Incomplete Contracts and the U.S. Economy: Q&A with Nobel Laureate Oliver Hart

Oliver Hart, the recipient of this year’s Nobel Prize in economics, sits down for a special interview with ProMarket in which he discusses his life’s work and the state of the U.S. economy. 



Left to right: Oliver Hart, Guy Rolnik
Left to right: Oliver Hart, Guy Rolnik

Oliver Hart, the recipient of this year’s Nobel Prize in economics, is mostly famous for his decades-long work on contract theory. Together with his co-recipient Bengt Holmstrom, Hart’s work has effectively revolutionized the study of contracts, profoundly influencing both business practices and policymaking.


Hart, the Andrew E. Furer Professor of Economics at Harvard University, has been teaching at Harvard since 1993. In his work, he has laid the foundations for the theory of incomplete contracts, asserting that contracts are by their very nature incomplete since they cannot take into account every possible contingency, and explored the implications of this theory on policy areas such as private prisons.


We recently sat with Hart for a special interview, in which he discussed his work on contracts at length, as well as the state of the U.S. economy. Watch it here: 





The following is a transcript of the interview, slightly edited for clarity:


Guy Rolnik: Contract theory is probably your main contribution to our understanding of the world. How did this interest in contract theory develop, and what is it that we know today about contracts’ role in the economy and society that we didn’t know 30 or 40 years ago?


Oliver Hart: Those are big questions. My interest developed over time. I didn’t start off with an interest in contracts and started off as a general equilibrium theorist.


But I was interested in incomplete markets and I got interested in what firms should be maximizing in an economy where you don’t have a complete set of futures markets. What is the generalization of profit maximization in a world of uncertainty? What happens if different owners of the firm disagree about what the firm should do?


Over time, that evolved into an interest in conflicts between managers and shareholders, rather than between different shareholders, because I began to realize that was more important. The next step was to look at that in detail. The natural way to do that was to consider a contract between the manager and the shareholders.


I started working with Sanford Grossman on that. We weren’t the first. There was literature. This is the principal-agent literature, which was about the optimal arrangements between the principal, the shareholders and the agent, the manager, but that also evolved. Then, the next step was incomplete contracts. What happens when you can’t write a contract with your agent which anticipates all future events?


But, actually, the way we came at that was not through considering the manager and the shareholders, but rather we became interested in a separate topic and that led us to this. The separate topic was also a topic that had been studied a lot already, but this one was, “What’s the difference between a transaction inside of a firm and a transaction between two different firms?”


Again, there you have to think contractually. The transaction between two different firms is naturally regulated through a contract that the two firms devise. Is a transaction within a firm also a contractual transaction? Well, it seems to be in some sense. You have an employer and an employee. They have a contract. What’s the difference? We realized, at some point, in 1983, that the difference was what happens given that the contract is incomplete, who has residual control or residual decision rights?


It was an evolution, which led me to this. Then I did [more] later, where I just should say, my two chief collaborators in this endeavor have been Grossman, first, and then John Moore, second.


Oliver Hart receiving the Nobel Prize
Oliver Hart receiving the Nobel Prize

GR: What can this literature teach us about the role of contracts in today’s world?


OH: I think the principle thing is we’ve learned the importance of control. Other people who’d studied or thought about the firm and what’s different about within-the-firm transactions and between-the-firms transactions, like Ronald Coase and Oliver Williamson and people like that. They had talked about authority, so the idea that an employer can tell an employee what to do as opposed to having to negotiate with the employee, which is what happens between firms. It’s a negotiation rather than authority.


I think what Grossman, John, and I have added to the mix is the idea of control over non human assets, so ownership and control. The idea is that when Firm A buys Firm B and turns the inter firm transaction into an intra firm transaction that a key thing that’s changing is that Firm A acquires control over Firm B’s physical and other non human assets.


Forget about the people and don’t think in terms of being able to tell them what to do. I’m not saying that isn’t an issue, but we focused on a different aspect of it, which is that if I’m Firm A, I’m acquiring control over all the non-human assets that Firm B had, which might be machines, land, buildings, but also less physical things like patents, copyrights, existing contracts that Firm B had with other firms. The name of Firm B, all sorts of things like that.


To the extent that the initial contract was incomplete, and will always be incomplete, whatever contract we write will be incomplete. Having, owning those things now means that I can get to decide how they are used to the extent that the contract was silent about that. Whereas previously, it was the owner, Firm B, that wasn’t me, who had those rights. That’s a real change. That, we would argue, is one of the key reasons that Firm A is acquiring Firm B, to get those residual control rights.


GR: This is where the issue of bargaining power comes in. Why is bargaining power so important to our understanding of contracts?


OH: I’m going to give the example I often give. Let’s suppose I’m an electricity company, a power company, and you’re a coal mine. I want to use your coal to burn to make electricity.


The most interesting scenario is one where I’ve actually located my electricity plant right next to you, this happens in practice, in order to save on transportation costs of the coal.


Let’s consider two cases. One is where you’re a separate company, and now something comes up which wasn’t anticipated in the contract between us. Let’s say I needed a different type of coaI.


I would come, and we would negotiate about this. We’d renegotiate the contract. But the threat point in this case is if you own the coal mine, you can always say no. So I’m going to have to pay you quite a bit possibly in order to persuade you to change the coal.


Now consider the case where I bought your mine earlier on. It’s my mine. Now you’re simply the manager of the mine. Same situation, I want a different kind of coal.


This may seem a bit far fetched but suppose you didn’t really want to change the coal. Your bargaining position would be quite weak now because I can always replace you with some other manager. Presumably I can find one who’s willing to do what I want. If you’re human capital, if I really want to keep you on as the manager because you’re a really good manager, you might have a little bit of bargaining power. I might have to.


Let’s suppose mining this different coal was kind of unpleasant for you for some reason. I might have to pay you a bit to get you to do it. That might be better for me than replacing you. But replacement is always a pretty powerful threat point, much more powerful than the threat point in the case where the mine is yours.


Then my next best alternative is to go and ship the coal in from somewhere else, which can be extremely expensive. So you have much more leverage in that case. In most of my work, not all of it but certainly the main stuff, this is what actually drives things, that we’re going to be bargaining in all cases, but our threat points will be very different depending on who owns the mine or, more generally, who owns which asset.


GR: We tend to think that markets are usually the best way to allocate resources, but when we look at the world, we tend to forget that companies are responsible for a large part of the economic activity. Nowadays we have very large companies that operate like small countries of command and control structure. Why is it that so much activity is done in these huge corporations, and not in markets?


OH: I think the work on the firm, starting with Coase, continuing with Williamson, continuing with my work with co-authors, has shed some light on it. I’ll tell you what I think those theories say in a minute. But I’m not sure I can say, “Oh yeah, that explains why we have these giant companies.”


Sometimes these companies seem incredibly large. You read about the merger of AT&T and Time Warner. Is that really a good thing? Is it necessary on efficiency grounds? I have no idea because I don’t know enough about the particular case, but they seem like giants. Maybe there are other reasons they’re doing it than pure economic ones.


I would say two things. First of all, the classic market economy is the way economists used to think about it. That’s one that applies when there really aren’t frictions, that is, there aren’t lock in effects. I can trade with you today and someone else tomorrow. There’s nothing special about our relationship.


This applies to quite a few transactions. If I buy shares on the stock market from you, there’s nothing special about you. If I go back into the market and want to buy some more shares, I’ll be buying them from somebody else. These are sort of anonymous trades. In this kind of situation, a standard market setting is the natural way to carry out the transaction.


But then there are other cases like the electricity company and the coal mine, where the electricity company is actually locating a plant next to the coal mine. Once it’s done that, it wants to deal with that coal mine. We’ve turned something which might have been a frictionless ex ante into a situation ex post where the costs of going somewhere else are very great.


It’s not at all like me buying shares from you or somebody else. There are no costs at all of switching partners in that situation whereas there are huge costs in the electricity coal situation.


There are many cases in between. Think of a worker who works for a firm. They get a job, both parties had to search. They had to figure out whether there was a good match. Now that they’ve decided there is a good match, there are going to be costs for each party if they switch partners.


Whenever you have situations like that where there are switching costs, contracts become important. We have to recognize that the classical market view is actually quite a special one. There are many other ways that transactions take place.


GR: How can we explain the outcomes that we see in markets, both private and public?


OH: I think one has to think about it the same way. We’re talking about a situation where the government is going to pay for a service. It’s not difficult to understand from an economic point of view why that might be the case.


Many economists have long talked about public goods. These things will not be financed at an efficient level just through laissez-faire. We’re going to have to have a government who’s going to come in and finance them.


But it doesn’t follow from that that the government actually has to produce them. The government can pay for them but could get a private company to actually produce them.


In my work with Andrei Shleifer and Rob Vishny, we picked an example. It wasn’t the only example we talked about, but the one we focused on was prisons. It’s very obvious, I think, that the market is not going to pay for prisons, because having people incarcerated is something that lots of people benefit from. It’s a public good. There are free rider problems.


The government will be the party that will finance a prison. But should it run the prison itself? Should the prison be a government-run operation where the people in the prison are employees? Or should the government write a contract with a private company to run the prison?


That choice is very similar conceptually to, “Should the electricity company contract with the coal mine or should it buy the coal mine and turn it into an intra-firm transaction?” Conceptually, it’s the same, so you can use the same ideas to analyze the trade off.


GR: When it comes to prisons, your theory implies that usually prisons should be owned and run by the government. You cannot outsource it to the private market.


OH: Actually we had a slightly more nuanced answer. Actually, if you look at the contractswe did in this paper talk about the contracts that are actually writtenbetween a government and a private contractor to run a prison, they’re quite elaborate.


But we argued that there are some things it’s difficult to specify in the contract. One is the quality of the guards. We argued that the private company might have an incentive to hire lower quality, cheaper, less well trained guards and yet still be within the contract. This would be an example of if you could write a perfect contract, this wouldn’t be an issue because you could specify exactly the type of training the guards should receive. But we argued that in fact that was difficult to do. There was a gap there which a private contractor could take advantage of.


Then we went on to consider, “When is it going to be a serious problem to have these lower quality guards?” We argued that this was going to be the case more with high security prisons, the most dangerous prisoners. [With] those prisoners, we argued, this could be an issue, and the case for public ownership was quite strong. But with lower security prisons with less dangerous prisoners, it was going to be less of a problem. So there a private outcome might not be bad.


GR: Privatization is a very contentious issue. You developed a clean and simple model of what can be outsourced or privatized, based on what we can monitor, enforce, or design in contracts. Are those really the most important issues, or perhaps we should take into account other factors like corruption, culture, civic capital? Can contract theory shed some light on those questions?


OH: Well, that’s an excellent question. I’m not sure I know the answer to it. I think this approach is helpful.


I never think that economics provides the full answer to almost anything actually. I think this is one of the mistakes people make about economics. This is why they’re often disappointed.


Why can’t economists predict the Great Recession? Why didn’t they? That’s I think misunderstanding what economics is about. Economics is a helpful tool. There are many, many problems or questions where having economists give their views on what should be done is going to be extremely helpful, but it doesn’t mean you only want to listen to them. That’s my general view. That may be true here.


I do think though that where this is useful is that, as you said, a lot of this private-public ownership discussion is almost entirely political, that is there are going to be people on the left who think that it’s outrageous to do these things through a private contractor because they think private is bad, anyone who makes money, profits are badthe extreme case where you have socialists or communists who think everything should be run by the government. On the other hand, you have people on the right who think all government is bad. Government’s bad so the government shouldn’t be running anything.


I think both of them are wrong. The truth is somewhere in the middle. I think this approach is helpful to tell us where it should be in a particular situation. It’s a useful framework because what it says is we can just think about, “What are we trying to achieve here? What’s the best way to achieve it?”


Sometimes private is good because actually having people pursue profit gives them good incentives to be efficient and to reduce costs. That actually can often be a good thing for everybody. It can also give them the incentives to innovate. Those are also good. But it can also sometimes have them reduce costs in socially undesirable ways. That can be bad. So it’s a trade off. I think this is a useful perspective, but I don’t deny that sometimes other things are important as well.


GR: Can we say that your contract theory foreshadowed the recent decision to phase out the use of private for-profit prisons?


OH: I even read an article in a newspaper, a very good newspaper, which suggested that this paper, with Andrei Shleifer and Robert Vishny, was responsible for what happened. Which was news to me.


I’d be happy in a way to think that was true. I didn’t realize I had that much influence until I read that. I think it could be said that we anticipated some of the problems.


GR: Some people say the answer to the issue of corruption is “let’s privatize.” Others say, “government ownership.” You have a more nuanced view on this.


OH: We did have a nuanced answer. The question is, “Can I remember it?” I think we said that sometimes if you privatize things, but there’s corruption there, you might end up selling to the wrong person at a low price because there’s some kickback. The auction is rigged in some way. This could be a serious issue in some contexts. That might push you towards not going with a private contract.


More generally the private contractor could get a very sweet deal. If you keep it in the hands of the government, you’re going to avoid that problem. But then there are other forms of corruption which go the other way, where having it in government hands means that you can have a lot of patronage, people being employed in the prison to get votes, and this kind of thing. A lot of inefficiency that way. Again, it’s not a simple answer unfortunately, but life is not simple.


GR: Is this where the ideas of civic capital, social capital, norms in society, play out tremendously?


OH: Absolutely. Yes.


GR: When we look at a contract we have on the one hand what public choice people would say concentrated interests. On the other side we can have dispersed consumers, investors, and taxpayers. How would your framework about contracts, specifically about the enforceability and the monitoring of contracts, help us understand the implication of this? How would that play out in your models?


OH: I would say it hasn’t played out very well so far. I would say that’s a gap in the theory. I’ll tell you why. According to classical theory, and that’s even true of contract theory, if I’m a big company and you’re a little consumer, let’s say I have the bargaining power. Basically, I set the terms of the contract. The implication of that is that you won’t do very well.


I will go to use economics jargon: I will extract most of the surplus from the transaction. But I should still have an incentive to do something very efficient. To take a concrete example, suppose we’re talking about a product. The question is, “Should there be a warranty for it?”


Even though I have all the power in the relationship, in terms of having all the bargaining power, if a warranty makes sense on efficiency grounds, I should often use a warranty. I should include it. I would simply charge you a lot for it. If warranties are efficient, they will be offered. The only thing that will be different, because I have a lot of power, is that the price will be different.


GR: Or you can offer me a warranty, and when the time comes, it will be difficult for me to enforce that warranty, because there are going to be caveats.


OH: That’s a good point. Let me just say that I think in reality sometimes I won’t offer you the warranty. Maybe the reason is that if I did offer it to you, the price for that, or more generally the price for the product, would be so high that it would look bad or something, and so instead I keep the price lower and I simply shade on the quality, that is, I don’t offer you…We actually get an inefficiency.


That’s something that’s outside the scope of standard theory. Because it would have to do with perhaps psychology and so on, how you could react to a very high price or something like that.


Now the other thing though that you mentioned, that’s important. Which is if we have a dispute, what happens? That gets into the legal system, how it functions, and whether there are ways around that. In principle, the legal system could be set up in such a way that a little guy…In fact we do have a bit of this, with class action suits and so on. There’s some attempt to provide you with a way to actually fight me.


Another way would be if you sue and you’re successful, maybe the damages would be very great. We could even, by the way, write that into the contract.


If you’re worried ahead of time when I sell the product to you that, “OK, it comes with a warranty, but you’re never going to be able to enforce it. Because I have lots of lawyers working for me,” we could have actually written into the contract that we could write that you could sue. If successful I pay your legal fees. That could be part of our contract. We could agree that if you’re successful, given that you have to fight an army of lawyers, then the damages I will agree to pay are enormous to compensate you for that. These are things which, actually in a way, a contract could deal with.


Then the question is, “Why don’t we see that?” I don’t think we see very much of that. I’m afraid again it may come back to some psychological issues or something. I would say I don’t have very good answers to those questions. They’re important questions though.


GR: For many years, the top of the agenda when we talked about economics was productivity, growth, and taxes, big government vs. small government. In the last few years, inequality has risen to the top of the agenda. As an economist, do you think inequality is something that should be left to politicians?


OH: I think it’s important for economists too, because there’s a very well developed branch of economics called welfare economics where people spend a lot of time thinking about why we need taxes, what’s their purpose. One of the purposes is to redistribute income. I think many economists think that some income redistribution is good.


At one extreme, you would have people on the right who think that you shouldn’t go any further in providing a safety net for the very poor. You just want to have a floor that people don’t fall below, but that’s it. We don’t worry about inequality beyond that. Then you will have people on the left who say, “We’re very concerned about the rich not being too rich,” and that sort of thing. I’m not sure where I fall on that spectrum, but that is certainly a debate within welfare economics.


Utilitarians would think that there’s a big role for doing more than providing the safety net. Actually if you look at the optimal income taxation literature starting with James Mirrlees’ work, that also puts everybody behind the veil of ignorance.


It says, let’s imagine we were designing an optimal tax subsidy scheme before we know who we’re going to be, whether you’re going to very talented or very untalented. What’s the optimal arrangement in that situation? That’s going to involve some concern with not having too much inequality.


GR: Due to this veil of ignorance, the arrangements we’ll probably see will be pretty different than those we see now.


OH: Yes, although I’m not an expert in that area at all, but I don’t think you get extremely progressive tax rates out of that actually. Let me say, from my point of view, I think tax rates in this country are too low. I would be in favor of raising tax rates for the rich.


GR: You mean taxes on personal income?


OH: Yes. As everybody says, the system is mind bogglingly complicated with all sorts of special divisions and so on. It would be wonderful to clean that up. Of course, every time it gets cleaned up, then it gets dirty again. 


GR: You spent a lot of time understanding why firms are the way they are, and why they’re so important. One of your students, Professor Luigi Zingales, believes we need to reconsider the theory of the firm, because in many large firms and other special interest groups are able to design the rules of the game. Do you agree?


OH: I think it’s very possible. I think it’s a very important extension. It’s not part of my work, but I think when we go and look at the world, it’s plausible that it’s a part of what’s going on out there. I never thought my work was the full story.


I think the fact that these political elements, which are important, is not only important but intellectually very interesting. The fact that you might have a merger not for efficiency reasons at all, not for empire building, not because the managers just want to control more, but because the bigger firm will have an opportunity to get the rules changed in Washington or wherever elsethis is a new element, a new part of the story, and an interesting and important part. I don’t know much about it but I follow Luigi’s work in this area with interest.


Disclaimer: The ProMarket blog is dedicated to discussing how competition tends to be subverted by special interests. The posts represent the opinions of their writers, not those of the University of Chicago, the Booth School of Business, or its faculty. For more information, please visit ProMarket Blog Policy. 

One comment

Leave a Reply