When assessing performance, one of the key--but often overlooked--components is the time period under consideration.
Much has been made in the financial media about Apple's AAPL contribution to stock market gains over the past three months. Much less highlighted has been the fact that three-year annualized returns have now replaced the dismal first quarter of 2009 with decent (for most securities markets) first-quarter 2012 results. Three-year annualized returns are important, if only because many investors believe that long-term investing requires at least a three-year view. In fact, many of our own, proprietary metrics start with a three-year look-back period. But what is sometimes overlooked is the composition of that three-year time frame.
This is really nothing new. Every rolling period presupposes a "yesterday" and a "tomorrow"--a time frame that rolls off and another that will roll on. However, given the blowout bottom that some markets hit in early 2009, this is a good time to take a look at exactly how the composition of a time period alters the results. This holds true whether or not we are considering a fund's performance, a market trend, or an industry study. This is also why I prefer, when possible, to look at rolling periods over broad swaths of time on a quarterly or monthly basis. We may be able to draw out a couple of other suggestions as well.
Performance is one of the pillars by which we assign our analyst-driven, qualitative, forward-looking Morningstar Analyst Rating. In the methodology document, one of the points we make about performance is: "Trailing returns and calendar-year returns are of interest, but they are insufficient in themselves given end-point dependency in the former case and the arbitrary nature of the latter. We consider many periods and performance aspects to build as comprehensive a picture as possible." The change between three-year annualized performance for the period ended Dec. 31, 2011, and the period ended March 31, 2012, is a good case study for why this point is so important.
Before delving in too quickly, I do want to mention that this exercise only considers funds that were in existence on March 31, 2009, and on March 31, 2012. So, there is survivorship bias. We also are not considering funds with similar portfolios, but we are considering performance relative to other closed-end funds. This is simply because we are making a point about quarterly changes in annualized total return performance and not trying to compare funds against one another for an investment decision.
Below is a table showing the top-performing CEFs based on a three-year annualized period ended March 31, 2012. Typically, when looking at performance, we look at funds with similar holdings, comparing U.S.-equity CEFs with other CEFs, taxable fixed-income CEFs with one another, and so forth. Here, though, we are simply looking at all CEFs because we're not looking at relative performance; we're demonstrating the power of replacing a very bad quarter with a relatively decent quarter.
If you were to rank all 575 U.S.-traded CEFs that have at least a three-year track record, these would be the top 10 funds with the highest three-year annualized total returns on a net asset value basis. Two things jump out. First, half of the funds are real-estate-focused. Someone new to CEFs might look at this list and conclude, incorrectly, that about half of all CEFs are real-estate-focused. In fact, of the 575 CEFs under consideration, only nine are real-estate-focused, and five of them show up--somehow--on this list. That's remarkable. One might conclude that real-estate-focused funds must have had truly great performance, compared with other CEFs at least, over the past several quarters, but they would be wrong.
The second thing that jumps out is the disparity between the quarterly performances for the quarters ended March 31, 2009, and the quarters ended March 31, 2012. Cohen & Steer Quality Income Realty RQI, for instance, has a 61.1 percentage-point swing in total return performance by rolling off the first quarter of 2009 and rolling on the first quarter of 2012. Aberdeen Indonesia IF is the only fund on the list not replacing a double-digit loss with a double-digit, or nearly-double-digit, gain.
Looking at RQI as a proxy for the real estate CEFs, we can see that it has not been truly great performance over the past several quarters that has propelled the real estate funds to the top of the CEF pack. While RQI had top-10 total return quarterly performances in the second and third quarters of 2009, it fell back to the pack along with other real estate CEFs as other sectors predominated in the intervening period. To use a baseball metaphor: The fund essentially hit singles and doubles for the past 10 quarters, it had two home runs 11 and 12 quarters ago, and now that the absolutely atrocious 13th quarter has rolled out of the three-year window, it emerges as the top-performing CEF.
This is one lesson we can learn from this list. We've written about how funds that are consistently around the middle of the pack, quarter after quarter, typically have better performance than funds that occasionally hit the home run. In fact, when looking at MLP-focused CEFs, the funds managed by Tortoise Capital exhibit this trait of middling, yet consistent performance; it's one of the factors that I actually like about Tortoise MLP NTG and Tortoise Energy Infrastructure TYG. It's an investment strategy akin to the Oakland Athletics' baseball tactics in Moneyball.
The three-year list is what it is. But here is where looking at rolling periods and other time frames can be of real benefit. We've already noted how RQI--and the other real estate funds--had a blowout rebound in the second and third quarters of 2009. What happens when those quarters roll off? Well, to get a taste of what might be in store, we can look at performance from Sept. 30 2009, through March 31, 2012, excluding those two rebounding quarters.
We see that the only two funds remaining on the top performers list are RQI and Neuberger Real Estate Securities NRO. This suggests, at least to me, that those two funds are worth a deeper dive for investors interested in real estate funds because I suspect that, relative to other real estate funds, these two are consistently near the top in terms of performance.
We can also look at a broader time period to encompass the real estate funds prior to their meltdowns. Instead of looking at three-year annualized returns, we can look at four-year annualized returns. And here is where the deceptiveness of the three-year view becomes apparent.
Yes, someone who happened to purchase and hold the funds from March 31, 2009, through March 31, 2012, had great returns. But the investor who purchased RQI or Nuveen Real Estate Income JRS a year earlier is only now at break-even, while anyone who purchased NRO at the end of 2008's first quarter is still 6.2% underwater. In fact, to again use RQI as the example, while the fund has indeed had great performance over the past three years, all it really did was make up the ground that it had lost in the four quarters from March 31, 2008, through March 31, 2009. This is precisely the reason that rolling time period returns, various time frames, and comparative yardsticks are important when assessing performance. And we haven't even touched on the subject of risk-adjusted returns, which incorporate a fund's volatility.
Studies have shown that investors chase recent returns. This causes them--as a herd--to buy at or near the top of the underlying markets and to sell at or near the bottom of major sell-offs. Indeed, at a certain point it becomes tautological: Until the final marginal seller sells, the bottom will not be reached, and until the final marginal buyer buys, the top hasn't been achieved. Some deft investors can successfully play the momentum game, but most investors simply do not have the time or the discipline to win with that strategy. Instead, most long-term investors do best by buying what is out of favor and selling what the crowd is clamoring for. They do not chase returns. This is, in a nutshell, why past-performance measurements are only one of the variables we use in assigning our Morningstar Analyst Ratings.
Be careful when assessing a fund's past performance because a large influence on that performance isn't the performance itself but the time frame being considered.
Join Morningstar on Wednesday, April 25, 2012, at Capital Link's 11th Annual Closed-End Funds & Global ETFs Forum--the only premier industry, marketing, and networking event to combine closed-end funds and ETFs. Financial advisors, institutional investors, analysts, and financial media utilize this forum as a resource for sharing and evaluating the latest CEF and ETF products and trends. Morningstar's own Scott Burns will be delivering the keynote address. Paul Justice, director of North American ETF research, and Mike Taggart, director of U.S. closed-end fund research, will be moderating sessions. Click here for the agenda and registration.