Predicting the Next Enron

Via the Wall Street Journal, here’s further evidence that companies “tweak” quarterly earnings numbers. Joseph Grundfest and Nadya Malenko analyzed almost half a million earnings reports from 1980-2006. They discovered that when companies want to appear more successful than they are, they often massage their per-share earnings numbers upward by a tenth of one cent. The evidence? The number 4 appears significantly less often than expected in the post-decimal digits of earnings reports. In the U.S., per-share earnings are reported as pennies, so bumping that post-decimal digit from a 4 to a 5 results in the overall number being rounded up by a full penny. Grundfest and Malenko call the practice quadrophobia. While the tweaking may be legal in some cases, the authors also found that “quadrophobes are more likely to restate financials and to be named as defendants in SEC Accounting and Auditing Enforcement Releases (AAER).” Or, as Grundfest told the Journal, quadrophobia serves as “a leading indicator of a company that’s going to have an accounting issue.”[%comments]


Very interesting approach - I like the intuition they used behind the analysis. And having worked in corporate finance I don't doubt for a second that their findings are accurate.


Am I the only one who found it hilarious that "tweaking" is a gateway to harsher financial crime? Haha, Come on, there's got to be other underage Freakonomics readers.


And anyone wonder why the accountants get highly paid for those menial jobs?


Wha???? numbers get tweaked to make the finances look better? Call me Captain Renault: "I'm shocked, shocked to find that gambling is going on in here! "

Christopher Strom

I am not at all surprised either by the "tweaking" (I laughed at the word choice as well, Jib) or by the methods used to detect it.

The practice is not limited to finance.

In my engineering career, I have discovered enough instances of data massaging (to justify funding someone's pet project), that when I suspect that a project is being driven by factors other than "getting the job done", I often check the raw data for eveidence of tampering using methods similar to those used by Grundfest and Malenko.

Fortunately, most people who tamper with data do a very ham-fisted job of it, leaving evidence that anyone even modestly versed in statistical analysis can quickly detect.

What to do with such knowledge is another matter altogether...

Eric M. Jones

Hey, someday, somone is going to stand up and say, "That's illegal! People are getting degrees in business and finance so they can join in organized crime." When an emergency session of congress some years back deleted the "Loan Sharking Statutes"--this should have been a clue to honest people that the system broke.

Definition: " Certified Public Accountant", i.e., a person skilled at altering or destroying documents, ignoring or failing to investigate shell companies created by insiders who grotesquely enriched themselves while hiding mounting corporate debt in "off-balance-sheet companies"; Ignoring knowledgeable whistleblowers and accounting "red flags" that indicate massive fraud is taking place; misleading investors who continued pouring their money into failing companies. See "Enron".

Yes #2--Jib, I am non-plussed, in fact (expletive deleted).
I am now fighting with my bank AND Western Union to recover $806 stolen from my bank account by Westen Union and sent to ...oh I don't know...some anonymous terrorist overseas. WU won't even tell me WHERE they sent it.

But nobody is responsible....Hah. Perp-walk these people.



And this is also why a company's share price will dive if they miss estimates by even a cent or two. It is assumed that if they couldn't achieve their targets even with all the creative accounting techniques at their disposal, then they're really in the dog house.


when searching for a car here in Austria/Europe I found loads of cars with an offiicial average fuel consumption of x.7 liters/100km. Why seems obvious to me: We have to pay a tax (NOVA) based on round(2*average;0). Adversely there were less cars with x.8 liters...

Jack M

this is very late in coming, but in Chinese 4 is bad luck because it sounds the same as death. Clearly, this is an influence...