« Should we consider a gold standard? | Main | Winter reading »
Robbery fact of the day
In 2004, the total cost of all robberies in the United States was $525 million...every year, employees' theft and fraud at the workplace are estimated at about $600 billion.
That contrast is from Dan Ariely's Predictably Irrational. Even if you add in auto theft, burglary, and larceny-theft, the former sum does not exceed $16 billion.
Addendum: Some people are questioning the $600 billion number, see the discussion in the comments.
Posted by Tyler Cowen on February 9, 2008 at 01:38 PM in Law | Permalink
Comments
$600 billion sounds more like the sum of annual blue collar wages ... perhaps that's categorized as collective-bargained theft?
Who can believe such a preposterous contrast - a photocopy here, a notepad there versus the robbery figures read in the papers
every day?
Posted by: TomG at Feb 9, 2008 1:42:21 PM
Maybe that includes things like coffee-breaks, staring out the window, surfing the internet, and other 'thefts of the employer's time'.
I too find that figure very hard to believe.
Posted by: Tom West at Feb 9, 2008 1:49:06 PM
Tyler, is there a footnote for the $600 billion figure? I'll third the opinion that it's bullshit.
Posted by: Felix at Feb 9, 2008 2:03:29 PM
It is p.195 in the book, perhaps Dan can help us out with a more detailed back-up. Looking on my own I find this as the best source: http://books.google.com/books?id=bdUw3jZUt7AC&pg=PA370&lpg=PA370&dq=employees+theft+fraud+workplace+%24600+billion&source=web&ots=R6OTdl0zzZ&sig=5mU_Li-8LBWDpm6DsQ8v3uCb9HQ, from a book *Investigations in the Workplace*, by Eugene Ferraro, citing the Association of Certified Fraud Examiners. It is not clear what they are counting as fraud but when I wrote the post I did not find the number outrageous.
Posted by: Tyler Cowen at Feb 9, 2008 2:13:12 PM
Just to add in my skepticism: $600 billion is $2000 per capita. That's a lot of ballpoint pens.
Posted by: Bernard Yomtov at Feb 9, 2008 2:20:07 PM
This also raises the question of how many people are killed each year by ballpoint pen theft gone wrong.
Posted by: Greg Sanders at Feb 9, 2008 2:37:36 PM
This is better.
According to research conducted by the Association of Certified Fraud Examiners (ACFE), U.S. organizations lose an estimated 5 percent of annual revenues to fraud. Based on the estimated U.S. Gross Domestic Product for 2006 – $13.037 trillion – this percentage indicates a staggering estimate of losses around $638 billion among organizations.Any top-down estimate like that is nearly always bullshit. Start with GDP and then take some percentage of it? Ahem.
Posted by: Felix at Feb 9, 2008 2:37:56 PM
The 5% figure, by the way, came from a survey of Certified Fraud Examiners. Who of course (a) live and breathe this stuff, so they're likely to consider it more common than it actually is; and (b) have every incentive to exaggerate how common it is, because that way they can scare potential clients into hiring them.
Posted by: Felix at Feb 9, 2008 2:45:04 PM
Thanks, Felix. If that is how they calculated the number it is very likely wrong; for one thing they apparently are not distinguishing between gross and value added.
Posted by: Tyler Cowen at Feb 9, 2008 2:49:02 PM
A quick look at the methodology suggests the figure is nonsense. They asked 11,000 members or so to report on the largest fraud they investigated over a two-year period. About 10% responded and they apparently extrapolated some numbers from this, though they don't explain how.
Posted by: Bernard Yomtov at Feb 9, 2008 2:51:23 PM
The point being made is that internal losses are significantly higher than external losses. This is still valid even if the fraud number is off by an order of magnitude for either $16 billion number or the $525 million as the theft loss.
Posted by: PK at Feb 9, 2008 3:03:39 PM
600 Billion is pure garbage! Dan should be more careful with his facts. Maybe he should behave more like a scientist and less like a journalist out for fame and fortune
Posted by: bee at Feb 9, 2008 3:50:13 PM
Great comments. This is one reason why I love MR so much. By the way, note the "reciprocal nature" of theft: I wouldn't have to steal those ballpoint pens if my employers just gave them to me
Posted by: enrique at Feb 9, 2008 4:00:50 PM
What would be more interesting is how much is spent each year to prevent fraud and/or theft. This will give you an idea of the dead weight loss involved. The theft itself is just redistribution of wealth from rich to poor.
Posted by: Rob at Feb 9, 2008 4:02:27 PM
Isn't the spending on fraud/theft prevention just a redistribution to lawyers, security personnel, etc?
Posted by: at Feb 9, 2008 4:13:32 PM
Ballpoint pens etc are just (untaxed) implicit compensation.
Posted by: otto at Feb 9, 2008 5:24:49 PM
Rob is right: the true cost of a crime includes the measures that are taken to avoid it. People will impose costs on themselves to avoid street crime - more cops, security, guns but also spending more to live in a safer neighborhood. Same for businesses.
Posted by: tc at Feb 9, 2008 6:07:54 PM
I don't understand the problem people are having with the 600 billion dollar number. Especially with what appears to have been happening lately with home mortages and the bond rating agencies. I'd also guess fraud reduction accounts for a lot of the "savings" we see when a turn around artist takes over a company. If anything the 600 billion dollar number might be too conservative.
Posted by: Jason at Feb 9, 2008 6:09:14 PM
I agree with Rob. Becker talks about this in his Nobel Prize speech, I think, or either in his the original Crime and Punishment, I can't remember. But the costs can't be the actual transfers, since those are just redistributions of wealth. The costs are the resources used to prevent theft. Even if there were 600 billion dollars in losses, those aren't the costs of theft. all the spending on guns, security, dogs, and whatever else are though.
Posted by: jason voorhees at Feb 9, 2008 6:15:47 PM
"Isn't the spending on fraud/theft prevention just a redistribution to lawyers, security personnel, etc?"
No because those people could be doing something else with their time. Like producing more ball point pens.
Posted by: notsneaky at Feb 9, 2008 6:28:39 PM
I agree that this number seems outrageous, and iw as amazed to find it. But it is the best estimate I could fine (see the link posted by Tyler Cowen).
Regardless if the precision of this estimate (lets say it is ONLY 300 billion), the main point of the argument is the difference between this number and the cost of the "standard crime."
It seems to me that we are paying a lot of attention crimes with relatively small financial implications (the average income of a burglar is a few hundred dollars per house), and almost ignoring the actives that can have huge financial implications. We do this in the news, and also in the severity of punishment.
In my mind, this is the puzzle we should focus on
Dan
Posted by: Dan Ariely at Feb 9, 2008 6:57:18 PM
One more comment in reaction to Bee
I looked very hard for the best estimates on fraud, I talked to many people who are doing research in this field, but at the end this was the best estimate I could find.
Since we all FEEL this number is too high, I am wondering what number you would feel is the right one. In essence I am wondering what number you would intuitively feel is approximately correct (of course this would not make it accurate).
Posted by: Dan Ariely at Feb 9, 2008 7:23:52 PM
Assuming people are rational, we spend enough on crime prevention until the point where the marginal cost of preventing crime equals (or exceeds) the marginal cost of the crime. Why should I be surprised that our emphasis on preventing robberies leads to the activity not being all that profitable? Why ignore the possibility that we cannot prevent all robberies but we can suppress them to the lower number you cite? The numbers show a rational system successfully working to suppress criminal activity. Unless you want to test what the number would be if we stopped prevention measures.
Theft and fraud at the workplace is a different matter. Taking measures to lower the numbers may make the workplace less productive. Perhaps we tolerate a higher level of these crimes because the perceived costs - direct and indirect- are seen as more acceptable then creating less productive workplaces with more security. Since most companies are profitable, they must be keeping employee theft and fraud within some acceptable limits.
Posted by: DanC at Feb 9, 2008 8:01:46 PM
Regardless of how accurate the 600 billion number is, fraud certainly outweighs non white collar crime on a monetary basis. But that ignores the psychological costs--while the CEO may be frustrated by shrinking profit margins due to "lost" pens, imaginary sub-contractors and billed hours with the bartender, no one is traumatized.
The same cannot be said for the victim of a home invasion, mugging, etc. If there was some way to accurately monetize these effects, then maybe we could start making comparisons of where we spend our money doing prevention.
Posted by: Van at Feb 9, 2008 8:32:58 PM
"The theft itself is just redistribution of wealth from rich to poor." Nope, "Owner/Executives made up less than one-fifth of the perpetrators, but they accounted for the largest losses by far". Calling this "employee theft" is technically true (the boss is an employee, after all), but also fairly misleading.
Posted by: Anthony at Feb 9, 2008 8:48:45 PM
Dan-
My disappointment with the example stems from your seeking to frame a problem with little concern for the fact that you yourself are aware that it is a problematic assertion (i.e. your admission that you checked, were not satisfied, but proceeded forward because it told the story you wanted to tell) Why? Why not chose another example that can be better documented. Maybe the reason few besides the publisher of the information is concerned about this question is because it is a very questionable figure. This figure is just like many of the company based metrics that multiple a particular employee behavior or activity (e.g., online use in minutes at work) by the entire work force and conclude there is a massive loss of productivity (in dollars). The math fails to account for the fact that while people may be paid an hourly rate, it is the output that really matters. Productivity is not simply hours works divided by output. The math may be correct but it is not what is really happening.
As for my guess, I will say that Bernard's back of the envelop math is sufficient to say that sniff test is violated. In addition, I think that Van's point on the psychological impact of crime has not been imputed in your reasoning. Dralec's work on values impact on reasoning could also inform your thinking on this issue.
Posted by: bee at Feb 9, 2008 10:10:42 PM
When you use or take a company's ballpoint for personal use, that's theft. So what's it called when you bring in your own personal ballpoint pen or otherwise use it for your company's benefit? Unreimbursed business expense?
I'd like to see the flip side of this claim: how much (if any) do employers gain from their employees- bringing in their own office supplies, off the clock work (for hourly employees), etc.?
Posted by: Andrew at Feb 10, 2008 12:48:24 AM
Maybe that includes CEO pay(which is certainly a crime)and could easily mount up to 600 billion.
Posted by: lee at Feb 10, 2008 3:28:01 AM
Looking at the ACFE report, it looks like they are assuming that businesses that didn't report fraud had just as much fraud as businesses that did. That seems improbable
Posted by: Will McLean at Feb 10, 2008 10:05:31 AM
Even if you assume the occasional fraud at the scale of Societe Generale (which was in France, by the way, and only $7 Billion), I don't think you can get close to $600 Billion annually.
Posted by: Corbett at Feb 11, 2008 12:46:06 PM
Dan: It's not the number that I question, it's what you are comparing. You are comparing robberies (theft) with employee theft AND fraud.
You should be comparing non-employee theft and fraud with employee theft and fraud. Non-employee theft could include the various web and phone scams, corporate fraud (a la Enron), contractor scams (charging $40,000 for a renovation while only doing $10,000 of work or taking a deposit and leaving town), and so on.
Further does the cost of robberies include additional costs such as having doors and windows repaired, installing doors and windows with better security, or installing security systems? Does the cost of robberies include any health related costs - what if the robbery involved battery? Non-employee fraud could involve a year or more of credit checks to ensure that your identity has not been 'stolen'.
Let's compare apples with apples.
Then, and only then, should we look at the resource allocation issue.
As anyone who has had their house or business broken into knows, there are many additional costs beyond the replacement cost of goods stolen.
Posted by: Vincent Clement at Feb 11, 2008 2:21:04 PM
It isn't only the ballpoint pens and notepads. Supermarkets selling liquor for example, are notorious for massive amounts of stock disappearing. When I was a deli clerk, my boss used to tell me the stories: The liquor manager would toss a few cases in to the garbage can, then go to the compactor and would find ways to smuggle a case or two out eventually leading to a home-garage filled w/ cases. Or even the Mach-4 blades, those are hot items... Multiple that across the nation!
Posted by: RH at Feb 11, 2008 5:17:04 PM
Just in case someone is still checking this out, Felix Salmon at Porfolio profiled this post and the subsequent comments in the post
http://www.portfolio.com/views/blogs/market-movers/2008/02/11/blogonomics-even-great-commenters-cant-generate-wall-street-salaries
Kind of fun to have your comment sited in another blog post--first time for me, anyway.
Posted by: Van at Feb 11, 2008 9:33:23 PM
Looking at the ACFE report, it looks like they are assuming that businesses that didn't report fraud had just as much fraud as businesses that did. That seems improbable
Actually, very probable. Internal fraud is an issue at many corporations, which usually is resolved by mutual disassociation (and whatever recovery can be made). There are two main reasons for this. Most law enforcement agencies have fairly high thresholds to investigate and prosecute white-collar crime (due to difficulty and expense to prosecute, even with evidence provided by the employer). In addition, many corporations do not want their reputations sullied with having employed the perpetrator, even if failing to report prevents them from offsetting the loss on their taxes.
If the amounts lost become high enough, then the incentive to prosecute and report does become higher, but in most cases, single perpetrator net amounts under $50K are often kept in the family.
Posted by: Fraud Guy at Feb 11, 2008 10:43:28 PM
Maybe it's reasonable after all. That's about 4% of US GDP.
Posted by: CG at Feb 12, 2008 1:35:46 PM
Posted by: 深圳翻译公司 at Feb 23, 2008 9:07:07 AM
thank you for this information.sis jarMy local telecom is a monopoly, and it is out-of-control as far as wiretapping, eavesdropping, hacking, controling e-mail programs, phishing, spoof websites, etc.
No company should be immune from law suits and especially companies that control our communications.To give telecoms immunity will make "big brother"free nokia 6600 games"In this paper, we compare the incidence and extent of formal coauthorship observed in economics against that observed in biology and discuss the causes and consequences of formal coauthorship in both disciplines. We then investigate the economic value (to authors) of informal comments offered by colleagues. This investigation leads us naturally into a discussion of the degree to which formal collaboration through coauthorship serves as a substitute for informal collaboration through collegial commentary. Data on manuscript submissions to the Journal of PolzticalEconomy permit us to shed additional empirical light on this subject. Finally, we demonstrate that while the incidence and extent of formal intellectual collaboration through coauthorship are greater in biology than in economics, the incidence and extent of informal intellectual collaboration are greater in economics than in biology. This leads us to search for evidence (which we find) of quids pro quo offered by authors to suppliers of free nokia n70 games
Posted by: free jar mobile at Feb 10, 2010 3:44:08 AM
