Market Timing With The RPF Model: Spot Opportunities, Stay The Course, Avoid Mistakes

| About: SPDR S&P (SPY)


Proceed with caution. Market timing is risky, but the RPF Model can help spot rare opportunities.

Spot the truly rare event. The bubbles and opportunities in 1987, 2000 and 2009 – 2013 could be seen with the RPF Model.

Do no harm. Knowing whether the market is valued fairly can provide resolve to stay the course during troubled times. Often the best action is no action.

Market timing is risky and often dangerous to your portfolio's health. At its worse, it causes investors to buy high and sell low by jumping back into a hot market or abandoning the market after a fall. Periods of significant overvaluation (bubbles) and significant buying opportunities are rare. But these rare opportunities can be spotted.

I introduced the Risk Premium Factor Model in the Journal of Applied Corporate Finance in early 2010 after tracking it for more than a decade and after blogging about it in 2009. I began writing about it in Seeking Alpha in September 2010 and Wiley published my book in 2011. The RPF Model is simple and can enhance understanding for investors by demystifying the factors that drive valuation. In short the model says that price is driven by earnings and long term interest rates:

Intrinsic Value of the S&P 500 Index =

S&P Operating Earnings / (Long Term Treasury Yield x 1.48 - 0.6%)

The model shows that equity prices (NYSEARCA:SPY) move inverse to yield. In other words, when interest rates rise, equities must provide a higher proportionate return to remain competitive. This lowers P/E or increases its inverse, the earnings yield. In the equation, 1.48 is the Risk Premium Factor and 0.6% is the difference between long-term expected growth and real interest rates. I've written about the model many times, so rather than repeat my entire overview of the model, you can read about it here on Seeking Alpha.

The most frequent question that I get from friends, colleagues and readers is: "How can I use it to make money?" The purpose of this article is to summarize a few ways to use the model to invest by illustrating notable examples from the past 30 years.

As a Media, Friends & Family Filter

One use of the model is to help sift through the media chatter. In truth, most reporting is good if read carefully. The danger comes from those alarmists who espouse dire predictions with persuasive passion may substitute for facts as they make their case. This often includes friends, family and colleagues as well. In short, I suggest two things: 1) If the source does not support their argument with fundamentals like earnings, interest rates, economic fundamentals, then ignore them; 2) Read beyond the headlines. Headlines are often written to attract readers but may not be completely consistent with the conclusions or facts in an article. Take the time to understand the nuanced arguments.

1987 Crash

October 19, 1987 is referred to as Black Monday. The Dow fell a record 508 points or 22.6 percent. A crash must be unexpected, otherwise the market would have already reacted, but 1987 crash it could have been spotted.

Interest rates began climbing in March 1987, rising from 7.25 percent in March to 9.25 percent in October, driving down the predicted P/E and the predicted level of the S&P 500. Yet despite earnings remaining flat, the market grew by 12 percent from February to September (a total of 25% from December), creating a clear bubble shown below. With the market crash in October, the predicted and actual fell back into parity.

The model provided more than six months warning that valuations were out of line but having the resolve to act is difficult. If you were to have exited the equity markets or bought puts in May 1987, you would have had to endure the market's continued rise by another 13% by September - perhaps taking comfort that the model showed you were still following the correct strategy.

2000 Dot-Com Bubble

As shown in the chart below, the 2000 Dot-com bubble was long and persistent. Prior to the S&P 500 beginning its decline, 10-year treasuries rose from 4.6 percent in November 1998 to 6.2 percent in March 2000. This can be seen as the decline is predicted during that period. Like the 1987 crash, events played out over many months with the RPF Model providing plenty of time to react.

Still, deciding when to execute and sticking with that decision would have required tremendous resolve. What's more, the Fed pushed long-term rates back down below 5 percent, causing predicted value of the S&P 500 to increase and remain close to parity with actual. Getting the timing right would have required a good forward earnings forecast, since declining earnings continued to drag the market down.

Note that1987 looks tiny by comparison but percentage-wise was of the same scale as 2000.

2009 Meltdown

Unlike the other two major market crashes, the predicted levels for the S&P 500 tracked actual very closely. It is critical to understand that the meltdown was driven by the decline in operating earnings, falling 57 percent from 91.47 in July 2007 to 39.61 in September 2009. Since the RPF Model values the market based on trailing twelve month operating earnings, it was fairly valued from a historical perspective in 2009. However, real power of the model comes from using it to sensitivity test key variables.

The question was "Will earnings recover?" Given that earnings should almost certainly recover (they always have) and the market was fairly valued based on the extremely low level of earnings, it was relatively simple to conclude that 2009 was a tremendous buying opportunity for those with liquidity.

Likewise, it was a terrible time to sell. Yet too many people panicked, sold, and locked in their losses. In 2009, analysis using the RPF Model would have suggested, buy if you can, but certainly don't sell. Many of those that sold, did so out of fear. Unlike liquidity driven selling, fear driven selling is avoidable. If you see the value of your portfolio fall by 50% but don't understand what drives the market, it is reasonable to succumb to that fear. Many did. If they understood or their advisors could have convinced them that the market would recover with earnings, they would have avoided the loss.

In many ways, this was an easier situation to act on than the bubbles in 1987 and 2000, since the most important action was to hold tight and not lock in losses. The best action is often no action.

The Long Recovery

The S&P 500 Index began a long recovery beginning in February 2009. I wrote my first article for Seeking Alpha, in September 2010, suggesting that the market was 30% undervalued. The market continued to be undervalued until May 2013 and is now fairly valued.

It is important to note that I have been using a normalized long-term interest rate of 4.5% (2% real plus 2.5% inflation) to adjust for the Federal Reserve's artificially depressing long-term rates by keeping short-term rates near zero since August 2011. The RPF Model can also be used to calculate the implied yield on long-term treasuries. That is, based on current earnings, what yield brings the market to equilibrium between predicted and actual. The implied yield on July 3, 2014 was 4.21%.

The chart below shows the implied yield on the 10-Year Treasury from 1986 - present.

It clearly shows the effectiveness of Federal Reserve actions aimed to keep long-term rates low by keeping short-term rates near zero and QE policy.

The chart shows that the implied rate is well above actual today. Is this a potential trading strategy?

The answer is yes but it's risky. Rates have been out of equilibrium since 2010. Shorting anything is risky.

Remember the quote often attributed to John Maynard Keynes: "Markets can remain irrational longer than you can remain solvent."

A less risky strategy today is to continue avoiding long duration bonds, since when long-term rates rise, they will lose value.

The fact that you might not trade bonds along this strategy does not mean the information has no value. Significantly, this indicates that the market has already factored in higher rates and has ability to absorb them.

Also take note, in 1987 and 2000, a bold strategy in both those periods would have been to sell equities and buy bonds, since the implied yields were well below actual, as a way to hedge the interest rate risk implied by the model.

Forward Earnings and Sensitivity Analysis

As suggested above, perhaps the most frequent use of the RPF Model is to look at the sensitivity of valuation to changes in different variables. For example, the forward predicted value of the S&P 500 is shown below based on S&P's estimated operating earnings for companies in the index.

As shown above, with a 4.5 percent normalized 10-year treasury yield, the S&P is overvalued by about 7%. Considering the forward earnings forecast and implied yield on treasuries (discussed above) of 4.21%, valuations do not appear to be calling for selling.

You can also test sensitivities on your own. For example, if you believe that 4.5 percent is unrealistically high for a normalized long-term rate, you could calculate that the predicted value of the S&P at 4.0 percent is 2,100. You can link to free iPhone and Android apps from my profile page to perform these calculation on your own. If you think S&P's estimates are too high, you can easily test the impact of different levels of earnings on price.

While the RPF Model can help spot those rare events, solid market timing opportunities are rare. The many ups, downs or small corrections that happen with great frequency are not those rare opportunities. The RPF Model more frequently serves to hold the course maintaining comfort with current valuations and test sensitivity to the variables that drive the market.

Disclosure: The author is long SPY. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: And short long-term treasuries