Profits Without Cash? What Cash-Based Operating Profitability Really Tells Investors

Two companies each report $100 million of operating profit. One collected most of it in cash. The other booked much of it as a rise in receivables and inventory that has not turned into cash yet. The income statement treats them as identical. A 2016 paper by Ray Ball, Joseph Gerakos, Juhani Linnainmaa, and Valeri Nikolaev shows that, historically, the stock market did not: the profit that arrived as cash predicted higher future returns, and the profit that sat in accruals did not.¹

What accruals are, and why accountants use them

Accrual accounting separates when a company earns revenue from when the cash arrives. Deliver $100 of product in December and collect in January, and accrual rules book the $100 as December revenue, because that is when the work was done. Cash accounting would wait for January. The accrual approach matches activity to the period it belongs to, which is why it is the basis of financial reporting.¹

The side effect is that two firms with identical operating profit can have generated very different amounts of cash. Suppose both report $100 of operating profit. Company A collected its sales and held inventory flat. Company B produced $40 of that profit through a rise in accounts receivable and inventory. Company B might be growing normally, funding more inventory and extending terms to win sales. It might also be collecting more slowly, letting stock pile up, or recognizing revenue aggressively. The profit figure alone cannot tell those stories apart. Cash-based operating profitability moves one step closer to the cash the business actually threw off.

The puzzle the paper set out to solve

Two findings in the prior literature looked like they were in tension. More profitable firms had earned higher average returns. Firms with higher accruals had earned lower average returns, a pattern Richard Sloan documented in 1996 and that became known as the accrual anomaly.² The awkward part is that accruals are already inside accounting profitability. How can profitability point up while one of its own components points down?

The paper’s answer is that the market was not rewarding all reported profitability equally. The part that predicted returns was the part backed by operating cash. Once you convert operating profitability to a cash basis, the separate accrual signal has almost nothing left to add. In the authors’ phrase, cash-based operating profitability subsumes the accrual anomaly.

How the measure is built

Start with operating profitability, which the authors define as revenue minus cost of goods sold minus selling, general, and administrative expense. This captures operations without the effects of leverage or taxes.

Then remove the working-capital accruals embedded in that number. An increase in an operating asset (receivables, inventory, prepaid expenses) used cash, so it is subtracted. An increase in an operating liability (deferred revenue, payables, accrued expenses) supplied cash, so it is added.

Every profitability and accrual measure is scaled by the prior year’s total assets. Across the sample the average firm ran operating profitability near 12.9% of assets, accruals near −2.9%, and cash-based operating profitability near 11.7%.¹ The calculator below walks a single company through this bridge so you can see where reported profit and cash profit diverge.

Five measures it is easy to confuse with this one

Much online writing collapses every cash-related number into “free cash flow” or treats accruals as fake earnings. Both are wrong, and the distinctions matter for reading the research correctly.

• GAAP cash flow from operations. Reported operating cash flow is net of taxes and interest, which makes it a levered measure shaped by financing and tax structure. Cash-based operating profitability deliberately excludes those.
• Free cash flow. Free cash flow subtracts capital expenditures. Cash-based operating profitability does not; it stops at operating working capital.
• A valuation ratio. This measure is scaled by assets, not divided by the stock price. A highly cash-profitable company can still be expensive. A later study that divides cash profitability by price is testing a different, valuation-flavored idea.⁷
• A fraud detector. Low cash conversion can come from ordinary growth, seasonal working capital, or a business model that simply carries more inventory. It is a prompt to investigate, not evidence of manipulation.
• A universal quality score. A company can lift cash profitability for a year by stretching supplier payments or running inventory lean, without improving the underlying economics.

What the researchers found

The tests use CRSP returns and Compustat accounting data for U.S. common stocks on the NYSE, AMEX, and Nasdaq from July 1963 through December 2014, excluding financial firms, and assume annual statements become investable six months after the fiscal year ends.¹ In the cross-sectional regressions that exclude microcaps, operating profitability carried a t-value of 7.04 and accruals −3.9, both significant. Put cash-based operating profitability in the same race and it reached 9.69. Run cash-based profitability and accruals together and cash profitability held at 7.4 while accruals fell to 0.34, statistically indistinguishable from zero. Run operating profitability against cash-based profitability directly and ordinary operating profitability dropped to 1.56 while cash-based stayed at 5.27.

Portfolio sorts told the same story without the regression assumptions. The value-weighted high-minus-low decile spread was 0.47% per month (t=3.14) for cash-based profitability, against 0.29% (t=1.84) for operating profitability and −0.39% (t=−2.55) for accruals. The cash-based strategy’s three-factor alpha was 0.89% per month (t=8.48).¹

The signal was durable as well as strong. Cash-based operating profitability kept predicting returns up to ten years ahead, and its edge over ordinary operating profitability was statistically reliable out to about four years. Accruals, by contrast, predicted returns only about one year out.¹

Why the headline Sharpe ratio is not your expected return

The paper’s most quoted result is a Sharpe-ratio comparison. Over 1963–2014 the market’s annualized Sharpe ratio was 0.39. Building the ex-post mean-variance efficient portfolio from the market, size, value, and momentum factors lifted it to 1.06. Adding the accruals factor took it to 1.12; swapping in operating profitability instead reached 1.40; adding cash-based operating profitability reached 1.67. An investor who already held cash-based profitability gained almost nothing from also adding accruals, moving from 1.67 to 1.69, whereas adding both operating profitability and accruals to the base four factors reached only 1.54. Replacing operating profitability with the cash-based version raised the implied profitability strategy’s Sharpe ratio by almost 40% (t=4.6), and in a bootstrap the cash-based measure produced the higher Sharpe ratio in 99.4% of resamples.¹

That 1.67 is genuine evidence that the signal carried independent information. It is a poor forward-looking assumption for a real portfolio. It rests on historical data, an ex-post optimizer that chose factor weights with full hindsight, long-short factor portfolios, and gross returns before taxes, trading costs, fund fees, bid-ask spreads, market impact, and the cost of borrowing stock to short. Treat the number as a measure of information content, not as a promise about your account.

Mispricing, risk, or an investment effect?

The paper establishes the pattern more firmly than it explains the cause. Three readings remain live, and they carry different implications.

Investor underreaction. Investors may lean too much on reported profit and too little on whether cash backs it, with the gap closing as weak cash generation shows up later. The authors note this differs from Sloan’s original story: under simple fixation on earnings, accruals should still predict surprises after controlling for cash profitability, and they do not.¹

Compensation for risk. Firms with high cash profitability may bear a risk investors demand to be paid for, in which case the return difference is a fair premium rather than a mistake. The ten-year persistence of the signal is at least consistent with a risk-based, slow-moving determinant.

An investment effect. Building inventory and receivables consumes cash, so cash-based profitability mechanically penalizes firms making large working-capital investments. The measure therefore carries information about growth and payment timing, not profitability alone, and can overlap the documented tendency of high-investment firms to earn lower subsequent returns. The authors say as much.¹ This last reading is where a major practitioner disagreement begins.

What later research adds

The core U.S. result has held up. A 2020 study in the Pacific-Basin Finance Journal independently replicated it: in the United States, cash-based operating profitability again subsumed both accruals and ordinary operating profitability.⁴ A 2022 decomposition study in Accounting & Finance estimated that cash-based operating profitability accounted for roughly 94% of the accrual anomaly in its framework, and found cash-flow and investment explanations more important than limits to arbitrage or information asymmetry.⁵ The signal is also catalogued in the Open Source Asset Pricing project as “CBOperProf,” where the value-weighted long-short spread is recorded at 0.47% per month with a t-value of 3.17 and the definition and replication code are open to inspection.⁶

The evidence is not uniform across markets. The same 2020 study found the pattern reversed in China: there, ordinary operating profitability subsumed the cash-based version, and the cash-based long-short spreads were not statistically significant.⁴ Accounting conventions, market structure, the investor base, and the link between working-capital growth and future performance all differ across countries, which is a reason to treat the U.S. result as evidence rather than as a law of markets.

The strongest disagreement: Dimensional’s view

Dimensional Fund Advisors examined this question directly and reached a more skeptical conclusion. Using its own definitions and controls, it argues that a cash-profitability adjustment does not reliably improve the prediction of future profitability or returns once investment is handled separately. Its reading is that much of the apparent benefit comes from avoiding small firms that combine high accruals, high asset growth, and historically weak returns. Because subtracting accruals mechanically penalizes firms investing in receivables and inventory, Dimensional treats cash profitability as partly bundling two things, profitability and investment, and prefers to keep operating profitability while controlling for investment on its own. It says it began excluding the highest-asset-growth companies from small-cap strategies around 2019.⁸

This is not a clean refutation, because Dimensional uses different profitability definitions, scaling, and portfolio construction than the paper. It is a serious counterpoint, and it means the academic conclusion should not be presented as settled.

Auditing the backtest

A signal can be statistically strong and still disappoint in a real portfolio. Several standard checks apply here.

• Microcaps. Anomalies often concentrate in tiny stocks. The authors address this by separating microcaps, which were 55% of firms but only about 3% of total market value, and the result survives in larger stocks.¹
• Long-short versus long-only. The headline spreads come from buying the best and shorting the worst. A long-only investor captures only part of that, and much of an anomaly can live in the short leg, which is expensive or impossible to hold.
• Publication decay. Across 97 predictors, returns were about 26% lower out of sample and 58% lower after publication.⁹ More broadly, Hou, Xue, and Zhang found that 65% of 452 anomalies failed a t > 1.96 hurdle under NYSE breakpoints and value-weighting, rising to 82% under a multiple-testing threshold.¹⁰ Cash-based profitability is among the sturdier findings, but the base rate for published anomalies is humbling.
• Costs, taxes, and the end date. The paper reports no net, tax-aware return for a retail portfolio, and its sample stops in 2014, before a decade of heavy factor adoption.

One paper, two implementation philosophies

The paper reaches DIY investors mostly through systematic funds, and the two best-known families read the same evidence differently.

Avantis embraces the finding directly. Its white paper “A Scientific Approach to Investing” discusses Ball et al. and the cash-based operating profitability factor, repeating the 4.88% historical premium, and states that Avantis ranks securities on a joint measure of adjusted book-to-market and cash-based profitability, then separately excludes small companies that combine a high price relative to book, low profitability, and high asset growth.¹¹ So Avantis uses a cash-based profitability proxy informed by the paper and also controls for investment, rather than relying on the accrual subtraction to do both jobs. Its public materials do not disclose the exact denominator, accrual definitions, industry adjustments, or weights, so Avantis is best described as informed by the research rather than as mechanically reproducing the academic factor.

Avantis makes its case with the same kind of evidence. In the white paper, U.S. stocks sorted from 1940 to 2023 on both profitability and book-to-market put the high-profitability, high-book-to-market corner at 1.50% average monthly return against 0.45% for the low-low corner, a gap above one percentage point a month, and Avantis reports joint value-and-profitability spreads above 1.2% per month in large caps and 1.8% in small caps across non-U.S. developed and emerging markets. It also strips goodwill out of book value, so firms that built book equity through acquisitions are not mistaken for cheap ones.¹¹ These are the sponsor’s own figures from a marketing document; treat them as Avantis explaining its reasoning.

Dimensional takes the other route. It keeps ordinary operating profitability, defined as operating income before depreciation and amortization minus interest expense, scaled by book equity, and adds a separate investment control, having argued that the cash adjustment mostly repackages an investment screen.⁸ Both approaches end up tilting toward profitable companies while managing investment exposure; they disagree on whether subtracting accruals is the better way to get there.

Morningstar’s portfolio data put numbers on the resulting tilts. As of the dates on its portfolio page, the Avantis U.S. Small Cap Value ETF (AVUV) traded at a price/book of 1.41 against a small-value category average of 1.62 and a price/cash-flow of 5.67 against 8.66, with value and quality factor loadings above its peers.¹² The Avantis All Equity Markets Value ETF (AVGV) is a fund-of-funds holding seven Avantis ETFs rather than stocks directly, so its characteristics are a blend of its underlying sleeves.¹³ These are dated, quarterly-changing snapshots, not fixed properties of the funds. For the broader RMW profitability factor and the Avantis-versus- Dimensional comparison, see our RMW guide.

Knowing that Avantis uses the Ball signal is informative, and it does not establish that Avantis funds will beat comparable Dimensional funds. Realized returns also depend on how aggressively each signal is weighted, its interaction with value and size, sector controls, turnover, tax management, securities lending, execution, and fees. Two managers can start from different profitability definitions and hold heavily overlapping portfolios.

What this means for DIY investors

If you hold broad index funds, change nothing on the strength of this paper. A globally diversified low-cost portfolio does not need every documented characteristic optimized. Use the research to understand why reported earnings and cash flow can diverge, why two “profitability” funds can own different companies, and why a strong backtest is not automatically a good allocation.

If you already own systematic value or profitability funds from Avantis or Dimensional, the message is reassuring rather than actionable. You are already getting integrated profitability exposure with an investment control, so bolting on a separate cash-profitability ETF or screen would mostly duplicate what you hold. The useful work is due diligence on the funds you own: how they define profitability, whether they scale by assets or book equity or price, whether they handle investment separately, how much sits in microcaps, and what turnover, spreads, and fees cost you.

If you pick individual stocks, use cash-based operating profitability as a diagnostic, not a decision rule. Compare it with ordinary operating profitability, find the reason for any gap in the working-capital accounts, then read it alongside capital spending, free cash flow, leverage, valuation, and several years of history for economically similar firms. A large accrual adjustment is a question worth asking, not an automatic sell.

A direct cash-profitability strategy is a poor fit for anyone who lacks point-in-time data, cannot properly handle restatements and reporting lags, would end up trading small illiquid names, cannot short cheaply, carries meaningful taxable gains, or is likely to abandon the tilt after a few years of underperformance. A professional quant can run it as one input in a broader model. Sorting free screener data once a year is a different exercise with a different success rate.

How Summitward helps

The question that matters for your money is whether adding an exposure changes your actual portfolio enough to be worth the cost. Summitward’s portfolio dashboard regresses your holdings against market, size, value, profitability, and momentum factors, so you can see whether a profitability tilt you think you have is actually there and whether a new fund would duplicate exposure you already own. When you model a factor tilt in retirement projections, run the Monte Carlo stress test with a no-premium scenario rather than wiring in the paper’s historical factor return; a plan that only works if the premium repeats is a fragile plan. And for a taxable switch, weigh the expected improvement against the capital-gains tax, spreads, and ongoing tax drag, because the hurdle to justify selling appreciated shares is higher than a better signal on paper.

Frequently asked questions

What is cash-based operating profitability in plain terms?

It is operating profit (revenue minus cost of goods sold minus SG&A) after removing the parts that have not turned into cash: increases in receivables, inventory, and prepaid expenses are subtracted, and increases in deferred revenue, payables, and accrued expenses are added. The result is scaled by the prior year’s total assets. It measures how much of a firm’s operating profit is backed by cash.

Is it the same as free cash flow?

No. Free cash flow subtracts capital expenditures and is affected by taxes and interest through operating cash flow. Cash-based operating profitability stops at operating working capital and leaves out financing and taxes, so it isolates the cash content of operations.

Should I buy the stocks with the highest cash profitability?

For almost all DIY investors, no. The paper’s returns come from diversified, frequently rebalanced, long-short portfolios and are gross of costs and taxes, with hindsight-optimized Sharpe ratios. A concentrated long-only bet on high cash-profitability names is a different and riskier proposition. A diversified fund that weighs profitability alongside price and investment is more defensible.

Do Avantis and Dimensional use this measure?

Avantis says it uses a cash-based profitability proxy informed by Ball et al., combined with adjusted book-to-market and a separate asset-growth screen. Dimensional keeps ordinary operating profitability and controls for investment separately, having argued the cash adjustment mostly repackages an investment effect. Neither publishes a formula detailed enough to reproduce its live portfolios, and neither result proves one firm will outperform the other.

Does the finding hold outside the United States?

Not uniformly. A 2020 study replicated the U.S. result but found the pattern reversed in China, where ordinary operating profitability subsumed the cash-based version and the cash-based spreads were not significant. Accounting practices and market structure differ enough across countries that the U.S. evidence should not be treated as universal.

Key takeaways

• Cash-based operating profitability removes working-capital accruals from operating profit; in the 1963–2014 U.S. sample it predicted returns better than gross profitability, operating profitability, or net income, and it absorbed the accrual anomaly (t-value 9.69 alone; accruals fell to 0.34 beside it).
• The value-weighted long-short spread was 0.47% per month and the factor earned 4.88% annualized, but these are gross, long-short, hindsight-optimized figures, so the 1.67 Sharpe ratio is not a forward-looking expectation.
• The result replicated in the United States and in an independent decomposition (about 94% of the accrual anomaly), but reversed in China, so it is strong evidence rather than a universal law.
• Dimensional argues the cash adjustment largely repackages an investment effect and prefers operating profitability plus a separate investment control; Avantis adopts a cash-based proxy plus an asset-growth screen. The disagreement is about implementation, and neither side is proven superior.
• For most investors the lesson is a habit, not a trade: ask how much of a company’s or fund’s profitability is backed by cash, and weigh it with price, investment, diversification, and costs. Holders of broad systematic value funds already get this exposure.

Related guides

• RMW Explained: The Profitability Factor the broader profitability factor and the Avantis-versus-Dimensional comparison in depth.
• Is Factor Investing Dead? how premiums behave out of sample and after publication.
• Fama-French Factor Analysis what your portfolio is actually loading on, profitability included.
• The Case for Small-Cap Value decades of evidence and years of pain in one of the funds that uses profitability.
• Why I Like AVUV, AVDV, and AVGV turning factor research into a real Avantis portfolio.
• Compensated vs. Uncompensated Risk why some risks pay and diversifiable ones do not.

Sources

1. Ball, R., Gerakos, J., Linnainmaa, J. T., & Nikolaev, V. (2016). Accruals, Cash Flows, and Operating Profitability in the Cross Section of Stock Returns. Journal of Financial Economics 121(1), 28–45 (SSRN 2587199). All paper statistics quoted here are from the working paper and published version.
2. Sloan, R. G. (1996). Do Stock Prices Fully Reflect Information in Accruals and Cash Flows About Future Earnings? The Accounting Review 71(3), 289–315. The original accrual anomaly.
3. Novy-Marx, R. (2013). The Other Side of Value: The Gross Profitability Premium. Journal of Financial Economics 108(1), 1–28. Background on profitability as a return predictor.
4. Du, Q., Wang, Y., & Wei, K. C. J. (2020). Does cash-based operating profitability explain the accruals anomaly in China? Pacific-Basin Finance Journal 61, 101336. U.S. replication; reversed result in China.
5. Hu, G., et al. (2022). Have existing theories explained the accrual anomaly? An evaluation based on the decomposition method. Accounting & Finance, 10.1111/acfi.12900. Cash-based operating profitability explains roughly 94% of the accrual anomaly in its framework.
6. Chen, A. Y., & Zimmermann, T. Open Source Asset Pricing, signal “CBOperProf.” SignalDoc.csv. Value-weighted long-short 0.47%/month, t=3.17.
7. Wang, B. (2024). A New Value Strategy. Review of Asset Pricing Studies 14(1), 40–83. Cash-based operating profitability divided by price (COP/P); value-weighted long-short about 0.78%/month.
8. Dimensional Fund Advisors (2020). Do Accruals Adjustments Help Capture the Profitability Premium? dimensional.com. The skeptical practitioner view and DFA’s profitability definition.
9. McLean, R. D., & Pontiff, J. (2016). Does Academic Research Destroy Stock Return Predictability? Journal of Finance 71(1), 5–32. Returns 26% lower out of sample, 58% lower post-publication, across 97 predictors.
10. Hou, K., Xue, C., & Zhang, L. (2020). Replicating Anomalies. Review of Financial Studies 33(5), 2019–2133. 65% of 452 anomalies fail t > 1.96; 82% under a multiple-testing hurdle.
11. Avantis Investors / American Century Investments. Our Scientific Approach to Investing (white paper). americancentury.com (PDF). Cites Ball (2016) and states Avantis uses a cash-based operating profitability proxy plus adjusted book-to-market; a marketing communication.
12. Morningstar. Avantis U.S. Small Cap Value ETF (AVUV), Portfolio. morningstar.com. Portfolio metrics as of the dates shown on the page (mid-2026).
13. Morningstar. Avantis All Equity Markets Value ETF (AVGV), Portfolio. morningstar.com. AVGV holdings and fund-of-funds structure.

Editor’s note

Educational content, not investment advice, and not a recommendation of any fund. Statistics from the paper were verified against the SSRN working paper (abstract 2587199) and its Journal of Financial Economics publication; other figures against the cited journals, the Open Source Asset Pricing project, Dimensional and Avantis materials, and Morningstar, as of July 2026. Fund characteristics change every quarter; the Avantis white paper is a marketing communication, and named funds are cited to illustrate implementation choices, not as recommendations.