Risk and Return Benchmarks
What is a Benchmark?
A benchmark (usually an index) represents a “neutral” portfolio. It is a “default” investment that the client might otherwise make. For example, a U.S. client investing in stocks might use the S&P 500 Index as a benchmark. That index includes the stocks of 500 of the largest U.S. corporations. A U.S. client investing in bonds might use the Bloomberg Aggregate Index, which includes all investment grade, US dollar denominated, fixed-rate taxable bonds. A client investing in both stocks and bonds could blend these two indexes into a single blended benchmark using a constant ratio, for example, 60/40.
A good benchmark is “investable.” That is, it can be easily and readily invested in. Most exchange-traded funds (ETFs) are index funds, so they often make good benchmarks. For example, Vanguard S&P 500 ETF (VOO) and Vanguard Total Bond Market ETF (BND) are both excellent benchmarks. They both have very low expense ratios of only .03% per year. This makes them “efficient” (low-cost), which is another desirable characteristic for a benchmark.
Why Uses Indexes?
The alternative to using indexes and index funds would be to use actively managed funds as benchmarks. However, actively managed funds are not neutral relative to an investment universe. Actively managed funds generally seek to outperform a benchmark index. And they generally fail in that attempt in the short run overwhelmingly fail in the long run. (See “Why Invest in Index Funds?”) An actively managed fund by definition owns only a sub-set of a broader investment universe.
In fact, even the S&P 500 is somewhat active relative to the overall U.S. stock market because it includes only 500 of the largest companies. For full exposure to the U.S. stock market, Vanguard Total Stock Market ETF (VTI) would be a better choice because it includes not only large-capitalization stocks, but also mid- and small-capitalization stocks. VTI currently includes 3,608 stocks.
Including international stocks in a portfolio provides broader exposure and enhanced diversification over the long-term, which tends to lower volatility or risk. Vanguard Total World Stock ETF (VT) includes both foreign and U.S. stocks. Currently, there are 9,809 stocks included, 64.2% U.S. and 35.8% foreign.
Why We Use Vanguard Funds as Benchmarks
I have always been a big fan of Vanguard index funds. Over the years, when friends and family have asked for investment advice, I have generally steered them toward Vanguard. At Sapient Investments, we invest in a lot of Vanguard funds, and we often also use their funds as benchmarks. Here’s why:
- Vanguard is the largest provider of index funds
- Vanguard is a familiar and trusted fund company
- Vanguard funds generally have the lowest expense ratios
- Vanguard funds are broad and all-inclusive
That last attribute is particularly distinctive of Vanguard global balanced funds because they include both international stocks and international bonds. However, their international bonds are hedged back to the U.S. dollar, reducing foreign currency risk. We often recommend including allocations to Vanguard LifeStrategy Funds in clients’ benchmarks. These funds offer a wide range of return/risk objectives:
Absolute vs Relative Risk and Return
Selecting the right benchmark for a client is vitally important, since the absolute level of long-term risk and return will be more-or-less determined by the benchmark. As shown in the graph below, an asset mix with more stocks means higher expected risk and higher expected return.
The absolute level of risk and return for a given asset mix will change over time. For example, at the end of 2020 (blue dots below), the expected return for bonds was low because yields were low. At the end of 2024 (red dots below), the expected return for bonds was considerably higher. By comparison, the expected return for stocks moved down slightly, but stayed close to around 8%.
“Benchmark-driven” portfolios (such as Sapient Investments’ “Custom Core Portfolios”) are expected to have risk and return characteristics similar to their benchmarks. Absolute return and risk have been specified by the benchmark. The focus of portfolio management then becomes “relative” return and risk compared to the benchmark.
Benchmarks Define Relative Risk
A benchmark is a “neutral” portfolio. It is reasonable to expect that actual portfolio return will be close to benchmark return in the absence of information that would lead the portfolio manager to deviate from the benchmark portfolio. Deviating from the benchmark introduces the possibility of underperforming the benchmark. Staying close to the benchmark minimizes this benchmark relative risk:
Portfolio return
-Benchmark return
=Relative return
A portfolio manager’s objective is to provide positive relative return over the long-term. The portfolio manager is not responsible for the absolute return of the benchmark—that is determined by whatever the stock and bond markets provide.
Benchmarks Facilitate Performance Evaluation
Relative return reveals the marginal impact of the active decisions of the portfolio manager. Did the portfolio outperform or underperform the benchmark? That is the most important question. The next question is, what contributed to this relative performance?
At Sapient Investments, we use four broad risk factors to measure and control those risks that tend to explain the majority of the returns for the exchange-traded funds (ETFs) in which we invest:
Risk Factor Risk Description Risk Index
MKT Stock market risk S&P 500
LTB Interest rate risk 10 Yr. Treasury
DLR Currency risk US Dollar
OIL Oil price risk WTI Crude Oil
We measure the sensitivity of each ETF to these four risk factors using exponentially-weighted multiple regression analysis over 36 months. One of the ways that we control risk relative to the benchmark is by making sure that the exposures of our portfolios are close to those of the benchmark.
For example, MKT risk is usually the most important risk factor in explaining performance. Using our proprietary risk model, we measure the MKT beta of both the portfolio and the benchmark. However, we don’t always seek to keep the MKT betas of our portfolios equal to their benchmarks. Instead, we carefully manage the MKT betas of our portfolios relative to their benchmarks using our proprietary factor return forecasting model. If we have a higher than benchmark MKT beta, that is an intentional decision, and it means that our models indicate that we expect a positive return from MKT beta exposure.
Using our four risk factors, “performance attribution” is straightforward. For example, over any past time period, we know what the return of the S&P 500 was. From our proprietary risk model, we know what the MKT betas were for both the portfolio and the benchmark. If we had an above-benchmark MKT beta and the S&P 500 was up, we know that MKT beta contributed some amount of positive relative return. We can calculate the amount contributed by MKT beta down to the basis point. (A basis point is 1/100th of 1%.)
In a similar fashion, we can calculate the positive or negative contribution to relative return from interest rate risk (LTB beta), currency risk (DLR beta), and oil price risk (OIL beta). Whatever relative return is left after we account for the return effects of our four risk factors we call “residual return,” or return not explained by our four-factor risk model. Sometimes we can point to other factors that contributed to residual return, such as growth stocks vs value stocks (“style”), large cap stocks vs small cap stocks (“size”), Treasury bonds vs corporate bonds (“bond sector”), or emerging market stocks vs developed market stocks (“economic development”). These secondary factors are also part of our proprietary factor return forecasting model. We expect to garner some amount of positive relative return through our active management of these factors.
Most retail advisors do not provide detailed performance attribution for their returns. First, unless they use a benchmark (very rare among retail advisors), they have no objective basis for making any comparisons of their portfolios’ risks or returns. Instead, they often speak in generalities regarding the causes of their returns. Because they cannot provide complete and accurate performance attribution, they can’t say how much of their return was contributed by what factors. They don’t have any way of even knowing that themselves.
Retail Advisors Don’t Offer Benchmarks
In fact, most retail advisors don’t even use benchmarks. Why would they? That could prove embarrassing if they underperform the benchmark. Retail clients generally don’t know enough to demand the use of a benchmark, so the subject never comes up.
Instead, many retail advisors provide a “range” of benchmarks. The S&P 500 is almost always included. Perhaps a broader U.S. equity benchmark, such as the Russell 3000, might also be included. Most retail advisors do not include a specific international benchmark index even if they have international stocks. If the portfolio includes bonds, they may also include a bond index, such as the Bloomberg Aggregate Bond Index. However, it is rare that there has been agreement with the client on a particular fixed mix of indexes. Consequently, listing a variety of indexes individually helps to obfuscate performance, diffuse difficult questions, and avoid accountability.
Institutional Clients Require Benchmarks
I spent most of my career in institutional money management. Institutional clients, such as pension plans, foundations, and endowments, almost never hire money managers without a benchmark. (“Absolute return strategies” such as long/short hedge funds, are an exception to this general rule. I was a hedge fund manager myself for a time. I still manage some client assets in absolute return strategies, such as Sapient Investments’ Global Macro Portfolio, but the vast majority of our client assets are in long-only, benchmark driven Custom Core Portfolios.)
I just wouldn’t feel good about managing a client’s long-only money without a benchmark. A benchmark tells me a lot about the client’s absolute risk and return preferences. It offers an alternative investment that they could have made instead of entrusting me with their assets. Unless I am able to show that I provided at least a modest level of added return (after all fees) compared to the benchmark over the long term, I will not be satisfied.