Lesson 4 of 7
Dreman: the contrarian multiple
David Dreman argued that forecasts miss by wide margins and that the lowest multiples already price the bad news in. Here is what a miss does at two multiples.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- Dreman's case rests on forecast error: analyst and investor expectations miss, and they miss by wide, measured margins.
- On Dreman's evidence, a low multiple has already priced in disappointment, so on average the same miss does less damage.
- A high multiple is priced for the forecast to come true, so the same miss can cost far more than the earnings alone.
David Dreman's argument starts from a measured fact: forecasts are wrong more often and by more than investors expect. His conclusion is that the stocks priced for the rosiest forecasts have the most room to fall when one misses, and the stocks priced for trouble have the least.
Same 10 percent earnings miss, two starting multiples
- Priced for growth, started at 35x−35.7%
- Priced for trouble, started at 9x−10.0%
Forecasts miss, and miss by a lot
Dreman's Contrarian Investment Strategies documents how far consensus earnings forecasts stray from what companies actually report, across large samples of companies and years. Lakonishok, Shleifer and Vishny's 1994 study found that low-multiple "value" portfolios earned more than high-multiple "glamour" portfolios in their sample. Their reading was that investors extrapolate past growth too far, so glamour prices assume more than later arrives.
What a 10 percent miss costs at two multiples
Why the low multiple absorbs the surprise
A multiple of 9 already reflects a market that expects little. There is less optimism left to remove. A multiple of 35 has priced in a specific, hopeful future, and a miss is evidence against that future, which the market re-prices on top of the earnings shortfall itself. Neither company's business outcome was better or worse than the other's here. Only the price each started from was different. The same arithmetic runs the other way on good news: a company priced for trouble that merely meets a low bar has more room for its multiple to rise than one that was already priced to be right.
What Dreman bought
Dreman's own portfolios held the lowest-multiple stocks across a broad set of industries, held with patience rather than turned over quickly.
Priced for a bright forecast, 35x
- The multiple already assumes the growth story continues.
- A small miss is a large surprise to that story.
- The multiple itself has room to fall further.
Priced for trouble, 9x
- The market already expects little.
- A miss confirms what the price assumed.
- The multiple has little room left to fall.
The multiple, not just the earnings, decides how much a miss costs.
Check yourself
Dreman's argument rests mainly on what measured fact?
- Forecasts are, on average, very close to what companies report
- Forecasts miss, and miss by wide, systematic margins
- Low-multiple stocks pay higher dividends
Show the answer
Answer: B.
Dreman documents wide, repeated forecast errors. The rest of his case follows from that.In the worked example, why did the 35x stock fall by more than the size of its earnings miss?
- Its dividend was cut at the same time
- Its multiple compressed further, on top of the lower earnings
- Its earnings actually fell by more than the 9x stock's did
Show the answer
Answer: B.
Both missed by 10 percent on earnings. The growth stock's price fell more because its multiple fell too, from 35x to 25x.On Dreman's argument, which kind of stock has less room left to be re-rated down on bad news?
- A stock priced for a bright, hopeful future
- A stock already priced for a weak or troubled future
- A stock with the highest trading volume
Show the answer
Answer: B.
A low multiple has already discounted disappointment, leaving less room for the price to fall further on confirming news.Sources
- Dreman, Contrarian Investment Strategies: The Next Generation, 1998: overreaction, forecast error, low-multiple strategies.
- Lakonishok, Shleifer and Vishny, "Contrarian Investment, Extrapolation, and Risk," Journal of Finance, 1994.
What this leaves out
- Both companies and every figure in the worked example are round invented numbers, not any filed company's history.
- Overreaction is an average tendency measured across large samples. It is not a rule that holds for any single stock or quarter.
- A low multiple is not automatically cheap: it can also be correctly pricing a business that is genuinely getting worse.
- Low-multiple strategies have gone through long stretches of lagging the market; see Value's long drawdown.
- Nothing here is advice about any security.