Wall Street loves slogans. Everyone knows that correlation does not necessarily imply causation. Having uttered the magic words, most proceed to sin again.
My mission is to help investors to greater understanding and better financial results. Much of this comes from identifying the best sources – the true experts.
I am often frustrated in this quest because it is difficult to explain errors in conventional Street wisdom – what I call Wall Street Truthiness. It is so much easier and more profitable to garner page views by pandering to the preconceptions and worries of the readers.
At first glance, today's post might not seem to have "actionable investment advice." That conclusion would be a mistake, since the point I am making is at the foundation of the current debate over the role and impact of the Fed. It is going to take too long to get there for one post, so this is the start. If you stick with me you will see how sloppy thinking on this point can be very costly.
Background – Some Wise Advice
My favorite professor from my college days is still inspiring both students and alums. I pay serious attention to his observations, including this gem:
I'll bet you are well aware of the difference between correlation and causation. And yet, I'll also bet that if we followed you with a recording device, it would not be long before the warning bell I placed on the device that would clang every time you jumped from correlation to causation would soon be playing a discordant hymn to your human desire to tell make-sense stories where causation is packaged in a neat box with an orderly ribbon decorating it. Me too.
To perhaps reduce the frequency of our doing so, take a look at the critique of popular science writing, here, and click on the link to "strong rebuttal" in the 3rd paragraph
This link describes how an admitted plagiarizer embellished his writings. You might want to compare the following segment to the reactions of the modern doomsday gurus:
I think one of the clearest was Seth Mnookin's analysis of Lehrer's retelling of psychologist Leon Festinger's famous original story of "cognitive dissonance," based on Festinger's experience of infiltrating a doomsday cult in 1954. Of the moments after an expected civilization-destroying cataclysm failed to start, Festinger wrote, "Midnight had passed and nothing had happened ... But there was little to see in the reactions of the people in that room. There was no talking, no sound. People sat stock still, their faces seemingly frozen and expressionless." Lehrer narrated the same event as follows: "When the clock read 12:01 and there were still no aliens, the cultists began to worry. A few began to cry. The aliens had let them down." Do you see the difference? Lehrer's version is more dramatic: people worry, they cry, they feel let down. It's more human. Each one of these little errors or fabrications makes the story work a little bit better, makes it match our expectations more closely, and thus gives it greater influence on our beliefs.
When you read the recent writings of those who have been consistently wrong about the Fed, the economy, and the market, keep this in mind. These sources use symbols like training wheels, sugar high, bubbles, manipulation, and stall speed. This is not analysis. It is an effort to explain their own past errors and to deliver to their own following.
A Jarringly Different Example
It would be easy to pick (yet another) misleading statement from Zero Hedge or Santelli channeling ZH. Since that would be unproductive, let me take an example from a source I have often highlighted with great respect: Mark Hulbert. I admire Hulbert as an entrepreneur and a source of information not found in other places. He is someone who found a need for a service and created jobs by filling that need. Well done!
As much as I love Hulbert's work when he is analyzing and reporting on newsletters, I worry when he strays beyond his expertise. Here is a recent example. Let us start with the title:
Yet another reason to be scared
Commentary: Consumer confidence index is a contrarian indicator
This is certainly an attention grabber. The rebound in consumer confidence is seen as positive by most (including me). Let us look deeper. Is this warning justified?
The article suggests the following themes:
- Hulbert has done his own research, showing that the market does poorly after big jumps in consumer confidence;
- He suggests that consumer confidence trails the stock market, rather than predicting it;
- He points out that the biggest market moves upward came after confidence lows and the biggest declines after highs;
- He cites academic research that allegedly reaches the same conclusions.
I have worked with consumer confidence as an indicator in my own research on many different models. It is pretty good as a coincident indicator of economic activity, but it can be distorted by things like political events (the debt ceiling debate in 2011) and gasoline prices (2008 and the rebound last year). Most of the time it is a good read on employment. It has little to do with stocks (at least directly) since few people own stocks. This simple fact is lost on most pundits.
The stock market correlation is what we call a spurious relationship. My old classroom example was a guy in civilian clothes, standing at a corner, and waving to traffic to move at various times. He coordinated his moves with the changes in a traffic signal. The correlation was perfect. Causation was non-existent. Now suppose that the traffic signal was burned out and the guy was wearing a uniform while standing in the middle of the intersection. See the difference?
The current relevance is that many indicators move with the economy (the guy in street clothes on the corner). Most of the things that the punditry sees as correlated are simply responding to changes in economic growth. This is the technical definition of a spurious relationship.
My own research makes me deeply suspicious of Hulbert's findings. Please note that those attempting to find a link to consumer confidence can do three different things:
- Look at levels of confidence (the approach of the academic study he cites);
- Look at changes in confidence (which Hulbert says he did);
- Look at a second derivative, the acceleration of change (often selected by those unable to prove their point with methods 1 or 2).
I might be the only one who followed the link to the study cited by Hulbert. It does not reach the conclusion that he claims, and it is not even close. This was a 2002 paper, not a citation to a peer-reviewed journal. On Hulbert's key point, the authors write as follows:
The negative relationship between consumer confidence and subsequent stock returns is useful to investors even if it is too weak for robust tactical asset allocation is useful. The fact that low levels of consumer confidence predict high rather than low subsequent stock returns should reassure investors that consumer confidence and stock returns do not follow each other in an endless downward spiral. Indeed, when people lose confidence as consumers, they should regain it as investors.
Please note the "too weak" statement. They also produce a table of effects. The relationship cited has no substantive significance. This would be determined by looking at the R-squared results, showing the amount of variance explained. This is about 1%. You can get "statistical significance" with enough cases even when there is no substantive significance. This means that we can be confident that the relationship is not zero, but we may not conclude that it is important. In fact, the data show that it is probably not important, as the authors conclude.
Hulbert's errors in interpreting the academic study make me even more suspicious about his own findings. This suggests a request. As we all learned from the Rogoff and Reinhart incident, there is merit in peer review. Perhaps Hulbert will put his data and analysis online for others to consider. What can be the harm?
And Finally
And finally, it should be obvious that the biggest market drops come from times of high confidence, high economic growth, high stock prices, low financial stress --- etc. The biggest gains come from the opposite conditions. This has been a constant theme in my work. You get the biggest rewards when the wall of worry is highest.
Since none of these elements – growth, confidence, employment, and other worries – is currently at an extreme, we are not seeing the signs of a market top. Increases in consumer confidence continue to show coincident evidence of improving personal consumption and employment.