The Beneish M-Score: spotting numbers that look too good
The M-Score compares this year’s financial statements with last year’s and asks whether the numbers have shifted in the ways that companies caught bending their results often showed. This lesson explains each of its eight clues in one sentence, works part of the math for a small bike company, and shows why a high score is a reason to read closely, not proof of anything.
Why profits can be bent
A company’s profit isn’t counted like coins in a jar. Accountants make judgment calls: when a sale counts, how long a machine will last, how many customers won’t pay. Most of these calls are honest. But each one gives managers a little room, and a manager under pressure can use that room to make a year look better than it was.
Messod Beneish, an accounting professor, published the M-Score in 1999. He studied companies that had been caught manipulating their earnings and compared them with companies that hadn’t. He found eight ratios that tended to look different in the caught group, then weighted them into one formula.
The eight clues
Seven of the clues are an index: this year’s ratio divided by last year’s. An index of 1.0 means nothing changed. Above 1.0 usually means the number moved in the direction the clue worries about.
- DSRI (receivables index). Are unpaid customer bills growing faster than sales? That can mean sales were booked too early or made to customers who may never pay.
- GMI (gross margin index). Is the profit on each product shrinking? A squeezed company has more reason to dress up its results.
- AQI (asset quality index). Is a bigger share of the assets made of things you can’t touch or easily sell? That can mean ordinary costs were recorded as assets to keep them off this year’s profit.
- SGI (sales growth index). How fast are sales growing? Growth isn’t wrong, but keeping up a fast pace creates pressure to fill gaps creatively.
- DEPI (depreciation index). Is the company now spreading equipment costs over more years? Slower depreciation lifts profit without any real improvement.
- SGAI (overhead index). Are selling and office costs rising faster than sales? It’s a sign the business is getting less efficient.
- LVGI (leverage index). Is the company leaning more on borrowed money? Loan agreements often require certain results, which adds pressure.
- TATA (accruals to assets). How much of this year’s profit isn’t backed by cash? It’s net income minus operating cash flow, divided by total assets. This one is a plain ratio, not an index, so zero means profit and cash matched.
The formula, once
Here is how the eight clues are weighted:
M = −4.84 + 0.920 × DSRI + 0.528 × GMI + 0.404 × AQI + 0.892 × SGI + 0.115 × DEPI − 0.172 × SGAI + 4.679 × TATA − 0.327 × LVGI
Two things stand out. TATA carries by far the largest weight, so profit that isn’t backed by cash moves the score the most. And SGAI and LVGI have minus signs, so a rise in them nudges the score down a little. The weights came from fitting past cases, and not every clue behaved the way you might guess.
A handy anchor: a company where nothing changed (every index at 1.0, cash matching profit) scores about −2.48.
Reading the score
The scores are negative numbers, which takes a moment to get used to. Closer to zero is more worrying. So −1.5 is a warning, while −3.0 looks clean. A Filing.fyi card reads:
| M-Score | Card reading |
|---|---|
| Above −1.78 | High manipulation likelihood |
| Between −2.22 and −1.78 | Elevated |
| Below −2.22 | Clean |
Pebble Creek Bikes Inc. is the made-up bike maker from the statements lessons. Let’s compute four of its clues by hand.
Step 1, DSRI. Receivables as a share of sales: this year $250,000 ÷ $2,000,000 = 0.125. Last year $150,000 ÷ $1,500,000 = 0.10. Index: 0.125 ÷ 0.10 = 1.25. For each dollar of sales, the amount still unpaid rose by a quarter.
Step 2, SGI. $2,000,000 ÷ $1,500,000 = 1.33. Sales grew by a third.
Step 3, GMI. Last year’s gross margin over this year’s: 42% ÷ 40% = 1.05. Above 1.0 means the margin shrank.
Step 4, TATA. ($120,000 net income − $20,000 operating cash flow) ÷ $2,000,000 = 0.05. Most of the profit didn’t arrive as cash.
Step 5, the rest. The other four clues use numbers from the earlier lessons (equipment, depreciation, overhead and debt). None of them moved much:
| Clue | Value |
|---|---|
| DSRI | 1.25 |
| GMI | 1.05 |
| AQI | 0.88 |
| SGI | 1.33 |
| DEPI | 0.95 |
| SGAI | 1.00 |
| TATA | 0.05 |
| LVGI | 1.03 |
Step 6, the total. Plugging all eight into the formula gives about −1.76. That’s a hair above −1.78, so the card would read High manipulation likelihood.
Notice where the push came from: fast sales growth, slow-paying customers and thin cash flow. Is Pebble Creek cooking its books? Nothing here says so. It looks like a growing company whose customers pay slowly. But the score tells you exactly where to look next: the note on receivables (who owes the money, and for how long) and the cash flow statement.
A flag, not a verdict
A high M-Score means the numbers moved in ways that often came before trouble in the past. It does not mean anyone broke a rule. Common innocent reasons for a high score:
- Fast growth. Young, fast-growing companies sell more on credit and build up stock to meet demand, which pushes up DSRI, SGI and TATA. Pebble Creek is a good example.
- Buying another company. An acquisition adds receivables, debt and hard-to-touch assets overnight.
- Banks and insurers. Their statements don’t fit ratios like gross margin, so their scores mean little.
The opposite error happens too. A determined fraud can fake cash or hide debt in ways these ratios never see, so a clean score is not a guarantee either.
Where to see it on Filing.fyi
Each company page has a Beneish M card with the score and its reading: Clean, Elevated or High manipulation likelihood. If you see Elevated or High, open the latest 10-K from the filing timeline and read the notes on revenue and receivables, plus the cash flow statement. The score also feeds the page’s health score; see how to read a company page. For background on the numbers involved, see the cash flow lesson and financial ratios.
- Earnings manipulation
- Using accounting choices, or breaking the rules, to make profit look better than it really was.
- Index
- This year’s ratio divided by last year’s; 1.0 means no change.
- Receivables
- Money customers owe for things they already received.
- Accruals
- The part of profit that comes from accounting entries rather than cash collected.
- Depreciation
- Spreading the cost of equipment over the years it will be used.
- False positive
- A warning that goes off when nothing is actually wrong.
- The M-Score compares two years across eight clues that past manipulators often shared.
- Closer to zero is more worrying: above −1.78 is high, below −2.22 is clean.
- Profit that isn’t backed by cash moves the score the most.
- Fast-growing companies often trip it innocently.
- A high score is a reason to read the filing closely, not proof of fraud.
This lesson is education, not investment advice. Examples use made-up companies and round numbers unless a real filing is named.