Op-Ed: Using the Risk-Free Rate as a Basis to Optimize Investment Costs (Part 2)
In my last column, I demonstrated why cost controls in a low rate/low return environment are paramount. While this point may seem intuitive or even obvious, in today’s column, I explain how and why allocators should examine their investment costs by making a direct connection between investment fees and expenses and the yield of the 91-day Treasury Bill (T-Bill).
As a reminder from the prior column, and importantly for this article, recall that we reviewed the components of a portfolio’s expected return, which are simply the combination of beta and alpha expectations. Beta can be further decomposed into the risk-free rate + the risk premium. Therefore, the expected return of a portfolio is:
ER (portfolio) = Risk-free rate* + risk premium + alpha
Beta
ER (portfolio) = Risk-free rate* + risk premium + alpha
Beta
ER (portfolio) = Risk-free rate* + risk premium + alpha
Beta
ER (portfolio) = Risk-free rate* + risk premium + alpha
Beta
Because the risk-free rate is an input to beta and beta an input to a portfolio’s expected return, the correlation between the risk-free rate and a portfolio’s expected return is obvious. As the yield on the risk-free rate declines, so too does the expected return for both beta and the entire portfolio. The opposite is also true.
As we did in the last article, we can review historical returns to see that the relationship between the risk-free rate and beta has held up well. Consider the data below, which shows the average annualized returns for a passive 60/40 S&P 500/Barclays Aggregate portfolio (beta) and the 91-day T-Bill (risk-free rate) during the past 40 years and 20 years.
Long-Term Returns
Average annualized returns ending June 30, 2020
| 40-year | 20-year | |
| 60/40 Portfolio (Beta) | 10.2% | 5.9% |
| 91-Day T-Bill | 4.2% | 1.5% |
| Risk Premium (Beta-RF) | 6.0% | 4.4% |
Source: Marquette Associates
As would be expected, the decline in the average annualized return of the passive 60/40 portfolio coincides with the decline in the yield of the risk-free rate. The decline in the risk-free rate from 4.2% to 1.5% from the last 40 years to the last 20 years can explain 270 basis points (bps) of the 430 bps decline in the average annualized return of the 60/40 portfolio across the same two time periods.
Notice too that the risk premium compressed from 6% to 4.4% between the 40- and 20-year time periods. A compressing risk premium is significant to investors (it lowers the “passive” portfolio expected and realized return); however, many factors can contribute to this compression, and most factors are unpredictable. Although the expansion or compression of the risk premium is important, investors will not know or be able to reliably predict its level or direction, which is one reason that I encourage allocators to focus only on the risk-free rate, not the risk premium, when examining investment costs.
Connecting Investment Program Expenses and Manager Fees to the Risk-Free Rate
The idea of underwriting managers/products through the lens of “net” returns is certainly not new. And the importance of negotiating the split of the gross return is becoming more widely understood. However, drawing a specific connection between investment costs and the yield on the risk-free rate is less common, but, in my opinion, a more useful framework upon which to assess program fees and manager expenses.
Let’s explain by looking back to the past 40-year average annualized returns in which we observed that the annualized return of the 91 day T-Bill was 4.2%. As we did in the last article, let’s assume that the investor has TOTAL investment expenses and manager fees of 1% and that this 1% is a direct offset to the yield of the risk-free rate. In other words, an allocator calculates the portfolio expected return by netting the investment costs against the risk-free rate and then adding the gross risk premium and the gross alpha expectations. In the example above, the investor would net the 1% of total expenses against the 4.2% return of the risk-free rate, meaning the investor expects to capture 3.2%, or 76%, of the available risk-free return after fees.
Doing the same for the 20-year period, the investor would net the 1% in total expenses against the 1.5% return of the risk-free rate, meaning the investor would capture a risk-free return of 0.5%, or only 33% of the available risk-free return after fees.
Investor Retained Portion of Risk-Free Rate
40-Year Period
Average annualized returns ending June 30, 2020
| RF Rate | Risk Premium | Total Return | Retained Risk-Free Return | |
| 60/40 Portfolio (gross) | 4.2% | 6.0% | 10.2% | |
| 60/40 Portfolio (net) | 3.2% | 6.0% | 9.2% | 76.19% |
20-Year Period
Average annualized returns ending June 30, 2020
| RF Rate | Risk Premium | Total Return | Retained Risk-Free Return | |
| 60/40 Portfolio (gross) | 1.5% | 4.4% | 5.9% | |
| 60/40 Portfolio (net) | 0.5% | 4.4% | 4.9% | 33.33% |
Source for Gross Returns: Marquette Associates
Connecting fees to the risk-free rate does not change the total “net” portfolio return expectations, but it does give an investor a definitive way to measure and ultimately decide how much of the “risk-free” return he/she is willing to give up in exchange for the pursuit of alpha. Plus, as noted earlier, the specific risk premium is unknown at any given time, whereas the risk-free return is always known and measurable. Fees are also always known and measurable, so, intuitively, it makes sense to connect the two “knowns” when developing expected returns. Plus, fees are not a market “risk factor,” so it also intuitively makes sense to remove fees from the risk premium and alpha forecast. By viewing fees as a reduction of the risk-free rate, an investor is connecting the two “non-risk” inputs of the expected return (risk-free rate and fees).
By evaluating fees using this framework, an investor will observe that as the yield on the risk-free rate goes down, it pulls the expected returns for the passive portfolio down with it, meaning that unless fees come down at the same rate, the investor must be willing to give up a higher portion of the available market return in order to pursue alpha.
Allocators may also consider the drivers of changes to the risk-free rate, which are typically inflation expectations and the Fed’s stance to these expectations. Interestingly, as inflation expectations change, investors accept changes to the risk-free rate, and therefore changes to their portfolio’s expected return; however, we allow fees to stay static. This means that investors accept the fact that fees will float as a percentage of the available passive return (representing a higher percentage as rates come down and vice versa). As allocators, it would make sense for us to hire managers on a variable fee basis in which the management fee is directly tied to the risk-free rate so the investor pays the same percentage of the available risk-free return at all times. Although this is sensible, good luck implementing that!
Today
For investors today, connecting investment costs to the risk-free rate can be especially useful.
As of July 10, the yield on the 91-day T-Bill was 14 basis points. Using the framework described above, an investor paying 1% in total fees has decided to not only give up all 14 basis points of the “risk-free” return, but also to accept an 86 bps deficit to the risk-free rate in order to pursue alpha. And this decision has significant implications!
Over the past 20 years, the risk premium has averaged 4.4%. If we assume this risk premium remains constant (it will not, but it will likely stay within a relatively tight range of 4% to 6%), then based on today’s risk-free rate and assuming no alpha, an investor’s gross expected return is 4.54% (0.14% + 4.4% +0%). Assuming the 1% in total fees, on a “net” basis, the investor’s expected return is 3.54% (-0.86% + 4.4% + 0%).
In order to achieve alpha under today’s conditions (with an expected return for beta of 4.54% and program fees of 1%), the investor must generate a return of 5.54%, or a return 1.22 times higher than the return of the passive 60/40 portfolio (1%/4.5%). Although 5.54% seems low on an absolute basis, a 22% return premium over beta is impressive if you can get it.
While the relative expense for the pursuit of alpha is higher when the risk-free rate is lower, so too is the penalty for poor performance. Paying high fees and NOT achieving alpha, or even worse, generating below beta returns, is much more costly in a low return environment because it requires the investor to produce a significant return premium above the passive portfolio in the future to “make up” the performance shortfall.
In the examples throughout this article, we have assumed that TOTAL fees are 1%. This includes all program fees (staff, consultants, custodian, etc.) and manager fees (aggregate passive and active fees). In the third and final installment of this article, I will illustrate how “program expenses” and “manager fees” can be isolated for a more focused cost assessment vis-à-vis the risk-free rate.
Until then, I hope that the suggestion to compare investment costs against the risk-free rate helps allocators make optimal decisions regarding one of the inputs within their control.
Tony Waskiewicz has nearly 30 years of financial services, investment advisory, and CIO experience and most recently served as chief investment officer for Mercy Health in St. Louis.
This feature is to provide general information only, does not constitute legal or tax advice, and cannot be used or substituted for legal or tax advice. Any opinions of the author do not necessarily reflect the stance of Institutional Shareholder Services or its affiliates.
Related stories:
Op-Ed: Using the Risk-Free Rate as a Basis to Optimize Investment Costs (Part 1)