Cheating to Be Hot

Is Freakonomics too cynical?

I don’t think so, but some people do. Occasionally we hear from readers who say it’s a shame that we’ve called attention to so much deceit, trickery, and cheating among sumo wrestlers, school teachers, tax filers, and online daters. I could argue back and say, “Hey, don’t we also call attention to people who don’t cheat, like the office workers who eat Paul Feldman‘s bagels?”

The point isn’t that you can divide people into piles of good people or bad people, cheaters or non-cheaters. The point is that people’s behavior is determined by how the incentives of a particular scenario are aligned.

So it was interesting to see this article on Salon‘s Machinist by Farhad Manjoo about a contest run by the website Fishbowl DC to decide Washington’s two hottest media folks. While agreeing that the winners were indeed a comely pair, Manjoo reports that the contest was a total rig job:

[The winners] Capps and Andrews acknowledge that they won only because their online friends — without their express encouragement, they both say — built software “bots” that voted thousands of times for each of them. The bots were distributed on Unfogged, a humorously wonky blog and discussion site popular with D.C. types, within a day of the poll’s opening. If you downloaded and ran the software, your machine began tallying up votes for Capps and Andrews faster than a Diebold rigged for George W. Bush.

Which makes me say:

1. The stakes don’t have to be very high for people to cheat.
2. When no punishment exists for cheating, it’s pretty damn appealing.
3. We have been accused of stuffing a ballot box or two ourselves, although there were no bots involved (that I know of).
4. Can you please point me in the direction of the Diebold folks who rigged those machines? I would love to interview them.

Mario Ruiz

Hello Stephen.

"The point is that people's behavior is determined by how the incentives of a particular scenario are aligned." You said in your post.

This can be a endless debate with profound implications. We can even end up in religion and politics.

My opinion is that there is a relationship between incentives and behavior to a point. My professor of Business Ethics in my MBA (a fellow with a PhD from Harvard) used to tell us, that "in a meeting a gay approaches a very distinguished lady: would you make love to me for 5 million dollars? And the lady answered, of course! The guy asked again: For 500? The lady immediately replied: do you think I am a h...? The gay finalized saying, yes, we are just fixing the price."

We always have a choice, no matter how good the reward is.

Mario Ruiz

Greg Palast


For more information on Bush's voter fraud, check Greg Palast's excellent work "The Best Democracy Money Can Buy." The research is meticulous and the analysis is on point.


I once saw a poll about the Boston Mooninite scare. It had some serious options and a joke option, "What's a lite-brite?"

Someone constructed a few lines of code to repeatedly vote for this option and posted it to an internet forum; members of the forum ran the code until (after rounding) 100% of the vote was for the joke option and 0% for other options. This entailed casting around 100,000 votes.

Why do this? I gather mostly because it was funny, and cost the participants nothing. I've got to say, the benign absurdity of the nonsensical attack did tickle me.


We actually had a gentleman Andrew Appel come down last semester to give a talk on election machines. His angle is a bit computer sciency, as that's what we do, but I hear his latest research is something you might very well appreciate: fraud detection using probabilistic means.

As an aside, it turns out that possessing the software run on voting machines is illegal in the state I live in, so he was perhaps skirting the edge of the law when showing voting equipment and flaws in voting technology. They have a web site you can hit up at Princeton.

No 1 of Consequence

Meticulously researched article about voter fraud in the 2004 election.

Love the blog.

No 1 of Consequence


Is stuffing online ballot boxes really "cheating"? Do voters actually agree, implicitly or otherwise, to vote only once?


I've been hoping for the chance to express this since I first read the book, and probably before that since one outstanding sociology professor I had asserted (quite convincingly) that the economics is the most influential sociological institution for determining behavior. This seems as good a time and place as any. I really dislike the term "incentive." It implies that the behavior occurs because of the result. The cause and the result of behavior cannot be the same, that's circular. Manipulating the incentive (independent variable) does not change the behavior (dependent variable). Manipulating the "incentive" changes the future probability of the behavior. Since behavior occurs because of past experiences, the results of those experiences, and the extent that the environment resembles the environment that those experiences previously occurred in. Incentive like "reward" is an arbitrarily defined term (in which observer bias or other errors is implicit). We could change the definition of incentive (not that I think I've ever heard one offered up from the authors), but why discard what is already the well established, and studied phenomena of reinforcement. Reinforcement is an operationally defined term, a consequence that increases the likelihood of future occurrence. It also comes with a whole vocabulary that might help better describe many economic relationships. Punishment, extinction, generalization, discrimination, stimulus control, etc.; that might better describe, compare and contrast events. If human behavior is your interest why not use the vocabulary of behavior. The results would help demonstrate a robust field of factual, significant, relationships for both fields.



To me this mostly suggests why the penalties for cheating in public office elections need to be absolutely draconian, to balance the high incentive for cheating, and the low risk of being caught.


The article about the bagel man keeps his last name secret -- but today's post gives it away!


@ jldugger
A few years ago I read an article about voting machines and they showed that the software used was Microsoft Excel or Access (don't quite remember which one) to count the votes. So strictly speaking, if those machines are still used... you're not allowed to own MS Office in your state? LOL, Microsoft cannot be happy about that. Unless the law knows the difference between software and a spreadsheet or database.


having only ever been a victim of "bots" myself, I find the fact that they can actually work to one's advantage to be quite intriguing...

Zach B.

It's funny that Freakonomics shows us the ways in which people cheat, but not the ways in which they develop and follow rules. And following rules, being honest, desiring to be perceived as honest is putatively also a powerful motivator to human behavior. That is, (Fr)economics is good at showing why people are selfish egotists, but not why they are altruistic moralists.


Wow. An oblique comment from the authors about voting irregularities. Asking for evidence. How novel. Hey here's an idea - why not do back-testing on exit polls and then apply the work to the 2000 and 2004 presidential elections reported results? Have never gotten a response from the authors why this would not be desired or possible. Still waiting.


For the record, I love Freakonomics. But I think the authors sometimes forget that they are "only" economists (as am I). Economics is by construction deeply a-moral: It is only concerned with what people do, but not what they should do. In many circumstances this moral vacuum comes in handy to make a clear point, but we have to be careful when trying to explain actual behavior, which does have a moral component. Saying that people just act on their incentives is confusing the simplified model with reality. Economics simply has nothing to contribute when it comes to morals and ethics, and many economists don't seem to be aware of this.

I am absolutely certain that there are many people who would never stuff a ballot box, even with the incentives aligned in favor of doing so. (We only hear from the stuffers, because the others don't make the news.)

Ken H

Responding to 'Beat': Adam Smith described himself as a moral humanist, the farthest place away from amoral. The deepest underpinnings of our free market economics and individual political freedoms lay in the Age of Enlightenment and its direct antecedent, the democracy of ancient Greece. Smith argued (and I agree) that enlightened self interest guides people to moral decisions in that they lead to the greatest good for the greatest number.

The big question: what did Smith truly mean by 'enlightened' and what meaning does that have in the 21st century global economy? If by enlightenment, he meant a concern for how we each fit into a greater society (drawing upon Donne from the early-17th century: "No man is an island, entire of itself; every man is a piece of the continent, a part of the main."), then economics is perhaps the most moral of the social sciences.

Rita: Lovely Meter Maid

I like that Freakonomics exposes what is normally kept hidden. I am still as cynical (and as idealistic, in certain ways) about human nature as I *ever* was, even after reading their book (and this blog). Trust no one, really, but still have some faith in the basic goodness of human nature, as well. Life is just hardly ever (and maybe never) the black/white deal that helps people to sleep easy (but also dulls things down considerably).


I agree with Ken H that Smith was a moralist, but Smith himself saw the immorality of the emerging global economy- that is, as capital had freedom of movement, while the labor force does not, this would inherently create an imbalance in the labor market- i.e. the laborers are exploited- the irony, then, is that while Smith warns of the immorality of the global market, his name is often used to justify deregulation, which further advantages the giant hegemonies of capital which have developed over and against the world's labor market (the rest of us)


This article , published a few days ago in this very same publication, reminded me of the Bagel Man. It's about people who pilfer office supplies, from pens and post-its to furniture and computers, from their workplaces, and the conditions that make us feel that we are entitled to just take things sometimes. Especially if no one will know that it's us. It has little to do with needing anything. It's a little scary to think that one might begin with staples but be able to rationalize all the way up to a new laptop or aeron chair.


Your fellow Times Magazine columnist Randy Cohen ("The Ethicist") had an (IMO) very well-done TimesTalk last year on the relative importance of character and circumstance. I think this is both a useful lens, and a persistent attribution error, in many cases where we analyze or judge human behavior (not just when we decide on motives for cheating). (The talk was posted on the TimesTalks podcast on 9/7/06 and should still be available on the feed... if not, he's actually quite good about responding to emails.)

Daniel Merritt

Beat: While I do think what we might call scruples has a part to play - even in pure economic terms, we could call it a disincentive due to cognitive dissonance - I doubt there are many, if any, people who would not /ever/ stuff a ballot box. If you define incentives in purely monetary terms, you might be right: there are doubtless people who place a much lesser value on even a massive sum of cash than they do on fair elections.

But defining the incentives in purely monetary terms is a mistake, especially in a political context. Incentives can also take the form of bringing on policies you believe would be beneficial, or avoiding policies you think would be disastrous. While I doubt many people would steal elections over a 1% change in the tax rate, or federal subsidies for the arts, or the government's policy toward the importation of monkeys, it becomes much more plausible in more serious cases. Vast numbers of people would probably stuff a ballot box to prevent a fascist dictator from coming to power; a lesser but still substantial number to stop what they believe to be a catastrophic mistake (Democrats to avoid a war with Iran; Republicans to avoid a truly socialist economic system); and then a surprising number just to elect their favored candidate in an important election (as noted here, some Democrats suspect the Republicans of doing such a thing in Ohio in 2004, while I suspect many Democrats would have secretly rigged things in the other direction given the opportunity).

Whether such ballot stuffing actually occurs would also have to take into account more substantive, material costs, opportunities, and risks. I tend to view most accusations of a massive conspiracy of vote-tampering in 2004 with strong suspicion, considering how difficult it would be to keep quiet such a large, multi-state operation, but I don't doubt for a second there are many people who /would/ do it.

It seems to be a widespread mistake - from some segments of economists, but mainly from those 'outside' trying to invalidate economics - to think that economic models, theories and thinking can take no account of anything but money.