By Paul Walker 05/02/2018


At the 2018 ASSA meeting Luigi Zingales delivered a lecture honouring Oliver Hart, co-winner of the 2016 Nobel Prize for economics.

In part Zingales said,

In the 1970s, this question [the make-or-buy dLuigi Zingaleecision] started to receive attention in the so-called transaction-cost literature. The key contributions during that period were Alchian and Demsetz (1972), Williamson (1971 and 1975) and Klein, Crawford, and Alchian (1978). Alchian and Demsetz identified the unique nature of the firm in the team production and the associated cost of metering individual contributions to the collective output. This is the line of research the other Nobelist we are celebrating today, Bengt Holmstrhom, pursued. By contrast, Williamson and Klein, Crawford, and Alchian focused on the role of the firm in mitigating the cost of opportunism. Imagine a printing press owned and operated by someone different from the publisher. If alternative users have much lower valuations for the services of the printing press than the current user, there exists an appropriable quasi rent. Klein, Crawford, and Alchian conclude that “if an asset has a substantial portion of quasi rent which is strongly dependent upon some other particular asset, both assets will tend to be owned by one party.” They are quick to add that “this advantage of joint specialized assets … must of course be weighed against the cost of administering a broader range of assets within the firm.”

In these few lines there are all the key insights of the pre-Hart literature, but also all its limitations, among them:

  1. If integration has to do with joint specialized physical assets, to what extent does an employee “belong” to a firm? What makes an employee different from an independent contractor? Are Alchian and Demsetz right in claiming that “I can fire my grocer by stopping purchases from him”? (Not just a theoretical discussion, think about Uber.)
  2. What aspect of integration produces all these benefits? Why can’t they be achieved through long-term contracting?
  3. Can you have too much integration? Are the limits of the firm determined only by “administrative or bureaucratic costs” or can integration also be detrimental? Does communism not work only because of bureaucracy?

It is these limitations that Hart along with Sandy Grossman and John Moore sort to deal with.

The main source of these limitations is the lack of a formal model, the lack of a language to express clearly the driving forces. The work by Grossman and Hart and Hart and Moore answers all these questions.

  1. First of all, the emphasis moves from the specialization of physical assets to that of human capital. In this way, they are able to explain how a firm has “power” over its employees. When my colleague “fires” United Airlines, he does not deprive United employees of any asset they have specialized to. When the CEO of United fires a pilot, he does deprive her of assets (physical and organizational) she specialized to. A pilot (especially an older pilot) is more valuable inside United than outside of it. Thus, a specialized pilot will continue being employed by United. Nevertheless, United’s ability to control the pilot’s quasi-rent is the source of the power an employer has over its employee.

One of the many merits of Oliver’s contribution is to have brought back the concept of power inside economics. This is a concept pervasive in political science and sociology, and pervasive in Marxian economics, but completely absent from neoclassical economics. In fact, Oliver’s view of the firm is very reminiscent of the Marxian view, but where Marx sees exploitation, Oliver sees an efficient allocation.

  1. Having identified the source of power, Grossman and Hart help us understand why this can only stem from integration.

To cut this Gordian knot, Oliver and Sandy introduce a new concept: the “residual right of control,” i.e. the right to dispose of an asset in all the situations that have not been explicitly contracted out. They identify this residual right of control with ownership.

With regard to ownership being identified with residual rights of control it is interesting to note a footnote in the Grossman and Hart paper,

Richard Posner, whose opinion on the legal definition of ownership we solicited, has referred us to the following statement by Oliver Wendell Holmes (1881/1946, p. 246): “But what are the rights of ownership? They are substantially the same as those incident to possession. Within the limits prescribed by policy, the owner is allowed to exercise his natural powers over the subject-matter uninterfered with, and is more or less protected in excluding other people from such interference. The owner is allowed to exclude all, and is accountable to no one but him.”

Zingales continues,

Of course, this opens the question of why some contingencies cannot be written in a contract, a question Oliver has spent a great deal of time on and a question I will return to momentarily, if time allows.

But if we accept that contracts are incomplete, then it is easy to see how the residual right of control can be used opportunistically to reduce the share of the surplus of other parties in a relationship.

This insight applies in all walks of life. For example, I just launched an economic podcast. In preparation for this event, I have spent a lot of time and effort. Much of this effort is specific to this particular podcast. So how does ownership of this podcast, which currently is in the hands of the University of Chicago, affect my incentives? Not only I, but most of the economics profession would not have a framework to think about this important question without the work of Oliver.

This is an inconsequential example, but it illustrates how rich the Grossman-Hart-Moore framework is in addressing a fundamental problem of entrepreneurs: how to allocate cash flow and control rights to maximize the commitment of all the key players to a new venture. I regularly teach this in my entrepreneurship class and I would not know how to frame this problem if it wasn’t for Grossman-Hart-Moore.

Note that—unlike Williamson—their results do not rely on friction in the renegotiation process. Grossman-Hart-Moore assume costless renegotiation ex post. Still, ownership (and the residual right of control it confers) matters because of the way it affects the out-of-equilibrium outside option and, through it, the share of ex post surplus each party appropriates. This share impacts not only the distribution of the quasi-rents, but also the ex ante incentives to make firm-specific investments, and thus efficiency itself.

  1. Having identified how integration works, Grossman-Hart-Moore can also explain the costs of integration. Since control is zero-sum, control given to party A takes control away from party B, reducing B’s incentives to make firm-specific investments. Thus, integrating a supplier reduces the incentive of the supplier to invest in his human capital. As a result, who should own what, the famous question of the boundaries of the firm, finds a very simple answer: it depends upon the relative importance of the contribution to the various parties.

Hart’s work has affected many areas of economics.

Before Oliver finance scholars had focused on the allocation of cash flow rights. Yet, it was difficult to explain why capital structure mattered purely on the basis of cash flow right allocation, since all the effects produced by financing decisions could be undone by contracts. It is thanks to Oliver’s model that researchers could start thinking in terms of control allocation: capital structure mattered because it provides a contingent way to allocate control. In other words, Grossman-Hart-Moore changed the way corporate finance theory was done and did so in an irreversible way.

Unlike poets who only allow people to express their feelings, economic theorists also provide a framework for empirical researchers to study new phenomena. The work of Kaplan and Stromberg (2003) on the allocation of control rights in venture capital contracts or the work of Michael Roberts and Amir Sufi on debt covenants would be inconceivable without Oliver’s work.

Similarly, it is very difficult to understand corporate governance without Oliver’s contribution. The famous survey by Shleifer and Vishny (1997) that incorporates Oliver’s work on control rights, changed the literature on corporate governance. Without Oliver’s seminal contribution that change would not have taken place.

While finance has been the main area of application of ICT, there is hardly a field in economics that has not been impacted. One of the first applications, pioneered by Oliver himself with Tirole, is to industrial organization. ICT provides a way to rationalize the famous market foreclosure argument. Another natural area of application is to the costs and benefits of public ownership. This also was pioneered by Oliver with Shleifer and Vishny in the famous prison paper. It is also not surprising that ICT has been applied to internal organization (Aghion) and to international trade (Antras). Finally, issues of power and contract incompleteness are essential to political economy, and in fact in recent years we have seen a proliferation of applications in this direction.

There are a lot of other areas where Oliver’s theory of incomplete contracts can be profitably applied, as family economics. There is hardly a contract that is more incomplete than marriage and one where relationship specific investments (like the children) are more important.

Oliver is a role model not only for his intellectual achievements, but also for the integrity with which he has achieved them. For more than a decade he argues back and forth with two other Nobelist Erik Maskin and Jean Tirole abut the foundations of his theory of incomplete contract. In spite of the enormous amount of reputation he had a stake, he had the courage to recognize that Maskin and Tirole’s critiques were valid and he went back to the drawing board and produced with John Moore a new foundation of incomplete contracts [this work is the ‘reference point’ approach to incomplete contracts].

The incomplete contracts theory has been one of the major theoretical advances in economics in the last 30 years and Hart’s role in it fully justifies the award of the Nobel Prize.