Save Me From Myself

(Photo: davidd)

Season 3, Episode 1

Sometimes we have a hard time committing ourselves – whether it’s quitting a bad habit or following through on a worthy goal. In this episode of Freakonomics Radio, we share stories about “commitment devices.” They’re a clever way to force yourself to do something that you know will be hard. Host Stephen Dubner talks to a struggling gambler who signs himself up for a program that bans him from state casinos – only to return, win a jackpot, and have it confiscated. We’ll also hear from a new father trying to shed bad habits. So he makes a list of things he wants to change and vows to pay a penalty if he can’t shape up in 30 days. The penalty? He’d write a $750 check to someone he really dislikes: Oprah Winfrey. Freakonomics co-author Steve Levitt offers a few of his own off-the-wall commitment devices, and the Brown economist Anna Aizer talks about using commitment devices to fight domestic violence.

Then we’ll take a look at some misadventures in baby-making.  First, the story of how China’s one-child policy was inspired by a couple of scholars having a beer in the Netherlands.  Also: Levitt discusses his controversial research showing that legalized abortion lowered the U.S. crime rate. We’ll also talk to Mara Hvistendahl, author of the Pulitzer Prize-nominated  Unnatural Selection: Choosing Boys Over Girls, and the Consequences of a World Full of Men , which looks at how the introduction of the ultrasound led to the disappearance of tens of millions of baby girls.  Finally: Stanford professor Stephen Quake ponders the consequences, intended and otherwise, of a new genetic test he has developed.

This episode is a compilation of two earlier podcasts: “Save Me From Myself” and “Misadventures in Baby-Making.”

Audio Transcript

Tony BALANDRAN: Well, it was in late August of ’09. I went to a Harrah’s Casino and played pai gow. And I was doing fairly well. I couldn’t believe the luck I was having.

Stephen DUBNER: That’s Tony Balandran. This was on one of those riverboat casinos in Kansas City, Missouri. Balandran is a journalist, in his late forties.

BALANDRAN: The big hand that I was dealt was a seven-card straight flush.




BALANDRAN: And a seven-card straight flush is pretty amazing, pretty rare. Of course it had a joker involved so there was a wild card there. And I believe that was 750-to-1 one on the odds. So my $5 bet on it netted me, I think, $3,750.


DUBNER: And what happens then, did you keep playing for a few hours or were you jazzed? Did you decide I’m going to cash out? How’d it work?


BALANDRAN: No, they spread the cards out for the camera of course. They did a card count.




BALANDRAN: And the supervisor congratulated me and asked for my ID because this was going to be a big payout, a significant payout. And I just thought, wow, here it goes.

DUBNER: Here’s the thing: that beautiful hand Balandran was dealt, that 750-to-1 straight-flush hand, it actually turned out to be bad luck.



ANNOUNCER: From WNYC and APM, American Public Media, this is Freakonomics Radio, the show that explores the hidden side of everything. Here’s your host, Stephen Dubner.

DUBNER: Tony Balandran wasn’t supposed to be gambling in that casino, or any other casino in the state of Missouri. It all began when he realized that his gambling had gotten out of control.

BALANDRAN: Sometimes after work and on the weekends when the casinos did not close, there were a couple of times when I stayed there until I had to go to work the next day.


DUBNER: Wow. And at what point did you say to yourself, “Tony, this is not good, and I need to do something?”


BALANDRAN: Well, when I was borrowing money from friends and I wasn’t completely honest with them about why I needed the money. And I knew I was in a pattern of living paycheck-to-paycheck despite, you know, a decent income. You know, when I was trying to figure out how many dollars I had to get my next meal, I thought this is not good, and I got to stop.

DUBNER: Balandran knew he did not have the will to stop himself. So he decided to do something drastic. Missouri, like many states with casinos, offers what’s called a “self-exclusion” plan. You sign yourself up for a registry that effectively bans you, for life, from all casinos in the state. If you ever do come back, you can be arrested for trespassing. It’s like signing a contract with yourself, against yourself. For a while, it worked. Balandran stayed out of the casinos, until he started going back. He discovered a loophole. He found he didn’t have to show his I.D. to get into the casino, or even to play the table games. In fact, he wouldn’t have to show his I.D. at all unless he won a certain amount of money, at which point the IRS would have to be notified. And that brings us back to Tony Balandran’s seven-card straight flush.

BALANDRAN: I just kept a smile on my face thinking, I don’t know, maybe I’m going to be lucky in another way. Maybe the computer glitch occurred, or maybe somebody accidentally erased records from six years ago. But all the time I thought, they’re taking too long.


DUBNER: And then what?


BALANDRAN: I believe it was highway patrol, I believe that’s the state agency, the gaming entity, they tapped me on the shoulder and asked me to step away from the table and to gather my chips and come with them. And they didn’t need to explain why. And I followed them.


DUBNER: And is at least a little bit of you thinking, Holy cow, why did I do this? I did this to myself! I signed up so that the casino…


BALANDRAN: Yeah, I was beating myself up.


DUBNER: I mean, no offense, I don’t blame you. Because, I mean as bets go this was a bad bet. You could only lose; there was no way you could win once you signed that ban.


BALANDRAN: Well if I’d won several more four-of-a-kind hands, I could probably win under the radar. But what I’m really doing is hoping I don’t win big. And you’re right—the logic doesn’t follow. Was it dumb? Yes. And I even kind of joked about it with the officer. I said, a seven-card straight flush, come on, at least give me some compliment for that while you write that citation.

DUBNER: Balandran was charged with criminal trespass, paid a small fine, and -- worst of all -- he had to surrender his $3,750 jackpot. So the casino got to keep the money he lost and the money he won. Which makes a “self-exclusion” ban seem like a pretty good deal at least from the casino’s end.

DUBNER: And so if you had it to do all over again, if you could rewind back those years ago, would you sign up again?


BALANDRAN: Well, that’s a tough question. If everything was going to be repeated in hindsight I would not want to go through all that again. I know I had to get away from those casinos for a while. So in the time that it kept me away I think that was worth the ban, worth putting my name on that list.


DUBNER: If there were a national casino ban registry, a national, self-imposed casino ban registry would you sign up for it?



DUBNER: So economists have a name for the self-imposed ban that Tony Balandran signed up for: it’s called a commitment device. And that’s what we’re going to talk about in the first part of the hour today—clever ways to trick yourself, or trap yourself, into doing something that you want to do but, for whatever reason, aren’t able. In order to understand how a commitment device works, if it does work, you have to picture two versions of yourself: the current you and the future you.  

Steve LEVITT: Sometimes it’s the case that people know that their future version of themselves will want to follow a behavior that their current version of themselves is not comfortable with.

DUBNER: That’s Steve Levitt, my Freakonomics friend and co-author. He’s an economist at the University of Chicago.

LEVITT: So I’m on a diet and I would like to stick to that diet. But I know that when someone puts a chocolate cake in front of me I will lose my willpower and I will eat that chocolate cake. A commitment device is an attempt on the part of a person to set up constraints so that the future self isn’t able to take advantage of the situation and do what the future self wants, but instead requires the future self to behave in a way that the current self would like the future self to behave.

DUBNER: You remember how Odysseus had himself lashed to the mast of his ship so he couldn’t succumb to the sweet song of the Sirens? That is a commitment device. Have you ever bought an expensive gym membership to force yourself get in shape? That’s a commitment device. For some people, marriage is a commitment device. Steve Levitt, being Steve Levitt, has some singular ideas about commitment devices.

LEVITT: So bariatric surgery is one of the ultimate examples of a commitment device. It’s a case where you know you can’t control how much you eat so you actually have surgeons go in and, you know, put rings, or cut your stomach out so that you just don’t want to in the future eat at all, which has always struck me as a really, really extreme solution to not being able to lose weight. I think I have at least two far superior ways of losing weight that I haven’t been able to convince anyone else to go with. Now, the one that’s maybe more civilized is if you’ve ever had really bad canker sores or kind of cut your gums it’s so unpleasant to eat. So why not just slice up your gums a little bit, you know, cut up your mouth so you just don’t feel like eating at all. I think that would be a great diet approach. But people say, no, no, no, too violent, I couldn’t cut myself. One thing I know would work is just take a little can, like say a baby food jar and fill it with vomit. OK? And wear it around your neck. And every time you decide that you’re hungry just open the jar and take a little sniff. And I guarantee you you will lose weight, guaranteed.


DUBNER: And we can’t figure out why people don’t like economists. I don’t understand. It’s such a pleasant idea.


LEVITT: But how is that worse than bariatric surgery? Wouldn’t you rather sniff vomit once in a while than risk death going under the knife to have someone cut open your body and make it so you’ll never be able to reverse it?


DUBNER: Yeah, but it’s a one-time thing as unpleasant as it is for the one time, as risky as it is for the one time, you figure, If I get through it then I’m done, as opposed to just walking around with a baby food jar of vomit on a string my whole life? I mean, come on.


LEVITT: Think about what a great conversation starter that would be.


DUBNER: With what kind of people? Who are you trying to start a conversation with?


LEVITT: The ladies at ladies’ night at the bars.

DUBNER: Okay, so maybe Levitt’s methods aren’t for everybody. Or anybody. But let’s say you do want to change your life in some way large or small. Maybe you find yourself questioning the choices you make. Maybe you’re looking for answers to life’s big, big questions.

ADAM Scott: I was talking to my cousin Jimmy, who asked the question that really kind of summed up the whole thing. He said, “is life without steak and porn worth living?”

DUBNER: That’s Adam Scott. He works in telecom policy for the Canadian government in Ottawa. He’s 35 years old. He and his wife had their first kid a year ago. His perspective on life has shifted.

ADAM: I’ve always been a weak-willed person, but now I’m becoming a weak-willed fat old person. So I’ve definitely had my health on my mind lately. Recognizing I’ve only got so many years, so many hours left to live. Am I spending them in a valuable way?

DUBNER: So he took his cousin Jimmy’s question kind of seriously: what if he did give up steak and pornography, and while he’s at it, all the other habits he considered unhealthy?  

ADAM: I’ll give you the full list is: hamburgers, hot dogs, french fries, onion rings, chicken wings...

DUBNER: Adam decided to go cold turkey, for thirty days. He committed himself to a life of health and virtue. His cold turkey list ended up with forty-two items.

ADAM: ...soda pop, illegal drugs, which I don’t really do anyway....

DUBNER: I remember, when I was a kid, how hard it was to give up one thing for Lent—like ketchup. I loved ketchup! How do you live without ketchup for the entire Lenten season? And here’s Adam Scott trying to give up not one thing he loves but fort-two!

ADAM: These are things that I have been doing habitually, you know, foods that I’ve been eating for the last twenty years, television that I’ve been watching every night for the last two or three years, and I’ve never really made a conscious decision about them. I’ve never sat down and assessed the value of watching “TMZ” every night. It just kind of crept into my life and now it’s there. So I don’t necessarily want to give it all up, but I do want to evaluate it and see if I think my life is better with beer and entertainment television in it, I’ll be drinking in front of the TV on day thirty-one.

DUBNER: Adam knew how hard this was going to be. And he knew that he was going to need some kind of commitment device.

ADAM: So what I came up with is, I’ve written a $750 check made out to Oprah Winfrey, who my wife adores and who I -- “despise” might be too strong a word, but who I’m not a big fan of.


HEATHER Scott: Yeah, it’s true, I do love Oprah. I think she’s great, I think she does great things.

DUBNER: That’s Heather, Adam’s wife.

HEATHER: I couldn’t see anything bad coming out of her having more money. I’m sure it would go to something charitable and good. But Adam didn’t see it that way.


ADAM: I’m convinced she’ll spend it to advance her evil empire one way or another. My hope would be that she’d spend it on herself. A new pair of shoes. Or I mean, something she had in abundance already.

DUBNER: Adam needed one more thing: a referee. So he gave the $750 check to his friend Scooby.

ADAM: And if he gets any credible evidence that I’ve broken my contract, he has instructions to mail that check immediately to Oprah Winfrey.

DUBNER: There was one last layer of the commitment. Adam decided to post a series of videos on YouTube, called the “Cold Turkey Diaries,” in order to make public his progress.

ADAM: Hi, my name is Adam Scott, and this is the Cold Turkey Diaries.

DUBNER: We talked to him on Day 5.

ADAM: The commitment contract is really working. The combination of guilt and fear of getting caught really has been effective so far.

DUBNER: But by Day 10, the novelty seemed to be wearing off. …

ADAM: Like yesterday when I got home from work after a really stressful day, I was tired, kind of annoyed, fed up with my job. It would have been a great day to sit down, crack a cold beer, open a bag of pretzels and sit in front of the TV for half an hour, just to decompress a little bit. I don’t get to do that anymore.

DUBNER: But Adam presses onward. He certainly didn’t make it easy on himself, giving up so much, all at once. He has to get going in the mornings without the benefit of caffeine. He has to avoid TV even in other people’s homes. At poker nights with his buddies, he drinks Gatorade while they’re all pounding beer and chips. But Adam Scott, hearty Canadian that he is, doesn’t quit and he doesn’t cheat. But then, on Day 21…

ADAM: There was a slip-up.

DUBNER: Adam and Heather had gone to a diner for breakfast.

ADAM: … which I knew was risky at the time.

DUBNER: He ordered a decaf coffee, which is not on the Cold Turkey List. Heather also ordered a coffee.

HEATHER: And there were two coffees on the table and one was black and one was sort of creamy looking and I hadn’t put anything in my coffee and I was like, well that’s kind of strange that that one’s kind of creamy looking, maybe there’s something wrong with it, I don’t know. And I turned to Adam and I said, “What’s in the coffee?”


ADAM: I didn’t know what she was talking about, and then she just yelled, “milk!”

DUBNER: Milk. Not allowed. Adam had only put a couple little containers of 2% milk in his coffee and had a few sips. It was, as he would later insist, a total accident. But it didn’t matter. After three weeks of perfect compliance, he had broken his commitment.

HEATHER: Oh, it was an immediate freakout. It was like, such a blow. The wind was just completely knocked out of our sails, like, damn.


ADAM: So I was really wondering, is this what I’m going to go down on? Something so stupid?

DUBNER: Adam knew what he had to do. He contacted Scooby.

SCOOBY: Yeah so he texted me, and I think my first instinct was, who puts milk in their coffee, first of all? That’s just sick. So I thought about it, and said, ok well, I said if it was half and half, yes. I’ll let 2% slide, don’t let it happen again.

DUBNER: So Adam keeps going. And: he makes it all the way to Day 30 with no further slip-ups. No sleepwalking into the pantry for a bag of chips. No plopping himself down in front of the TV. He had committed himself to a cold turkey regimen and, as hard as it was, as unlikely as success may have seemed, he made it! But when we talked to him, on Day 30, he didn’t sound so happy.

ADAM: I’m kind of feeling guilty about those two little sips of milk. Right, it’s like the record book is always going to have the asterisk beside it. It’s a bit of a letdown, to be honest.

DUBNER: So he comes up with a workaround. Call it a two-percent solution.

ADAM: So maybe the best thing to do is just read you the letter I’ve written. It starts off, Dear Oprah, attached you’ll find a check for $750. Please allow me to explain. One month ago...

DUBNER: He mails the check to Oprah Winfrey. And his letter explains the whole project , what happened with the milk, how Scooby forgave him but how Adam couldn’t forgive himself… and then, he just leaves it up to her.  

ADAM: I throw myself on the mercy of the Oprah and ask only that you adjudicate this matter fairly. Tonight I will sleep guilt-free.

DUBNER: I’ll be honest with you, I think Adam Scott was way, way too rough on himself. Forty-two banned items and one slip up in thirty days? I mean c’mon! But I also find myself thinking: Look how much effort he expended, look how much angst he put himself through, simply to get himself to do the things he already wanted to do. That's why some people aren't such big fans of the commitment device.

LEVITT: I really think commitment devices are kind of a farce.

DUBNER: That’s Steve Levitt again. He tried a commitment device on himself once.

LEVITT: It was actually in Thomas Schelling’s undergraduate class in economics at Harvard. So, Thomas Schelling was one of the greatest economists of all time, he won the Nobel Prize, he was a wonderful lecturer.

DUBNER: Schelling was also one of the first scholars to study what we now call the commitment device.

LEVITT: But I had this problem which is it was a nine a.m. class. As soon as I was settled into a seat I would immediately go to sleep. And I really wanted to see these lectures, I just couldn’t keep my eyes open. And so I had this brilliant idea. Now this was an economic class on strategic thinking. And what better way of using strategic thinking than to use the commitment device of putting myself in the very front row in the very middle seat? How could I possibly fall asleep if Thomas Schelling was looking right at me as he delivered a lecture? And unfortunately, it couldn’t have been more than two minutes into class when I found myself dozing off already. And I couldn’t keep my eyes open the entire lecture. And I decided, well, that commitment device didn’t work, I’m going to sit in the back of the class. At least it’s not half as embarrassing. The problem with commitment devices is that as clever as your current self is at trying to devise ways to keep your future self from getting around it, the future self just desperately wants whatever it’s being denied and it finds ways to get around it. I mean, while people pretend to want commitment devices, I think deep down real people don’t really want them.

DUBNER: All right, so personal commitment devices are imperfect, to say the very least. But what if we raise the stakes a bit? What if a commitment device could be applied to help fight, say, domestic violence?

Anna AIZER: There are essentially two commitment devices available to women who are the victims of domestic violence. And one of them is essentially to kill him.

DUBNER: That’s coming up, on Freakonomics Radio. And: if you like this show, why don’t commit yourself to it? Subscribe to our free, weekly podcast on iTunes or visit for an archive of all our shows. Thank you!


ANNOUNCER: From WNYC and APM, American Public Media, this is Freakonomics Radio. Here’s your host, Stephen Dubner.

DUBNER: So commitment devices don’t always work, in part because your future self rebels against your current self, and in part because the devices may simply not be powerful enough. And there’s another problem: sometimes you just change your mind. Anna Aizer is an economist at Brown University. She studies commitment devices in an unusual context.

AIZER: Which is the context of domestic violence. And the reason why this, we believe, is sort of a nice context for studying this is that in domestic violence what you witness is a large degree of cyclicality, a large degree of women leaving abusive relationships and returning multiple times.

DUBNER: If you think domestic violence isn’t a big problem, you are wrong. In the course of a lifetime, an estimated one in four women is beaten or raped by an intimate partner.

AIZER: In the U.S., every day roughly 14,000 women are the victims of domestic violence, and four are killed by their intimate partners.

DUBNER: One big problem is that many domestic-violence victims don’t report their abuser to the police. And there’s another problem: a woman who does call the police might end up dropping the case —her boyfriend or husband might sweet-talk her out of it, or maybe threaten her. Back in the 1980s, a vast majority of domestic-abuse cases were dropped before final judgment. So across the country, lawmakers tried to come up with solutions to these problems. One was a mandatory-arrest policy, which meant that the police were required to arrest a suspect even if the victim had already changed her mind by the time the police showed up. Another, similar solution, is called the “no-drop policy.”

AIZER: And this policy states that once a case has been presented for prosecution, a woman can choose not to follow through, but the prosecution will continue anyway. And I should say I believe that the justification for that was prosecutors believed that women were being threatened by their abusers to drop the charges. And so once they remove that from her power they felt that they were actually making her safer in that sense.


DUBNER: So both of these solutions that you’re talking about, both of those are commitment devices then, yes? They’re both bells that cannot be unrung.


AIZER: That’s correct.

DUBNER: Most big American cities now have these mandatory-prosecution policies. Aizer and her co-author Pedro Dal Bó, who happens to be her husband, wanted to know what kind of effect these policies had on domestic violence. They began by looking at the data from California. During the 1990s, the seven biggest cities in California adopted the policies at different times, which made for a nice “natural experiment” to help study cause and effect.

AIZER: So we could essentially look at a county before and after it adopted one of these policies and look to see what happened to both arrests for domestic violence and calls reporting domestic violence. And what we found was an increase in both on the order of fourteen to twenty percent. So after these policies were adopted it seemed to be that there was an increase in reporting of domestic violence.


DUBNER: And what was your response to that, as an economist? Did you say, you know, this means that the policy makers got it right and that they came up with a kind of rejiggered incentive that produced some good pro-social change? Or?


AIZER: Essentially yes. You know, this was very much consistent with women demanding or wanting a way to commit themselves to prosecution.

DUBNER: So that sounds good, but let me play devil’s advocate. You could imagine that in some circumstances, a commitment device like this might backfire, that is, some women might be more reluctant to call the police knowing they won’t be able to change their mind later. And men, now that the stakes are higher for them, might have greater incentive to force women to not call the police. And how about this: Aizer tells us that domestic-violence calls and arrests rose as a result of these new laws, but what about domestic violence itself? Do more arrests necessarily mean that less violence is happening?  

AIZER: That’s less clear. So, that’s hard to look at also. So you can’t get survey data on this at that kind of level, at the level we’re looking at, this very local level. What we could do is look at hospitalizations for assault among women over this period, and we did not see that they declined significantly. So, you know, I won’t sort of hang my hat on that.


DUBNER: So you’re saying that the law works well, that the commitment device works really well in that it gets more people to report, but you can’t really say that it works well in lowering the actual, you know, symptom that it was trying to address in the first place, right?


AIZER: Right, the data’s just not there to allow us to answer that question well.


DUBNER: So you could say well that’s good that this law’s been arrived at, this commitment device that works. On the other hand, we can’t really prove that it produces anything more than kind of legal and prosecutorial activity.


AIZER: So, if we sort of take this sort of a fundamental principle of economics that you look at people, if you want to know what people want, what people like, you look at what they do, and not what they say. So the fact that you see an increase in reporting suggests that this is actually a policy that they prefer.

DUBNER Aizer did find some convincing data of a slightly different sort when she went beyond California and looked at the national figures.

AIZER: So what we did find, and I have to admit that we tried as hard as we could to make these results go away but we couldn’t. What we found was that we saw a significant decline in intimate partner homicide with men as the victim.

DUBNER: Did you catch that? What Aizer’s saying is that after these domestic-violence policies were adopted, fewer women murdered their husbands and boyfriends. Now granted, that’s not a very big category of murder in the first place. But still, how do you explain that?

AIZER: Well, there are essentially two commitment devices available to women who are the victims of domestic violence. And one of them is what we call the extreme device, which is essentially to kill him. Once she kills him there is no way to return.


DUBNER: That is a commitment device isn’t it?


AIZER: That is a commitment device. And the second is the one offered by this policy, which is obviously a considerably cheaper device.


DUBNER: It makes it seem so much less draconian compared to murder now.


AIZER: Doesn’t it though?


DUBNER: Because if you were just to offer the one choice of, you know, if you are going to report your partner as abusing you then it is a bell that cannot be unrung, that might sound very stiff. On the other hand, if you consider that the other option is that if you don’t report you may end up killing him, that’s a much more costly commitment device to the woman to say nothing of to the man.


AIZER: That is correct.


DUBNER: As an economist who looks at, you know, the way people make decisions for themselves how does that sit with you I guess philosophically? Do you like the idea of a commitment device that helps that much?


AIZER: Yes and no. I mean what we’re essentially doing is we’re preferencing one state of being over another. We’re saying with this policy that we believe that the woman, right when she is attacked and reports her abuser, that we are going to follow that woman’s preferences and not the second woman who shows up a couple weeks or a couple months later and decides she no longer wants to prosecute and actually wants to return to him. So, you might argue, who are we to preference that woman’s self as opposed to her other self if we can call it that? And you would be right. I think the way to look at it is to say, A) is there any collateral damage? Are there externalities? And the answer to that is yes. There are children who bear the costs of these decisions and that the policy maker should consider the welfare of those children as well. And the second part of the response is well let’s see what women do. Let’s see whether they in fact increase their reporting. And if they do that would suggest that they have a preference for this as well and that the policy choice was correct.


DUBNER: So, what’s interesting, to me at least, is that with these commitment devices, it’s somebody else who’s doing the committing for you. Now, maybe that’s how it’s got to be. Self-imposed commitment devices, as we’ve seen, are rather imperfect. Tony Balandran signed himself up for a gambling ban and he couldn’t keep himself out of the casino, and then got busted when he won a huge jackpot. He lost by winning! Adam Scott, meanwhile, he won by losing. That is, he considered his cold turkey experience a failure because of those few wayward sips of coffee with the milk in it. But in the end, the cold turkey thing really worked out for Adam. He lost eight and a half pounds during that month. He may, of course, be out the $750 that he mailed to Oprah Winfrey.

ADAM:  So I picture my check, you know, traveling through the mail, landing in Oprah headquarters, she opens it, sees the $750, and decides that she can go shoe shopping.


Oprah WINFREY: Stop the car! Stop the car! I see a shoe in the window!

DUBNER: But that’s not what happened. One day, Adam got in the mail an envelope from Chicago, from Oprah Winfrey’s office. It contained nothing but his letter to Oprah, which had been opened, and his $750 uncashed check.

DUBNER: Coming up: one of the few laws you can’t escape, no matter how hard you try: the law of unintended consequences. You’ll hear, among other things, about how a conversation over a few beers helped shape a policy with huge repercussions.

Geert Jan OLSDER: I hesitated at a moment, shall I talk about this population planning paper?

DUBNER: And how one piece of technology led to 160 million missing women.

Mara HVISTENDAHL: In some countries, where sex selection has taken off, people see this machine as really a way to ensure them a boy.

DUBNER: That’s next on Freakonomics Radio. We love hearing from our listeners, so: if you’ve got something to say, or a question to ask, or a strange fact to share, visit us at or drop us a line at Thanks.


ANNOUNCER: From WNYC and APM, American Public Media, this is Freakonomics Radio. Here’s your host, Stephen Dubner.

DUBNER: Most of us, when we’re about to do something, whether it’s trivial or significant, we think at least a little bit about its consequences. But no matter how smart you are, or hard-working or well-meaning, some actions end up having consequences that nobody could have foreseen. That’s what we’re talking about in this part of today’s show: the unintended consequences of… what’s the right word?… how about: baby-making.
Let’s start in the Netherlands.

OLSDER: Ok, my name is Geert Jan Olsder, last name is Olsder spelled as O-L-S-D-E-R. My age is sixty-seven, which in the Netherlands means that you are retired.

DUBNER: Before he retired, Geert Jan Olsder was a university professor in applied mathematics. In 1975, Olsder was thirty-one years old. He and his colleagues at Twente University were very bright. They thought they could do… anything.

OLSDER: It simply popped up suddenly: why don’t we work on population planning?

DUBNER: Of course: population planning! What else should a mathematician be working on? Olsder’s crew imagined an island nation with no emigration or immigration—just births and deaths.

OLSDER: It looked like a nice mathematical problem.

DUBNER: The essential riddle was this: as the population aged, and as longevity increased, what was the right birth rate to prevent the island from becoming overpopulated? Olsder and his colleagues worked hard on the problem, and came up with an elegant equation. Their research paper was called “Population Planning: a Distributed Time-Optimal Control Problem,” was published in a university report. One day, Olsder was in his office with a colleague…

OLSDER: And then we got a telephone call from the central, or main office of the University and they said, Oh we have a group of Chinese people here, scientists, and apparently something went wrong in the organization, we didn’t know about this group coming, but they would like to be entertained this afternoon. And there are two mathematicians among them. Would they be welcome in your department?

DUBNER: Okay, so there were a few visiting Chinese scientists on campus who needed babysitting. Would Olsder take care of one of them? Sure, he said. He took him to a café; they ordered beers. They chatted about the university, the math department…

OLSDER: But after one hour, or one and a half hour or so, the conversation stopped somewhat, because I don’t know from lack of interest, I don’t know, but, but...


DUBNER: Tell me, how did the conversation turn to the topic of your paper about the mathematics of population?


OLSDER: That took a while (laughs), because in the beginning we were talking in general terms. I hesitated at a moment, shall I talk about this population planning paper?

DUBNER: Olsder hesitated because his paper had to date only been published internally. He didn’t want to risk getting scooped by another scholar. But, well, math is math and beer is beer, so Olsder told this Chinese guest about his population paper.

OLSDER: He took a lively interest. I mean, it seemed as if he thought at that moment that it might be something for us. I did not realize that at the time, but, you know, once I look, now that I look back, and I see what the consequences were, I think that he was triggered by this kind of scientific work. And of course he was just, “just,” I mean between parentheses, a domino in a long series of dominoes. And maybe I was also somewhere a domino quite in the beginning and they all started falling. And ultimately of course the consequences were the one-child policy in China.

DUBNER: What Geert Jan Olsder (didn’t know was that his visitor, Song Jian, was a leading Chinese government scientist. His background was in missile science, but his research portfolio now included population control. In fact, he would go on to become one of the architects of the one-child policy that China introduced in 1978. And he later credited Olsder’s research as an influence. Now, it’s not as if the one-child policy wouldn’t have happened had those two men not met back in Holland, but it’s still pretty sobering to consider the unintended consequence of that chance encounter over a beer.

OLSDER: I was this butterfly who moved its wings, you know, and that causes a fly to move, and then you get a little pebble that moves, et cetera. And then ultimately you have a hurricane. Maybe I was one of the starters of this process, but purely by accident. And no, I couldn’t have realized at the time, so I mean, if we could redo history, probably the same would happen again. On the one hand, you feel flattered, of course, if people take an interest in your research. My feelings about these consequences, I mean, how should I put it, some people ask me do you feel guilty, but I think that’s not the right question. Because, I mean, as I said, the same could happen again nowadays under the same circumstances.

DUBNER: The one-child policy that Olsder’s research helped inspire is thought to have prevented roughly 250 to 300 million births in China since 1980. And given what you remember from biology class, you’d probably think that those never-born children would be about equally divided between boys and girls, right?

Mara HVISTENDAHL: In most human populations, the natural sex ratio at birth is around a hundred and five boys for every a hundred girls.


DUBNER: And what is the most accepted explanation for why that ratio exists?


HVISTENDAHL: It seems to be nature’s way of ensuring a balanced population later on, because it turns out males die at higher rates throughout their lives. And so by the time we reach adulthood we have an equal number of males and females.

DUBNER: That’s Mara Hvistendahl. For those of you keeping score at home, that’s H-V-I-S-T-E-N-D-A-H-L, because, she’s originally from Minnesota.

HVISTENDAHL: I’m a Beijing based correspondent with Science magazine, and the author of “Unnatural Selection: Choosing Boys Over Girls and the Consequences of a World Full of Men.”

DUBNER: When Hvistendahl first moved to China, in 2004, she noticed something strange.

HVISTENDAHL: The fact that you could go to a school and look at a classroom, an elementary school usually, and you just see many more boys than girls.


DUBNER: And what did you do about it at first? Did you talk to the principal or the teachers and say, ‘Hey what’s going on? Are the girls being kept at home?’ Or did you kind of know right off what was going on?


HVISTENDAHL: Well, it crops up in the news from time to time that sex-selective abortion has become very common in China. So that when women are pregnant, they get an ultrasound exam, and if the fetus turns out to be female, they abort. That’s not all women by any means, but enough that there’s a pretty significant gap in male and female births. So, that appears in the news every time a new census comes out, or a new survey, you see another article, and the Chinese press reports on it as well as the foreign press. But the reasons why that gap existed aren’t typically very well explained.

DUBNER: Hvistendahl dug into the scholarly work that had been done on the boy-girl ratio. In China, the overall ratio is 121 boys for every 100 girls. In one city, Lianyungang, the ratio was 163 to 100. And it wasn’t just in China. In India, the overall ratio was 112 to 100. According to one estimate, there are now more than 160 million missing women throughout Asia. That’s about the same size as the female population of the U.S. Now, this gap is surely not all the result of sex-selective abortions. There’s a lot of fatal violence against girls and women; girls tend to have worse economic and educational and medical opportunities. But still, no one can deny that in a lot of places, new parents have an overwhelming “son preference.” Now, why? There are probably a lot of reasons: male children carry on the family name, they’re considered more valuable to the parents, because they can perhaps provide for the parents later in life. So, I asked Hvistendahl: is this decision typically an economic one?  

HVISTENDAHL: You know, not entirely economic. I think there may be a more emotional reason as well. In my book I actually don’t go into these different reasons so much, because I was really interested in how you get the same trend of many more boys being born than girls in this very wide variety of cultures. More boys than girls are born in China, India, also South Korea, Taiwan, Azerbaijan, Armenia, Albania, Vietnam, and you know there are places that have very different political traditions, some shared religions and cultures, but there’s nothing that really binds them all together.


DUBNER: Nothing that binds these countries together except for maybe one piece of technology, yes, one piece of medical technology?


HVISTENDAHL: That’s right, ultrasound.

DUBNER: As we know, an ultrasound machine can be used to monitor a fetus in the womb to make sure the pregnancy is coming along all right. But since it was introduced in Asia in the 1980s, the ultrasound has also been used to determine the sex of a fetus and, if it’s a female, have an abortion. Over the past half-century, the female-to-male gap in Asia has more than tripled. So what happens in a world with a surplus of men? For starters, there’s more sex-trafficking, more AIDS, a higher crime rate. In fact, if you want to know the crime rate in a given part of India, one surefire indicator is the gender ratio: the more men, the more crime. Now, you might assume that sex-selective abortion is a plague of the lower classes, where the economic penalty of having a girl might have the most sting. But that’s not what Hvistendahl discovered.

HVISTENDAHL: It’s high-income people who have access to new technology first, so when the ultrasound machines arrive, they are the first to use them. But at the same time also the birth rate has fallen pretty dramatically among upper classes and among educated people. And so, when the birth rate drops, that puts pressure on a woman to make one of her children a son.

DUBNER: Now, the ultrasound machine didn’t create this kind of problem; but it does enable it. “Son preference” already existed; but along came a new birth technology that let mothers do something about it. Technology has consequences—often unintended ones. So do laws.

LEVITT: So, probably the most controversial finding that I’ve ever had in economics was the argument John Donahue and I made that legalized abortion led to decreases in crime.

DUBNER: That’s Steve Levitt again, my economist co-author.  He was in the library one day, just leafing through the Statistical Abstract of the United States...

LEVITT: And I was shocked to see the number: a million abortions a year. And I thought to myself, you know, a million of anything is a lot, but abortion. I never would have imagined that abortion was so prevalent. But I didn’t really have any sense of the scale, so I thought well how many births are there? And luckily this book had everything in it. So, I flipped a few pages forward, and I saw that there were only about three million live births in the United States each year. And I thought to myself one million abortions and three million live births, that means one out of every four pregnancies is ending in abortion. And that just seemed shocking to me. And I thought to myself that’s got to affect something.

DUBNER: Abortions spiked in the U.S. after Roe versus Wade, the 1973 Supreme Court decision that made abortion legal in all 50 states. As Levitt thought, a number that big – one million abortions a year, one in four pregnancies – that’s got to affect something. Now, he’d spent the past several years doing research on crime. His mind immediately mashed up his old research topic with this new one.

LEVITT: So, what does this legalized abortion have to do with crime? Well, the argument’s really simple, that there’s enormous volumes of scholarship going back fifty years that suggests that unwanted children are at risk for crime. Basically if your mother doesn’t love you, nothing very good is going to happen to you in your life. It’s also pretty clear that after legalized abortion became available, the number of unwanted children plummeted, so we see that the number of domestic children put up for adoption went way down. And in surveys, if you ask women whether they had unwanted births, those went way down as well. So, those two simple pieces of the argument are all it takes. Unwanted children are at risk for crime, and after legalized abortion the number of unwanted children went way down. Therefore, after legalized abortion, crime should go way down if you wait sixteen to eighteen years to the point where that cohort exposed to legalized abortion actually becomes old enough to be in the criminal ages.

DUBNER: That was the theory, at least. And as Levitt found, the data backed it up: Roe v. Wade, a decision meant to increase a woman’s reproductive control, was never intended to decrease crime in the U.S. But it did! Again, like the ultrasound, a natal development that had the most unintended of consequences. So you have to wonder: what’s next? What’s the next baby-making law or technology that we’ll be talking about in 20 years? And who’ll come up with it?

Stephen QUAKE: Well in this case the impetus was becoming a parent. You know, I think like everyone who becomes a parent, one can only marvel at how amazing it is to create a new life and watch it grow, but before your baby’s born it also can be a very nerve- wracking time.

DUBNER: That’s Stephen Quake, a professor of bioengineering at Stanford. I talked to him a while back about a new prenatal test he was developing, inspired by his own impending parenthood.

QUAKE: And in my case, because my wife was over thirty-five and only had her first kid, the doctors recommended amniocentesis as a form of prenatal testing to look for Down syndrome and things like that. And there’s some risk associated with the test because a large needle goes into mom’s belly, right up next to the baby to grab a few cells and to pull them out. And both sort of going through the procedure and the kind of the fretting while one waits for the results made a big impression on me. And we went through this whole thing twice, with two kids, and so that got me interested in perhaps trying to think of ways to develop non-invasive prenatal diagnostic tests.


DUBNER: Now, just walk me through your thinking on this. You were thinking you know what the procedure of an amnio is; it’s taking a big needle, it’s putting it into the mom, getting into the amniotic fluid very close to the baby, right? So there’s some risk there, but then the reward is some early notification of potentially, you know, abnormal circumstances. How did you, as a scientist, and as a father-to-be go about thinking about the tradeoff between the risk and reward there?


QUAKE: Yeah, you know, it’s a really difficult question. And the way the doctors explain it to you they say that the risk of there being something wrong is about one percent. In other words that you’ll learn something from the test, that there’s a genetic abnormality in the baby, that’s about a one percent chance. And the risk of losing the baby because of the test is also about one percent.

DUBNER: That risk of losing the baby as a result of amniocentesis made a big impression on Quake. He thought: there’s got to be another way. So he got to work on a simple blood test. It turns out that when a woman is pregnant, DNA from the fetus is floating in her bloodstream.

QUAKE: And so each molecule was voting for a chromosome, and we’re essentially looking for voter fraud, for slight overrepresentation of one chromosome relative to another. And so if baby has Down syndrome, they’ll be slightly more chromosome twenty-one molecules in the mother’s blood than any other chromosome.


DUBNER: Very good. So, what you’re talking about in a nutshell is, instead of amniocentesis, which is an invasive, and risky, and expensive procedure, relatively expensive procedure, you’re talking about a simple blood test to determine pretty much exactly what amniocentesis currently discovers, is that about right?


QUAKE: That’s about right.


DUBNER: Okay. So, if there’s a test, a blood test that’s non-invasive, not dangerous, presumably just about anybody who would be giving birth to a baby with Down syndrome could or would find out about it in time to do something about it. When you say do something about it, that’s really a euphemism for abort. Talk to me a little bit about that framework, and what your test contributes to change that framework.


QUAKE: Sure. Well, let’s take the euphemism for a moment, because I’ll take issue with that. You know, it’s a very personal and challenging question about whether to keep or terminate a pregnancy with Down syndrome. And there have been studies that have shown that for people who decide to take a Downs baby to term and deliver the baby, the earlier they know the baby has Down syndrome, the more prepared they are to deal with it, and sort of the lower the stress, and the better the outcomes both for the kid and the parents. And so, my argument would be that this sort of test adds value whichever side of the debate you’re on, whether you’re going to keep the baby or terminate it. The earlier you know the better, and it’s better for everyone.


DUBNER: And there are a lot of people who would argue that there is absolutely no reason in the world to terminate a Down syndrome pregnancy. That there’s a strong population of Down syndrome advocates, parents and people with it who say, yes, this is syndrome, it’s a set of challenges, a set of barriers, but not cause for termination. Do you have a position on that?


QUAKE: Yes, I do. And again, it’s sort of a very personal one, because it turns out that one of my wife’s cousin has Down syndrome, and I’ve known him for nearly twenty years. I think that time I first met him he was four years old, and so, kids with Down syndrome, are very interesting, they’re warm, and open, and loving, and I can completely understand and sympathize with the argument to having a baby with Down syndrome and deciding not to terminate having seen this kid grow up and what a wonderful kid he is.


DUBNER: Can you just blue sky and imagine what some of the surprising or counterintuitive consequences might be of having such an easier availability of a blood test that can detect Down syndrome? Do you think there will rise up to be part of the population that says, you know what they easier it is, then the more likely I am to not want to have it, because I want what life, and fate, and order, and God are going to give me? I don’t know. I’m just wondering what you thought through on that dimension?


QUAKE: Yeah, you know, it’s…The over all goal here I think is to lower stress. You know, impending parenthood is a very stressful time, and that was kind of my motivation here. And so the hope is to make these tools available to people who want them and to lower their stress. And for people who don’t want them, it’s fine, it’s not something I’m going to impose on a larger world. That being said, you know, if you want to think about blue sky things, we’re not very far away from being able to sequence the better part of the fetal genome non-invasively. And so you could learn many other things very early in pregnancy. And the question there is do you want to turn that loose on the parents?


DUBNER: And on society, really, right?


QUAKE: And on society, exactly. And the answer is I don’t know. It would be horrible to think that people are going to take some sort of action because, you know, their baby’s eyes weren’t going to be the color they wanted, or they’re not going to be as tall as they wanted. These are things that you could presumably figure out pretty early on. And in practice I think it is a sort of self-correcting phenomenon. You know it is challenging enough to conceive, especially these days, that people I’m hoping will not want to end it for trivial reasons. That’s where I hope it’s all going to equilibrate.

DUBNER: Since we first spoke to Stephen Quake, his blood test has been put into wide use. He estimates that well over 10,000 women have used it. Also, when Quake said that “we’re not very far away from being able to sequence the better part of the fetal genome non-invasively,” well, we indeed weren’t very far. In July, the journal Nature published a paper in which Quake and his co-authors describe how they’ve done just that. Which means, as Quake said, that expectant parents will soon be able to quite easily learn even more about their baby-to-be, from the color of her eyes to the probability that she may one day get colon cancer. It is, of course, impossible to predict how we’ll act on this knowledge. There are a lot of powerful laws in the universe, but the law of unintended consequences may be one of the most powerful, especially when applied to something as intricate, as intimate, as important as ushering a small new life into this big old world.



Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. robyn says:

    I have beem using such devices for yrs just never callef em that.havenot found the right one yet for article amd book and besides havent been looking of late. Here is one more place where the separation of public and private matters. Like ‘separation of church and state.

    Thumb up 0 Thumb down 2
  2. Brian Schack says:


    In your radio episode “Save Me From Myself,” you talk about the impact of domestic violence where men are the attackers and women are the victims [1]. I would like to suggest that you use more neutral language out of respect for men who may also be the victims.

    NYC Health says, “An estimated 93,000 women and 48,000 men in NYC report that they are afraid of their partner,” [2]. In particular, I am afraid that men may not report domestic violence because of stigmas created in part by lack of recognition in your radio program and elsewhere [3].




    Brian Schack

    Thumb up 0 Thumb down 0