Identifying The Tapestry Of A Market Top

• The importance of identifying the direction of the primary trend
• The Shiller P/E
• The four-way test

The importance of identifying the primary trend

Sometimes it’s a good idea to step back from the everyday trading of daily charts and short-term activity and look at the big picture. If you are a long-term investor that obviously makes complete sense. However, short-term traders, with as little as a 2-3-week time horizon, also need to pay attention to the big trend.  The reason is that false short-term technical signals almost always develop when triggered in the opposite direction to the prevailing primary trend. In a primary bull market when the incoming tide lifts all boats, it’s easy to be a hero. But for traders, navigating the treacherous waters of a bear market, it is a totally different matter entirely. In a bear market, oversold readings reach greater extremes and remain there for much longer periods, for example. Contra-trend short-term rallies often fall short of expectations.

Having some understanding on the direction of the main trend is therefore of importance to everyone, with the possible exception of swing and intraday traders. With that in mind, it makes sense to make a checklist of what the next top might look like, and whether recent action could qualify. I call it a “tapestry” because each top is different, which means that we cannot rely on one or two indicators, but need a broader consensus. If too many come under consideration the process becomes complicated, too few and we lose some accuracy.

The Shiller P/E

Chart 1 compares the Shiller P/E ratio to the inflation adjusted S&P Composite. I think of a P/E as more a reflection of confidence than as a valuation measure. For example, people were obviously pretty optimistic in 1929, otherwise why would they be willing to pay $40 or so for $1 worth of earnings. Similarly, at the 1982 low, investors must have been extremely negative, as that same dollar’s worth of earnings could be purchased for less than $7, a bargain.

The pink shading shows that this indicator has rarely been at a higher level, only once in fact. That was for a few months prior to the 2000 tech bubble peak.  The 1929 experience delivered a reading slightly below the current one. Both instances though, were followed by a secular bear market for inflation adjusted US equities.

I am sure that those high levels in 1929 and 2000 contributed to the severity of the ensuing declines. However, the P/E was much lower at the 1972 and 2007 peaks, both of which were followed by price drops greater than 50%. Clearly identifying a top purely by observing the level of the P/E is not the way to go. A better approach is to identify a reversal in its trend.

Chart 1

Chart 2 attempts to do that by flagging peaks in the KST. Solid lines show signals that were associated with a bear market, though a couple, like 1962 were late and another, 2007 was unduly early. The dashed arrows tell us where this technique failed because there was no bear market.

Chart 2

Chart 3 is a short-term chart that only goes back 18-years! Here we can see that the KST is still rising and the P/E itself (32.77) is above its 12-month MA. No sign of a break here. The chart also shows that trend line breaks in the P/E itself have often signalled reversals in the price. In that respect the trend is still positive. Updated data and charts for the Shiller P/E can be followed here.

Chart 3

The four-way test

Chart 4 shows us what the previous three important NYSE Composite ($NYA) market tops looked like. Most of them met the four-way test. That test comprises, a KST sell signal, a negative 12-month MA crossover, completion of a trading range top and an up trend line penetration. By definition these indicators do not turn bearish at the actual peak, but since tops take some time to complete the signal usually lags. When an extended bear market unfolds though, this delay is usually a price worth paying. Let’s compare these three previous experiences with the present situation.

The most objective is a KST reversal and sell signal. I say the most objective, because there is usually little argument when this smoothed long-term momentum reverses trend. It’s not always accurate, as we can see from the early 2005 false negative, but usually when it reverses it’s a signal that lasts for many months.

The negative 12-month MA crossover is another objective signal, but less so, because it is more subject to whipsaws than the KST.

A trading range can be quite extensive, as it was at the 2000 top, or simply take the form of a declining peak and trough, as was the case in 2007. That was a tricky example, because the Index dropped below its 12-month MA and trendline at approximately the same time as the KST went negative. However, it wasn’t until the Index broke down from the blue ellipse that a declining peak and trough was confirmed. Note the absence of a trading range distinguished the three false KST sell signals in 1998,2004 and 2011.

The final requirement, involving a trendline violation, is the most subjective of all. Sometimes it is fairly obvious, such as the 2003-07 line, as that had been touched or approached on numerous occasions and was fairly lengthy. The clincher, lay in the fact that the line also intersected with the MA. When both were violated it offered a major signal of support penetration. A line and the MA also intersected at the 2015 mini-top. The red arrows on both Chart 4 and Chart 5, the latter of which features the total market cap in the form of the Wilshire 5000, signify when all four conditions were met.

As we begin the month of April, the Index is close to its 12-month MA and the KST has begun to stabilize, but neither has gone bearish. The trendline is not the greatest in the world, because it has only been touched twice so far. It nevertheless remains intact. Finally, there is no trading range development, at this point just a nasty drop and a partial recovery. Certainly no indication of a series of declining peaks and troughs.

Chart 4

Chart 5

None of this means that future price action will not trigger some bear signals, but as it stands right now the trend of the Shiller P/E is up, and none of the four-way test requirements for either the $NYA or the Wilshire 5,000 have been met.  Should we get a strong retracement rally from here, subsequently followed by new lows on a monthly closing basis, that would be another matter entirely, but why worry about something that has not yet happened?

One final note, please remember, that these charts are based on month-end closing prices, so all plots until those based on April 30 are really "estimates".

Good luck and good charting,
Martin J. Pring

The views expressed in this article are those of the author and do not necessarily reflect the position or opinion of Pring Turner Capital Group of Walnut Creek or its affiliates.