Blog : More Winter; Speculating Against Speculation

by Ed Zwirn on February 14th, 2014

Plus, How to use Google stats to bet on the stock market:

Thursday's retail sales report is one of those economic indicators that, depending on whom you listen to, is either an anomaly due to the unseasonably cold weather being experienced through much of the United States or a further indication of a genuine economic slowdown.

The consensus had expected sales to remain flat for January. Instead, in a shock registered by almost no one, the indicator actually fell 0.4% last month after declining a downwardly revised 0.1% (from +0.2%) in December. Although the automotive sector was the weakest point in January sales trends, sales excluding autos were flat after increasing a downwardly revised 0.3% (from 0.7%) in December. The consensus had expected these sales to increase 0.1%.

As I pointed out in an earlier penny stock investment blog, the "winter of our discontent" has been blamed for a host of weak economic readings, including the December and January jobs reports and the downturn seen over the same period for auto and truck sales.

And with each episode of biting cold and heavy snows, it is easy to believe that this explanation holds water (or maybe ice). Although the retail sales report is, like many other government economic releases, seasonally adjusted, these adjustments are based upon statistics derived from "average" seasonal changes, and so far this winter has been anything but.

On the one hand, sectors that are normally affected by weather conditions such as motor vehicle sales, down 2.1%, and restaurants, down 0.6%, saw significant pullbacks. On the other hand, sales were generally weaker across the board, and the fear is that this could signal that the spending spurt that occurred in December, which saw consumers dip into savings to make purchases, was a one-time event related to the holidays and not the start of a new trend.

In fact, core sales -- which exclude motor vehicle dealers, building materials and supply stores, and gasoline stations -- were down 0.3% in January after increasing a downwardly revised 0.1% (from 0.7%) in December. Also, sales at clothing stores fell 0.9% and demand at sporting goods and hobby stores declined 1.4%. Sales at grocery stores were up 0.4%, possibly a result of consumers stocking up from the winter conditions.

If the weather is to blame for this downturn, then spending growth should rebound in the near future. Usually, explanations like these fall flat, but I'm guessing that I'm not the only one to give credence to the weather excuse as I shovel the snow between my house and my car, whatever the groundhog may have to say about the duration of this winter.

Speculating Against Speculation

Here's one you can try at home.

It is a truism among contrarians that investor sentiment is a strong counter-indicator for stocks. The greater the interest on the part of individuals in speculative investments, the argument goes, the more ripe the stock market as a whole may be for a downturn.

Three researchers from the U.K.'s Bangor Business School set out to test this proposition. In so doing, they measured online queries as tabulated by the Google Search volume index and used queries for "penny stocks" as an indicator of the type of upsurge of speculative demand that would tend to lead to a short-term reversal.

Using the Google trend, O. ap Gwilyn, A. Kita and Q. Wang devised a trading strategy that would have the investor sell a stock index when speculative demand, as measured by the Google system, is high, and buy into or hold the index when this demand figure goes lower.

What they found that, using this strategy, they were able to generate annual excess returns (over a buy-and-hold strategy) of up to 20%.

"Individual investors are often perceived to have behavioral biases and trade on noise," they write. "If their aggregate demand is random, it should have no predictable and persistent influence on stock prices."

But during periods in which this erroneous demand is "unpredictable and systematic, noise trade limits the arbitrage and enables individual investors to earn higher expected returns than sophisticated investors by bearing more risk."

Not surprisingly, the efficacy of this trading strategy varies depending upon which stock market index you use. The Google factor is in fact much more of a factor when assessing the outlook for the NASDAQ Composite (which includes many "penny" stocks) than it is for the Dow Jones Industrial Average or the S&P 500.

While this paper highlights only one tool you can use to decide whether speculative interest in the stock market is reaching an unsustainable level (and, if so, what to do about it), it could prove to be a useful tool all the same. For centuries market analysts have recognized that excessive public interest in "hot" stocks could mean bad news, and have usually relied on their "gut" to assess this. Now that everything is searchable (including searches themselves) we may at last have a tool to measure whether this is the case.