The third quarter of 2011 was impressively bad. The S&P 500 Index lost 13.9% for the quarter. The VIX, the standard measure of market volatility, repeatedly closed above 40 during this quarter. To put this in perspective, the average daily closing value of VIX from the start of 1990 through the end of September 2011 was 20.5. The average daily closing value for VIX during Q3 of 2011 was 30.6.
Many critics of Target Date funds felt that these funds lost too much during the bear market in 2008. Special attention was focused on 2010 Target Date funds, funds designed for investors planning to retire in 2010. The poor performance of these funds even got the attention of the SEC, which proposed new disclosure standards. Market observers (including the SEC) noted that 2010 Target Date mutual funds lost an average of 24% in 2008. In light of 2008, many funds redesigned their asset allocations to be more resistant to massive market declines.
Now, let’s flash forward three years. Q3 of 2011 provides an excellent opportunity to revisit this discussion, not least because of the enormous growth in the Target Date Fund industry. In fact, it’s projected that Target Date Funds will comprise 48% of all assets in self-directed (a.k.a., Defined Benefit) retirement plans by 2010.
Performance of 2010 Target Date Portfolios in Q3
Bearing in mind that a 2010 Target Date fund is designed for investors retiring in the year 2010, how do things look?
The hypothetical investor who invests in these funds has been in retirement for somewhere between 0.8 and 1.8 years (assuming that he or she retired sometime during calendar year 2010). The three largest Target Date fund families hold approximately 75% of all of the assets invested in Target Date funds. The average Q3 return for 2010 Target Date funds from these three fund families was -8%.
Folio Investing developed its own Target Date solution in 2007 called Target Date Folios (Disclosure: I am a paid consultant to Folio Investing and I worked on developing the Target Date Folios). The Target Date Folios are portfolios of ETFs that investors can invest in and manage as pre-built portfolios. There are three 2010 Target Date Folios, which were designed for investors who retired in 2010. The Moderate 2010 Folio was designed for the average investors. There is also a Conservative 2010 Folio and an Aggressive 2010 Folio for investors who, by reason of the specifics of their personal financial situations, prefer a more conservative or more aggressive portfolio. The 2010 Moderate Folio returned -2.5% in Q3 2011, and the average of the conservative, moderate and aggressive 2010 Folios was -2.9% for Q3 2011. (Please note: The results above do not include brokerage fees).
Did 2010 Target Date Portfolios Perform Better This Time?
How do we view Target Date Funds and Folios in light of Q3 2011? This is not a simple question.
The standard argument in favor of maintaining a substantial exposure to equities and other high risk asset classes in portfolios designed for people within a few years of retirement is that they will need the growth potential of stocks in order to provide sufficient return to live during an extended retirement. Many Americans can expect to live 20 to 30 years or more in retirement. While Q3 reminds us of the loss potential in equities, it is also true that the anemic 2.0% yields on 10-year Treasury bonds will provide sufficient income only for the wealthiest (or most frugal) retirees.
If you are a recent retiree invested in a 2010 fund and have experienced a decline of 8% in your retirement portfolio during the last three months, you will, no doubt, worry. For a recent retiree, an 8% decline equates to a substantial reduction in either income or retirement security. To a person retiring in 2010 who was planning on following the “4% rule” for drawing income, the 8% decline in Q3 represents a loss of about two years of retirement income.
How Risky Did These Funds Look Before The Fall?
Let’s step back to the period before volatility really picked up again—May 2011. In May, market volatility was still very low (see chart below) and stayed between 15% and 20% as measured by the VIX.
VIX Index over the Past Six Months (Source: Yahoo!)
From May to August, the VIX doubled and equity returns turned negative for the year. In thinking about the loss potential in 2010 Target Date funds, it is very informative to ask whether the levels of losses we have experienced were predictable or not before they occurred. In other words, was the risk level something knowable before the losses happened in Q3?
In an article in late May of 2011, I noted that the long-term market implied volatility (effectively the market’s forecast of risk) for the S&P 500 was at 21%. This number is an important input for risk projections. You can use this value (combined with Monte Carlo Simulation) to come up with projected risk levels for a portfolio or a single mutual fund. I am going to estimate risk from the perspective of the end of May of 2011 for this article because I noted the long-term market implied volatility in an article at that time.
When I run my Monte Carlo Simulation tool (Quantext Portfolio Planner) using data available only through May of 2011, I get the following results:
Source: Author’s calculations
Standing at the end of May, the projected 10th percentile loss level for a 91-day period (e.g. one quarter) was -7.7% for the composite of the “Big 3” 2010 mutual funds. This means that the simulations indicate that there is a 10% chance of losing 7.7% or more over any 91-day period. Given this projection of potential loss, the 8% loss in the composite of the largest 2010 Target Date funds is bad, but not outside of fairly common risk levels. Q3 of 2011 was, in other words, no worse than about a 1-in-10 event. The 2010 Moderate Folio, which lost 2.5% in Q3, has a projected 10th percentile loss of 6.0%. We can understand the difference in losses by looking at the fourth column in the table above, which shows a standard portfolio metric called R-squared (also written as R^2). This measures the fraction of the variability in returns from a portfolio that can be explained by moves in the S&P 500. By definition, R-squared is 100% for the S&P 500.
What is really notable here is that the R-squared for the composite of 2010 Target Date funds is 96%, which means that increases or decreases in the value of these portfolio’s are almost entirely determined by moves in the S&P 500. Putting this another way: the substantial losses in the 2010 mutual funds in Q3 are a result not just of the amount of risk in these funds, but also the type of risk. The dominant source of risk in these mutual funds is the stock market index. In the 2010 Moderate Folio, more of the risk budget is allocated to other asset classes such as commodities and low-Beta equity classes such as utilities, healthcare and telecommunications.
The main result of this analysis is that it was clearly evident back in May that the level of losses experienced by 2010 mutual funds and Folios in Q3 were well within the realm of possibility—far from a lightning strike or shark attack kinds of odds. One-in-10 events happen pretty often.
What Have We Learned?
Asset allocations designed for recent retirees (retirement in 2010) sustained substantial losses in Q3 of 2011. For the composite of the three 2010 Target Date funds from the three largest fund complexes, the Q3 return was -8%. In addition, Monte Carlo simulations suggest that this level of loss was not outside of the realm of losses that will happen periodically in a retiree’s lifespan.
As in 2008, the key question is whether investors had sufficient information to understand the risks that they were assuming by investing in a given Target Date solution (or any other portfolio, for that matter). Even if investors did have meaningful projections of potential losses, what would their best recourse have been? If you retired in 2010 and you want a simple investment solution, but the risk in your 2010 Target Date fund looked too high what could you have done?
One alternative, of course, would be to look for more conservative asset allocations across a range of fund families. This dilemma is why Folio created Target Date Folios in multiple risk levels. The 2010 Conservative Folio returned +0.6% in Q3 of 2011. The projected 10th percentile, 1-quarter loss for this portfolio using data through May 2011 is -3.7%.
Debating whether the risk levels in the various 2010 funds is not likely to be highly productive. Each firm that develops Target Date funds has well-educated practitioners who choose the target risk levels and create the resulting asset allocations. Professionals in the field have widely varying opinions as to the appropriate risk level for the average investor at each stage in his or her life, but there is not (nor will there ever be) universal consensus on this issue. The real question is whether a given investor has a risk tolerance that makes a given fund (or Folio) suitable. I would imagine that many investors in 2010 funds were surprised that their portfolios dropped as much as they did in Q3. That is the core of the problem.
Even through many Target Date Funds have changed their investment policies since 2008, the challenge that we face in responding to Q3 of 2011 is same one that we faced in 2008. First, we need to provide meaningful estimates of loss potential to investors. Second, investors must be able to compare these loss estimates across a range of investment solutions to determine which portfolios provide the best match to their risk profiles.
- Sanity Checking Estimates of “Expected Returns in Retirement Planning
- Risk Budgeting: A Critical Tool for Portfolio Management
- Target Date Strategies Weather the Storm
SPONSORED BY Folio Investing The brokerage with a better way. Securities products and services offered through FOLIOfn Investments, Inc. Member FINRA/SIPC.