Cheap for a Reason: Value Traps, the Value Premium, and Why Bad Companies Can Be Good Investments

A value trap is a stock that looks cheap and stays cheap, or gets cheaper, because the business behind it keeps deteriorating. The standard warning is that such stocks are “cheap for a reason.” That warning is accurate, and it applies to nearly every value stock. Fama and French showed in 1995 that stocks with high book-to-market ratios are less profitable than expensive stocks for years before and after they get cheap.¹ A low price is how the market records bad news.

What matters is whether the price already reflects the reason, and that has a different answer for one stock than for five hundred. The research on value traps explains why they sink stock pickers more often than diversified value funds, and where profitability and quality screens help.

What a value trap is

Price multiples divide a price by a number from the most recent financial statements. A stock at $80 with $10 of earnings trades at 8 times earnings. If earnings then fall to $3 and stay there, the buyer paid about 27 times the earnings the company actually went on to produce. The multiple looked low because the denominator was about to shrink.

A value trap is a stock whose fundamentals decline by more than its price already assumed. The price can be low because the market expects falling profits, impaired assets, a dividend cut, dilution, a debt problem, or a product that is losing to competitors. A low multiple tells you the market is pessimistic. It does not tell you whether the market is pessimistic enough.

Every value stock is cheap for a reason

Fama and French followed the profitability of cheap and expensive stocks around the date they were sorted. High book-to-market firms were less profitable “for four years before and at least five years after ranking dates,” and they described high book-to-market as “typical of firms that are relatively distressed.”¹ Piotroski used the same language about his sample.²

French’s portfolio data show how rare a cheap, highly profitable company is. Sort US stocks each June into fifths by book-to-market and, separately, into fifths by operating profitability. Inside the cheapest fifth, the least profitable group held an average of 490 firms from 1963 to 2026. The most profitable group held 28, and 17 over the last ten years.³ A cheap company with strong profits is uncommon, because the market prices strong profits.

A company with a clean balance sheet, rising margins and admired management rarely sells at a deep discount. When a stock does sell at one, something about the business is worrying investors. The value premium, where it has existed, came from buying those worries at a price that turned out to overstate them on average.

One stock can be a trap; a value portfolio holds hundreds

Joseph Piotroski studied 14,043 cheap-stock observations from 1976 to 1996, taking the highest book-to-market quintile each year. Fewer than 44% of them earned positive market-adjusted returns over the next two years. The portfolio still beat the market, because a minority of big winners outweighed a majority of disappointments. In his words, the strategy “relies on the strong performance of a few firms, while tolerating the poor performance of many deteriorating companies.”²

Fama and French later traced where value’s return comes from. The largest part came from cheap stocks that improved enough to migrate into the neutral or growth groups, or that were acquired. Expensive stocks that fell back to neutral or value added to the spread as well. Cheap stocks that stayed cheap contributed the least.⁴ A value portfolio holds many stocks that never recover, and it does not need them to. It needs the recoveries and takeovers to be large enough to cover them.

We tested that on French’s data. The chart below follows the cheapest, least profitable fifth of US stocks, the corner where you would expect value traps to cluster, from July 1963 to August 2026.

A basket of several hundred cheap, weak businesses beat the market by about two percentage points a year over 63 years. It also fell 63% from May 2007 to February 2009, and it compounded at 0.9% a year from 2007 through 2020. Over rolling ten-year windows it beat the market in 59% of them.³

That extra return was not skill. Regressed on the five Fama-French factors, the corner has a market beta of 1.14, a value loading of 0.72, a negative profitability loading of −0.38, and an alpha of 0.12% a year with a t-statistic of 0.12. Its return is what its factor exposures predict. A diversified pile of stocks that are cheap for a reason earned the value premium, and a value fund captures the same premium.

“Unprofitable” here means the bottom fifth by operating profit relative to book equity, so it includes low-margin businesses as well as money-losers. These are research portfolios with no fees or trading costs, and the cheapest small stocks are expensive to trade.

Why value has paid

Three explanations compete, and the evidence has not settled on one. The HML guide and the small-cap value guide cover the long-run record in detail. Here is how each explanation treats a value trap.

Risk. Value stocks may be riskier in ways that matter to investors, so they must be priced to earn more. Campbell and Vuolteenaho found that value stocks carry more exposure than growth stocks to news about the market’s future cash flows, a risk their model prices more heavily than news about discount rates.⁵ Lettau and Wachter built a model in which value firms have shorter-duration cash flows that “covary more with cash flows” in the aggregate economy, while growth firms move more with discount rates.⁶ Under this view, traps are part of the risk investors are paid to bear.

Mispricing. Lakonishok, Shleifer and Vishny argued that investors extrapolate past growth too far, so they overpay for glamour stocks and underpay for stocks with a poor record. They found that value strategies were not riskier in the ways they tested.⁷ Piotroski and So sharpened this idea. From 1972 to 2010, value-minus-glamour returns were concentrated among firms whose fundamentals disagreed with their prices: cheap stocks with strong financial statements and expensive stocks with weak ones. That combination earned 22.64% in the following year, size-adjusted. Where price and fundamentals agreed, the spread was 0.14% and insignificant.⁸ Under this view, a value trap is a stock whose low price was correct.

Investment and profitability. In Fama and French’s five-factor model, value was redundant in US data from 1963 to 2013 once profitability and investment factors were included. They warned the result “may be specific to this sample.”⁹ Hou, Xue and Zhang’s q-factor model explains many anomalies with market, size, investment and return on equity, and has no book-to-market factor.¹⁰ These models do not say value is useless. They say price, expected profitability and investment carry overlapping information, which is why a cheap price alone says little about a stock’s prospects.

Distress is a separate risk from value

A tidy story links the two: value stocks are distressed, distress is risky, and risk is rewarded. The evidence on distress itself points the other way. Dichev found that “bankruptcy risk is not rewarded by higher returns” and that firms with high bankruptcy risk earned lower than average returns after 1980.¹¹ Campbell, Hilscher and Szilagyi built a better failure-prediction model and found that since 1981 “financially distressed stocks have delivered anomalously low returns,” despite higher volatility, higher market betas and higher loadings on the size and value factors. They concluded the results are “inconsistent with the conjecture that the value and size effects are compensation for the risk of financial distress.”¹²

Griffin and Lemmon split the question by price. From 1965 to 1996, the book-to-market spread was largest among the most distressed firms, 14.44% a year against 3.87% among the least distressed. The low returns came from distressed firms that were also expensive. The effect was strongest in small stocks with little analyst coverage, which they read as mispricing.¹³

Our French data point the same way. The expensive and unprofitable corner returned 3.4% a year from 1963 to 2026, below Treasury bills, and beat T-bills in only 32% of rolling ten-year windows. A struggling company priced for a turnaround has been a worse bet than a struggling company priced for failure. A stock screener sorted on low P/E will not separate the two, because a firm with collapsing or negative earnings can show a low, high or meaningless P/E.

Where profitability and quality help

If some cheap stocks deserve their price, the next step is to look for signals that separate them. Four lines of research point the same direction.

• Piotroski’s F-score. Nine accounting signals on profitability, debt and liquidity, and operating efficiency. Among cheap stocks from 1976 to 1996, buying the high scorers raised the market-adjusted return by at least 7.5% a year, and high minus low earned 23%. The gains were concentrated in small and mid-sized firms, thinly traded firms and firms without analyst coverage.² Walkshäusl found high-score firms beat low-score firms by about 10% a year in developed markets outside the US and in emerging markets from 2000 to 2018.¹⁴
• Gross profitability. Novy-Marx found that gross profits relative to assets predicted returns about as well as book-to-market from 1963 to 2010, and the two were negatively correlated (−0.57). A 50/50 mix of value and profitability had a Sharpe ratio of 0.85, against 0.34 for the market. He framed the goal as avoiding stocks that are “more unprofitable than cheap.”¹⁵ Our RMW guide covers the profitability factor.
• Quality. Asness, Frazzini and Pedersen found that high-quality stocks trade at higher prices, but not by enough to offset their higher returns, in the US and 24 countries. Their working paper described quality as a complement to value that helps “avoid the ‘value trap,’ namely the trap of buying securities that look cheap but deserve to be cheap.”¹⁶ The QMJ guide shows the factor has been close to flat since 2016.
• Size with junk removed. Small firms tend to be lower quality. Controlling for quality made the size premium more stable over time and across 24 markets, and less concentrated in microcaps and in January.¹⁷

The limits matter as much as the findings. McLean and Pontiff studied 97 published return predictors and found returns 26% lower out-of-sample and 58% lower after publication.¹⁸ The F-score and gross profitability are both widely known now, and funds trade on them.

Our own sorts could not add much here. Cheap, highly profitable companies are too rare. Among small stocks in French’s 2x4x4 sort, the cheap and most profitable quarter returned 17.8% a year from 1963 to 2026, against 13.4% for the cheap and least profitable quarter, and it came out ahead in 70% of rolling ten-year windows. That portfolio averaged 41 firms and at times held one, so treat it as consistent with the research above and not as separate proof. The same cell among large stocks averaged eight firms, too few to read anything into.³

Screening has a cost of its own. Filter hard enough for clean balance sheets and steady earnings and you remove most of the cheap stocks, and the value exposure goes with them. Value and quality work together because they measure different things.

Book value and intangibles

Book-to-market has a second trap built into its denominator. Accounting rules expense most internally generated research, software, brands and training, so the balance sheet of an intangible-heavy firm understates the capital it runs on. A software company can look expensive on price-to-book because its investment never became an asset. An industrial firm can look cheap because its recorded assets are worth less than their book value.

Researchers disagree about how much this explains value’s 2007–2020 slump. HML fell 57.8% from December 2006 to September 2020 and was still 32% below that peak in August 2026, despite averaging 8.6% a year from January 2021 through August 2026.³ Arnott, Harvey, Kalesnik and Linnainmaa found that the widening valuation gap between cheap and expensive stocks accounted for “well over 100% of the cumulative shortfall,” and that the rate at which cheap stocks migrate to growth was essentially unchanged. Adding intangibles to book value improved HML by about 2.2 percentage points a year after 2008 in their tests, a partial fix.¹⁹ Lev and Srivastava put more weight on accounting that misidentifies value and on slower migration between value and glamour.²⁰ The two papers disagree on migration, and both are plausible readings of one period.

Either way, a single ratio built on one accounting number is a weak definition of cheap. Most systematic value funds now combine several measures, such as earnings, cash flow, sales and book value, and rank stocks against each other instead of using a fixed cutoff.

What a value trap looks like

None of these signs means a stock should be avoided. The market sees most of them too, and a heavily indebted cyclical company with falling earnings can still be underpriced. They are the common ways a multiple misleads.

Checking one company against this list is what the stock valuation guide calls asking what has to be true for the price to be right. It is worth doing. It is also what professional analysts do all day, and it is why picking individual cheap stocks is hard to do better than they do.

Small-cap value and small-cap junk

Value traps concentrate among small companies. Small firms have less analyst coverage, thinner trading and a wider spread of business quality, and the gains from Piotroski’s screen and the mispricing that Griffin and Lemmon found were both largest there.²¹³ A fund that pushes toward the smallest, cheapest stocks with no other filter can end up holding a lot of small-cap junk.

This is the main difference among small-cap value funds. An index fund such as VBR sorts on valuation and size. Avantis and Dimensional add a profitability measure. The VBR vs. AVUV guide compares what the screens cost and what they deliver, and the small value vs. small growth guide shows how badly the junkiest small stocks, tiny growth companies, have done since 1927.

When a value tilt fits

No law says value must beat growth. Fama and French found the US value premium was lower from 1991 to 2019 than from 1963 to 1991, and that the difference was too noisy to call statistically reliable either way.²¹ A market-cap-weighted index fund is a complete portfolio. A value tilt is an optional bet on a long-run premium, and it comes with years of trailing the market.

The HML guide’s Value Tilt Pain Test shows how far a tilted portfolio can fall behind, and Is Factor Investing Dead? covers how large a tilt the expected premium can justify after costs.

What I recommend

I prefer a diversified value tilt to buying individual stocks because they look cheap. A single cheap stock asks you to be right about one company’s future. The evidence above says most cheap companies disappoint, a few recover sharply, and the market has sometimes priced the group too low. A diversified fund collects that average. A stock picker has to find the few, and professionals with more data are competing to do the same.

For investors who want the tilt, I prefer funds that combine several valuation measures with a profitability screen, keep costs low and hold hundreds of names. I would keep the allocation modest enough to hold through a drawdown like 2007–2020. For investors who do not want it, a total-market index fund is a sound choice, and I would not push anyone into a tilt they are likely to abandon.

When someone says a stock is cheap for a reason, the answer is almost always yes. The question for a value investor is whether the price has already paid for the reason, and diversification is how a systematic fund copes with not knowing the answer for any single company.

Key Takeaways

• Nearly every value stock is cheap for a reason. High book-to-market firms are less profitable for years before and after they are sorted. The price discount records that.
• Value traps are a stock-level problem. Fewer than 44% of Piotroski’s cheap stocks beat the market over two years, and the group still earned a premium. The cheapest, least profitable fifth of US stocks returned 13.1% a year from 1963 to 2026, against 10.9% for the market, with no five-factor alpha.
• Distress has not been rewarded. Highly distressed stocks have earned low returns since the 1980s, and expensive, unprofitable stocks returned 3.4% a year from 1963 to 2026, below Treasury bills.
• Profitability and quality help select among cheap stocks. The F-score, gross profitability and QMJ all point that way, with the caveat that published signals weaken after publication.
• Value is best owned as a diversified, long-term allocation. It can trail for more than a decade, as it did from 2007 to 2020, and it is optional.

Frequently Asked Questions

What is a value trap?

A stock that looks inexpensive on a measure such as P/E, price-to-book or dividend yield, where the business then deteriorates enough that the low price was justified or too high. The multiple was low because the earnings, assets or dividend it was based on were about to shrink.

Are value index funds full of value traps?

Yes, and they are built to tolerate them. In Piotroski’s sample, most cheap stocks trailed the market over the next two years. A value fund holds hundreds of them, and its return depends on the average outcome, which has historically been positive because a minority recover or are acquired.

How can I tell whether a single stock is a value trap?

Not reliably. Signals such as rising debt, falling margins, earnings that exceed cash flow, and an industry in structural decline raise the odds. Piotroski’s nine-point F-score formalizes several of them. None of these signals is secret, though. Across 97 published return predictors, returns were 58% lower after publication on average.

Does a profitability screen avoid value traps?

It lowers the share of weak businesses in a value portfolio, and the research supports combining value with profitability. Some traps will still get through, and a screen strict enough to exclude every struggling company would also remove most of the value exposure.

Is distress risk the reason value stocks earn more?

The evidence says no. Dichev and Campbell, Hilscher and Szilagyi found that the most distressed stocks earned low returns despite their higher risk, and Griffin and Lemmon found the lowest returns among distressed stocks that were also expensive.

Should I buy value after a long stretch of underperformance?

Only if it fits your plan anyway. HML fell for almost 14 years, from December 2006 to September 2020, and the bottom was only visible afterward. If you hold value, hold a fixed allocation and rebalance to it, instead of adding after bad years or cutting after good ones.

Related Guides

• Fama-French HML Explained for the value factor’s construction, record and the Value Tilt Pain Test.
• The Case for Small-Cap Value for the long-run small value evidence.
• RMW Explained for the profitability factor and how Avantis and Dimensional use it.
• Quality Minus Junk for the quality factor and which funds already hold it.
• VBR vs. AVUV for the cost and benefit of a profitability screen in small value.
• Stock Valuation for DIY Investors for valuing a single company and why it rarely justifies stock picking.
• Most Stocks Lose to T-Bills for the base rate behind single-stock outcomes.
• Conviction vs. Diversification for the arithmetic of concentrated portfolios.

Sources

1. Eugene F. Fama and Kenneth R. French, “Size and Book-to-Market Factors in Earnings and Returns,” Journal of Finance 50(1), 1995, 131–155, p. 132. doi.org
2. Joseph D. Piotroski, “Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers,” Journal of Accounting Research 38 (Supplement), 2000, 1–41. Tables 1 and 3. doi.org
3. Kenneth R. French Data Library: 25 portfolios formed on book-to-market and operating profitability; 32 portfolios formed on size, book-to-market and operating profitability; Fama/French 3 and 5 factors. Files built from the 202608 CRSP database. Author’s calculations, script and data.
4. Eugene F. Fama and Kenneth R. French, “Migration,” Financial Analysts Journal 63(3), 2007, 48–58. doi.org
5. John Y. Campbell and Tuomo Vuolteenaho, “Bad Beta, Good Beta,” American Economic Review 94(5), 2004, 1249–1275. doi.org
6. Martin Lettau and Jessica A. Wachter, “Why Is Long-Horizon Equity Less Risky? A Duration-Based Explanation of the Value Premium,” Journal of Finance 62(1), 2007, 55–92. doi.org
7. Josef Lakonishok, Andrei Shleifer and Robert W. Vishny, “Contrarian Investment, Extrapolation, and Risk,” Journal of Finance 49(5), 1994, 1541–1578. doi.org
8. Joseph D. Piotroski and Eric C. So, “Identifying Expectation Errors in Value/Glamour Strategies: A Fundamental Analysis Approach,” Review of Financial Studies 25(9), 2012, 2841–2875. doi.org
9. Eugene F. Fama and Kenneth R. French, “A Five-Factor Asset Pricing Model,” Journal of Financial Economics 116(1), 2015, 1–22. doi.org
10. Kewei Hou, Chen Xue and Lu Zhang, “Digesting Anomalies: An Investment Approach,” Review of Financial Studies 28(3), 2015, 650–705. doi.org
11. Ilia D. Dichev, “Is the Risk of Bankruptcy a Systematic Risk?” Journal of Finance 53(3), 1998, 1131–1147. doi.org
12. John Y. Campbell, Jens Hilscher and Jan Szilagyi, “In Search of Distress Risk,” Journal of Finance 63(6), 2008, 2899–2939. doi.org
13. John M. Griffin and Michael L. Lemmon, “Book-to-Market Equity, Distress Risk, and Stock Returns,” Journal of Finance 57(5), 2002, 2317–2336. doi.org
14. Christian Walkshäusl, “Piotroski’s FSCORE: International Evidence,” Journal of Asset Management 21, 2020, 106–118. doi.org
15. Robert Novy-Marx, “The Other Side of Value: The Gross Profitability Premium,” Journal of Financial Economics 108(1), 2013, 1–28. doi.org
16. Clifford S. Asness, Andrea Frazzini and Lasse Heje Pedersen, “Quality Minus Junk,” Review of Accounting Studies 24(1), 2019, 34–112. The value-trap passage is from the October 2013 working paper. doi.org
17. Clifford S. Asness, Andrea Frazzini, Ronen Israel, Tobias J. Moskowitz and Lasse Heje Pedersen, “Size Matters, If You Control Your Junk,” Journal of Financial Economics 129(3), 2018, 479–509. doi.org
18. R. David McLean and Jeffrey Pontiff, “Does Academic Research Destroy Stock Return Predictability?” Journal of Finance 71(1), 2016, 5–32. doi.org
19. Robert D. Arnott, Campbell R. Harvey, Vitali Kalesnik and Juhani T. Linnainmaa, “Reports of Value’s Death May Be Greatly Exaggerated,” Financial Analysts Journal 77(1), 2021, 44–67. doi.org
20. Baruch Lev and Anup Srivastava, “Explaining the Recent Failure of Value Investing,” Critical Finance Review 11(2), 2022, 333–360.
21. Eugene F. Fama and Kenneth R. French, “The Value Premium,” Review of Asset Pricing Studies 11(1), 2021, 105–121. doi.org