Incomplete contracts.
Hart, Oliver D.
Economists have a very well-established theory of market trading, and
are on the way to a similarly well-developed theory of contractual
transactions. However, the economic analysis of institutions is in a
much more rudimentary state. This article discusses a recent literature
that tries to provide a framework for thinking about economic
institutions such as firms. The basic idea is that firms arise in
situations in which people cannot write good contracts, and in which the
allocation of power or control is therefore important.(1)
The starting point of this recent literature - which is sometimes
called the "incomplete contracting" approach - is that it is
prohibitively expensive for parties to write a contract governing their
economic relationship that is all-inclusive, that is, that anticipates
all the many things that may happen. Instead, parties will write a
contract that is incomplete, and that will be revised and renegotiated
as the future unfolds. The contract they write can be seen as a backdrop
or starting point for such renegotiations, rather than a specification
of the final outcome. Thus, the parties look for a contract that will
ensure that, whatever happens, each side has some protection, both
against opportunistic behavior by the other party and against bad
luck.(2)
In a world of incomplete contracts, the ex post allocation of power
(or control) matters. Here power refers roughly to the position of each
party if the other party does not perform (for example, if the other
party behaves opportunistically). These two ideas - contractual
incompleteness and power - can be used to understand a number of
economic institutions and. arrangements. I discuss some of these in the
remainder of this article.
The Meaning of Ownership
Economists have written a great deal about why property rights are
important, and in particular why it matters, for example, whether a
machine is privately owned or is common property. However, they have
been less successful in explaining why it matters who owns a piece of
private property. To understand the difficulty, consider a situation in
which I want to use a machine that you own. I can buy the machine from
you or rent the machine from you. If there are no contracting costs,
then we can sign a rental agreement that is as effective as a change in
ownership. In particular, the rental contract can specify exactly what I
can do with the machine; when I can have access to it; what happens if
the machine breaks down; what rights you have to use the machine; and so
on. Given this, however, it is unclear why changes in asset ownership
ever need to take place.
In a world with contracting costs, though, renting and owning are no
longer the same. If contracts are incomplete, not all the uses of the
machine will be specified in all possible eventualities. The question
then arises; who chooses the unspecified uses? A reasonable view is that
the owner of the machine has this right; that is, the owner has residual
rights of control over the machine, or residual powers. For example, if
the machine breaks down or requires modification and the contract is
silent about this, then the owner can decide how and when it is repaired
or modified.
Now it is possible to understand why it might make sense for me to
buy the machine from you, rather than to rent it. If I own the machine,
I will have more power in our economic relationship, since I will have
all the residual rights of control. To put it another way, if the
machine breaks down or needs to be modified, I can ensure that it is
repaired or modified quickly, so that I can continue to use it
productively. Knowing this, I will have a greater incentive to look
after the machine, to learn to operate it, to acquire other machines
that create a synergy with this machine, and so on.
The Boundaries of Firms
A long-standing issue in organization theory concerns the
determinants of the boundaries of firms. Why does it matter if a
particular transaction is carried out inside a firm, or through the
market, or via a long-term contract? To put it another way, given any
two firms (A and B), what difference does it make if the firms transact through an arms-length contract, or merge and become a single firm?
It is difficult to answer these questions using standard theory for
the same reason that it is hard to explain why asset ownership matters.
If there are no contracting costs, then the two firms can write a
contract governing their relationship that specifies the obligations of
all parties in all eventualities. Since the contract is all-inclusive,
it is unclear what further aspects of their relationship could be
controlled through a merger. This is true whether firm A is buying an
input from firm B, or firms A and B sell complementary products and want
to save on some duplicative production costs.
But since contracts are incomplete, it is possible to explain why a
merger might be desirable. Consider the well-known example of Fisher
Body, which for many years supplied car bodies to General Motors (GM).
For a long time, Fisher Body and GM were separate firms linked by a
long-term contract. However, in the 1920s, GM's demand for car
bodies increased substantially. After Fisher Body refused to revise the
formula for determining price, GM bought Fisher out.(3)
Why did GM and Fisher Body not simply write a better contract?
Arguably, GM recognized that, however good a contract it wrote with
Fisher Body, situations similar to the one it had just experienced might
arise again; that is, contingencies might occur that no contract could
allow for. GM wanted to be sure that next time around it would be in a
stronger bargaining position; in particular, it would be able to insist
on extra supplies, without having to pay a great deal for them. It is
reasonable to suppose that ownership of Fisher Body would provide GM
with this extra power by giving it residual control rights over Fisher
Body's assets. At an extreme, GM could dismiss the managers of
Fisher Body if they refused to accede to GM's requests.(4)
Of course, although the acquisition increased GM's power and
made GM more secure in its relationship with Fisher Body, it arguably
had the opposite effect on Fisher Body. That is, Fisher Body may have
had more to worry about after the merger. For example, if Fisher
Body's costs fall, GM is now in a stronger position to force a
reduction in the (transfer) price of car bodies, hence reducing the
return to Fisher managers. Anticipating this, Fisher managers may have
less incentive to figure out how to reduce costs. Thus, there are both
costs and benefits from a merger.(5)
Together with Sanford J. Grossman and John Moore, I have developed a
theory of the firm based on the idea that firm boundaries are chosen to
allocate power optimally among the various parties to a transaction.(6)
This work argues that power is a scarce resource that never should be
wasted. One implication of the theory is that a merger between firms
with highly complementary assets enhances value, and a merger between
firms with independent assets reduces value. If two highly complementary
firms have different owners, then neither has real power, since neither
can do anything without the other. It is then better to give all the
power to one of the owners through a merger. On the other hand, if two
firms with independent assets merge, then the acquiring firm's
owner gains little useful power, since the acquired firm's assets
do not enhance their activities. The acquired firm's owners lose
useful power, though, since they no longer have authority over the
assets they work with. In this case, it is better to divide the power
between the owners by keeping the firms separate.
Financial Securities
A. Debt
Suppose you have an interesting idea for a business venture, but do
not have the capital to finance it. You go to a bank to get a loan. In
deciding whether to finance the project, the bank is very likely to
consider not only the return stream from the project, but also the
resale value of any assets you have or will acquire using the
bank's funds; in other words, the bank will be interested in the
potential collateral for the loan. In addition, the durability of your
assets and how quickly the returns come in are likely to determine the
maturity structure of the loan. The bank will be more willing to lend
long-term if the loan is supported by assets such as property or
machines than if it is supported by inventory, and if the returns arrive
in the distant future rather than right away.
These observations fit in well with the ideas of incomplete contracts
and power. The bank wants some protection against worst-case scenarios.
If there is very little collateral underlying the loan, then the bank
will worry that you will use its money unwisely or, in an extreme case,
disappear with the money altogether. Similarly, if the collateral
depreciates rapidly or the returns come in quickly, then the bank would
be unhappy with a long-term loan: it would have little protection
against your behaving opportunistically when the collateral was no
longer worth much, or after the project returns had been realized (and
"consumed"). Basically, the bank wants to ensure a rough
balance between the value of the debt outstanding and the value
remaining in the project, including the value of the collateral, at all
times.
Moore and I have developed a theory of debt finance based on these
ideas, and derived results about the kinds of projects that will be
financed.(7)
B. Equity
Investors who finance business ventures sometimes take equity in the
venture rather than debt. Unlike debt, equity does not have a fixed set
of repayments associated with it, with nonpayment triggering default.
Rather, equity-holders receive dividends if and when the firm chooses to
pay them. This potentially puts equity-holders at the mercy of those
running the firm, who may choose to use the firm's profits to pay
salaries or to reinvest rather than to pay out dividends. Thus
equity-holders need some protection. Typically, they get it in the form
of votes. If things become bad enough, equity-holders have the power to
remove those running the firm (the board of directors) and replace them
with someone else.
However, giving outside equity-holders voting power brings costs as
well as benefits. Equity-holders can use their power to take actions
that ignore the (valid) interests of insiders. For example, they might
close down an established, family-run business or force the business to
terminate long-standing employees. Philippe Aghion and Patrick Bolton
have analyzed the optimal allocation of power between insiders and
outsiders.(8)
Dispersed Power
So far I have supposed that those with power wield it. That is, I
have assumed that owners will exercise their residual control rights
over assets; for example, equity-holders will use their votes to replace
a bad manager. However, if many people hold power, then no one of them
may have an incentive to be active in exercising this power. Then it is
important that there be automatic mechanisms for achieving what those
with power are unable or unwilling to do by themselves.
A leading example of dispersed power is the case of a public company
with many small shareholders. Shareholders cannot run the company
themselves on a day-to-day basis, so they delegate power to a board of
directors and to managers. This creates a free-rider problem: an
individual shareholder does not have an incentive to monitor management,
since the gains from improved management are enjoyed by all
shareholders, whereas the costs are borne only by those who are active.
Because of this free-rider problem, the managers of a public company
have a fairly free hand to pursue their own goals: these might include
empire-building, or the enjoyment of perquisites.
Two "automatic" mechanisms can improve the performance of
management: debt (in combination with bankruptcy) and takeovers. Debt
constrains managers. If a company has a significant amount of debt, then
management is faced with a simple choice: reduce slack - that is, cut
back on empire-building and perquisites - or go bankrupt. If there is a
significant chance that managers will lose their jobs in bankruptcy,
then they are likely to choose the first option.
Takeovers provide a potential way to overcome problems involving
collective action among shareholders. If a company is badly managed,
then there is an incentive for someone to acquire a large stake in the
company, improve performance, and make a gain on the shares or votes
purchased. The threat of such action can persuade management to act in
the interest of shareholders.
The view of debt as a kind of constraint can explain the types of
debt that a company issues (for example, how senior the debt is, or
whether it can be postponed).(9) The possibility of takeovers can
explain why many companies bundle votes and dividend claims together -
that is, why they adopt a one share-one vote rule. One share-one vote
protects shareholder property rights: it maximizes the chance that a
control contest will be won by a management team that provides high
value for shareholders, rather than high private benefits for
itself.(10)
Bankruptcy
Of course, if a company takes on debt, then there is always the
chance that it will go bankrupt. If there were no contracting costs,
then there would be no need for a formal bankruptcy procedure, because
every contract would specify what should happen if some party could not
meet its debt obligations. However, in a world of incomplete contracts,
there is a role for bankruptcy procedure. A bankruptcy procedure should
have two main goals: First, a bankrupt company's assets should be
placed in their highest-value use. Second, bankruptcy should be
accompanied by a loss of power for management, to ensure that management
has the right incentive to avoid bankruptcy. Aghion, Moore, and I have
been working on a procedure that tries to meet these goals, and at the
same time avoids some of the inefficiencies of existing U.S. and U.K.
procedures.(11)
Our basic idea of the procedure is that a bankrupt company's
debts are canceled and the company is put up for auction (the auction
would be supervised by a judge, say). However, in contrast to a standard
auction, noncash bids are permitted. A noncash bid allows existing
management, or any other management team for that matter, to propose a
reorganization plan. As an example of a noncash bid, incumbent
management might offer former claimants shares in a new (debt-free)
company managed by the old management team. Alternatively, managers
might offer former claimants a combination of shares and bonds in the
post-bankruptcy company.
At the same time that bids are being made for the company, an
automatic debt-equity swap takes place. Senior creditors of the company
receive equity in the new (post-bankruptcy) company, while junior
creditors and former shareholders receive options to buy equity (the
exercise price of the options issued to each class is set equal to the
amount owed to classes senior to that class).(12) After the bids come in
- three months might be allowed for this - another month or so is
allowed for option-holders to exercise their options. The final step in
the process is that the company's equity-holders (that is, those
people who hold equity in the company at the end of the fourth month)
vote on which of the various cash and noncash bids to select. Once the
vote is completed, the winning bid is implemented and the firm emerges
from the bankruptcy process.
This procedure has the advantage relative to Chapter 7 of the U.S.
Bankruptcy Code that it allows for the firm to be restructured as a
going concern if this is efficient. It has the advantage relative to
Chapter 11 of the U.S. Bankruptcy Code that bargaining between different
claimant groups with possibly conflicting interests is avoided: instead
the firm's future is decided by a simple vote by a homogeneous
class of shareholders.
1 For a more extensive discussion of this literature, see O. D. Hart,
Firms, Contracts, and Financial Structure, Oxford, England: Oxford
University Press, forthcoming in late 1995. This article is based on the
introduction to this book.
2 The incomplete contracting approach borrows a great deal from the
earlier transaction cost literature. See, in particular, R. H. Coase,
"The Nature of the Firm," Economica 4 (193 7), pp. 386-405; B.
Klein, R. Crawford, and A. Alchian, "Vertical Intergration,
Appropriable Rents, and the Competitive Contracting Process,"
Journal of Law and Economics 21/2 (1978), pp. 297-326; and O.
Williamson, The Economic Institutions of Capitalism, New York: Free
Press, 1985.
3 For interesting and informative discussions of the GM-Fisher Body
relationship, see Klein, Crawford, and Alchian, op. cit.; and B. Klein,
"Vertical Integration as Organizational Ownership: The Fisher
Body-General Motors Relationship Revisited," Journal of Law,
Economics and Organization 4/1 (1988). pp. 199-213.
4 There has been some debate about whether GM did in fact increase
its power over Fisher Body by buying Fisher Body out. See R. H. Coase,
"The Nature of the Firm: Influence," Journal of Law, Economics
and Organization, 4/1 (1988)p. 45.
5 Sometimes the costs of a merger will exceed the benefits. This may
explain why GM did not merge with A. O. Smith, which has supplied a
significant fraction of its automobile frames for many years. For a
discussion of the A. O. Smith case, see Coase, "The Nature of the
Firm: Influence," op. cit.; and Klein, "Vertical Integration
as Organizational Ownership . . .," op. cit.
6 See S. J. Grossman and O. D. Hart, "The Costs and Benefits of
Ownership: A Theory of Vertical and Lateral Integration," Journal
of Political Economy 94 (1986), pp. 691-719; and O. D. Hart and J.
Moore, "Property Rights and the Nature of the Firm," Journal
of Political Economy 98 (1990), pp. 1119-1158. See also Hart, Firms,
Contracts, and Financial Structure, op. cit., Chapters 2-4.
7 O. D. Hart and J. Moore, "A Theory of Debt Based on the
Inalienability of Human Capital," NBER Reprint No. 1963, April
1995, and Quarterly Journal of Economics 109 (1994), pp. 841-879; and
"Default and Renegotiation: A Dynamic Model of Debt," MIT Working Paper No. 520, 1989. For related work, see P. Aghion and P.
Bolton, "An 'Incomplete Contracts' Approach to Financial
Contracting," Review of Economic Studies 59 (1992), pp. 473-494;
and P. Bolton and D. Scharfstein, "A Theory of Predation Based on
Agency Problems in Financial Contracting," American Economic Review
80 (1990), pp. 94-106.
8 See Aghion and Bolton, "An 'Incomplete Contracts'
Approach. . .," op. cit.
9 O. D Hart and J. Moore, "Debt and Seniority: An Analysis of
the Role of Hard Claims in Constraining Management," NBER Working
Paper No. 4886, October 1994, and American Economic Review, (June 1995),
pp. 567-585; and O. D. Hart, "Theories of Optimal Capital
Structure: A Managerial Discretion Perspective," NBER Reprint No.
1806, September 1993, and in The Deal Decade: What Takeovers and
Leveraged Buyouts Mean for Corporate Governance, M. Blair, ed.
Washington, DC: The Brookings Institution, 1993, pp. 19-43. See also M.
Jensen, "Agency Costs of Free Cash Flow, Corporate Finance and
Takeovers," American Economic Review 76 (1986), pp. 323-329
10 S. J. Grossman and O. D. Hart, "One Share-One Vote and the
Market for Corporate Control, "Journal of Financial Economics 20
(1988), pp. 175-202. See also M. Harris and A. Raviv, "Corporate
Governance: Voting Rights and Majority Rules, "Journal of Financial
Economics 20 (1 988), pp. 203-235.
11 P. Aghion, O. D. Hart, and J. Moore, "The Economics of
Bankruptcy Reform," in The Transition in Eastern Europe, O. J.
Blanchard, K. A. Froot, and J. D. Sachs, eds. Chicago: University of
Chicago Press, 1994; "Improving Bankruptcy Procedure,
"Washington University Law Quarterly 72 (1994), pp. 849-872; and
"Insolvency Reform in the United Kingdom: A Revised Proposal,"
Insolvency Law and Practice 11/3 (1995), pp. 4-11.
12 The use of options in the debt-equity swap is based on an idea of
Lucian Bebchuk. See A. L. Bebchuk, "A New Approach to Corporate
Reorganizations," Harvard Law Review 101 (1988), pp. 775-804.
Oliver D. Hart is an NBER research associate in corporate finance
and, since 1993, a professor of economics at Harvard University. A
British citizen, Hart received his B.A. in mathematics from King's
College (Cambridge University), his M.A. in economics from Warwick
University, and his Ph.D. in economics from Princeton University.,
He taught at Cambridge University (in England) from 1975-81; was a
professor of economics at the London School of Economics from 1981-5;
and was a professor of economics at MIT from 1985-93. He also has been a
visiting professor at the Wharton School and Harvard Business School,
and a visiting scholar at Harvard Law School.
[INCOMPLETE TEXT FROM ORIGINAL PUBLICATION]
nomic theory, mathematical economics, theory of the firm, and law and
economics. He is a fellow of the Econometric Society and the American
Academy of Arts and Sciences, and his work has been published in a
number of prominent journals.
Hart is married to Rita Goldberg, who teaches in the literature
concentration and the extension school at Harvard, and who is a
contributing writer at the Boston Book Review. They have two sons,
Daniel (18), who will be a freshman in communications at the University
of Miami (Florida) next year, and Benjamin (12), who is at the Clarke
Middle School in Lexington, MA. Hart enjoys swimming, playing tennis,
and going to the theater.
