Understanding How to Analyze Market Metrics

Tuesday, July 22, 2008 | 11:58 AM

Longtime readers of mine -- going back to the email days 10 years ago -- will note that at various times I have been bullish or bearish (or both, depending upon different sectors). Regardless of our investment posture, there is always a good faith effort to maintain an analytical rigor and intellectual discipline in our investment approach.

This means sometimes discarding attractive theories, or embracing counter-intuitive ones. It also means questioning your assumptions, stress testing your theories, and whenever possible, having your ideas "peer-reviewed" to identify weaknesses or possible theoretical holes. Blogging has been very helpful in that regard.

That approach also means reading a variety of different viewpoints, theories, and people -- most especially those who I may be in disagreement with; It forces a rethink of current principles, and keeps the reasoning process sharp.

Which brings me to a column on CNBC.com today. In it, Vince Farrell mentions towards the end:

"Jason Trennert, my oft quoted strategist, calculates that dividends on stocks exceed the yield on 3-month Treasury bills. That has happened two other times in the last 50 years and both coincided with major market bottoms."

That seems to be fairly reasonable on the surface . . . but after thinking about it for a few minutes, it struck me as potentially having some hair on it. 

A little digging, and the flaws in the reasoning surfaced:

Start with where Real (not nominal) Rates are today: They are currently negative at 3.0%. That makes the 3 Month T-bill rather artificially low. That becomes a second variable worth reviewing: Does this buy metric -- SPX dividends > the yield on 3-month Treasury bills -- work when real interest rates are negative?

The quick test of this theory is to see when the last time Fed engineering might have pushed the 3 month rate below the dividend yield.

It turns out that they had done so recently. Indeed, the dividend yield on the S&P500 has been above the 3 month T-bill for quite a while now -- since February 2008. (Recall the emergency Fed cuts made just prior to that). If you made any purchases on the basis of SPX Yield exceeding 3 Month T Bill, you ended up with some jumbo losers on your hands.

Hence, we must repeat our prior admonishment against relying on Single as opposed to Multiple Variable Analysis in making any market forecasts or buy decisions. More often that not, you can find varying degrees of correlation, but end up lacking a causation sufficient to make an informed investment decision.

I would suggest those who like to use this metric to rerun their analysis -- but run a cross check with a 2nd variable of real versus nominal rates. You may find a rather a different set of results . . .

>
S&P500 Yield (blue) Vs 3-Month Treasury Yield (red) against SPX (gray)

Dividendandtbillyield
chart courtesy of Michael Panzner


>

Previously:
Single vs. Multiple Variable Analysis in Market Forecasts (May 2005)
http://bigpicture.typepad.com/comments/2005/05/single_vs_multi.html

“A Whiff of Panic . . .”  (January 22, 2008)
http://bigpicture.typepad.com/comments/2008/01/a-whiff-of-pani.html

Source:
Farrell: Touching Bottom in this Bear Market?
Vince Farrell
Jul.22 2008, 7:23 AM ET
http://www.cnbc.com/id/25794408/site/14081545

Tuesday, July 22, 2008 | 11:58 AM | Permalink | Comments (17) | TrackBack (0)
de.li.cious add to de.li.cious | digg digg this! | technorati add to technorati | email email this post

bn-image

TrackBack

TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8341c52a953ef00e553ccdd9c8834

Listed below are links to weblogs that reference Understanding How to Analyze Market Metrics:

Comments

....yes and if you add in a few more variables like:

1. Consumer Debt
2. Uprecedented Real Estate Declines
3. Ongoing War
4. $10 trillion deficit
5. Etc.

You get an even different story. Yep, single variable correlations are dangerous.

Example: I asked a mortgage broker in 2006 at a small gathering: How can banks make loans without checking income?

"Well, they did an tudy and found that people with a high FICO rarely defaulted, so now we don't need to spend all that time verifying incomes." Righto buddy....

I guess when things are not going your way, single variable correlations are the only straws you can clutch at.... Barry, you are absolutely right, lots of money lost on that methodology....lots..

Posted by: Rich Shinnick | Jul 22, 2008 12:22:21 PM

The comments to this entry are closed.



Recent Posts

December 2008
Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      

Archives

Complete Archives List

Blogroll

Blogroll

Category Cloud

On the Nightstand

On the Nightstand

 Subscribe in a reader

Get The Big Picture!
Enter your email address:


Read our privacy policy

Essays & Effluvia

The Apprenticed Investor

Apprenticed Investor

About Me

About Me
email me

Favorite Posts

Tools and Feeds

AddThis Social Bookmark Button

Add to Google Reader or Homepage

Subscribe to The Big Picture

Powered by FeedBurner

Add to Technorati Favorites

FeedBurner


My Wishlist

Worth Perusing

Worth Perusing

mp3s Spinning

MP3s Spinning

My Photo

Disclaimer

Disclaimer

Odds & Ends

Site by Moxie Design Studios™

FeedBurner