Reversion to the Mean for Equities 33 comments
-
Font Size:
-
Print
- TweetThis
The following is quoted from yesterday’s 5-Minute Forecast (a daily newsletter published by Agora Financial).
From "Mortgage Rates, Scary Jobs Details, Investing in 2009, Russian Gas Dispute, and More!":
By one metric, the future shouldn’t be TOO terrible for U.S. equities. Check out this chart, sent over by Rob Parenteau of The Richebacher Letter:
“The contraction of the total value of the equity market relative to GDP,” notes Rob, “has reversed nearly the entire premium introduced during the New Economy bubble years.
"If there is a reversion-to-the-mean process under way with respect to the equity market capitalization-to-GDP ratio, the most violent part of the move must be behind us. Given the severe recession developing before our eyes, however, we are in no rush to be buried beneath a landslide of earnings shortfalls, employment reductions and bankruptcy announcements.
"A fiscal push in early 2009 may help stabilize or improve the near-term earnings growth expectations held by professional equity investors, which are already much lower than those offered by brokerage house equity analysts. But the larger question remains: If financialization is not going to be the growth driver for the U.S. economy, what will take its place? If credit booms and busts are going to be restrained by a stripped-down financial system, especially one that is heavily regulated, what will drive earnings growth?"
Not noted by the the folks at Agora Financial, but obvious on the graph, is the overshoot of the mean in the two previous super cycles (1930’s - 40’s and 1970’s – 80's). If history is any guide, this cycle has still some way to go down to reach bottoms similar to 1941, 1974 and 1982. This graph is normalized to GDP. If we do not turn GDP around in the next 1-2 years (end the recession), the outlook for stocks is grim. It is my opinion that the recession will not become ingrained (if it does, it’s called depression) and will end within the 1-2 year limit, hopefully well within.
One other comment about overshoot relates to the often observed effect of overshoots to the downside tend to mirror the preceding overshoot to the up side. If that were to obtain to the current cycle, the downside is much lower than in the previous two super cycles.
Can massive, coordinated monetary and fiscal response by governments around the world make this cycle turn out better than past cycles? This question is a challenge to commenters to add some value to the discussion. Please try to limit political invective and finger pointing. Can we have a discussion looking forward, as painful as that may be?
Related Articles
|
-
The Seeking Alpha 100
See allThe top 100 stock
market authors
selected for publication
Seeking Alpha 100 » -
Fastest Climbers
Authors with the largestSee all
increase in followers
in the last 7 days
Fastest Climbers » -
Top Commenters
Top commenters,See all
as ranked by their peers
Top Commenters » -
Top Instabloggers
The 100 most active
Instabloggers.-
1
David Fry
-
2
TraderMark
-
3
Don Dion
-
4
Cliff Wachtel
-
5
Mike Havrilla
Top Instabloggers » -
1
-
Top StockTalkers
TopSee all
StockTalkers.
Top StockTalkers »
This article has 33 comments:
Right after I submitted this article I read an excellent article by John Hussman: http:seekingalpha.com/... John provides information from a completely different process of analysis which leads one to similarconclusions as discussed here.
A troubling thought to me is the make-up of the GDP, which today is composed of much more intangible components (ie prominince financial of service sector) than it was in previous recessions.
Ther are a lot differences in today's markets that might lead to faster discovery. You mention market capitalization and electronic trading exchanges. I would add more liberal short-selling rules and stricter margin requirements. All these factors should, in theory, lead to more rapid price discovery and argue against overhoots.
However, then we should not have has the extremes in market behavior to the upside that have occurred twice during the past 10 years, if all these changes really did "smooth the markets".
You raise a very good point, but I think the debate in this area is ongoing. Perhaps some other readers have research or ideas and we will hear from them.
The excellent comment from Keithco suggests that the total equity market valuation, after adding the NASDAQ, is now at an even higher percentage of GDP and thus more overvalued relative to historic norms.
On Jan 13 08:10 AM Royal Scot wrote:
> Good point Keithcho, but wouldn't considering these new exchanges
> make the graph look even worse? I assume the respective market caps
> would be additive.
>
> A troubling thought to me is the make-up of the GDP, which today
> is composed of much more intangible components (ie prominince financial
> of service sector) than it was in previous recessions.
I've never been too fond of this 'indicator,' since it can go up just because more companies are going public, as in the tech IPO boom, and go down during an IPO slow period. How many IPO's are being done now?
Is there any REASON to suggest a relationship between market capitalization and GDP? Given fluctuating levels of equity use by corporations and the expansion of the number of firms publicly traded, the huge burst in the 1990s might just have been a 'counting' phenomena. We counted market cap where before the firms were private or at least not publicly listed.
Second: As a statistician: You have a grand total of '5' data points. Keep in mind you are trying to 'forecast' MEAN REVERSION. You have four examples of 'excess valuation upon which you have based your premise-- roughly 1929, 1937, 1959 and 1966. Not a tremendous amount of data upon which to base a premise. BAsed upin your data I would have concluded, LOOKING FORWARD, that the market was overvalued and due for MEAN REVERSION....in about 1995...when the dow was at 4000.
Not very helpful with timing.
cyclingscholar
the 401K vehicles offered in the past by companies have already been severely modified, curtailed, or eliminated.
Decades worth of market psychology has popped along with the paper wealth that drove it. It's now a traders market, not buy and hold,
Might the average Joe come back in someday, certainly, but I think it will be a long time.IMO
Critical to drawing the mean is how one calculates adjusted inflation. Doug Short does a better job explaining and displaying this concept than I can:
dshort.com/articles/re...
Briefly, if one adjusts as the BLS has been doing since 1919 we still have some more overshooting to do. But if one incorporates CPI data, generated since 1982, the mean has already been overshot substantially. Which side of the fence do you stand?
So if another domino falls things could get much worse: otherwise, S&P 750 shoule be the low point.
DaveW - - -
I have studied inflation adjusted market returns in the past (AAII Journal feature article in Feb., 1996) and have said in other comment streams that I want to return to the subject. I have generated such a long list of interesting research ideas that inflation adjustment hasn't gotten to the top yet. I think it is important to get to it because the possiblilty of big deflation/inflation swings is increasing.
Thanks for the reference link. I have bookmarked it.
Tom Armistead - - -
I appreciate your opinion. Many hold the same view. I think there is better than 50% probability that S&P 500 752 (Nov. 20 closing) will not hold. Absent another domino, the ultimate bottom may be no lower than 640 - but may not occur until second half of 2009. Obviously, I also think that there is a chance that 752 will hold, just less than 50/50 chance.
While another domino (if it happens) may come from an unexpected source, my eye is on Citi. I think the entire implementation of the first TARP tranche was aimed at supporting Citi and hiding their weaknesses. If there is an Achilles heel, I would vote for C.
cyclingscholar - - -
I admit that the article makes a statistically shakey observation. With only five data points, the relationships could correctly be called possible coincidence.
I also must admit that I look at the chart in this post and the Hussman article I referred to in the comment stream with prejudice. I published an article in SA in August that used 10-year and 10-year rolling returns to make some projections about multi-year stock market expectations. The article link is: seekingalpha.com/artic.... This article needs an update this quarter. Ah, more congestion at the top of my research list.
Thanks again to all commenters.
One other point to consider is the implied change to the market's structure which has occurred or might occur in the future.
Several companies have all but shed original stockholders in favor of USTreasury preferred share ownership. What is the true 'market cap' of AIG now? How about the TARP banks who without those bailout funds would be insolvent? What is their true contribution to Market Cap? Ditto for the Big 3. Without financial blood infusions they are technically dead.
Several pertinent definitions have changed in the past 6 months. Whether that renders John's chart less useful or not remains to be seen, but it might be important that we're not really comparing apples to apples any more.
It would also be interesting to see a comparable chart for the same time period showing total debt (personal + public) as a percentage of GDP for comparison. What contribution to market cap did the borrow and spend mentality have?
A massive shift to saving and paying down debt would surely amplify any downside overshoot just as the borrowing and spending amplified the upward surge. One would think that a measure of the overshoot and reversion of debt would give some insight to any reversion in the market.
Very astute observation - the mean does include the "bubble" years. I noticed that, but failed to point it out in the article.
User55065 - - -
Excellent questions. I don't have answers but agree with your point about the giants of NASDAQ being missing from the NYSE. This definitely is different than the world before 1990.
Maybe a reader yet to come will have some knowledge of research in the areas you pointed out.
this is the crux of my argument about recovery. no recovery without driver. no driver without healthy financial system.
we have a different and faster communication / information system in 2009, than previously. we do not have to overshoot to the extent of previous recessions. i am hoping we emerge with a more realistic view of the earnings and profit potential of the market (although somehow i doubt it).
There is no big money to be made in traditional banking.The death of Glass-Stegall was the death of the stable banks that kept the economy going and the basis for the money supply.All this other conversation is the aftermath of the death of the traditional bank....BTW,its still alive in the small regional banks and credit unions,if they don't get wiped out by this mess...
1) The Wal-Mart-ization of America is usually seen as an analogy of big business swallowing small business, among other themes. This is one possible explanation of the last spike, as more and more public companies today account for GDP than private ones.
2) Many financial institutions (were) not only a lot bigger than before, they used to be private, and wouldn't have registered through an NYSE market cap basket in the past. Even after this grotesque deleveraging, firms like BAC and WFC still retain market values reminiscent of the past 20 years (that has changed these past two weeks, but your chart may also have changed dramatically these past three months). I would use Wall Street firms to accurately demonstrate my point, but Wall Street doesn't exist anymore.
My point is that perhaps more of our economy is "for sale" in the public marketplace than before. Perhaps this deviation from mean expressed in your chart is a natural byproduct of this reality. In my mind, a figure that reflects "total return" on all US assets would be more useful than taking a cross-section of our economy that may or may not have changed dramatically over the years.
If this "total return" number was grossly over the mean, you'd really have something there. It would include more rental income and housing-related data as well, which would not only give a more accurate diagnosis, it would also more accurately reflect the nature of this downturn. My guess is that more than likely, such numbers would be even more contorted, as housing really warped into something economically incomprehensible.
Regardless, you make a very good point. It complements your prior outstanding article about the housing market. I just find public equities to be a bit unwieldy as a foundation for this argument.
You both make good arguments regarding the overshoot and suggest different possible outcomes. There are probably some factors that none of us has thought of in this discussion which could also mitigate or aggrivate an overshoot to the downside. Maybe we are just going to have to imitate our government and try to figure it out on the fly.
Ricard - - -
About 15 years ago I started a study comparing inflation adjusted equity growth and real GDP. I didn't finish the project and don't know where to put my hands on any of the material, but my recollection was that the general trend to that date was that equity value (discounted for inflation) had grown faster than real GDP for decades. If that is true, it might be explained by increasing P/E ratios - I don't remember if I got far enough to consider that.
It is something I will go back to sometime because the entire question of total equity value and GDP seems intuitively to be related.
As you point out, not only is equity value important, but the income spun off (diividends) should also be accounted for.
It just seems to be a nice package: Value of public equity + value of private equity + value of income produced = Constant x GDP. If the constant is much different than 1, the possible reasons would be a great research project.
I'll try the John Hussman link again: seekingalpha.com/artic...
Ricard - - -
My research suggestion is misstated above. A better statement is:
Value of public equity + value of private equity + value of income produced (i.e. dividends, current plus present value of future dividends) = Constant x (GDP + present value of future estimated GDP)
I've written a lot about cleantech as a likely market driver for the next 50 years and believe there is a lot of opportunity in the sector. But I remain cautious about the neurotics who build fairy castles in the sky and the psychotics who have already scheduled the movers. To quote Dylan "the times they are a changing."
money.cnn.com/magazine.../
I am not one to critique Buffett, so I'll let this stand as it is.
Thanks for the link - very interesting article.
Are you a principal of the Seeking Alpha website?
No.
Sorry, I hit publish too soon. John is a neighbor of mine and as far as I know has no relationship to SA except as a contributor and commenter.
sleepless - Thanks for noting the enquiry and making a comment.
What about the earnings yeild compared to the interest rates? That comparison suggests that stocks have never been this undervalued. It seems more logical b/c it suggests where money will be best treated.
Very good point about the change in variables over the last 80 years. Perhaps we should only look at the last 20 years when some of the practices and conditions you describe are more constant. But then I would suggest that maybe the analysis quoted should not have been done at all because the last 20 years have been much stronger than many previous periods and we shouldn't expect that strength to go on forever.
Using earnings and discounting for interest rates ("risk free rate of return") is indeed a logical way to value stocks. The problem for me is that I don't know what to use for earnings and the "risk free rate".
I have never liked to use trailing earnings and trailing earnings growth for valuations. It makes much more sense to try to use forward earnings and forward earnings growth because that is where my investment will "live". Forward visibility is (and has been for the past 9 months or so), for me, non-existant.
I have the same problem with interest rates. Do I dare discount 3-5 years of forward earnings (which I just said I don't know how to estimate) with the current 5-year treasury rate? I have little hope that the current "risk free rate" will remain for 3-5 years. If it does, I doubt that my earnings projections (wild guesses would be a better descriptor) will be met.
I can't invest right now. I do a little bit of trading, but even less investing.
Thanks for the comment. It was very succinct but touched on a lot of questions.