Why Not 100% Equities - From Paper to Practice
"Why Not 100% Equities" was originally published in The Journal of Portfolio Management in 1996.
In this post, I will provide a brief summary of the article and then tie these ideas into the real-world applications for how I invest.
As this article was written nearly 30 years ago, the main idea was essentially theoretical or only realistic for institutional investors like pensions or hedge funds. But fast forward to today, and now there are ETFs that allow virtually all investors to put this article’s ideas into practice and ultimately benefit your financial plan.
An Overview of Why Not 100% Equities
Why Not 100% Equities was written in 1996 by Cliff Asness, founder of investment firm AQR, who at the time was a director of quantitative research at Goldman Sachs.
Background & Overview
Why Not 100% Equities is a response to a 1994 paper entitled "College and University Endowments Funds: Why Not 100% Equities?" which questioned whether College and University endowments, with virtually infinite time horizons, should invest 100% in stocks instead of a more balanced allocation, like a 60/40 stock & bond portfolio common for retirees.
While Asness agrees with the idea that an endowment likely can take more risk than a 60/40 portfolio, he pushes back on the idea that the best way to take more risk is to invest entirely in stocks.
“An investor willing to bear the risk of 100% equities can do even better with a diversified portfolio” and “this is shown most clearly when an investor is willing to lever.”
Asness then explains how, to maximize risk-adjusted returns, an investor should (1) take the optimal combination of assets, which yields the highest return per unit of risk (or highest Sharpe Ratio), and then (2) mix this portfolio with cash to reach their desired risk and return.
So identify a diversified portfolio with a high Sharpe Ratio (or high return per unit of risk) and, if the expected return and risk of this portfolio is below the investor’s targets, borrow (or use leverage) to increase your exposure to this diversified portfolio instead of offloading diversifying assets and concentrating into the highest-expected returning asset (stocks).
So Asness is pushing back against the conventional wisdom of holding 100% stocks if you have a high risk tolerance and long time horizon. The alternative he proposes is to apply leverage to a diversified portfolio in pursuit of higher returns.
This idea can be expressed graphically by taking a "High Sharpe" portfolio (red dot) and borrowing at the risk-free rate of Cash (blue dot). In theory, applying this kind of leverage should result in a portfolio that moves essentially in a straight line from the cash return, through the High Sharpe portfolio, and beyond.
Conventional investment wisdom suggests you should hold most or all of your investments in stocks if you have a long time horizon. But this investment approach is highly concentrated and, as a result, can be subject to deep drawdowns and long underwater periods.
As we look at combinations of global stocks and U.S. Treasury bonds from 1996 through 2024, you can see how this line flattens out as you approach 100% stocks. What this demonstrates is how you are losing the benefits of diversification and, as you approach 100% stocks, you are eking out a little bit more return for a lot more risk.
Data from May 1996 through December 2024. Data source: PortfolioVisualizer. All portfolios are rebalanced annually. Stocks are represented by 60% Vanguard Total U.S. stock market (VTSMX) and 40% Vanguard Total International Stock Market (VGTSX). Bonds are represented by Vanguard Interm-Term Treasury Fund (VFITX). Cash is represented by 3-month Treasury Bills (FRED).
We would like to be better compensated for this additional risk, so instead of how this curve flattens out we would like to move as much up this graph as to the right. More aggressive portfolios that follow along the red dotted line in the graph below would be preferable.
The paper goes on to compare a levered, diversified portfolio to 100% stocks.
Performance Comparison 1926-1993
Asness reviews the long-term data used in the 1994 article and points out how stocks outperformed bonds and the 60/40 portfolio. But he notes how this is not an apples-to-apples comparison because the risks are dramatically different.
As you can see, stocks returned 10.3% per year, outperforming the 60/40’s 8.9% per year. But stocks had a risk of 20 compared to 12.9 for the 60/40. So in order to make this an apples-to-apples comparison, you would need to lever up the 60/40 portfolio to the same risk level as stocks, or 155% leverage (12.9 * 155% = 20).
The paper adds this Levered 60/40 portfolio to the chart to demonstrate how levering a diversified portfolio to the same risk of stocks produces greater returns at the same level of risk. After levering up the 60/40 to the same risk of stocks, the Levered 60/40 returned 11.1% annually vs. 10.3% for 100% stocks.
The paper also compares the worst months and years of both portfolios. The timeframes are broken into the entire 1926-1993 period as well as isolating the post-World War 2 era. As you can see, among the worst one-month and one-year periods, the Levered 60/40 portfolio was down less than the 100% stock portfolio, with the exception of the worst year since world war 2, where stocks were down 39% and the levered 60/40 declined 44%.
So despite the use of 155% leverage, the Levered Portfolio had less severe drawdowns than the unlevered stock portfolio by three of four of these metrics. A leveraged, more diversified portfolio could be less risky than an unlevered, concentrated portfolio.
With this data in support, Asness reiterates how an investor should (1) construct the optimal portfolio (the highest Sharpe portfolio) and then (2) apply leverage in the pursuit of higher returns, rather than concentrating into 100% stocks.
Investor Limitations in 1996
Asness closes the paper by acknowledging that for the vast majority of investors, this is just a theoretical exercise and not practical. This is because most investors, except for large institutions, would be unable to borrow at competitive interest rates, a key assumption of this theory.
While this was true in 1996, the investment landscape has evolved and now virtually all investors may employ this strategy through low-cost ETFs.
Extending the Data (1997-2024)
Before explaining the industry developments that allow people like you and me to employ this strategy, let's first review the updated data since this paper was published in 1996.
1997-2024 Returns & Risk
Let’s compare the returns of 100% stocks vs. a 60/40 portfolio from 1997 through 2024.
Data from 1997 through 2024. 100% stocks represented by Vanguard S&P 500 Index Fund (VFINX). 60% Stocks/40% Bonds represented by 60% Vanguard S&P 500 Index Fund (VFINX) and 40% Vanguard Intermediate-Term Treasury Index Fund (VFITX), rebalanced annually. Data source: PortfolioVisualizer. Sharpe Ratio = (Portfolio return – Risk-free rate)/(Portfolio Standard Deviation). You cannot invest in an index. Past Performance is not indicative of future results.
A similar story has played out since the paper was published in 1996. Stocks have outperformed the 60/40 portfolio, but this isn’t an apples-to-apples comparison because stocks assumed a lot more risk. Stocks had a standard deviation of 15.5 compared to 9.2 for the 60/40. And stocks also had far more significant worst years and drawdowns. And notably, the Sharpe Ratio of the 60/40 at 0.64 is superior to the 0.53 Sharpe Ratio of 100% stocks.
To equalize the risk of these portfolios, we would need to lever the 60/40 portfolio by 70% (9.2% * 170% = ~15.5%). The result is the data in the following table.
Data from 1997 through 2024. 100% stocks represented by Vanguard S&P 500 Index Fund (VFINX). 60% Stocks/40% Bonds represented by 60% Vanguard S&P 500 Index Fund (VFINX) and 40% Vanguard Intermediate-Term Treasury Index Fund (VFITX), rebalanced annually. Levered 60/40 represented by 102% Vanguard S&P 500 Index Fund (VFINX) and 68% Vanguard Intermediate-Term Treasury Index Fund (VFITX), -70% 3-month Treasury Bill Rate, rebalanced annually. Data source: PortfolioVisualizer. Sharpe Ratio = (Portfolio return – Risk-free rate)/(Portfolio Standard Deviation). You cannot invest in an index. Past Performance is not indicative of future results.
Again, a similar story has played out. With equalized risk, the Levered 60/40 has now outperformed the 100% stock portfolio. Not only that, but the more diversified levered 60/40 also had a better best year and not as poor worst years or maximum drawdowns when compared to the 100% stock portfolio.
You can also see how applying leverage allows an investor to preserve the higher Sharpe Ratio of the unlevered 60/40.
Graphically, you can see how these return and risk numbers bear out the theory.
The Levered 60/40 experienced similar risk to the 100% stock portfolio, but achieved a higher return.
Paper to Practice: How to Implement this Strategy through ETFs
When this paper was published in 1996, this strategy was out of reach for most investors. But developments in the ETF landscape in the last 10 years have opened this strategy to virtually all investors.
Removing the Roadblocks for Individual Investors
A key tenet (and major impediment to individual investors) of this strategy is the ability to borrow at or near the "risk-free rate" to employ leverage. The "risk-free rate" is generally the short-term government bond rate, or 3-month Treasury bill rate.
The risk-free rate can be visualized by the blue "cash" dot along the y-axis. For this strategy to make sense, you need to be able to borrow at a competitive rate below the return you can expect from your investment portfolio. If you are only able to borrow at 10% and you expect your portfolio to return 8%, this strategy won't work.
For individual investors using margin, the borrow rates have always been and continue to be well-above the risk-free rate. But institutional investors have had access to lower borrow rates near the risk-free rate, or they may use futures contracts embedded with competitive borrow rates. Today, there are ETFs using these futures contracts which offer competitive borrowing near the risk-free rate, making the theory outlined in the 1996 paper a practical option.
How ETFs Allow Investors to Implement this Strategy
Beginning in 2018, ETFs have launched that allow investors to employ this strategy. One ETF, for example, is a 150% levered 60/40 portfolio (very similar to the Levered 60/40 portfolio from Asness’ paper).
Below is a comparison of this ETF launched in 2018, compared to 90% S&P 500 Index, 60% government bonds, less -50% 3-month Treasury bill rate. These portfolios track one another very tightly, demonstrating how individual investors can now access the borrowing rate required to put this theory into practice.
Data source: PortfolioVisualizer. 90/60 Benchmark represented by 90% Vanguard S&P 500 Index Fund (VFINX), 60% iShares U.S. Treasury Bond ETF (GOVT), -50% 3-month Treasury Bill Rate, rebalanced monthly. You cannot invest in an index. Past Performance is not indicative of future results.
But while an investor could invest 100% in this type of levered ETF instead of stocks, that’s not what I am suggesting nor how I invest. Instead of investing entirely in this type of ETF, an investor can use these types of ETFs as a building block in their portfolio.
In Search of the "Optimal" Portfolio to Leverage
Asness’ paper was a relatively quick and simple demonstration of the basic idea that you are better off levering up a diversified portfolio compared to concentrating into 100% stocks. But this is a simple demonstration and Asness does not suggest the 60/40 is the optimal portfolio that you should leverage.
For example, an investor could add more diversifiers beyond stocks and bonds in an effort to combat risks to these traditional assets and increase the portfolio’s Sharpe Ratio, then use these capital efficient ETFs to lever up that portfolio.
This chart compares 100% stocks and the 60/40 portfolio to a Risk Parity style portfolio that includes allocations to gold and managed futures. This Risk Parity-style allocation has a notably higher Sharpe Ratio at 0.80 compared to 0.64 and 0.53 for the other portfolios.
Data from 1997 through 2024. 100% stocks represented by Vanguard S&P 500 Index Fund (VFINX). 60% Stocks/40% Bonds represented by 60% Vanguard S&P 500 Index Fund (VFINX) and 40% Vanguard Intermediate-Term Treasury Index Fund (VFITX), rebalanced annually. Risk Parity represented by 30% Vanguard S&P 500 Index Fund (VFINX) and 55% Vanguard Intermediate-Term Treasury Index Fund (VFITX), 7.5% Gold, 7.5% BTOP50 Index, rebalanced annually. Data source: PortfolioVisualizer. Sharpe Ratio = (Portfolio return – Risk-free rate)/(Portfolio Standard Deviation). You cannot invest in an index. Past Performance is not indicative of future results.
So while Asness’ paper provides an example of levering a 60/40 portfolio, this was a simple demonstration and not necessarily the optimal portfolio to leverage.
Final Thoughts
So this post has certainly been on the nerdier side of things, but it is meant to demonstrate some of the theory and research behind my investment strategy, which involves combining risk parity-style investing with capital-efficient ETFs.
The goal with this approach is to aim for the highest returns for the risk we are willing to take, which should ultimately translate into better financial planning outcomes.
To learn more about Kardinal Financial's investment strategy and financial planning services, visit the Contact page to schedule an introductory call.