| |
 March 2006 Past as Prologue Backtesting P6's Benchmarks
The first simply mimics the way a lot of investors handle their own portfolios: It's just buy-and-hold. When we considered returns from 1979 - 2003 we found that 25% corporate bonds, 48% domestic large cap equities, 21% domestic small cap equities, and 6% foreign stocks would have been the "best" mix. This was adopted as the buy-and-hold benchmark.
As time passes, however, this initial "best" mix will give way to other changing combinations. As soon as the underlying assets begin to trade, the initial mix is altered. Presumably, it should be necessary to periodically rebalance back to the original mix to capture optimal performance. That's P6's second, or "static", benchmark. Oddly enough, in the two-plus years that P6 has been out of sample, the two benchmarks have never strayed more than 5% -- the trigger we set for rebalancing -- from the original allocation. As a result, the two remain identical, something we clearly did not anticipate.
Is this normal or is this just the result of a relatively trendless market? One way to address this question is to observe the same mixes in the time before P6 came into being. Did both benchmarks behave the same then?
Similar But Different The results are illustrated by Chart 1's cumulative return graph. Despite appearing to track almost identically, the reallocated benchmark actually managed to beat out the buy-and-hold approach. After the full 25 years, a dollar invested in the rebalanced benchmark would have grown to $17.89 while it would only be worth $16.55 in a buy-and-hold scenario. That's a difference of 8.1%. This review doesn't reveal when the first rebalancing would have occurred or how frequently thereafter, yet the difference in cumulative return suggests rebalancing did occur. Just looking at Chart 1, it would appear the buy-and-hold mix followed stocks up in the late-1990s while the other was rebalanced. This makes sense given that domestic large cap stocks were far and away the major market winners. The exact opposite occurred in 2001 and 2002 as stocks plunged and rebalancing allowed the static benchmark to regain the lead over the buy-and-hold mix.
You can clearly see this from Chart 2. Here we've divided the 25 years up into 5-year periods. In three of the five, the rebalanced mix leads the buy-and-hold strategy, and the fourth is basically a push. In the fifth, however, it trails badly. That's no surprise since stocks were running from January 1994 through December 1998. Every time the static mix rebalanced, it sold high-performing stocks for low-performing bonds. In this case, rebalancing caused it to fall behind. Even so, looking back at Chart 1, you'll notice the static mix didn't fall behind its counterpart until the late-1990s. In essence, it took that long for the latter to overcome the lead established in the earlier periods. Rebalancing helped then, and again when stocks stumbled in the early part of this decade. Although it's hard to tell from Chart 1, rebalancing must have occurred in each of the 5-year periods or else the returns would not have differed to such an extent in each period. Not only did rebalancing occur, it really made a difference. This is exactly what Modern Portfolio Theory predicts. Also as predicted, the static mix is more efficient than buy-and-hold. This is illustrated on Chart 3 which plots the benchmarks as well as their constituent indexes in risk/return space. Here both have roughly the same return but the static mix has less risk. As a result, its risk-adjusted return is higher. So in the past, these two approaches have had textbook results. Asset allocation actually prevailed over simple buy-and-hold. But what about now?
Recent Past That's not to say the buy-and-hold mix hasn't begun to stray. As you'd expect, that started in its first few months. As you probably recall, stocks reawakened in 2003 with small caps taking the lead. The Russell 2000 was up 17% in 2004 and just over 3% in 2005. Bonds, on the other hand, have been flat to slightly down, victims of the Fed's tightening policy. By December 2005, bonds had fallen from the initial 25% to 21% of P6's benchmark mix. Had they fallen just one percent more, the first rebalancing would have occurred and the two benchmarks would have started 2006 moving independently
Instead, small stocks faltered briefly as the year got underway, allowing the bond position to move back to 22%. By mid-February, however, fears of rising inflation and a prolonged monetary tightening again sent them back down to 21%. Even if bonds continue to decline as a percentage of the benchmark, no rebalancing will occur until P6 is reoptimized. This will occur in the first few weeks of the second quarter. If markets continue their current trend, the benchmarks could easily diverge in April. It's not an issue of if the two benchmarks will separate, it's simply a question of when. Chart 2 clearly illustrates how differently the mixes behaved in the past. There's no reason to believe they won't -- and in fact there's every reason to believe they will -- act in the same manner over the next 25 years. Search this site! Just enter you key word or words: Get current quotes or follow your own custom portfolio,
courtesy of E-Line Financials:
|
||||||||||||||||||||||||||||||||||||
UANTITATIVE PORTFOLIO 6 WAS launched a little over two years ago, effective January 1, 2004. As a balanced portfolio with both stocks and bonds, a pure equity or bond index wouldn't be the appropriate benchmark. Instead, it needed a "blended" benchmark -- in fact it got two.
