Systematic Relative Strength

If your approach to providing asset allocation advice to clients is heavily influenced by mean-variance optimization, the following commentary from BNY Mellon Asset Management will probably make you a little weak in the knees.

Investors who fail to adjust their asset allocations to changing market conditions are likely to achieve disappointing returns, according to a recent report by Mellon Capital Management Corporation, part of BNY Mellon Asset Management.

“We believe the inability of a static asset allocation mix to accept new information was the main culprit behind the unrealized return expectations for many institutional investors,” said Jonathan Xiong, director, global asset allocation, for Mellon Capital. ”Investment managers need to dynamically change their asset allocations within a portfolio to reflect the most recent changes in expectations.”

Most public, corporate and endowment portfolios over the last decade have adhered to a static asset allocation, with the only changes in asset class exposures driven by market movement, according to the paper.

“A fallacy of the static allocation approach is that it assumes return expectations will not change, regardless of capital market or macro economic conditions,” said Xiong. ”Credit spreads and equity risk premiums can be volatile, and our studies indicate that changes in these factors have affected returns. Our research concludes that a five percent change in expected equity returns has the potential to shift the optimal asset allocation by more than 80 percent.”

The reality is that mean-variance optimization lacks the flexibility to deal with paradigm shifts. So asset X has generated an annualized return of Y over the past 80 years. What guarantee do you have that its annualized return over the next 10+ years won’t be +/- five percent from its long-term average? As pointed out by the research cited above, if it is +/- five percent from its long-term average then your asset allocation could be wildly off the mark. For example, if your strategic asset allocation model assumed a ten percent annualized return for U.S. equities as of the year 2000, then you ended up being off by roughly ten percent for that input over the past decade (and your portfolio suffered greatly as a result). And yet oddly enough, confidence in strategic asset allocation remains high throughout much of the industry. Is having confidence in strategic asset allocation really all that much different than having confidence in your ability to correctly predict the next four presidents of the United States?

The alternative to strategic asset allocation is tactical asset allocation which attempts to better deal with the dynamic nature of the financial markets. Furthermore, our preferred method of tactical asset allocation is strict adherence to a relative strength model. Research demonstrates that relative strength can be an effective method of asset allocation over time. It is a pragmatist’s dream because it keeps a very open mind about which asset classes are going to deliver the best returns going forward and simply keeps the portfolio fresh with those asset classes that have the best intermediate-term relative strength.

When clients have a choice between entrusting their retirement savings to the promise of an elegant, but unproven, theory of strategic asset allocation or to go with pragmatism and long-term research, I believe they will choose the latter.