Investing Knowledge

So far in this series on asset allocation I have focused on variants of strategic asset allocation; choosing a mix of assets to suit the time scale and risk tolerance of the investor. That mix is then held, or rebalanced periodically to the original percentages. 

Tactical asset allocation introduces an element of market timing, or forecasting, to enhance returns. In this respect it is similar to an all equity investor who eschews an index fund to create his own portfolio of shares. I reckon an element of market timing can help performance so at the end of the article I provide a simple tool for deciding when to increase or decrease the equity portion of a portfolio.

Tactical asset allocation has always had its high priests and followers. However the crash of 2008, when all assets bombed except cash and government bonds, led many investors to demand an element of tactical advice.

How does it work?

The investor who uses tactical asset allocation takes large or small over or under weighted positions within a traditional asset allocation framework. 

For example, in a 60% equity/40% bond portfolio, if an investor felt nervous about high equity valuations, he would trim back equity exposure to 40% and increase bond or cash exposure to 60%. From a personal standpoint, I don't know if equities are going to rise 30% over the next year or dive 30%. I do know my equity portfolio has increased by something north of 50% since March. I'm locking in some of that gain and I'm reducing the equity portion of my portfolio to below my usual holding.

An extreme, rare and successful example of tactical asset allocation can be seen in the management of Jupiter Financial Opportunities. Anticipating problems in the sector, the fund manager started to raise cash in 2007. By September 2008 his fund was 80% cash. By April 2009 he was back to 80% in equities.

Tactical asset allocation depends on identifying and exploiting market inefficiencies. Furthermore, these inefficiencies (for example over or under valuation or mispricing) must be recurrent so they can be identified and predicted. Trading systems can then developed to exploit them.

Sophisticated trading firms like Goldman Sachs (NYSE: GS - news) build extremely sophisticated models to assist in buy/sell decisions and they actually publish their tactical adjustments on their website. Most private investors are unable to build such models.

One approach you can use

I have therefore scoured the investment universe to see if I could find an approach to tactical asset allocation that was both useful and implementable by the average private investor. Not only did I find one, I'm going to share it with you. 

It's not perfect -- in fact it's nowhere near perfect -- but it is useful. Its originator is a guy called Mebane T Faber and it was published in the Journal of Wealth Management (2007, updated February 2009)

Faber set out to develop a quantitative market timing model to manage risk. The result is a risk-reduction signal that indicates when an investor should exit a risky asset in favour of government bonds. He developed it based on the US market but has proved the model on foreign shares, commodities and real estate.

The author has examined the timing model from 1973 to 2008 during which time it has delivered equity-like returns with bond-like volatility and draw down (peak to trough loss).

The system is simplicity itself:

1. Buy equities when the monthly index is greater than the 200-day moving average.

2. Sell and move to cash when the monthly index is below its 200-day moving average.

The following caveats, rules and assumptions are made:

1. All entry and exit prices are on the day of the signal, at the close. The model is only updated once a month on the last day of the month. (Price fluctuations during the rest of the month are ignored -- this is not a short term trading system.)

2. All data series are total return series including dividends, updated monthly.

3. In proving the system he estimated cash returns to be comparable with 90-day T bills.

The results of applying this system from 1900 to 2008 are in the table below.

It's important to note that these figures exclude taxes and trading costs. That said, the system would have kept the investor invested 70% of the time, which will mitigate their impact.

So any market timers out there might want to give it a go -- or at least set up a monitoring system.