Update on Our Formularized Asset Allocation Processes

Share this article

The majority of Mission’s clients have chosen to have us employ one of the two asset allocation processes that we introduced in 2012. Both processes have gotten off to productive starts. In the ten months since the first purchases last December, assets invested with the processes have produced low double-digit returns, although just a small portion of that was earned in the last few months. Assets have been exposed to equity risk only 43% of the time since that December start. Those who chose to employ one of the processes in the early months have experienced the strongest returns.

Ten months of real time experience have led to quite a number of questions, which indicate a need to repeat several of the points we made in our introductory meetings and to clarify the reasons behind some process signals. I will address some of those questions here, and we encourage you to contact your portfolio administrator with other questions you may have.

A quick refresher. The 2-Mode Equity ETF model attempts to own stocks when equity markets are advancing and to keep portfolios safely in cash equivalents when markets are declining. The 3-Mode Equity ETF model attempts to own stocks in the most attractive market environments, to keep portfolios safely in cash equivalents in uncertain market environments, and to short stocks in the weakest market environments. While there are some situations in which we would employ individual stocks, in most instances we use highly liquid exchange traded funds (ETFs), which move in step with major market indexes.

We began to search for an alternate approach to equity investment a few years ago, when the extraordinary actions of the Federal Reserve markedly changed conditions in our domestic securities markets. Our deep discount, value-based equity selection process that had produced outstanding equity returns for more than 20 years suddenly identified very few purchase candidates. At times, not one was coming through the identical screens that had formerly produced many dozens of attractive buyable stocks. Market conditions had clearly changed.

Last year we finally identified a multi-dozen collection of economic and market measures that could be combined into a formula that had outperformed the S&P 500 for more than three decades. Most importantly, the formula had done so with fewer and smaller losses than the broad market index had experienced. While successful past performance can never point with certainty toward similar future performance, we have a high degree of confidence in the two equity allocation processes for three reasons: 1) they have produced excellent returns with considerable consistency for nearly a third of a century; 2) the factors measured are numerous and diverse, leading us to expect continued success even if some of the measures become less well correlated in the future with market directions; and 3) price movement and price momentum are heavily weighted in the formula, which is designed to keep us from overstaying in a long or short position that is losing money.

Let me address a few questions. One client who started with the processes several months into 2013 had the unfortunate experience of two small losses marking his first exposure. He asked whether we were reevaluating our support for this approach. At each of our introductory meetings we have attempted to point out the worst periods in the nearly 33 years of back-testing. The 2-Mode and 3-Mode processes have lost money in 16% and 22% respectively of the calendar quarters over that time span. The S&P 500, however, has lost money in 30% of those quarters. Losses can’t be avoided completely. The worst performance quarters for the 2- and 3-Mode processes have been -10.5% and -8.6% respectively compared to -22.5% for the S&P 500. The worst 12-month performances for the processes were -7.9% and -11.7% respectively compared to -43.3% for the S&P 500. Over the years there have even been extended periods in which the market has done considerably better than either of the processes. In the long run, however, both processes have outperformed the market far more often, leading to their long-term outperformance. Notwithstanding occasional missteps, we expect positive long-term performance.

When another client received his March quarter report, he noticed that the process was up more strongly than his full portfolio. I reminded him that we only apply the process to that portion of the portfolio that he wants invested in equities. While we have a few clients that have asked us to apply the process to their entire portfolios, most have restricted it to lesser percentages. Given the volatility of any equity approach, a reduced exposure is appropriate for most investors.

A client who employs the 3-Mode process asked whether we had made any money on the short side in the three-week decline from mid-September to early-October. We did not. In fact, in the first 10 months the 3-Mode process has not yet gone short, although historically it has been short about 20% of the time since the early 1980s. While some signals last a short period of time, in an extended uptrend it is not typical to see a bearish signal interposed. This year has seen a larger than normal number of signals, because the formula has vacillated between bullish and neutral. While prices have largely risen, justifying the bullish signals, significant uncertainty is apparent in several of the measured criteria, which has resulted in several trips back into the neutral zone. It is also important to recognize that the processes typically will not outperform the market in its strongest phases. The assets are not invested in stocks all the time. The 2- and 3-Mode processes have gained their advantage over the S&P over time by participating in equities about 60% and 40% of the time respectively, then avoiding most of the major declines (2-Mode) and profiting by shorting many of the extended declines (3-Mode).

As you experience these newly implemented processes over a longer period of time, we encourage your questions. We hope to provide a solid understanding of how the processes identify positive and negative market environments.