Thursday, 25 November 2010

When Is a Bubble Not a Bubble?

The following article has been adapted from "Value: The Four Cornerstones of Corporate Finance" by McKinsey & Co.'s Tim Koller, Richard Dobbs and Bill Huyett. The publisher is John Wiley & Sons.

It's easy to construe all instances of sharply rising, then falling, stock prices as bubbles, but most of the time they're not.

True stock-price bubbles are essentially nonexistent at the level of the aggregate economy, very rare in specified industry sectors and not common for individual companies. The fact that they are so rare makes it all the more important for the value-minded executive to be able to spot them.

What, then, is a bubble? Bubbles occur when one group of investors irrationally pushes a stock price far above a level that can be justified by its potential financial performance, and other, often more sophisticated investors, aren't able to offset the actions of the less rational investors due to structural constraints and the liquidity risks of shorting stocks.

Financial crises are often described as bubbles, but they're not the same. Bubbles are the rise and fall of company share prices. An unlike debt, equities don't have maturity dates or covenants that can allow the holder to immediately demand cash from the company. Therefore, when bubbles burst, they don't have a drastic effect on the economy (unless they're accompanied by large amounts of debt).

Financial crises, on the other hand, do have a dramatic and far-reaching effect on the real economy, because they're brought about by excessive financial leverage, which has a negative domino effect when the value of the underlying assets falls and those who owe the debt can no longer service it. The debt crisis causes an economic downturn, which then causes the stock market to decline. But we can't call this a bubble if stock prices were reasonably valued in light of the economic conditions before the crisis.

A good example of a stock bubble was when 3Com(COMS_) spun off its Palm subsidiary 10 years ago. Immediately after the share sale, the market capitalization of Palm was $45 billion. At the same time, 3Com's market cap was only $28 billion, even though it owned 95% of Palm (presumably worth $41 billion). The only way that 3Com could be worth 60% of Palm would be if the rest of 3Com's businesses were worth negative $13 billion!

So why didn't rational investors exploit the mispricing by going short in Palm shares and long in 3Com shares? Because they couldn't. The free float of Palm shares was too small after the carve-out because 95% of all Palm's shares were still held by 3Com. Establishing a short position in Palm would have required borrowing the shares from a Palm shareholder, but there weren't many. Therefore, the bubble remained until the supply of shares available to borrow increased steadily over the months following the carve-out.

Bubbles that affect the broad market in developed economies are rare. Over the past 50 years in the U.S., two periods (1967-1972 and 1997-2001) might be considered market-wide bubbles. But as we dig deeper, we find that even these bubbles weren't broad based; they were concentrated in certain segments of the market.

During the technology bubble of the late 1990s, the aggregate S&P500 price-to-earnings ratio was greater than 30 times for several years versus an expected level of about 16 times (given the level of interest rates and inflation). Looking deeper, the bubble was concentrated in large-cap companies with high P/E ratios clustered in three sectors: technology, media, and telecommunications (known as TMT). P/E ratios were significantly lower in most other sectors.

One would expect rational investors to question the apparent mispricing -- and some did. Julian Robertson, one of the leading investors of the 1980s and 1990s, said: "Well, we've had a movement away from value investing to momentum investing, where price is not a factor ... everybody -- day traders, hedge fund operators, LBO people, right down the line -- is piling into the same stocks, which is, in effect, an inadvertent Ponzi scheme. And it will eventually blow up."

On the other hand, the media found commentators who could explain the high prices with new theories of economics and finance. One idea was the "new economy," although definitions of it were vague. But vagueness and hope drove herds of people into buying overpriced Internet stocks, so much so that those who questioned the new economy were said to "not get it." At this time, there weren't enough rational investors with the resources and risk appetite to prevent prices from rising too high.

Sector bubbles are more frequent than market-wide bubbles, but still rare. The biotechnology sector experienced a bubble between 2005 and 2006, when the sum of listed biotech companies' market capitalization was about $450 billion (excluding the traditional large pharmaceutical companies that were investing in biotech). Making some assumptions about future margins, one can estimate that these companies would need to earn $600 billion in revenues in 2025 (in 2006 dollars) to justify these prices. To put that into perspective, all the listed pharmaceutical companies combined earned about $600 billion of revenues in 2006.

Company-specific bubbles are hard to find too, and sometimes misnamed. Imagine a sharp share-price drop for a pharmaceutical company that announces its promising new drug has failed in clinical trials. The steep decline doesn't indicate that the price was originally at a bubble level; the original price might have reflected a reasonable estimate of the company's value given a reasonable estimate of the drug's probability for success.

Brinker International(EAT_), however, is an example of a company that experienced a bubble in the early 1990s. At that time, Brinker, which has 1,700 restaurants (including the Chili's chain) in 27 countries, increased its revenues by more than 20% per year, with improving margins. In 1993, its market capitalization reached $2.1 billion with a P/E ratio of greater than 40 times. But eventually same-store sales began to decline, and the market realized the company's economics were vulnerable. In 1994, Brinker's market cap dropped by 61% to a more reasonable P/E ratio in the high teens.

Although not common, the fact that bubbles do sometimes occur -- that is, prices do sometimes deviate from the fundamentals -- makes it even more important for corporate managers and investors to understand the true, intrinsic value of companies. They may be able to exploit any market deviations if and when they occur, for example, by using shares to pay for acquisitions when those shares are overvalued by the market. But more importantly, they will not make decisions based on what may be the unreasonably high price of their stocks, only to be surprised when those prices return to more normal levels, which they always do.

Newscribe : get free news in real time