Active investment managers seek to improve rates of return and/or lower risk in investing. These are lofty goals. Over time, some managers are successful in achieving those goals. Having been the client of several advisors over the years I can speak from experience that having the goals of improved rate of return and/or lower risk is admirable, but hard to achieve.
Backtests are flawed
When I first began my journey into active management, I believed that a wonderful looking backtest for a long period of time was to be admired and that somehow that backtest implied that the future would look somewhat like the past. Not true, the markets change character over time, but my investment models were static adhering to past performance patterns. Each time I created a backtest and executed its paraments in real time, it eventually broke down. I became so disenchanted with backtests that I believed they were useless – at least for the way I build active investment models. I still believe that.
I found that backtests in general were over-optimized. A backtest is basically an attempt to prove an investment theory by fitting a mathematical formula to a specific set of historical data. The market is dynamic and keeps changing, and the future is truly not like the past. It seemed that the longer the period of data I used, the more flawed the output was going forward. I began to notice that older historical data was irrelevant to current market conditions for the way I was developing my models.
What is the answer?
In his book, Quantitative Trading Systems, Dr. Howard Bandy, retired professor of applied mathematics, describes a method he believes is the optimal way to design active management strategies, so they have the best chance of working effectively into the future. He also discredits the value of backtests. Instead, he proposes the use of a simulated methodology to help validate one’s thesis. After becoming frustrated with the use of backtests, I decided to give Dr. Bandy’s methodology a try.
He calls this process of development “Walk Forward Out of Sample Testing”. Time does not permit me to describe it, but I highly recommend his book if you are interested in developing active management strategies. After adopting and testing the walk forward methodology, it seemed that Dr. Bandy’s method was what I was looking for.
How to Measure an Active Manager
Global Investment Performance Standards (GIPS) is the highest standard for investment performance verification. The use of GIPS is fairly expensive but is considered highly credible in the industry.
As a less expensive alternative, Theta Research offers a platform that verifies an active manager’s performance by publishing their performance output on the Theta website. They offer metrics to help quantify output and compare that to other active managers. I recommend the Theta site for this kind of analysis. Of course, not all active managers publish their data on the Theta site, but if a manager has historical and verifiable data, this is always an option.
Live Accounts
There is nothing like the real thing – performance, that is. Performance from live accounts tells a story of the true client experience, measured by all the pertinent metrics as it unfolded in real time. The longer the real time performance periods the better.
Does active investment management make sense?
Active management provides a way for an investment manager to potentially lower risk during periods of high volatility– and conversely, a way to potentially participate in the market when the probability of returns is more favorable. This does not mean that active management is always preferred to a more strategic buy and hold allocation. Each has its place in a well-rounded and tax efficient client portfolio. A manager’s performance, as measured in live accounts, should make the case for an actively managed model’s suitability for a given client. If the active model adds value to the overall client experience, then yes, it makes perfect sense.
What about a buy and hold portfolio?
Nearly every buy and hold portfolio that I have been shown is not something in which I would invest. They typically consist of a well-diversified group of issues that consistently underperform the benchmark. Just as with active management, buy and hold portfolios should be strategic in their composition, and management measured over time on a risk/reward basis. If they cannot improve upon the benchmark on a risk adjusted basis, why bother? Just buy a surrogate ETF which tracks the benchmark.
Conclusion
Actively managed and strategic buy and hold trading models both have their place serving the client. Having a track record of live performance using a sound methodology is the key to confidence in potential outcomes. There are many active investment managers from which to choose. Do your due diligence on the active and strategic models you encounter before allowing them to find a place in you or your client’s portfolio. In my opinion, there are a few gems out there, but you need to dig to find them.
Ben Reppond is CEO and Investment Manager of Reppond Investments, Inc., located in the Flathead Valley of Montana. Reppond Investments, Inc. is a registered investment adviser. We may not transact business in any state where we are not appropriately registered, excluded or exempted from registration. Individual responses to persons that involve either the effecting of transactions in securities or the rendering of personalized investment advice for compensation will not be made without registration or exemption. You may contact Ben at ben@reppondinvestments.com.
Investment Advisory Services offered through Reppond Investments, Inc.