Manager of institutional portfolios debunks concern that asset allocation strategies and rebalancing won't work in coming years.
As an adviser to institutions around the world, Steve Hammers is hearing a lot these days about how Modern Portfolio Theory is failing investors.
“We heard the same thing in the late '90s,” said the Nashville, Tenn.-based chief investment officer at Compass Advisory Group, which oversees about $3 billion in assets for institutions and other advisory firms.
But just as complaints about the value of portfolio diversification and holding noncorrelated asset classes died to a dull roar after the go-go years of the '90s, Hammers is predicting a rebirth in MPTs during the next several years.
In fact, he finds it natural that more investors are questioning the prevailing wisdom of strategies first developed by Harry Markowitz some 50-plus years ago. “In an 18-month stretch, almost everything investors touched was falling and correlations across almost every asset class looked very similar,” said Hammers. “It was natural for people to question the nature of Modern Portfolio Theory and wonder if it could really serve them well going forward."
But he's stressing to clients that a nearly simultaneous collapse in credit markets and equities is a very rare occurrence. "We’ve run data showing that since 1926, asset allocations based on long-term correlations between multiple asset classes have produced superior risk-adjusted returns 95% of the time,” said Hammers.
That leaves some 5% open to periods such as what took place from September 2008 through early March 2009. Hammers views such a six-month time frame much like other past downturns when correlations were dragged down almost uniformly, and, as a result became tightly connected.
“There are going to be anomalies when differences in correlations become almost nonexistent,” he said. “But our research shows that happens very infrequently. As index-based asset allocators, when such anomalies do take place, we see it as an enormous opportunity to add more value to our portfolios through rebalancing.”
Big Rebound In Preferred Stocks
Beginning in late November 2008, Compass rebalanced portfolios by adding to asset classes that were down the most. Those included international, emerging markets, real estate investment trusts and preferred stocks. But they also included commodities, buywrite strategies and currencies.
Also, in late February of this year, Hammers again rebalanced as markets were still tumbling. He built client positions in the same ETFs.
“Those two rebalancing periods really have added to our performance this year,” he said. “Our more conservative and moderate portfolios were up slightly more than 10% this year through June 30. That compares to the S&P 500’s 3.16% gain and the Barclays Capital U.S. Aggregate Bond Index’s 1.91% return.”
A big part of those outsized returns came from bumping up exposure to the PowerShares Preferred Portfolio (NYSEArca: PGX). “The ETF’s underlying index from March 9, when markets bottomed, through the end of the second quarter has risen more than 100%,” he said. “That’s why rebalancing is so powerful. I don’t have to try to time anything—it does everything for me.”
Between last September 2008 and March 9 of this year, the ETF’s Merrill Lynch benchmark plummeted some 62% in value, according to Hammers.
“Our investments in the preferred index didn’t fall as hard as a lot of others, but they dropped enough to trigger our rebalancing red flags,” he said. “As markets have recovered since then, our positions in PGX have really helped bolster our returns—especially considering we replenished those allocations while prices were still quite low.”
Currencies & Hedging ETFs
Compass uses a methodology based on an index-of-indexes approach. It takes 15 ETFs representing 13 asset classes and more than 114 markets across the world. It doesn’t break out sectors, styles or specific countries. Instead, Hammers sticks to broad-based ETFs covering wide swaths of asset classes and geographies.
Hammers sets bands to rebalance around 13 different asset classes used in portfolios. Each client portfolio is set at a specific allocation to each using exchange-traded funds. If any of those numbers move up or down by more than 20% from the original target allocation percentage, then a rebalancing will take place.
So for example, if large-cap U.S. stocks have a 20% weighting of an overall portfolio, then it can go as high as 24% or as low as 16%. "That's a deviation of plus or minus 20% of the target allocation," said Hammers. "So you take whatever target in each asset class you've set and multiply it by 20% to get the bands."