Corporate governance chart of the day, Benford’s Law edition

By Felix Salmon
October 12, 2011
Jialan Wang, and it shows the degree to which companies' reported assets and revenues deviate from a Benford's Law prediction over time.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">


This chart was put together by Jialan Wang, and it shows the degree to which companies’ reported assets and revenues deviate from a Benford’s Law prediction over time. (If you want some good background on Benford’s Law and how it can uncover dodgy numbers from eg the Greek government, Tim Harford had a great column last month on the subject.)

Writes Wang:

Deviations from Benford’s law have increased substantially over time, such that today the empirical distribution of each digit is about 3 percentage points off from what Benford’s law would predict. The deviation increased sharply between 1982-1986 before leveling off, then zoomed up again from 1998 to 2002. Notably, the deviation from Benford dropped off very slightly in 2003-2004 after the enactment of Sarbanes-Oxley accounting reform act in 2002, but this was very tiny and the deviation resumed its increase up to an all-time peak in 2009.

So according to Benford’s law, accounting statements are getting less and less representative of what’s really going on inside of companies. The major reform that was passed after Enron and other major accounting standards barely made a dent.

This doesn’t necessarily mean fraud, per se; it could just be a chart of the degree to which companies are managing and massaging their quarterly figures over time. The kind of fraud that’s so respectable, Jack Welch got lionized for it. Once you start down that road, it’s easy to go further and further forwards, while it’s almost impossible to reverse course. So I can easily see how the natural tendency in this chart would be up and to the right.

Still, it’s worrying; all the more so because I can’t think of any way of reversing the trend. If Sarbox can’t do it, nothing will.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

“If Sarbox can’t do it, nothing will.”

Make you a deal: You pass SarbOx. I get to appoint Harvey “I fought for these guys so certainly I’ll do everything I can to keep them honest” Pitt, Bill “Made my fortune selling these guys” Donaldson and Chris “never met a crooked businessman I didn’t like” Cox as heads of the SEC.

Followed by Mary “We should deregulate everything so I don’t have to do jack at the CFTC and did I mention I created and ran FINRA” Schapiro.

You spend the next several years waiting for discoveries of fraud. I get to steal to my heart’s content (and then some), even after three years of turmoil produce blatant evidence of fraud (which my SEC gets to decide means nothing, and destroys).

It’s win-win, #OWS style.

Posted by klhoughton | Report as abusive

To the extent that the chart reflects “Welch managed accounting”, the root cause would seem to be the dramatic change over the time from a market based on “investment” to one that is built for “speculation”. We need to address the root cause to cure the sympton – and then identify and prosecute the real fraud.

Posted by G2G | Report as abusive

This is under the assumption that Benford’s law will always be correct. If, due to other reasons, the reporting of accounting figures changes such that it is no longer correct, no fraud or ‘massaging’ is necessary

Posted by gleisdreieck | Report as abusive