Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.

We all want to build the best possible portfolios for our clients. But we do it in different ways. Some use classic modern portfolio theory (MPT). Some prefer factor tilts. Others try to time the market. Whatever your approach, let me make a suggestion: Keep it simple.

Keeping it simple improves performance no matter what type of investment process you use. Our industry is in love with complexity, but complexity is costly to clients.

Here are some reasons why simple is good and some guidelines to help you stay simple.

The problem with optimization

MPT is a wonderful theory, but suffers from a fundamental problem. For it to work you need to know the future expected returns, volatilities and correlations for all the asset classes in your portfolio. Unfortunately, no one knows what they are. In addition to being mysterious, they are also ephemeral – they keep changing over time.

This is a problem for two reasons. Strict adherence to MPT causes us to build jumbo portfolios. We load up our portfolios with a laundry list of asset classes and sub-asset classes, many of which come along with high price tags. We think we are going to achieve diversification benefits that don’t materialize.

Our capital market assumptions don’t line up with what happens over the ensuing years. They are estimates (guesses) about the future that are often based on the past – a past that rarely repeats itself. Our assumptions are typically very long-term in nature. Even if they play out over 20, 50 or 100 years, they frequently don’t in a time frame that is meaningful to clients. Jeremy Siegel says the long-term real rate of return of the stock market is 7%, but it is unusual for that number to hold true for one-, three- or five-year time periods.

Meanwhile, we dissect the world into more asset classes, sub-asset classes and factors than we know what to do with. We bedeck our portfolios with these “diversifiers” like ornaments on a Christmas tree. Some of them are added solely for their theoretical diversification benefit, but have no long-term expected-return benefit. Each one has a cost and those costs add up.

Even Harry Markowitz, the father of MPT, knew better than to do this. He is reported to have allocated his retirement account at the RAND Corporation 50% to stocks and 50% to bonds, and left it at that. Was he also the father of elegantly simple portfolio theory?

The second problem is that once we’ve bought into the illusion that we have created an “optimal portfolio,” we can’t stop tweaking the poor thing, trying to reset it to its optimal mix. We revisit it weekly, monthly or quarterly trying to fine tune it like a prized race car. We can’t keep our hands off our perfect creation. Each tweak costs money. But does it add value?

The first step toward building elegantly simple portfolios is acknowledging the limitations of MPT. Do we really need all those ornaments on the tree? What would happen if we eliminated asset classes that have no long-term expected return, or only owned them when we had high conviction they were poised to provide a return benefit? Do we add value by owning separate funds to represent every nook in the style-box, or would a single all-cap fund do?

Then let’s acknowledge that no matter how hard we try, we can’t build a truly optimal portfolio, except in hindsight. If we admit that our diversified portfolios are not optimal, then we can be more relaxed in our fine tuning of them. What would happen if we let our portfolios drift a little more from their allocation targets? Probably nothing bad and we would save our client money each year.