…according to a speaker at the recent IMCA conference. The speaker was Arun Muralidhar, an economist trained at the Sloan School of Management at the Massachusetts Institute of Technology. He is the former head of investment research for the World Bank, a former managing director at J.P. Morgan Investment Management and currently head of AlphaEngine Global Investment Solutions LLC.
What is his specific criticism? It has to do with the static nature of traditional portfolio theory. According to an article in Investment News:
The foundation of the CAPM investing model is to construct an optimal desired portfolio based on the investor’s objectives. As the market moves, the portfolio allocations are re-balanced to get it back to its optimal state. For example, if stock prices go up, a portfolio with 60% stocks and 40% bonds will become over-weighted in stocks. To prevent drift in the portfolio, the adviser re-balances by selling stocks and investing the proceeds in more bonds.
The benefit of this process is that it’s simple, transparent and easy to execute. However, a static re-balancing process does not work well in falling markets, Mr. Muralidhar said. The biggest risks faced by investors are large draw-downs in asset values, as it requires more substantial gains to recover the lost value. The re-balancing does nothing to prevent further losses when asset values plummet — as they did in the financial crisis.
“It’s like tieing the rudder on your ship in place and ignoring the winds and currents that you experience,” Mr. Muralidhar said.
How does he believe assets should be managed? Well, I’m feeling quite validated today. In fact, it sounds exactly like the systematic relative strength process we use.
The solution is for advisers to focus on using optimal investment strategies rather than maintaining optimal portfolios, he said. These strategies involve managing allocations more actively, based on market conditions. Mr. Muralidhar suggested advisers employ a “systematic management of assets using a rules-based technique” — “smart” investing.
Essentially, that involves an informed re-balancing of the portfolio toward the more-attractive assets and away from the less-attractive ones.
I added emphasis to the good parts. I think his focus on optimal investment strategies rather than optimal portfolios is quite insightful. We’ve long argued that the path to constructing better portfolios is to mix strategies, not just assets. As it turns out, relative strength and value are both excellent strategies—and they work extremely well in combination because the excess returns are uncorrelated. Unlike asset cross-correlations which can and do swing wildly up and down, strategy correlations are likely to be much more stable. The reason is simple: trend-following and trend reversion are opposites; stocks and bonds (or pick any asset pair) are not.
It’s nice to see an economist discussing the theoretical flaws in MPT that have been evident to thoughtful practitioners for years and years.
Posted by JP Lee 
