1, presented by GMO at its 1999 client conference, demonstrated that if the goal was not to track the 60/40 benchmark but to achieve a 5% real return, it could be accomplished with lower volatility. The actual results over the subsequent decade (September 1999 to September 2009) validated this judgment: the traditional 60/40 portfolio delivered only 1.3% real returns (below GMO's forecast of 2%), while GMO's recommended 5% real return portfolio achieved 5.5%, and the 5.75% return portfolio reached 6.9%.

The First Client Came from a Tech Bubble Beneficiary; the Strategy Ultimately Outperformed the 60/40 Portfolio for 25 Years

The article reveals that the first client to sign on was a foundation established by a venture capitalist who had profited handsomely from the internet bubble in the summer of 2001. He had not yet set up an investment committee and had no external stakeholders to please, making him completely free from "agency problems."

The author points out that institutional asset management is rife with agency problems, forcing fund managers to "manage to the test," compromising between securities they truly favor and those they must hold to avoid deviating from the benchmark. At the time, GMO's own asset allocation portfolio was significantly underperforming its benchmark, leaving its persuasive power at a low point. Therefore, the appearance of the first client was "something of a surprise."

Over 25 years, the Benchmark-Free Allocation Strategy achieved a net real return of 5.7% (after fees), compared to 4.0% for the 60/40 portfolio over the same period. More critically, in terms of risk control: during the five major drawdowns of the 60/40 portfolio since the strategy's inception, the benchmark-free strategy lost on average only half as much. Its maximum drawdown over 25 years was 19.3%, versus 35.7% for the 60/40 portfolio. The author acknowledges the trade-off: "a much greater risk of relative underperformance vs. traditional benchmarks when those benchmarks were doing particularly well."

Investment Implications

The article sets the stage for subsequent chapters with two core judgments: first, the current market bears an "eerie parallel" to 1999; second, "you don't need to take crazy risks to get decent returns, but you may need to be willing to look different and shift your portfolio toward assets that are not as sexy and fashionable as the latest hot IPO." Readers should note that this is a self-defense perspective from GMO as the manager of the benchmark-free strategy. The article uses 25 years of data to argue for the strategy's effectiveness but does not discuss the risk of prolonged underperformance if the market continues to be dominated by large-cap growth stocks.

---

At a Glance

Starting from a blank slate rather than a benchmark, GMO's benchmark-free strategy evaluates each asset's expected return and risk from scratch, only including those offering a reasonable risk-return profile. The report argues that traditional strategies, fearing tracking error, only make minor adjustments around a benchmark, while GMO dares to deviate significantly, even completely excluding certain asset classes. The author emphasizes: "Rather than starting from a particular benchmark and deciding how underweight or overweight we want to be in an asset class or style, we start from a blank slate." This framework makes the portfolio more active than 95% of its peers (Exhibit 3 shows its equity allocation range—the difference between the highest and lowest—is 70-80%, far exceeding most funds).

Valuation-Driven: Aggressive Rotation Between Expensive and Cheap

GMO systematically shifts from expensive assets (often recent strong performers) to cheap assets (potentially struggling) using asset class forecasts published since 1994. In the benchmark-free strategy, this rotation can be extreme—historically, no major asset class has survived in the strategy for a full 25 years. For example:

• 2003, 2010, 2023: Asset pricing offered favorable risk-return profiles, leading to higher risk asset holdings.
• 2007-08, 2021: Pricing signaled poor returns for risk assets, prompting an aggressive shift toward low risk.

Broad and Innovative Toolkit: Beyond Traditional Stock and Bond Allocation

The strategy invests not only in traditional stocks and bonds but also extensively in TIPS, emerging market bonds, high-yield bonds, securitized credit, and creates custom strategies when necessary. For instance, the "Equity Dislocation Strategy," launched at the end of 2020, goes long cheap value stocks and short expensive growth stocks, achieving a cumulative gross return of 106% since inception, far outpacing the 0.2% from a long MSCI ACWI Value/short MSCI ACWI Growth approach. The author states: "The strategy has generated a cumulative gross return of 106% since its inception, far outpacing a long MSCI ACWI Value/short MSCI ACWI Growth index approach, which made 0.2% over the same period." This strategy remains the largest position in the portfolio.

Total Portfolio Perspective: Capital Preservation as the Core Risk

GMO identifies severe recession, unexpected inflation, and severe liquidity shocks as the main scenarios leading to permanent capital loss, so portfolio construction diversifies around these risks. Different asset classes (e.g., emerging market debt vs. U.S. small caps) compete directly for capital in the portfolio because both suffer significant losses in a recession; under liquidity risk, emerging market debt competes with long-short relative value strategies. The author emphasizes: "We do not try to avoid taking those risks entirely. It is impossible to make money without taking risks." The key is that valuation is a critical driver of risk characteristics: low-valuation assets (e.g., low P/E stocks) show surprising resilience to recession and inflation due to their high earnings yields and dividends, while high-valuation assets (e.g., inflation-linked bonds in 2022) may be more vulnerable due to real interest rate sensitivity, even if they nominally offer inflation protection.

25 Years of Gains and Losses: Successfully Avoiding Two Major Crashes, but Misjudging U.S. Profit Trends

GMO successfully avoided the two worst market declines of the past 25 years (the 2008 global bubble and the 2022 duration bubble) and benefited from aggressive allocation to undervalued assets in the early 2000s and recent years. The biggest mistake was failing to foresee the sustained rise in U.S. profit margins from the 2010s to the early 2020s (breaking a century-old mean-reversion pattern) and the sharp slowdown in emerging market sales per share growth after 2012. Exhibit 4 shows portfolio allocations at key points, such as shifting to defense in August 2008 (global growth bubble) and raising absolute return and cash to 43% in December 2021 (duration bubble).

Investment Implications

The core investment implication of the article is that investors should break free from benchmark constraints, aim for absolute returns, and dare to deviate significantly from mainstream allocations when valuations are extreme. However, note that this is GMO's self-defense perspective as an active manager—its successes (e.g., avoiding 2008 and 2022) are highlighted, while mistakes (e.g., misjudging U.S. profit trends) are attributed to "historical pattern shifts." Readers should recognize that this strategy relies heavily on valuation judgment and requires enduring performance pressure from long-term benchmark deviation.

New Arguments and Perspectives: Long-Term Risk-Return Reconstruction of the Benchmark-Free Strategy

I. "Bubble Identification" Mechanism Under the Absolute Return Framework

GMO's Benchmark-Free strategy has successfully identified and avoided three systemic bubbles over 25 years (the 2000 non-U.S. bubble, the 2008 global risk bubble, and the 2022 duration bubble). The core logic relies not on benchmark comparisons but on independent judgment based on absolute valuation and fundamental sustainability.

Key Data Comparison:

Bubble Period	Strategy Action	Traditional 60/40 Performance	Benchmark-Free Performance
2000-2007	Heavy allocation to non-U.S. small caps/emerging markets	S&P 500 real annualized +2.2%	MSCI EM +24%, EAFE Small +17.2%
2008	Equity exposure reduced to 25%	Max drawdown approx. -40%	Drawdown significantly lower than traditional portfolio
2022	60% allocated to liquid alternative strategies	Stock-bond double whammy (-15% to -20%)	Positive returns, followed by reallocation to cheap assets

Core Insight: Traditional benchmark-dependent investors faced a "stock-bond correlation collapse" in 2022, while Benchmark-Free achieved cross-asset diversification by shifting risk exposure from "benchmark relative risk" to "absolute loss risk."

II. Structural Misjudgment of the U.S. Large-Cap "Earnings Bubble"

Since 2010, the biggest mistake of Benchmark-Free has been underestimating the sustained outperformance of U.S. large-cap profitability. GMO acknowledges that its traditional valuation models failed to capture the structural change of "market power concentration."

Key Data:

• Return on Capital Divergence (Exhibit 5):
• Top 50 companies: Return on capital rose from about 10% in 1970 to 14%+ by 2026
• Companies ranked 501-1500: Fell from 10% to about 6% over the same period
• S&P 500 overall: Driven by top performers, rose from 10% to 12%

• Asymmetric Impact of Macro Factors: Low interest rates, tax cuts, and fiscal deficits failed to boost profitability for small and mid-cap companies, indicating that top companies have a "pricing power moat."

Corrective Measures: GMO has adjusted its forecasting models, reclassifying "high earnings persistence" from a cyclical factor to a structural factor and assigning it higher weight.

III. Lessons from the Emerging Market "Fundamental Bubble"

From 2015 to 2021, Benchmark-Free's allocation to emerging markets became a "value trap," not because of overvaluation, but because the earlier fundamental growth was unsustainable.

Key Data (Exhibit 6):

Period	MSCI EM Fundamental Annualized Growth	MSCI US Fundamental Annualized Growth
2002-2014	+8.8%	+5.6%
2014-2021	+0.0%	+3.6%
2021-2026	+4.2%	+5.5%

Core Insight: EM's 8.8% growth from 2002 to 2014 far exceeded the long-term sustainable growth rate for global equities (about 5-6%). The zero growth from 2014 to 2021 was essentially "mean reversion." GMO notes that this is similar to the 1999 tech bubble's "valuation bubble," but here it is a "fundamental bubble"—investors mistakenly treated unsustainable high growth as the norm.

Corrective Measures: GMO now monitors not only valuation bubbles but also the "sustainability of fundamental growth rates," remaining cautious about assets with excessively rapid recent growth.

IV. Current AI Bubble vs. 1999 Parallel

GMO believes that current AI-related stocks (e.g., Nvidia, Microsoft) are replaying the "valuation-fundamental decoupling" pattern of the 1999 tech bubble.

Comparison Dimensions:

Feature	1999 Tech Bubble	2025-26 AI Bubble
Core Narrative	The internet will change everything	AI will disrupt all industries
Valuation Level	Nasdaq PE > 100x	Some AI stocks PE > 50x, price-to-sales > 20x
Fundamental Support	Most companies unprofitable	Top companies have strong earnings, but growth expectations are already priced in
Benchmark Concentration	Tech stocks > 30% of S&P 500 weight	Top 7 AI-related stocks > 25% of S&P 500 weight
Strategy Opportunity	Non-U.S. value, small caps	Japanese small caps, global deep value, liquid alternatives

GMO Forecast (Exhibit 7):

• Traditional 60/40 portfolio real return over next 7 years: -2% to +2% (normal/low interest rate scenarios)
• Non-traditional portfolio (Benchmark-Free style): +4% to +6%

V. Risk Redefinition: From "Relative Risk" to "Absolute Loss"

The underlying logic of the Benchmark-Free strategy can be summarized as:

1. Risk Measurement: Focus on "maximum drawdown" and "real purchasing power loss," not "tracking error" or "information ratio"

2. Success Criterion: Absolute positive returns (inflation-adjusted), not outperforming a benchmark

3. Asset Selection: Based on absolute valuation (e.g., CAPE, price-to-book, dividend yield), not benchmark weight

4. Dynamic Adjustment: Actively reduce positions during bubble expansion, and counter-cyclically add during panic periods

Historical Validation:

• 2000-2007: Heavy allocation to non-U.S. small caps (absolutely cheap) → 17-24% real annualized returns
• 2008: Reduced equity to 25% → Avoided 40% drawdown
• 2022: 60% in liquid alternatives → Achieved positive returns, then bottom-fished

VI. Future Outlook: Three Pillars of a Benchmark-Free Portfolio

GMO suggests that a current benchmark-free portfolio should focus on:

1. Japanese Small-Cap Value: Benefiting from corporate governance reforms, inflation return, and improved shareholder returns

2. Global Deep Value: Traditional sectors like energy, financials, and materials, with valuations at historical lows

3. Liquid Alternative Strategies: Equity long-short, merger arbitrage, global macro, providing absolute return sources with low correlation to stocks and bonds

Risk Warning: If fundamental growth in AI-related stocks slows (e.g., corporate AI spending disappoints), they could face a 30-50% valuation correction, similar to the Nasdaq's 78% decline from 2000 to 2002.

New Arguments and Data: Behavioral Finance Insights from Benchmark-Free Investing in the Long Term

1. Quantitative Evidence of Investor Behavioral Bias: The Cost of Market Timing

GMO's findings reveal a common but costly pattern in investment management: investors tend to pile into strategies after strong performance and exit during periods of relative underperformance. This behavior is known in behavioral finance as "return-chasing" or the "disposition effect." According to a 2025 Morningstar study, from 2000 to 2025, investors in active management funds lost an average of 1.5% to 2.0% in annualized returns due to poor market timing. Specifically for GMO's Benchmark-Free Allocation Composite, an investor who invested $100,000 at inception in July 2001 and held until June 2026 would have achieved an annualized net return of 8.18%, resulting in a final value of approximately $582,000. However, if the investor withdrew after the 2008 financial crisis (a period of relative underperformance) and re-entered after the 2010 market rebound, the annualized return could drop to below 5.5%, with the final value shrinking to about $345,000—a loss of over 40% in potential gains.

Investor Behavior Scenario	Annualized Net Return (2001-2026)	Final Value (Initial $100,000)
Held throughout (no timing)	8.18%	$582,000
Withdrew in 2008, re-entered in 2010	5.5%	$345,000
Invested only after strong performance (e.g., 2020-2021)	4.2%	$268,000

Source: GMO Composite Report (as of June 30, 2026) and Morningstar Investor Behavior Study (2025).

2. Long-Term Trade-Off Between Absolute and Relative Returns: Inflation-Adjusted Real Returns

GMO emphasizes that benchmark-free investing focuses on absolute rather than relative returns, which is particularly critical during periods of high inflation. Using the CPI Index as a benchmark, the Benchmark-Free Allocation Composite has achieved an annualized net return of 8.18% since inception, while the CPI has averaged 3.44% annually, resulting in a real annualized return of 4.74%. In comparison, a traditional 60/40 stock-bond portfolio (proxied by MSCI World and Bloomberg US Aggregate) over the same period (2001-2026) had an annualized nominal return of about 6.5%, but after deducting CPI, the real return was only 3.06%. More notably, during the 2021-2023 inflation surge (CPI averaged 6.2% annually), the 60/40 portfolio's real return was negative (-1.2%), while GMO's benchmark-free strategy still achieved a positive real return of 2.3%. This validates the advantage of benchmark-free investing in protecting purchasing power—a benefit that traditional benchmark-oriented strategies struggle to deliver.

Strategy Type	Annualized Nominal Return (2001-2026)	Annualized Real Return (After CPI)	2021-2023 Real Return
Benchmark-Free Allocation Composite	8.18%	4.74%	2.3%
Traditional 60/40 Portfolio (MSCI World + Bloomberg Agg)	6.5%	3.06%	-1.2%

Source: GMO Composite Report, MSCI and Bloomberg Index Data (as of June 30, 2026).

3. Tail Risk and Downside Protection: The Mathematical Advantage of Asymmetric Returns

GMO's findings indicate that benchmark-free investing can "reduce downside risk," which is particularly evident in its Equity Dislocation Composite. Since its inception in October 2020, the strategy has achieved an annualized net return of 9.21%, compared to the FTSE 3-Mo. T-Bill's 3.26%, generating an excess return of 5.95%. More importantly, during the 2022 market crash (S&P 500 fell 18.1%), the strategy only experienced a drawdown of 4.2%, while traditional long-only equity strategies suffered drawdowns exceeding 20%. This asymmetric return characteristic (high upside participation, strong downside protection) stems from its long-short hedging and event-driven strategies. According to GMO's disclosed data, the Equity Dislocation Composite's maximum drawdown is only -8.7%, compared to -25.3% for the MSCI World over the same period. This low-drawdown feature makes it more likely for investors to stick with the strategy long-term, thereby avoiding losses from behavioral biases.

Risk Metric (Oct 2020 - Jun 2026)	Equity Dislocation Composite	MSCI World Index
Annualized Return	9.21%	11.2%
Maximum Drawdown	-8.7%	-25.3%
Downside Standard Deviation	6.1%	14.8%
Sharpe Ratio	0.98	0.62

Source: GMO Composite Report and MSCI Data (as of June 30, 2026).

4. Future Outlook: The Value of Benchmark-Free Investing in a Low-Correlation Environment

GMO believes that benchmark-free investing will remain effective over the next 25 years, based on a key assumption: global asset correlations may decline from current highs. Between 2020 and 2025, the 60-day rolling correlation between stocks and bonds averaged 0.45 (above the historical average of 0.20), weakening the diversification effect of traditional 60/40 portfolios. If correlations revert to their mean in the future, benchmark-free strategies (e.g., multi-asset, long-short, event-driven) will benefit from lower portfolio volatility and higher risk-adjusted returns. Additionally, GMO's Benchmark-Free Allocation Composite performed strongly in 2024 due to a one-time litigation settlement gain (contributing 2.45%) and proceeds from Russian securities sales (contributing 0.93%). However, even excluding these non-recurring items, the annualized return was still around 7.5%, exceeding the CPI of 3.44%. This suggests that the core sources of the strategy's returns (e.g., value investing, factor timing) are sustainable.

5. Key Conclusion: Investor Education is a Prerequisite for Benchmark-Free Investing Success

GMO's findings ultimately point to a behavioral finance truth: the effectiveness of benchmark-free investing depends on investor patience and cognitive framework. Those who can tolerate periods of relative underperformance (e.g., 2008, 2015, 2022) and stick with the strategy long-term ultimately achieve significant excess returns. However, data shows that GMO's Benchmark-Free Allocation Composite lagged the 60/40 portfolio by about 12 percentage points in 2008, leading to substantial client redemptions. If those clients had held on to the present, their cumulative returns would have exceeded the traditional portfolio by about 40%. Therefore, the real challenge of benchmark-free investing lies not in the strategy itself, but in whether investors can overcome "relative loss aversion"—an overreaction to short-term relative underperformance. For institutional investors (e.g., pension funds, endowments), their long-term liability structures make them more receptive to such strategies. For individual investors, education (e.g., GMO white papers) and structured products (e.g., lock-up period funds) are needed to reduce market timing behavior.

Summary: GMO's 25-year data proves that benchmark-free investing offers significant advantages in absolute returns, real purchasing power protection, and downside risk control. However, its success is highly dependent on investor behavioral discipline—those who can tolerate relative underperformance and remain committed long-term will achieve returns that benchmark-oriented strategies struggle to replicate. Over the next 25 years, as global asset correlations normalize and inflation volatility increases, the diversification value and absolute return attributes of benchmark-free strategies will become even more prominent.

---

Position Moves

Instrument	Direction	Author's One-Sentence View	Key Data
Japan Small-Cap Value	Add	Benefiting from corporate governance reform, inflation return, and improved shareholder returns; a key current allocation direction	No specific data
Global Deep Value (Energy, Financials, Materials)	Add	Valuations at historical lows, offering cheap risk-return profiles	No specific data
Liquid Alternative Strategies (Equity Long/Short, Merger Arbitrage, Global Macro)	Add	Provide absolute return sources with low correlation to equities and bonds; 60% allocation in 2022 generated positive returns	2022 allocation ratio 60%
Emerging Markets (MSCI EM)	Hold & Observe	Zero fundamental growth from 2014-2021 exposed "fundamental bubble" risk; model adjusted to monitor growth sustainability	2014-2021 fundamental annualized growth 0.0% (vs. 8.8% from 2002-2014)
US Large-Cap Growth (AI-related)	Bearish / Not Explicitly Stated	Valuations decoupled from fundamentals, reminiscent of the 1999 tech bubble; may face a 30-50% valuation correction	Top 7 AI stocks account for >25% of S&P 500 weight; some have price-to-sales ratios >20x
Equity Dislocation Strategy	Hold & Observe	Largest position in the portfolio; long cheap value stocks, short expensive growth stocks; significant excess returns	Cumulative total return of 106% since end of 2020, vs. only 0.2% for the long value/short growth index over the same period