This is the third in a series of posts in which our research team leverages new investment risk analytics in Stylus Pro to demonstrate how historical and forward-looking stress tests can provide deeper insight into fund performance across various market regimes and hypothetical scenarios.
Scenario analysis goes beyond traditional risk measures (i.e., standard deviation, VaR, max drawdown, etc.) by enabling you to pose the question, “what might happen to my fund or portfolio if…?” To fill in the blank and answer almost any query you have, you simply shock the fund or portfolio with the appropriate market index, rate, credit spread or economic factor. (Factor selection and thorough testing is an important step in this process; we explore this issue in a previous post: How to Ensure Your Portfolio Risk Analytics Are Working as Expected.)
In this post, we analyze a group of 338 target-date funds (TDFs),1 which offer a particularly good illustration of the value scenario analysis provides. TDFs, which are popular defined contribution investment offerings, have risk characteristics that are intended to become more conservative over time as the fund gets closer to its target retirement date. A large loss due to an extreme event could be disastrous to prospective retirees as was seen for some near-dated funds in 2008, the worst of which lost more than the S&P 500 index. (If you’d like to read more of our research on target-date funds, you can find it here.)
In the below chart, we have selected a universe of target-date mutual funds with a return history of at least five years. We then shock the group to plot estimated fund performance in the event of a 10% drop in the S&P 500 index (based on current fund factor exposures) against the funds’ five-year Morningstar Risk Rating.2
We can take away three main points from this factor shock:
- As we would expect, the TDFs closest to retirement date tend to display the lowest sensitivity to a market shock, with the funds meant to generate retirement income being the least sensitive.
- There is a visible relationship between the average shock estimate and the risk ratings, with the least sensitive corresponding to the lowest (least risky) rating of 1, and the most sensitive corresponding to the highest (most risky) rating of 5.
- There appears, however, to be a pretty wide range of sensitivities for each risk ranking within each category – signaling that using generic risk measures may not suffice to prepare for specific market downturns.
Focusing on near-dated 2020 TDFs, we see that those with a risk rank of 2 have shock estimates ranging between -3.2 and -4.8, meaning that the more sensitive funds could lose 150% of what the less sensitive funds do when subjected to the same shock.
The decrease in sensitivity as the risk rating decreases is not wholly consistent. For example, there are funds with sensitivities of approximately -4.8 with risk ranks of 2, 3 and 4. We can also observe that some funds are riskier than their rank might suggest, based on the market shock. For example, the most sensitive TDF with a risk rank of 2 has a sensitivity (again) of -4.8. The least sensitive fund with a risk rank of 4, however, demonstrates an estimated sensitivity of -4.5 to the same factor shock.
These differences may be due to allocation changes along each fund’s individual glide paths. For example, if a fund has 30% in growth assets today, and it held 50% five years ago, a risk rank based on 5 years’ historical returns (representing varying allocations) will probably make this fund appear riskier than it actually is.
The disparity may also be due to different asset mixes that happen to have similar recent volatility. For example, a traditional 60/40 portfolio of the S&P 500 Index and Bloomberg Barclays Aggregate Bond Index has the same historical standard deviation estimate as an 80/20 portfolio of Bloomberg Barclays High Yield Index and the Russell 2000 Small Cap Index. In this case, the specificity of scenario analysis will highlight differences in asset mixes without requiring a longer analysis history.
Whether due to differences in glide path or asset mixes, the estimate of loss due to a hypothetical shock helps to quickly identify current and specific risks, which could potentially go undetected with traditional, historical risk measures. This serves as a high-level, yet clear, example of how scenario analysis can complement traditional risk measures. In the next and final post of this series on risk analysis, we will isolate a universe of 38 near-dated funds to provide a more detailed demonstration of scenario analysis as a risk differentiator.
Footnotes
- 1Target-date funds follow a “glide path,” which starts with the bulk of the allocation to growth assets (primarily but not strictly equities). Allocation to growth assets is reduced over time in favor of a more balanced portfolio containing significant capital preservation assets (primarily, but not strictly, bonds) to a landing point at which point the allocation remains static.
- 2 Morningstar Risk Rating is a ranking within each Morningstar category based on downside variation.