Tuesday, May 25, 2010

Evaluating Market Commentary

Despite the efforts of a growing number of people who advocate a passive approach to investing, the bulk of investing commentary is focused on active trading. Will the problems in Greece cause North American stocks to rise or fall? Will the rising US debt start to drive up interest rates?

I don’t know the answers to these questions, and based on the track record of various commentators, it seems that they don’t know either. So why do they persist in making short-term predictions? The answer is that this is much easier than coming up with something fresh to say about passive investing. We can always come up with something to say about the effect of some recent event on financial markets, even if the commentary is useless gibberish.

Does this mean that we should ignore all commentary about the immediate future of financial markets? Well, ignoring all of it isn’t a bad approach. I do listen sometimes, but I tend to be very critical. I won’t believe anything just because it sounds plausible. I look for certain specific things as a signal that the commentary may be worth considering.

The main things I look for are long-term evidence along with a plausible explanation for the long-term trend. You may ask how can there be long term evidence for short-term predictions. I think this is best illustrated with some examples:

1. “With interest rates so low, they are bound to rise. When they rise, stocks will get hammered. It’s time to get out of stocks.”

To evaluate this advice, let’s break it down. Over the long term (say 50 years or more), when interest rates have been at current levels, how have stocks performed over the next year? Since I just made up the quote above, I don’t know the answer, but I just assume the quote is useless until I see the evidence. Even then, the evidence may not be statistically significant. I would then want some plausible explanation of why this trend would persist even in the face of all professional traders knowing about the trend. This is a tough test.

Let’s try one that I do believe:

2. “On average, stocks have outperformed bonds for a very long time and there is good reason to believe that this will persist because investors demand a risk premium.”

This statement satisfies my test. It points to long-term evidence and explains why the situation is likely to persist even though everyone knows about it.

Let’s try one from Jim Cramer:

3. “[Monsanto] is just one bad stock. It’s gone down. We’ve got the Justice Department taking shots at it. We’ve got Dupont taking shots at it ...”

The implication seems to be that companies whose stock has gone down recently and who are in trouble with the law and are in legal fights with competitors are bad investments. Until someone does a long-term study, I’m sceptical. Further, if this trend exists, is there any reason to think it will continue once the investment community learns about it?

Keep in mind that Cramer may very well be right about the short-term prospects for this stock. However, I’m looking for evidence of why he should be right. Without this evidence, I’m assuming the odds are close to 50/50 and I’ll ignore Cramer.

In summary, I look for two things:

1. Long-term evidence that a particular type of investment analysis works.
2. A plausible explanation of why the analysis should continue to work even when the investment community knows about it.

I rarely see this combination in financial market commentary.

6 comments:

  1. I agree with you that world events can have an impact on the short value of a stock and that news agencies will focus on that as headline. We, as investors, need to focus on the value and filter the noise (like Cramer).

    I receive a couple of analyst documents performed at the banks that have daily and weekly report and I find those to be very grounded.

    ReplyDelete
  2. @Passive Income Earner: I've read many analyst reports on stocks and almost all of them sound like they were written by a smart person with a lot of knowledge about the company. However, none has ever come close to meeting the test I described in this post. There is no long-term evidence that this sort of analysis can generate excess returns. I'm still willing to look at analyst reports occasionally on the off chance that I'll actually encounter one that passes my test, but none have so far.

    ReplyDelete
  3. @Michael: Thanks for the comment. I agree that they are not very long term focus and that we can't confirm that it generates long term return (let alone excess return). Someone I know is actually investing closely to one of the Canadian recommended stocks from one of the reports. I have just started looking at them over the past year as one source of information. I was actually curious if there was evidence for their 1 year target accuracy.

    ReplyDelete
  4. Before you assume Cramer's opinions are 50/50, take a look at his track record. 50/50 may be generous.

    ReplyDelete
  5. @Passive Income Earner: Until I see evidence, I'm sceptical that there is any 1-year target accuracy.

    @Patrick: Thanks for the pointer. It would certainly be nice if Cramer were right significantly less than 50% of the time. We would have a shorting opportunity. I suspect that even if the 47% figure at the link you give is statitically significant, there probably isn't any way to profit from the slight 3% edge.

    This reminds me of the story in a kids' puzzle book I read as a child. A wealthy woman always brings a wino to the casino and gives him money to gamble. He bets on black or red at the roulette table and always loses his money very fast. Why does the wealthy woman keep giving him money to waste? Of course, the answer is that the wealthy woman became wealthy by betting the opposite way as the wino with much larger bets. Sadly, Cramer doesn't seem to be a suitable wino.

    ReplyDelete
  6. Great post Michael. Filtering the noise is key to successful investing.

    ReplyDelete