Is market timing merely a myth or is it doable with enough intellect, discipline and desire?
A lot of people say they can time the market and a few have been successful at it for periods of time. Probably the best site for a long-time practitioner of market-timing is Mojena. The site clearly reports out-of-sample returns of the market timing system since 1990. And the signals are free.
For a moment, let us assume that we find a market timing system oozing with the elusive secret sauce that Warren Buffett puts on his hamburgers. Why should you still exercise extreme caution?
Most market timing signals are trend-based. I dreamed a dream that the market would move in beautiful and slow-moving trends. In my fantasy, we would use lagging indicators such as moving averages to gently ride the trend up and side-step the crash. The reality is closer to someone repeatedly pulling the chair out from underneath just before you sit down. By the time you figure out what happened, its too late to react.
Signals are hard to follow. It is insanely hard to liquidate 100% of your stock positions. And costly. My guess is that only a small fraction of people who subscribe to market timing signals will act on them. Even if you do act, it needs to be insanely fast. And what if you are wrong and lose 15% – will you be as quick to pull the trigger the second time around?
Too few data points. Market timing signals are generally infrequent. At least they should be – who wants to go to 100% cash a dozen times a year? But with so few signals, your confidence goes down in the system. With only a handful of testable signals, maybe the data is curve-fit, maybe it’s a fluke and maybe the specific factors that caused a market melt-down will never happen again. You can analyze more signals by increasing the length of the test, but the further back you go the less relevant the signals are. What worked with tulipomania in the 1600’s will likely be of no use to you in today’s markets.
Having said all of that…do you still want to try your hand at market-timing? After signing up to Portfolio123 through this link, you will be able to access two of my custom-made market-timers.
This timer has three components. S&P 500 forecast earnings, S&P 500 price action and the unemployment trend. At least 2 of these 3 signals needs to be bullish to give a buy signal. If more than 1 of these is bearish, the signal is ‘go to cash’. This simple indicator is all in or all out.
The three criteria are:
- The 5-week moving average of the S&P 500 forecast earnings for the current year needs to be above the 21-week moving average. When analysts begin to broadly cut earnings forecasts, this often precedes a major downturn.
- The S&P 500 price should be above its 100-day moving average.
- The unemployment rate in the USA should be below the 6-month average.
The whipsaw markets have not been kind to this market timer over the past year. As of November 7th, 2016 the system is still in cash.
This next market-timer is still based on the previous 3 principles but it doesn’t go ‘all in’ or ‘all out’. This model buys 100 stocks based on analyst sentiment. When the market-timing indicator gives a bearish signal, it progressively sells positions. The longer a signal is bearish, the more cash the portfolio will raise. This will lessen the impact of whipsaws and false signals.
For this signal, simply look at how many positions it holds out of a possible 100. As of November 7th, it holds 20 stocks. This means you should only be 20% invested and be holding 80% cash. Ignore the actual stocks it holds – that is not important. Focus only on the number of stocks held out of 100.
In the chart below, the top graphic shows the model performance. The second window shows the model drawdown as compared to the market. The bottom window highlights what percentage the model is invested.
Should you try to time the market? That is a personal choice. With each passing year I become increasingly skeptical – and not just whether or not timing is possible. Even a good system will have lots of false signals. Assume there are 20 false signals, where you neither gain nor lose, for every 1 that saves you from a bear market. Those 20 signals cost you dearly in transaction fees and slippage. You could lose 30% of your portfolio value simply by moving in and out of stocks that frequently.
On the other hand, you could get lucky early on and the market timer could save your portfolio from a major crash. That one early rescue could be all you need to propel your portfolio into retirement greatness.