INVESTMENT INTELLIGENCER

Jason Trennert Responds: High Profit Margins ARE Sustainable

A few weeks ago, I took issue with Jason Trennert's argument in Barron's that today's P/Es were "low."  I felt that, like many other analysts, Jason had not taken into account the likelihood that today's record-high profit margins would eventually revert to the mean.  Here, Jason responds directly to that criticism:

I very much enjoy this website but I hope you don't mind, as one of its subjects, my posting a comment or two on your critique. Your point on margins is well-taken, but your readers should understand that that there is limited time and space to discuss all the nuances of one's views in such interviews, especially when speaking about something as broad as the stock market. I was unfortunately not asked about profit margins in my interview with Barron's but, as any one of our clients knows, we have written extensively about the subject.

There are many astute market observers that have rightfully pointed out that profit margins have been mean-reverting in the past. Although it's always dangerous to say it's different this time, it seems clear to us, after nearly three years of profit bears claiming that the end was nigh, that something has changed.

In our view the the double-barrelled secular changes of technology and globalization have made margins stickier, longer than the consensus would have ever thought possible. This is simply because companies have a better chance of arbitraging unit labor costs today than they have ever had in history. Labor costs comprise about two-thirds of total costs for the average company. With a seemingly limitless supply of labor in India and China, there has never been a better time to be an owner of a business or for companies to maintain margins. This has also been true, to a remarkable extent, in the service sector. Goldman Sachs' third largest branch office, as an example, is now in Bangalore. The greatest risk to profit margins, therefore, is not labor unrest, as it might have been in years past when labor markets were less fluid, but political efforts aimed at slowing the forces of globalization.

In the hope of prompting an additional thoughtful response from Jason (or another reader), my concern with the argument above is that it doesn't explain why companies won't begin to pass through these labor savings to customers, in the form of lower prices.  For a temporary period, the companies that are the first to exploit lower labor costs will benefit from higher margins and lower prices than their competitors, but over time, the competitors should follow suit--until outsourcing labor to Asia and India becomes not a competitive advantage but a competitive necessity.  Also, companies should begin to take market share by reducing prices, which will also put pressure on margins. 

So, bottom line, I agree that the massive reduction in labor costs from globalization should drive temporary margin gains (which I think we're seeing), but I don't understand why this should be expected to continue indefinitely.

Siegel Rebuttal, Part 1:

As described in the prior post, Jeremy Siegel made several arguments intended to refute the theory that U.S. corporate profit margins will soon regress to the mean, taking stocks down with them.  One of these arguments was that the percentage of overall U.S. profits captured by corporations vs. private entities is very high, so overall U.S. profitability is not, in fact, extreme.  The following email (from an independent analyst) suggests that this argument, at least, is a hallucination:

It’s hard to know where to begin with Siegel, but I have to start someplace, so I’ll pick his point that looking at corporate profits alone gives an inflated picture of overall profit growth, since capital has shifted to corporate ownership.  Here’s a useful chart from the Federal Reserve Bank of St. Louis (click here and see the second row down: Selected Component Shares of National Income.  Also note the third row, which contains the data everyone else is focused on Corporate Profits as a percent of GDP):

Yes, proprietor’s income has grown more slowly than corporate profits.  But looking at the two together, it’s clear that overall profit (as Siegel defines it) is at or close to a high, at least since 1981.

Siegel Fires Back: Bullish And Proud Of It

Jeremy Siegel's latest column in Yahoo Finance responds to John Hussman and others, who argue that Siegel's irrational bullishness is setting him up to be come a modern-day Irving Fisher.  Siegel's response: The critics are wrong, profit margins are fine, valuations should be higher, and stocks are going up.

Siegel's main arguments are these:

• The hand-wringing about "record-high profit margins soon reverting to means" is misplaced because 1) a greater percentage of U.S. corporate profits are coming from international operations, which aren't affected by U.S. GDP, and 2) a greater percentage of overall U.S. profits are now captured by public U.S. companies instead of private ones.

• Innovations in the financial system, namely lower trading costs and smarter central banks, have reduced the equity risk premium.  Thus, stocks should now trade at an average P/E of about 20-times, instead of the long-term average of about 14-times.

Siegel, in other words, believes that "it's different this time."  And that's reasonable--because sometimes it is different this time (the most salient example being a permanent change in the relationship between bond yields and dividend yields after the U.S. went off the gold standard--a change that cost many "prudent" investors to miss decades of gains).  Of course, more often than not, it's NOT different this time, and those who argue that it is end up with egg on their faces (believe me--I know).

I'll wait for Hussman, Grantham, Smithers, DeLong and others to respond to Siegel's arguments before taking a strong stand here.  I will simply suggest that the big lesson from most major bull markets is this: anytime someone argues that "it's different this time," the burden of proof ought to be on them (as opposed to on bears who appear to be "wrong" because the market keeps going up). With this in mind, I'd like to see more data from Siegal backing up his profits argument--a chart showing the percentage of U.S. profits generated from international operations over time, for example, as well as a chart of global profits relative to global GDP (are profit margins at record highs globally, as well?). 

The "risk premium" argument obviously won't be settled until after the fact, but in my mind there's an easy counter to that one: Today's risk premium is low because stocks haven't been that risky recently--even with the crash of 2001-2002, global equities are sharply higher than they were 10 years ago.  Go through a couple of decades of stagnation, meanwhile, like the periods that followed market highs in 1929 and 1966, and investors won't give a damn about low transaction costs or more enlightened central banks.  Instead, because stocks will seem like the worst investment idea anyone ever thought of, investors will once again demand a huge equity risk premium.

Jeremy Siegel: The "Irving Fisher of the 21st Century?"

In his latest weekly letter, fund-manager John Hussman posits that Jeremy Siegel, the famous finance professor known as the "wizard of Wharton," is well on his way to becoming the Irving Fisher of this era.  Fisher, you may remember, was the well-regarded economist who opined in early September 1929 that stocks had "reached what looks like a permanently high plateau."  As Siegel himself noted in his excellent Stocks for the Long Run, Fisher made this remark two weeks before the start of The Great Crash.  Soon thereafter, his reputation was forever destroyed.

Siegel hasn't said that stocks will never again fall, but in Hussman's opinion, he's now stretching his bullish interpretations so far that he's making methodological errors.  Hussman cites a Financial Times piece in which Siegel argues that "Real returns can be estimated from the earnings yield, the reciprocal of the more popular price-earnings ratio. Since stock earnings are based on real assets, the earnings yield provides a good estimate of the real return on the stock market.”  Siegel then uses this logic to suggest that stocks may produce even higher returns in the future than they have in the past.

Hussman, also a finance PhD, has a number of problems with this logic, starting with the fact that Siegel is equating earnings with dividends.  He also ignores that 1% of stock performance over the past 80 years has come from multiple expansion, compares apples with oranges by conflating trailing and forward P/Es, and ignores that profit margins are at currently at record highs (thus producing artificially low P/Es).

This latter omission is the part of Siegel's bullish argument that I find inexplicable.  As Jeremy Grantham has observed, earnings are "one of the most dependably mean-reverting series in finance."  It would be one thing if Siegel were to acknowledge today's record profit margins and then argue that this time they won't revert to historical means.  But he doesn't even acknowledge them!  Instead, he just takes a simple current P/E and compares it to a century average based on average profit margins.

If Siegel weren't so smart and well-informed, this omission might be understandable.  In fact, he is one of the world's foremost market experts and a good friend of Yale professor Robert Shiller who popularized the "cyclically adjusted P/E".  So the only conclusions that one can draw are either that 1) Siegel doesn't think profit margins will drop (an argument radical enough that it deserves to be made explicitly), or 2) he is now so wed to his bullishness that can't let it go (wouldn't be the first time this has happened to a market guru...).  A third, more negative interpretation is that now that Siegel is an advisor to ETF vendor WisdomTree, he can't afford to be publicly bearish because it would be bad for business.  I'll give him the benefit of the doubt on that one.

Grantham: Short the World

Jeremy Grantham's Q1 letter (www.gmo.com) paints a startlingly gloomy picture, even for him: For the first time in history, every major asset class is experiencing a bubble at the same time.  Globally, economic conditions have gotten so good, credit has gotten so accessible, profit margins have gotten so high, and investors have gotten so diversified, that risky assets no longer provide any expected premium over safer ones and there is nowhere to hide.   Even forestry, Grantham laments, has become a mainstream asset class.  The resulting race to scarf up the world's trees has reduced expected timber returns to a paltry 5%-6%.

As before the 2000 crash, Grantham is convinced that it's NOT different this time--and, therefore, will end badly.  He dismisses the idea that, in order for the markets to correct, we will need a catalyst (and, because we don't know exactly what it will be, we're safe).  He ridicules the theory that "private equity" will drive up stock prices indefinitely (and suggests that, in hindsight, this latest bubble will be called "The Private-Equity Bubble").  He acknowledges that predicting the timing of such a correction is impossible and, therefore, that all asset classes might get even more expensive for years (this was what happened in the late 1990s: Grantham and other value-driven forecasters called the top too early and suffered years of predictable ridicule and scorn.  Then the market crashed, and they were heroes.).

Grantham's less-gloomy colleagues point out that bubbles usually peak in a final orgy of excitement, in which markets rocket to levels that previously seemed unimaginable (e.g., 1999 and 2000 for tech, 2005 for housing) and that, globally, we haven't seen such a blow-off yet.  So perhaps, Grantham admits, the party will go on for a while longer.  But, he adds, "bubbles always burst."

 

Hussman Bangs The Bearish Drum. No One Listens.

John Hussman of the Hussman Funds keeps banging away on the impending-market-crash drum, this time in a long interview with Kathryn Welling of welling@weeden.  Hussman's points won't come as any surprise to Investment Intelligencer readers, but they are so rare in the larger world of market commentary today that I'll keep repeating them.  (And when the market doubles from here and stays there for a generation, you can throw both of us to the dogs):

• The market is trading at 19-times "peak earnings" (Hussman's equivalent of the cyclically adjusted P/E).  The common bullish arguments that the market's P/E is 15-times (or less) have no basis in theory or historical fact.  Today's valuation is similar to that of many historical peaks, such as 1929.

• Profit margins have always been mean-reverting, and none of the theories floated to suggest that they will forever remain at today's nosebleed level (50% above their historical average) make sense.  These theories, moreover, are only floated by a handful of thoughtful bulls--the rest don't even bother to try to account for today's high profit margins.

• The long-term profit growth trend line in the U.S. is 6% a year.  The reason everyone thinks (and says) profits will grow at least 10% a year forever is because profits have grown that fast for five years (a.k.a., eternity).

• "Fair value" on the S&P is at least 40% below current levels, even if one makes allowances for the (unlikely) possibility that cheap labor in China and India have permanently improved the profitability of U.S. corporations.

• The market might go up another 10% before it has a typical 10% correction.  From now until after this 10% correction, however, the market will have been outperformed by T-bills.

Hussman does not go so far as to make a forecast, but, for him, he comes awfully close. And given these data points, who can blame him?  He also notes that rising interest rates usually suggest that mean-reversion will come sooner rather than later (which it certainly did in 2000).

After having lived through the 1990s, and been on the opposite side of this argument, I can confirm that long-term valuation metrics are worthless at predicting near-term market movements: In the late 1990s, bears like Hussman were not "cautious," they were "wrong," and year-after-year they were carried out on stretchers.  This said, after the market finally collapsed, they looked pretty good.

Flawed Logic: Jason Trennert Barron's Interview

This week's Barron's interview featured Jason Trennert of Strategas Partners.  Jason is bullish about the U.S. market and bearish about emerging markets.  Much of his logic is sound, and his conclusion is fine--the U.S. market goes up about two-thirds of the time, so predicting it will go up this year is rarely a dumb bet.  One of Jason's points is flawed, however:

We expect $93 for S&P 500 operating earnings in '07, and that puts the market at roughly 15 times earnings...The downside risk is low because P/Es are so low.

P/Es are not "so low."  In fact, they're not even "low."  To get to "15-times earnings", Jason is looking at "forward operating earnings" (an estimate).  He's then comparing it to last-twelve-months earnings.  Cliff Asness has written in detail about the folly of this apples-to-oranges comparison.  According to Cliff, the average "forward operating earnings" P/E is about 11-times, not the 15-times that Jason and other contemporary bulls throw around.  This 11-times, moreover, is an average--so half the time the market's P/E has been below this level. 

And then there's the profit-margin problem.  As discussed often in this forum, profit margins are 50% above their long-term average and at 50-year highs.  To compare today's P/E multiples to historical average multiples, you have to believe that today's profit margins will stay where they are.  Anything's possible, but this seems extremely unlikely.  At the very least, the profit-margin question is deserving of 1) acknowledgement, and 2) an explanation.  In Barron's, anyway, Jason didn't have either.

The market may go up, but, if so, it won't be because P/Es are low.

Easterling: Earnings Do Not Grow 6% Every Year In Perpetuity

The seemingly prudent consensus these days is that corporate earnings growth will slow from the torrid double-digit pace of the last few years and settle in at a more normal rate of, say, 6% per annum (the long-term average).  Alas, this consensus is anything but prudent.

As fund-of-funds' manager, analyst, and SMU adjunct professor Ed Easterling writes, earnings just don't behave that way.  Yes, they average about 6% a year (6.6% in Ed's 80-year model).  But they achieve this average by shooting up for 3-5 years, collapsing for 1-2 years, shooting up for another 3-5 years, and so on.  In other words, for all of recorded history, earnings have followed a 3-steps-forward, 2-steps-back pattern, not safe, smooth trend-line growth.

So, what's the problem?  Well, as previously discussed, profit margins--"one of the most mean-reverting series in finance" (Jeremy Grantham)--just hit a 50-year high of 14% of GDP, as compared to the 9% average.  As Ed Easterling points out, profits have also now grown strongly for 5 years in a row.  According to Ed's data, there has been only one business cycle since 1950 in which profits grew for more than 5 years in a row (in the late 70s).  In the sixth year of that expansion, moreover, growth was miniscule, and in the following two years, it collapsed--taking the stock market to a 16 year low. 

As of this writing, meanwhile, "prudent" forecasters are expecting profits to continue to grow for not only 2007 but 2008--what would be the 7th year in a row.  Is this possible?  Anything's possible.  Is it likely?  A review of the business cycles of the past half-century suggests the answer is an emphatic "no."

If, instead of setting a record for consecutive years of impressive growth, earnings behave the way they normally do--collapse for a year or two--what is the stock market likely to do?  Happily deliver another two years of low-volatility, near-double digit growth?  We could do that math, but then we might be accused of always being the bearer of bad news. 

 

 

Kasriel Blasts Rose-Goggled Economist, Defends LEI

Nothern Trust's Paul Kasriel, whose recession-warning indicator is flashing yellow, devotes an entire eContrarian to poking fun at another economist, Michael Lewis, who scoffed at Kasriel's advocacy of the LEI as a decent recession-predictor. 

Lewis appears to have taken the same position as most economists: The LEI isn't a perfect recession predictor, so even though it is suggesting that there will be a recession, there probably won't be one.  Kasriel argues that this interpretation is bass-ackwards: The LEI is a pretty darn good predictor of recessions, so there probably will be one.  (For good measure, Kasriel throws in his personal Kasriel Recession-Warning Indicator, which, when combined with the LEI, has never given a false signal.)

Who's right?  Time will tell.  But of the two economists, Kasriel appears to be deriving his thesis from the data, while Lewis appears to be doing it the other way around.

Hussman: "Fair Value" on S&P 40% Below Today's Level

Not to keep beating the same drum, but John Hussman seconds Andrew Smithers's points about the U.S. stock market's overvaluation.  Hussman's bete-noire is the common use of "forward operating earnings" as a means of concluding that the market is cheap--a practice whose many flaws merely start with the fact that the historical average P/E multiple on "forward operating earnings" is about 11-times, not the 15-times that many investors use (Hussman cites Cliff Asness here).  Hussman is quick (and wise) to say that this condition doesn't mean that stocks will tank--just that future returns for the S&P 500 won't look anything like 10% a year.   

After triangulating with a variety of valuation methods, including the dividend discount model, Hussman puts the fair value of the S&P 500 at 850, 40% below today's level.  So anyone counting on U.S. stocks to appreciate at 10% a year for the next decade had better 1) scale back expectations, and/or 2) pray that it's different this time.