Research Matters: Saul Levmore on “Credible Threats”

Research Matters is a biweekly feature in which a member of the faculty talks about some of his or her latest work and its impact and relevance to law and society.

Saul Levmore, William B. Graham Distinguished Service Professor of Law, wrote “Credible Threats,” with Ariel Porat, Greenbaum Distinguished Visiting Professor of Law, who is now working from Tel Aviv University. In the paper, the authors explore various kinds of threats, running from the criminal to the commercial, and the way the law handles or ignores them. They propose that there may be room for the law to make some threats more credible – that is, bound by law or otherwise valuable to those making and receiving these “signals.” Levmore spoke about the work.

Q. Why are threats interesting?

A. We’re accustomed to threats. When I say, “give me your money or I’ll kill you,” we understand the threat, and there is the generally interesting question about when we punish threats in criminal law, and when we wait for you to do something worse. But there’s literature on that, so let’s leave that alone for a minute. I think the novel thing about this work with Professor Porat is that we talk about situations where a threat is conveying information that is really useful. Think about commercial settings, such as buying a car. Let’s say I go to a car dealer, and the car is offered for $30,000, and I say, well how about $20,000? The dealer says, don’t be ridiculous. I say, OK, $25,000 and I’ll buy it right now. I’m leaving the showroom unless you sell it to me for 25. We all know this game. But sometimes, I really won’t buy the car for more than $25,000, and I wish the guy would know that. He thinks I’m just another one of these people trying to get it at a lower price but otherwise intending to return the next day in order to buy it at 30.

There might be room for a bargain. I want to say that I guarantee that if you don’t lower it to 25, you won’t be selling me this car. I’m threatening you, and it’s credible. Then let’s say I come back tomorrow and buy it for $28,000. The guy is happy to sell it to me for that; even if we had a remedy, he would have no interest in going to court. So what we’re suggesting is that there might be room for a strange remedy. I really want him to know I’m not coming back tomorrow because I want to get that low price today. If I do come back tomorrow, go look at this new Levmore/Porat law. The law is, that if I sign a threat on a blue piece of paper – we have to make it some formality – then tomorrow, if I come back and say 28 and you sell it to me for 28, I can sue you for three later on. Meaning, I really mean it that you better help me not buy this car for more than $25,000.

Q. Why would the onus be on the car dealer?

A. It’s hard to see how to enforce it any other way. And our point is that it’s possible this makes them both better off. Because he wants to know when I really am the sort of customer who only wants to buy it at 25, and I want to be able to make the credible commitment. And that’s what we mean by credible threats.

We suggest that maybe law should make room for credible threats, just as it makes room for credible promises. If I promise to give the Law School a million dollars, there is a way to firm that up. But, if I promise that if the Law School doesn’t start teaching a certain course, I will never give it a million dollars, it is not possible to make that credible. Our suggestion is that perhaps we can make such threats credible.

Q. You also write about threats in criminal settings, such as kidnapping. Why?

A. Have you ever kidnapped someone? Ever been kidnapped? It’s an interesting crime. Think about the kidnapper. He kidnaps Smith’s kid, and he says, give me a million dollars or you’ll never see the kid again. Look at what the target has to believe. Smith has to believe that, first of all, the threatener really has the kid, and that he’ll really harm the kid if Smith doesn’t pay. That’s not so hard to believe. But Smith also has to believe that if he gives up the money, the kid will be returned. And the kidnapper must believe that when Smith gives him a million dollars for the kid, it’s not traceable. It’s very, very tricky to do that kind of ransom kidnapping. That’s why kidnapping’s not such a popular crime. It’s really hard.

We talk about primary credibility and secondary credibility. Primary credibility is the extent to which you believe me that I have the kid and will harm the hostage if you do not pay the ransom. Secondary credibility refers to the plausibility of my not repeating the threat, or proceeding with the harm even if you pay. It’s like blackmail; I tell you I have a picture of you killing someone, and for a million dollars I’ll destroy it. But do I have secondary credibility, so that you can be sure I do not have another copy of the photo? We talk about ways to make threats more credible in commercial contexts and and less credible in criminal contexts, where credibility likely enhances criminal activity.

Q. What about threats in international relations? What did you learn there?

A. Let’s say I’m the leader of a country, and I want a disputed piece of land in the neighboring country. It will cost me $3 billion to invade, and the land is worth $2 billion. Would I spend 3 to get 2? No, so any threat is incredible. But I would spend $1 billion to get $2 billion. So maybe I assemble my army and go to the border and that costs me 2 out of the total 3 in invasion costs. And then I say to you, look, I’ve already spent the 2. It’s gone. Now, it’ll cost me just 1 more to invade and the land is worth 2, so I’m going to proceed. You might now see this as presenting a credible threat.  I made that threat credible by sinking costs.  And now I don’t have to incur the full costs of invasion.  

It’s a really good example of when you have a threat that can be divided into parts. There's usually a way of dividing it to make the threat more credible. I think it might cast some light on these international negotiations, and why sometimes parties have to take steps. Probably you’ll think, I’m not going to give in until you spend the first $2 billion. You might want me to spend resources or rely on international law or opinion once people see me amassing an army. Or maybe my population won’t stand for it, if I just give over the land before you even do anything. Ironically, when you think about it that way, you start to understand how bullies work. I show you I can destroy you – give me your lunch money! – and I build up a reputation. And after a while, you hand over the lunch money before I even have to take the first step.

Q. What made you want to study threats?

A. I’m very interested in asymmetries in law. Porat and I have written a few papers about asymmetries. For example, we have an earlier paper questioning why prosecutors can plea bargain, but defendants can’t. We find there’s usually something interesting going on in the law when the rules are asymmetrical. And so here’s another interesting asymmetry: generally, law does not enforce mere promises or mere threats, but there is a way to make your promise credible, enforceable. There isn’t a way to make your threat credible. And maybe, in some cases, there should be.