Sign Up
Enter your email address to follow ECNFIN.com and receive notifications of new articles by email for free. Be the first to read and do not miss future timely research publications.
Is there a preferred asset allocation that can consistently deliver positive risk-adjusted returns across all market environments? This paper analyzes 20 years of market data to answer that question.
In this research paper, I analyze 20 years of stock market returns, divided into four periods of five years each, spanning from 2005 to 2025. For each period, I derive the optimal asset allocation based on the highest Sharpe Ratio. The objective is to determine whether a particular allocation delivers superior risk-adjusted returns and to provide insight into how portfolios can be constructed using historical data.
Asset Allocation #1: May 2020 – May 2025
The most recent five-year period included inflation and significant interest rate hikes in 2022, which caused the S&P 500 to fall by 25%. Additionally, there was a pullback in the technology sector with the NASDAQ down 15%, and the bond market declined by 13% due to rapid Federal Reserve rate hikes.
The optimal asset allocation for this period was as follows (see Table 1):
• 16% in NASDAQ-100 ETF (QQQ)
• 53.5% in Gold ETF (IAU)
• 12% in U.S. Oil Fund (USO)
• 18.6% in MSCI India ETF (INDA)
(Table 1: Asset allocation for May 2020-May 2025)
This portfolio achieved a standard deviation of 13%, which was lower than that of any individual ETF in this allocation. Its expected return was a high 21%, with a Sharpe Ratio of 1.39 — the highest among all periods analyzed (see Table 2).
(Table 2. Portfolio statistics for May 2020-May 2025).
Asset Allocation #2: May 2015 – May 2020
This period included the COVID-19 crash in early 2020, when the S&P 500 dropped 34%, as well as the 2018 correction driven by rate hikes and trade tensions that resulted in a 20% market decline. The optimal asset allocation during this time was (see Table 3):
• 25.6% in NASDAQ-100 ETF (QQQ)
• 60.4% in U.S. Long-Term Treasury ETF (TLT)
• 14.0% in Gold ETF (IAU)
(Table 3. Asset allocation for May 2015- May 2020)
This portfolio had a Sharpe Ratio of 0.77, a standard deviation of 6.26%, and an expected return of 6.81%. The portfolio’s volatility was significantly lower than that of any individual ETF it included, illustrating the benefits of diversification. Despite the challenging environment, the return remained relatively attractive (see Table 4).
(Table 4. Portfolio statistics for May 2015- May 2020)
Asset Allocation #3: May 2010 – May 2015
This period featured the 2011 U.S. Debt Ceiling Crisis, during which the stock market fell by 19%. The optimal allocation for this period was (see Table 5):
• 61% in Long-Term U.S. Treasury Bonds ETF (TLT)
• 39% in NASDAQ-100 ETF (QQQ)
(Table 5. Asset allocation for May 2010- May 2015)
This portfolio achieved a Sharpe Ratio of 0.83 and a standard deviation of 6.31%, again lower than the standard deviation of either ETF alone. This demonstrates how combining equities and bonds can lower overall volatility without sacrificing return much. The expected portfolio return was 7.66% (see Table 6).
(Table 6. Portfolio Statistics for May 2010-May 2015)
Asset Allocation #4: May 2005 – May 2010
The global financial crisis defined this period, with the S&P 500 down 57% at its worst. Due to the limited availability of ETFs during this time, not all categories were tested. The optimal portfolio was (see Table 7):
• 83.6% in Gold ETF (IAU)
• 16.4% in MSCI Latin America ETF (ILF)
(Table 7. Asset allocation for May 2005-2010)
Despite limited diversification, this portfolio had a favorable risk-adjusted return (see Table 8), benefiting from gold’s strong performance during the crisis.
(Table 8. Portfolio statistics for May 2005-2010)
Summary and Insights
The four asset allocations analyzed each produced Sharpe Ratios between 0.8 and 1.4, indicating strong risk-adjusted returns over the past 20 years (see Chart 1). Each portfolio also had a lower standard deviation than the individual ETFs it contained, showing the power of diversification (see Chart 2 and Table 9).
(Chart 1: Sharpe Ratio of four asset allocation models during May 2005- May 2025)
(Chart 2. Standard deviations of four asset allocation models and their expected returns)
(Table 9: Summary of the four asset allocation models)
There is no single “perfect” asset allocation that works in all market environments. As shown in Table 10, optimal allocations shifted significantly from one period to the next. Even a seemingly ideal allocation today is based on historical patterns that may not repeat in the future.
(Table 10. Summary results of optimal asset allocation during May 2005- May 2025)
Weighted Average Allocation (2005–2025)
Using a weighted average of the four optimal allocations, I arrive at the following portfolio:
• 20.2% – Invesco QQQ Trust (QQQ)
• 37.8% – iShares Gold Trust (IAU)
• 3.0% – United States Oil Fund (USO)
• 4.6% – iShares MSCI India ETF (INDA)
• 30.4% – iShares 20+ Year Treasury Bond ETF (TLT)
• 4.1% – iShares Latin America 40 ETF (ILF)
This allocation is unconventional and may not be comfortable for all investors due to its heavy exposure to gold and long-term Treasuries. However, it reflects the asset classes that most consistently delivered strong risk-adjusted returns across multiple market cycles based on the past 20 years of market data.
Conclusion
One key takeaway is that the asset classes that performed best in the past may not continue to do so in the future. The ‘optimal’ allocation is dynamic, not static. The three asset classes that contributed most to historical risk-adjusted performance were:
• Invesco NASDAQ-100 ETF (QQQ)
• iShares Gold Trust (IAU)
• iShares 20+ Year Treasury Bond ETF (TLT)
This research supports the view that strategic allocation should remain flexible, rooted in history but responsive to change.
Disclosures:
This analysis is based on historical data and forward-looking estimates that may not materialize. This content represents the author’s opinion only and is not guaranteed to be accurate or complete. Please consult a qualified financial advisor before making any investment decisions. Neither ECNFIN.com nor the author assumes liability for any actions taken based on this information. Past performance is not indicative of future results.
Never miss an episode
Subscribe wherever you enjoy podcasts:
Apple
Spotify
Amazon
ECNFIN
1288 Kapiolani Blvd Apt 4003, Honolulu, HI 96814