A Warning About Warnings

"You'll shoot your eye out!" Think back to when you were a kid. Remember you used hear all these warnings? Looking back now, some of those just seem ridiculous. You know, like "stop jumping on the bed" or "sitting too close to the TV will ruin your eyes." Or silly ones like "don't run into traffic."

The truth is, warnings exist as guidelines to help keep us on the path, but they often have ulterior motives. For instance, the military once requested that the FDA do a study on drug effectiveness. The FDA studied 100 prescription and over-the-counter drugs and found that over 90% of them were still effective and safe 15 years after the expiration date. That's right: the dates don't actually indicate if a drug is less effective once it's "expired."

The obvious undertone is that the manufacturers sell way more drugs if you're fooled into believing that they are no longer safe and effective. It brings to mind the ongoing argument between my wife and I about food expiration dates. Either way – keep 'em shopping is the true mantra of expiration dates.

I know I sound like a broken record when I say that the media is fixated on delivering bad news to you. The fact is, unicorns and rainbows don't get people to react and interact – they find fear and loathing to be more effective.

So, that brings me to what I am noticing now. We have been talking about how the market has been overbought since the first day our ratio popped above 80% as of the close Feb. 6. It has now been 21 trading days with the MAP-IT ratio over 80%. This is when the number of unusual institutional buy signals divided by buy + sell signals is more than 80% on a 25-day moving average.

Yet the ratio hit its peak on March 1 at a level of 92.6%. This means that almost 93% of all signals in the trailing five weeks were buys! As the ratio has been declining, it doesn't convey what is really happening every day – the buying has been slowing, and the selling has been picking up. Look at this table here:

The green column represents the daily buy signals, while the red is the daily sells. The right-most column is the 25-Day Moving Average (25DMA) of this daily measure. As you can see, it has been dropping over the past few days. But when you look to the column to its left – the one-day ratio – you can see how swiftly the deceleration of buying has been in the past four trading days.

Observing this led to the question: how does this overbought period compare with prior periods in our 30-year history? There were 28 prior times that the market stayed overbought for 19 or more trading days. (We posed this question after the 19th trading day of being overbought.) Given the power of our oversold signal timing market bottoms, we were hopeful of finding the overbought signal to be promising in identifying market tops. What we found was interesting.

Long story short – we found the peaks of the 25DMA ratio in each instance and then looked for when the ratio declined for at least five days afterwards. We took the average one- through eight-week returns afterwards. The theory was that, when the ratio started falling from the peak reading, we would see lower market prices ahead. Over the 28 instances in 30 years that fit these criteria, what we actually saw was less high prices ahead. This summarizes the average forward one- to eight-week returns from the ratio dropping from its peak (after five days):

Now those returns are the average returns over all those 28 instances. But what we see now are lower prices. Look at the market and sector performances since the ratio peaked on March 1:

Now that's a pullback – no question. On a sector basis, we have seen a rotation from more growth sectors into defensive. Utilities and real estate saw a move higher (as of this writing), while the remaining nine sectors were down for the week thus far. Health care felt the biggest blow, down nearly 4%. Energy, financials, discretionary and info tech were the other worst performers. This is the first week of lower prices in most sectors since the end of 2018.

As you can see below, buying slowed significantly, while unusual institutional (UI) sell signals picked up slightly.

A giveback is a healthy thing, and now the worriers who fretted about going up too far too fast can begin worrying about going down. Such is the curse of the eternally negative thinker. Trade war fears are back in the headlines. Nonfarm payrolls showed a pretty poor number this Friday, and the political headlines are irksome. The point is, the news cycle has shifted negative gain, and we can see the algorithmic selling starting to react to it, so I expect some bumps ahead.

However, given the look back we did over all prior overbought periods, the expectation has to be tempered to expect less high prices. In other words, the bull market is still alive. The sales and earnings cycle is nearly complete, with 96% of companies reporting earnings: 70% beat earnings estimates, and 60% beat sales expectations. Those numbers are not discouraging – not in the least. So, as we exit this earnings cycle and hit a pocket of less market-intense news, expect a near-term market burp followed by a resumption of bullish action.

The thing about warnings is they are just that: warnings. Take heed in the warnings that come from pessimistic thinkers. Data are strong, but there will always be storm clouds on the horizon. Norman Cousins said, "History is a vast early warning system."

We (Mapsignals) continue to be bullish on U.S. equities in the long term, but we see a risk of a near-term giveback if institutional selling increases. Selling has started to pick up mildly compared to prior weeks so far this year. Any pullback is a buying opportunity.

Disclosure: The author holds no positions in any stocks mentioned at the time of publication.

Investment Research Disclaimer