Warning: benchmarking against traditional indexes can mean further carnage. Ron Surz is president of PPCA, a San Clemente, Calif.-based software firm that provides performance-evaluation and attribution analytics, a principal of RCG Capital Partners, a Denver, Colo.-based fund-of-hedge-funds manager, and a co-founder of target-date index maker Target Date Analytics.

This is a warning to my performance-evaluation friends; with luck it comes in time to be of use in third-quarter performance reviews. Name-brand index results for the year through 30 September 2008 are misleading and likely to result in bad decisions. And the problems spill over into the new and growing area of target-date funds.

U.S. stock indexes

Some would say one style index is as good as another: they're all about the same. This is generally true, but not for the first nine months of 2008. As shown in the following exhibit, the most popular indexes -- the Russell and the S&P -- show value and growth losing about the same amount; roughly 19%. But these benchmarks are missing an important component of the market: the stuff in the middle -- between value and growth -- the stuff I call "core."

With core in the picture, value outperforms growth and core outperforms both value and growth. This is important because investment managers who are compared to the Russell or the S&P will be misjudged at a time of high investor anxiety. Core typically outperforms when investors lack conviction, favoring neither value nor growth. Now more than ever it's important to get the benchmark right. Otherwise today's faulty decisions will undermine future performance.

|image1|

This kind of discord, while infrequent, is explained by differences in methodology that can be best understood by considering how the S&P and the Russell treat stocks in the "gray area" between value and growth. There are degrees of value and growth: some growth stocks are more aggressive growth than others; some value stocks are deeper value. And some stocks have characteristics that aren't clearly value or growth -- they're the stuff in the middle. The Russell deals with this issue by allocating these fuzzy stocks pro-rata into value and growth. The S&P ignores the problem altogether by drawing a hard line that divides the market in half between value and growth.

A better way to deal with this "medial stock" issue is to define a separate classification called: core. Surz indexes break out value, core, and growth stock groupings within each market cap by establishing an aggressiveness measure that combines dividend yield, price-to-earnings ratio, and price-book ratio. The top 40% (by count) of stocks in aggressiveness are designated as growth, the bottom 40% are called value, and the 20% in the middle are core. The result is a family of indexes that are mutually exclusive and exhaustive, which makes them perfect for returns-based style analysis. In the first nine months of 2008, large core performed substantially better than both large value and large growth.

In terms of performance, core usually falls between value and growth. About a third of the time, however, it doesn't. This year to date is such a period. And in such unusual periods an alternative to the Russell and the S&P provides valuable insights. Details of Surz index construction and behavior are available here, and a list of stocks classified as core is available on request.

|image2|

You can use the index results above to take advantage of this insight, and you can plot your fund's return against the appropriate group in the next exhibit. As you see, there's little overlap in the intra-quartiles, so misclassifying a manager will lead to erroneous evaluations. Also, classification bias in traditional peer groups is highly problematic in the nine months to date. For a discussion of this little known bias click here.

The universes in this exhibit are created using an unbiased scientific approach called Portfolio Opportunity Distributions (PODs). They represent all of the possible portfolios that managers could have held when selecting stocks from their respective indexes. In essence, hypothetical monkeys expand an index into a peer group by simulating all the portfolios that could have been formed from stocks in that index. By contrast, traditional peer groups are very poor barometers of success or failure because of their myriad biases. Everyone knows it's easy to find a peer-group provider that makes you look good, and that the industry tolerates, even condones this deceptive practice. PODs are bias free and are therefore a much more reliable performance-evaluation backdrop.

Target-date funds

Unlike the dozens of U.S. stock indexes from plethora of providers, there are only three target-date fund-index providers. Dow Jones struck first, in April 2005. Then my firm, Target Date Analytics (TDA), introduced target-date indexes in October 2007. And Standard and Poor's just announced the creation of their indexes. Plan Sponsor magazine adopted the TDA indexes in August 2008 and re-branded them as PLANSPONSOR On Target Indexes (OTI).

These three indexes differ substantially in composition and philosophy. Let's discuss philosophy first.

The S&P indexes are industry averages, reflecting common practice among current target-date fund offerings. S&P describes this construction: "Each index is representative of the investment opportunity available to investors for the corresponding target date horizon, with asset class exposures driven by a survey of available target date funds for that horizon."
The Dow indexes aim to "measure the performance of a lifecycle portfolio that seeks to grow and preserve real value over time."
PLANSPONSOR OTIs have two objectives: "(1) protect the purchasing power of contributions with a very high probability, and (2) grow assets with a reasonably high probability, without jeopardizing the primary protection objective."

The major difference between the Dow and OTI indexes' objectives is priority. Dow places equal emphasis on preservation and growth whereas the OTI indexes emphasize preservation over growth. In sum, the Dow and OTI indexes are standards -- the way target-date funds should be managed -- and the S&P indexes are common practice aggregates. Unfortunately, we find common practice woefully lagging both the Dow and OTI standards. Target-date funds are a great idea with awful execution, at least so far.

These differences in philosophy and objectives lead to materially different index compositions and results. The following exhibit summarizes composition differences at the broad equity allocation level. As you can see, as the target date approaches the S&P indexes are the most aggressive, followed by the Dow, and the PLANSPONSOR OTI indexes are the most defensive. All three indexes are quite similar in equity allocation at the more distant dates.

There's a debate in the target-date industry regarding the purpose of target-date funds. At TDA we believe that target-date funds should be limited to the accumulation phase of a participant's lifecycle, and that the distribution phase is best served by vehicles designed for this purpose, such as annuities. Accordingly, we believe that target-date funds should be entirely in safe non-risky assets at target date, waiting for the participant to move to the next phase, which they should be thinking about long before retirement. By contrast, the industry sees target-date funds morphing into distribution funds at target date and continuing on, in some cases to death. This is akin to viewing the target date as a small speed bump on the highway of life.

What do you think the role of target date funds should be? I'd like to hear from you.

|image3| As shown in the next exhibit, these compositional and philosophical differences impacted target-date index performance in the first nine months of 2008. This has been a wake-up call for the target-date industry -- and gives TDA an I-told-you-so opportunity I wish we didn't have.

Our greatest concern is for investors in near-dated funds, who are at or near retirement, and have the most at stake emotionally and financially. For the most part these investors are in target-date funds as a default option in their 401(K) plans, since target-date funds are one of three qualified default investment alternatives (QDIAs). Do you suppose any of these folks were prepared for the kind of disastrous loss that has struck the S&P and Dow indexes? Did they know the risks they were exposed to? Participants in longer-dated funds can better tolerate these losses because the chances of a recovery are reasonable and they have many more years to save.

|image4|

Your choice of index makes a big difference in your evaluation of target-date fund results. So choose the index that is most in line with the plan participants' understanding of what target funds should be.

But the OTI defensive posture at target date is not the only difference among the indexes. Diversification is also an important distinction. The S&P indexes are not very well diversified, comprising mostly U.S. stocks and bonds, because this is the current industry practice. The Dow indexes are somewhat more diversified, and the newer Dow real-return indexes are even better. But the most diversified indexes are the OTI, which represents the world market. The benefits of diversification are shown in the following long-term performances of 2010 funds.

|image5|

As you can see, the choice of target-date index makes a big difference over longer horizons, extending back beyond the current meltdown into better times. So now you can make an informed selection.

Conclusion

This is one of those unfortunate times when consultants and investment managers will try to console their clients by explaining how their pain is less, hopefully, than most others. This will be awkward and delicate, and is likely to bring forth the difficult questions about bailing or doubling down. The answers to these difficult questions are best formulated when accurate benchmarks are employed. Clients need to know who is succeeding and who is failing, rather than who can pick the benchmark that makes them look best. -FWR

Purchase reproduction rights to this article.