 Vitaliy Katsenelson, CFA is a portfolio manager with Investment Management Associates in Denver, CO. He teaches finance at the University of Colorado and most recently is the author of Active Value Investing (above), where he describes his strategy of investing in range-bound markets. He is a frequent contributor to the Financial Times, Dow Jones MarketWatch, has written for Barron’s and Business Week, and maintains a website ActiveValueInvesting. We interviewed Katsenelson on February 14, 2008.
Your central thesis is that we are now in a range-bound market and not a bear market. For the benefit of our readers, can you define a range-bound market and why you are so confident we are not in a bear market?
Markets have long term secular cycles and short-term cycles, and it is the former that I will focus on in this discussion. Traditionally we have been trained to look at long-term cycles from a bull and bear perspective. But half of the secular markets in the last 200 years have been going up, and the other half going up and down and not really going anywhere. I define these long-term sideways markets as range-bound. We assume when markets are not going up they were going down, but really these are periods of prolonged sideways movement.
A secular bull market begins with very low P/E ratios and average economic growth. P/E ratios expand, adding to earnings growth, and P/E ratios serve as a tailwind producing supersize returns.
But when P/E ratios expand to very high levels they become a headwind. At end of a bull market we can have a range-bound or a bear market. The only difference is what happens to the economy in long run. If the economy is growing - even if that growth isn’t spectacular (average will do) - we will have a range-bound market. If the economy is contracting we will have a bear market. Either way, we will have P/E compression.
In a range-bound market, the slope (overall direction) of the market is flat, but the market is volatile.
The best recent case of a bear market was Japan in late 1980s. Japan had both a long-term contracting economy and contracting P/E levels. Economic contraction added fuel to the fire.
"There has been only one secular bear market in the last century, and that was the Great Depression."
I don’t know what the economy will do in the next six months or the next two or three years. But, if the economy remains healthy in the long run, we are in a range-bound market – one which started in 2000.
There is some probability of a severe or long recession, in which case a bear market results. There has been only one secular bear market in the last century, and that was the Great Depression.
You argue persuasively (see Down to the Last Drop of Profit Growth) that today’s P/E multiples are inflated, primarily because earnings (the denominator) is above historical averages. Currently pre-tax margins are at 11.9%, but have averaged only 8.5% since 1980. Is this based on the S&P 500 or a broader market? You argue that shifting to a service-based economy might improve baseline margins to between 8.9% and 9.2%, but this still implies a 30% drop in the market. If we are about to see a contraction in earnings, shouldn’t the market have already discounted this into prices, and reflected this in current P/E ratios?
Today’s margins (measured as pre-tax operating earnings across all US corporations as well as across the S&P 500) are almost 40% above historical averages. Our economy is more service-oriented than in the past, adding 40 to 70 basis points to historical profit margins.
To normalize the broad market’s earnings, I look at them over more than just the trailing 12 months. I use 3, 5, 10 and 12 year historical averages, to insure earnings are not artificially skewed by one or two year’s worth of exceptional (good or bad) results. This is consistent with the approach advocated by Robert Shiller of Yale [Ed. Note: see our article on this subject].
Today’s P/E ratio (based on 2007 earnings) is about 17, but if you normalize the earnings, the P/E goes up to 22 or 24.
The market is not cheap, and today’s market levels do not reflect lower earnings. If it did, P/E ratios would be lower. Also, investors love drawing straight lines to illustrate market trends, and thus most expect the above average earnings growth we observed in the recent past to continue (in part because we came out of the 2001 recession and in part because of margin expansion). Unfortunately, they’ll be negatively surprised.
"The market is not cheap, and today’s market levels do not reflect lower earnings. If it did, P/E ratios would be lower."
Just because profit margins are high and will contract does not forebode a 30% drop in the market, at least not in the short run. Depending on the presence or severity of a recession, profit margins will decline and earnings will either lag GDP growth or decline.
This profit margin analysis does not forecast the (cyclical) short-term direction of the market, which would be market timing – a loser’s game. I am trying to identify risks. (As a side note: in my book I argue against market timing, instead urging investors to focus on timing individual stocks – buying them when they are cheap and selling them when they are fairly valued. I provide a detailed strategy of how to do just that.)
The current range-bound market started in 2000 when valuations were 50-60% higher than the starting valuations of previous range-bound markets. Seven years later, after plenty excitement and volatility, we are at the same level as where previous range-bound markets started. Whether you value this market with current or historical earnings, you’ll get the same result – it is not cheap!
Display article as PDF for printing.
Would you like to send this article to a friend?
Remember, if you have a question or comment, send it to
. |
