Our experience with the Ivy Portfolio (including the use of a 10-month moving average trend system) has been a tale of two extremes for the first 10 months of the year. The Yin has been marked by markets with fairly strong trends and no trades where the Yang has been marked by negative trades exacerbated by market movements relative to monthly trading rules.
Given the popularity of the mechanical system, which we view as an attractive way to provide maximum drawdown portfolio protection, we are providing our hands-on experience working with the trend system for those interested in managing their own portfolios this way. [For reference, we use a diversified portfolio that would be benchmarked against a global allocation fund.]
The Yin experience so far in 2012 has centered on investments in fixed income securities, domestic equities, and domestic REITs. As expected, lower-volatile fixed income classes and real-return strategies have been relatively stable. However, domestic equities and domestic REITs have also been above their respective 10-Month Moving Average (10MMA) for most of the year.
The strong run up by domestic equities in the spring, including indexes such as large cap growth stocks (iShares Russell 1000: IWF) which closed 11.5% above its 10MMA in March, helped buffer the May market pullback. The Russell 1000 Growth ETF ended May at just 2.8% above its 10MMA. (The long-run average is 6.9% when above trend.)
While many of the non-domestic equities have yet to reach their 2011 peaks, domestic equities continue to outperform and reach new highs on a total return basis.
Figure 1: Russell 1000 Growth (IWF)
The bliss in domestic stocks and fixed income has not been realized elsewhere. International (EFA) and emerging market equities (EEM) as well as hard assets such as commodities (DBC) and precious metals (GLD) have struggled to maintain a consistent direction.
From an Ivy Portfolio perspective, many of these asset classes were late to go above trend and did not generate the buffer to stay above trend when the May downturn occurred. The result was several losing short-term trades. It is important to point the trade losses are expected and a common occurrence in the trend following portfolio. This year the negative trades have been exacerbated by the timing of the January and June run-ups and the May decline.
Emerging market equities (iShares Emerging Markets ETF: EEM) is a perfect example of a short-term losing trade. The initial trade was at approximately $43-$44 (closing price at 2/29/12) and the investment was closed at the end of May at approximately $37-$38.
Figure 2: Emerging Markets Index
More Yin, Less Yang
Our diversified Ivy model portfolio has underperformed our designated global asset allocation benchmark by approximately 200 basis points through October. The negative Yang effect had much to do with the impact. The challenge we have is how to reduce the cost of the portfolio insurance and improve the performance in up markets.
A Risk-on/Risk-off (RORO) model is an appealing way to compliment the basic trend following system. Our RORO model started the year Risk-on and lasted through May 4. It was Risk-off until mid-June. The RORO compliment allows several decisions to be made intra-month that preserves the mechanical nature of the 10MMA system including a limit on the number of tactical trades, which would likely increase if we shifted to a 200-Day Moving Average based mechanical system.
When the RORO model is Risk-on, it allows the system to make investments ahead of the end-of-month decision. This helped improve the trade profile in January and in June. When the RORO model is Risk-off it allows investments to be liquidated prior to the end-of-month decision. This was beneficial in May.
The addition of the complimentary rule set has put the modified Ivy Portfolio's performance more in line with the global asset allocation benchmark. Furthermore, it has done it without increasing the trades - just improved the trade profile. The improvement has been twofold: It decreased the size of the losing trade and it improved the entry point of new investments.
Our Ivy 10MMA tracking list of ETFs as of the end of October is long most major asset classes except for commodities and the dollar. At the end of October, the RORO model was Risk-Off (as of October 23rd). Therefore, our intra-month complimentary rule set opens the door to trades based on breaking the 10MMA. On a real-time basis, we have cut our exposure to commodities by 50%. With the recent market downturn including Apple Inc. (AAPL), we are watching the domestic large cap growth space.
The market downturn in May could have been worse based on the European Monetary Union's credit crises and the unknown impact of a 'Grexit'. The reality is the market didn't go into an extended decline similar to last fall and the negative trades were therefore expensive insurance. Unfortunately, investors reviewing historical returns won't recall the safety built into the results. That is the challenge investors face when using a mechanical trading system (that is not always intuitive in real-time) despite the well documented research regarding its benefits, which often takes the form of outperformance in bear markets.
To lessen the cost impact of the portfolio insurance and improve the returns, we do believe a complimentary mechanical system is attractive. It maintains the core principal of the Ivy Portfolio's mechanical system while improving the returns through opening trading windows based on market conditions. In practical terms - the RORO system provides a much needed level of intuition for our management style.
If there are other approaches being used to compliment the Ivy portfolio strategy, and worth sharing - please do so in the comment section.