Risk Vs. Return: Why Relying Only On The Sharpe Ratio Can Limit Performance

|
Includes: VBMFX, VGTSX, VIPSX, VTSMX
by: Derek Heckman
Summary

Portfolio optimization tools are a quick and easy way to improve your portfolio's risk and return characteristics.

While portfolio optimization can significantly reduce the riskiness of the portfolio, doing this may also steal your returns in the process.

Relying solely on one analytical measure, such as the Sharpe ratio, may give you an incomplete picture of the portfolio's risk or even drastically alter prudent asset allocations.

Proper asset allocation should always involve establishing appropriate allocation constraints and carefully reviewing the results of any portfolio analysis tools that are utilized.

Background

Passive index fund investors, after selecting the most appropriate investment assets, must then turn their attention to the next most important factor contributing to investment returns: asset allocation. In my own portfolio, I periodically review the investments representing each individual asset class and make substitutions and/or changes as needed. However, after selecting the individual investments, I then focus on the weight that I will assign each holding.

During this process, it is important to consider both the prior and expected return of each portfolio holding as well as the risk characteristics of the portfolio as a whole for each individual allocation setup. (For example, how would the return and risk of my 60% stock/40% bond portfolio change if I changed to a 40% stock/60% bond portfolio allocation?) I have found that, while analyzing these risk and return characteristics are important, prudent portfolio analysis demands that you do not rely solely on these calculations when constructing an investment portfolio.

Optimizing Your Investment Portfolio

I recently discovered the Portfolio Optimization tool on Silicon Cloud Technologies LLC's "PORTFOLIO VISUALIZER" website. Using this tool, I was able to enter my own Ivy League Endowment portfolio and manipulate the asset allocations to determine the risk and return characteristics of each variation.

The optimization tool also had several pre-made portfolios available, one of which caught my attention: Bogleheads Four Funds (you can find this portfolio, as well as the Bogleheads themselves, by clicking here). The asset allocations for this portfolio are shown below:

Bogleheads Four Funds portfolio

As you can see above, this is a broadly-invested, passive index fund portfolio consisting of U.S. equities (using Vanguard Total Stock Market Index fund VTSMX as the anchor), international equities, regular bonds, and inflation-protected bonds.

The optimization tool offers several different options for comparing your portfolio to an optimized portfolio. In this test, I applied the option to find the "Maximum Sharpe ratio" to the Bogleheads' Four Fund portfolio. While the Sharpe ratio is a valuable tool and many investors rely on it to help them determine the risk vs. return characteristics of an investment, this is not the only data point that should factor into your investments decisions. To illustrate why the following graphic shows the resulting portfolio when our original asset allocations were adjusted in order to maximize the Sharpe ratio:

Sharpe Ratio-optimized Four Funds portfolio

The original 50% allocation to VTSMX has been obliterated and an unreasonable and inadvisable weight of almost 90% has been assigned to VBMFX. Of course, a prudent investor would utilize the portfolio optimizer's constraint options to set minimum and maximum target weights for each asset class. However, in this example, these outrageous weights will suffice.

The chart below lists the resulting return and risk characteristics for the original Bogleheads Four Funds portfolio, the optimized Four Funds portfolio, as well as a third option which assigns equal weights (in this case 25% each) to the four asset classes in the original portfolio:

Risk/return stats for Bogleheads Four Funds portfolio and variations

The variability of returns (listed on the chart as "Stdev" for standard deviation) drops over 4 percentage points by simply equally-weighting the original holdings. Also, quite dramatically, the variability of returns drops to 3.28% for the optimized portfolio (10% stock, 90% bond)!

However, as is true with most investment situations, there is no such thing as a free lunch. On the next line down, we see that the expected return (ER) of the Maximum Sharpe portfolio is almost 2% below the ER of our original portfolio. If that isn't bad enough, the next line down is a perfect illustration of why financial ratios should not be followed blindly. In its best year (for the time period July 2000-March 2019), the original portfolio would have returned 28.64% while the Maximum Sharpe and Equal Weighted portfolios would have maxed at 8.25% and 20.81%, respectively. In this example, you are literally purchasing a reduction in risk by giving up almost 20% of the returns in your portfolio's best year.

Would you be willing to trade 20% of your returns for a less-bumpy ride? Some investors would say yes. But for me and my long-term investing timeline, the answer is a resounding no!

Summary

While I have no doubt that the results provided by the portfolio optimizer are accurate, applying the Sharpe ratio optimization to the original Bogleheads Four Funds portfolio resulted in the following:

  • Good: Drastically reduced the variability (standard deviation) of the portfolio's returns, making them more reliable.
  • BAD: Provided an asset allocation that was inadvisable and would not meet prudent investment standards.
  • BAD: Paid for the reduction in risk by giving up almost 20% of returns in the portfolio's best year.

As this example shows, relying on any single measure such as the Sharpe ratio may improve your returns but might also result in an asset allocation that is well outside of a prudent investor's level of comfort. While portfolio and asset analysis tools like these can do great things for your investment goals, it is always advisable to compare them with other portfolio measurement tools and to carefully review any resulting investment allocations or suggestions that these tools may provide.

Note: You can view the full portfolio optimization report from the Portfolio Optimizer by clicking here.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: While I do not have positions in any of the stocks mentioned, I do currently hold long positions in similar stocks of the following asset classes: U.S. and international equities, bonds, and inflation-protected bonds. These long positions would provide similar risk and return characteristics as the stocks mentioned in this article.