Beneish M-Score: Definition, Formula, Calculation, Screener [+Template]

Recently, Finbox users have been asking us to add a metric called the Beneish M-Score to help them quickly identify companies that are likely manipulating their earnings reports. So I decided to investigate this financial ratio and its statistical foundation. I then translated the concepts described in the paper into a spreadsheet financial model.

Note there is a lot of misinformation on this topic around the web and in some of the top search results on Google. To be sure, I reconciled all the calculations I use in the post with the Beneish M-Score calculation posted by Dr. Beneish and the Sunbeam example provided.

You can view and download the model for free using the links below. Note the "Linked" tab will allow you to calculate this score with the latest data for any company using the Finbox Spreadsheet Add-on. You just need to have the Finbox Excel or Google Spreadsheets add-on installed correctly.

[ View Spreadsheet Calculator ]: To make a copy of the Google Spreadsheet, go to File > Make a Copy

[ Download Excel Spreadsheet Calculator ]

What Is The Beneish M-Score

I was happy to find that professor Messod D. Beneish at my Alma mater, Indiana University, Kelley School of Business, invented the score. In his now famous paper, Dr. Beneish sets out ambitiously to "profile a sample of earnings manipulators" and identify the characteristics that distinguish them from their public peers. He then develops and outlines a statistical model to detect such manipulation by proposing eight key variables designed to "capture either the effects of manipulation or preconditions that may prompt firms to engage in such activity."

It's important to note, just because a company has been flagged by the model as an earnings manipulator does not necessarily mean that it's true. Like any statistical model, the Beneish M-Score is subject to false positives. The inverse is also true — the model does not identify all earnings manipulators. In Dr. Beneish's holdout sample tests, the model identified approximately half of the companies involved in earnings manipulation before the public market realized. Considering the catastrophic consequences of owning stock in a company that's involved in accounting trickery can have on a portfolio, the model's hit rate is very impressive. After all, some bright Cornell students used the Beneish M-Score to identify Enron Corporation as an earnings manipulator correctly.

Beneish M Score Formula, Methodology & Calculation Details

So let's get into the methodology and calculation steps. The paper outlines eight financial ratios required to calculate the score:

1. Days Sales in Receivables Index (DSRI): DSRI is the ratio of days sales in receivables in the first year in which earnings manipulation is uncovered (year t) to the corresponding measure in year t-1. This variable gauges whether receivables and revenues are in or out-of-balance in two consecutive years. Dr. Beneish explains, "I thus expect a large increase in days sales in receivables to be associated with a higher likelihood that revenues and earnings are overstated."
   
   

   DSRI = (Net Receivables_t / Sales_t) / (Net Receivables_t-1 / Sales_t-1)

2. Gross Margin Index (GMI): GMI is the ratio of the gross margin in year t-1 to the gross margin in year t. It appears Dr. Beneish is building on the work of Lev and Thiagarajan (1993) here that suggested gross margin deterioration is a negative signal about firms' prospects.
   
   

   GMI = [(Sales_t-1 - COGS_t-1) / Sales_t-1] / [(Sales_t - COGS_t) / Sales_t]

3. Asset Quality Index (AQI): Asset quality in a given year is the ratio of non-current assets other than property plant and equipment (PPE) to total assets and measures the proportion of total assets for which future benefits are potentially less certain.
   
   

   AQI = [(Total Assets - Current Assets_t - PP&E_t) / Total Assets_t] / [(Total Assets - Current Assets_t-1 - PP&E_t-1) / Total Assets_t-1]

4. Sales Growth Index (SGI): SGI is the ratio of sales in year t to sales in year t-1. I was surprised to find this variable included since investors typically love growth. Dr. Beneish notes, that while growth does not imply manipulation, management at growth companies are more likely to face investor pressure to achieve earnings targets.
   
   

   SGI = Sales_t / Sales_t-1

5. Depreciation Index (DEPI): DEPI is the ratio of the rate of depreciation in year t-1 vs. the corresponding rate in year t. When DEPI is greater than 1, it indicates that the rate at which assets are depreciated has slowed down. This can signal that aggressive accountants may be revising estimates of assets useful lives upwards or adopting a new depreciation method to reduce expenses.
   
   

   DEPI = (Depreciation_t-1/ (PP&E_t-1 + Depreciation_t-1)) / (Depreciation_t / (PP&E_t + Depreciation_t))

6. Sales General and Administrative Expenses Index (SGAI): SGAI is calculated as the ratio of SGA to sales in year t relative to the corresponding measure in year t-1. Dr. Beneish again builds on the work of Lev and Thiagarajan (1993) here, assigning a positive relationship between increasing profit margins and the probability of manipulation.
   
   

   SGAI = (SG&A Expense_t / Sales_t) / (SG&A Expense_t-1 / Sales_t-1)

7. Leverage Index (LVGI): LVGI is the ratio of total debt to total assets in year t relative to the corresponding ratio in year t-1. An LVGI greater than 1 indicates an increase in leverage. By including this variable, the model captures incentives for earnings manipulation to meet debt covenants.
   
   

   LVGI = [(Current Liabilities_t + Total Long Term Debt_t) / Total Assets_t] / [(Current Liabilities_t-1 + Total Long Term Debt_t-1) / Total Assets_t-1]

8. Total Accruals to Total Assets (TATA): Finally, Dr. Beneish total accruals to total assets as a proxy for cash earnings underlying the reported earnings. Higher positive accruals (less cash) can logically be associated with a higher likelihood of earnings manipulation.
   
   

   TATA = (Income from Continuing Operations_t - Cash Flows from Operations_t) / Total Assets_t

The Beneish M-Score

After computing the eight variables outlined above, they can be weighted together using the following multivariate model to calculate the score:

M-Score = 
-4.84 
(+) 0.92 × DSRI 
(+) 0.528 × GMI 
(+) 0.404 × AQI 
(+) 0.892 × SGI 
(+) 0.115 × DEPI 
(+) -0.172 × SGAI 
(+) 4.679 × TATA 
(+) -0.327 × LVGI

Here are the summary stats from the paper:

When an M-Score is greater than -1.78 (e.g -1.1), the model flags the company as a likely earnings manipulator. If of interest, you can also use the score to determine the probability of manipulation using this free calculator. Just enter the value of the Beneish M-Score in the "Calculate probability Q from z" section.

So in summary:

• When the Beneish M-Score is greater than -1.78, the company is likely an earnings manipulator
• When the Beneish M-Score is between -1.78 and -2.00, the company is a possible earnings manipulator
• When the Beneish M-Score is less than -2, the company is not likely an earnings manipulator

Beneish M-Score Stock Screener

We've done the heavy lifting of calculating the Beneish M-Score score for every company and keep an updated dataset. You can look up the score for any company using the Data Explorer tool: See Apple's Score

To find stocks that are flagged by the Beneish M-Score, you can use the Finbox Stock Screener tool. Add it as a filter in the screener, and you'll get the latest candidates.