| |
 November 2008 Cyclicality Is P6 Making Any Real Progress Relative to its Benchmarks?
That's true for our equity models, too. Check out the graphs page for proof. At this point, only Portfolio 3 remains noticeably below its benchmark. But is an upward sloping long-term chart truly a mark of a successful investment? After all, if the market itself tends to rise, is it so remarkable that any given portfolio would, too? That's why we compare each model against a benchmark index. Then again, is simply outpacing the index a mark of success?
Golfers will tell you it doesn't matter how you get the ball in the hole, the only thing that matters is how many strokes it takes. Some believe the same holds true for investments: It doesn't matter how you get ahead of the index, all that matter is that you do. But is that really the case? We suspect it's not. Investors, being human, tend to prefer gains that are steady and predictable, even if they're ultimately less than those attained by more volatile alternatives. Long-term charts tend to mask short-term volatility by providing a snapshot of performance at one particular point in time, years after inception. In such instances, it's advisable to consider the investment's short-term trading pattern (how it "got there") as well as the overall return. Quantitative model P6 is a good case in point.
So Far, So Good P6 has two custom benchmarks because P6 is a blended model, able to hold domestic and foreign stocks, as well as domestic bonds. Because its holdings can be so diverse, it was necessary to create blended benchmarks rather than just relying upon established equity or fixed income indexes.
Based on historical performance, we created two benchmarks, both starting with the same blend of equity and fixed income: 25% Domestic Bonds, 48% Domestic Large Cap Stocks, 21% Domestic Small Cap Stocks, 6% Foreign Stocks. The first benchmark is simply a "buy-and-hold" blend which is never rebalanced. The second is rebalanced every calendar quarter when at least one asset class strays 5% or more from its original allocation. This allows P6's performance to be compared to both a typical buy-and-hold investor's return as well as one who regularly rebalances back to a static allocation. P6 itself is based on exchange traded funds (ETFs) rather than individual stocks. Its universe of potential holdings is limited to S&P 500, Russell 2000, Lehman Government Bonds, Lehman Aggregate Bonds, and the EAFE Index ETFs. P6 is reoptimized in the middle of the first month of every quarter. This a forward-looking process and takes current market and economic conditions into consideration. As opposed to the static allocation benchmark, this is a "dynamic" allocation and can change every quarter. (For more on the difference between static and dynamic asset allocation, click here.) It's hard to gauge the difference from Chart 1, but from January 1, 2004 through September 30, 2008, P6 was up 11.8% while its buy-and-hold benchmark and static benchmark gained 7.4% and 8.1% respectively. That's not a tremendous difference, but it is in P6's favor. But here's the concern: Although P6 is ahead of its benchmarks, the lead isn't so great that it could easily be lost in a short period of time. Again remember, Chart 1 is a long-term graph showing the snapshot of the series relative results as of September 30. While these are the relative returns since 2004, they're not so large that they couldn't easily reverse themselves. In fact, as you'll notice from Chart 1, P6 actually trailed its benchmarks from April 2006 through October 2007. This came after having leads similar to today's back in 2004 and 2005. In other words, P6 has had and lost similar leads before.
What this also indicates is that P6 has not been steadily building its current lead, but instead has fluctuated vs. the benchmarks. This is clearly illustrated in Chart 2 which shows the relative difference between P6 and its buy-and-hold benchmark. The range is almost symmetrical around the index: +5.7% to -5.2%.
Does It Matter? Chart 2 shows how quickly P6 can build a lead over its benchmarks, but also how quickly it can lose it. More importantly, it also shows that there have been prolonged periods of time when P6 traded both above and below it's benchmarks. More precisely, P6 has been ahead of its benchmarks just a little over 54% of the time, so essentially half the time it's been above it and half the time it's been below it. It's tempting to say that at any give time an investor in P6 would be just as likely to be ahead of the benchmark as behind it, but that's not really true. P6 doesn't fluctuate that wildly. Instead, it does trade for those prolonged periods above and below, so it's not exactly a crapshoot as to where you'll land. But this is also a problem for P6 because you're just as likely to need to sell when it's up as when it's down. Chart 2 indicates that P6 oscillates around the benchmark. If you wanted to sell in 2006 or 2007, you probably would have underperformed even the buy-and-hold standard. You would have done better (relatively at least) this year, but the point is unless you can time your exit -- or your snapshot of performance -- you run the risk of ending up under the benchmark. That's not what P6 was designed to do. All of our quantitative models were created with the hope that they would consistently and incrementally outperform their respective benchmarks. If successful, they would build larger and larger leads over time, gradually moving away from the index. Under that scenario, after the first few years or so, the investor would always benefit from the model. So far that's not happened with P6. The current long-term snapshot is favorable, but the way it got there isn't. It's not clear why P6 has fallen into this pattern, but it's something that needs to be investigated and perhaps altered. This is a concern not only because P6 is not establishing an sustainable lead, but because it's incurring transaction costs along the way. As opposed to the buy-and-hold investor as represented by the benchmark, anyone actually investing in P6 would incur quarterly transaction charges whenever the model is rebalanced. There's little reason to pay for trades when gross returns spend almost half the time below the buy-and-hold benchmark. In this case, it really does matter how you get there. Search this site! Just enter you key word or words: Get current quotes or follow your own custom portfolio,
courtesy of E-Line Financials:
|
||||||||||||||||||||||
ONG-TERM CHARTS HAVE A WAY of making just about anything look good. Perhaps it's because the market tends to go up more than it goes down or simply because more time lends a better perspective. Regardless, even the most volatile portfolio has an impressive long-term chart.
