X-Ray Vision: The M-Score Test to Detect Fraud

In a prior post I discussed using the Altman Z-Score to predict business bankruptcy. Today, I highlight another analytical tool but this time we review one whose purpose is to help detect earnings manipulation.

The Beneish M-Score

The Beneish M-Score was developed in 1990s by Messod Beneish, then an Associate Professor at Indiana University. Interestingly, the Beneish model was one of the tools used by a group of six Cornell students who accurately concluded that Enron might be manipulating its earnings before its collapse.

Mathematically, the M-Score consists of eight ratios to capture financial statement distortions and uses two years worth of data for comparison. These ratios focus on financial statement distortions that can result from manipulating earnings and consist of the following indices:

1. Days sales outstanding in receivables (DSRI)
2. Gross margin (GMI)
3. Asset quality, as measured by non-current non-PP&E assets divided by total assets (AQI)
4. Sales growth (SGI)
5. Rate of depreciation, as measured by depreciation expense divided by depreciation plus net PP&E (DEPI)
6. The ratio of SG&A expense to revenues (SGAI)
7. The ratio of accruals to total assets, as measured by working capital excluding cash less depreciation divided by total assets (TATA)
8. Lastly, total leverage, or debt divided by assets (LVGI)

The M-Score formula that Professor Beneish developed based on his sample study of 50 known earnings manipulators and 1,708 non-manipulators with individual probability weightings for each index is:

M= 4.84+0.92*DSRI + 0.528*GMI + 0.404*AQI + 0.892*SGI + 0.115*DEPI – 0.172*SGAI + 4.679*TATA – 0.327LVGI.

An M-Score greater than -2.22 indicates that there is a high likelihood of earnings manipulation. In backwards testing, Beneish found that his formula accurately predicted earnings manipulation 76% of the time, while showing false positives 17.5% of the time.

Although the formula and its eight indices consist of a myriad of numbers, the gist of the formula’s predictive power is that changes in gross margins, and relative sales growth, depreciation expense, SG&A expenses, accruals, and non-current non-PP&E assets, can point to financial statement shenanigans.

A cautionary note: although the Beneish M-Score is a powerful predictive tool, it should not be used in a vacuum. Instead, it should be combined with other analytic techniques such as ratio analysis and correlation of net income with cash flows to corroborate the possibility of financial mis-statement.

So there you have it: an eight step test that will provide the X-ray vision needed to see through the reported  numbers to help determine if a company’s financials reflect manipulated results.