Oliver Hart y Bengt Holmström
Estaba en último año de la carrera, me acuerdo como si fuera ayer, cuando salió publicadoFirms, Contracts, and Financial Structure, de Oliver Hart. Impulsado por una reseña muy favorable del mismo, lo compré casi de inmediato. El libro no era muy largo, un poco más de 200 páginas, pero su lectura me causó un efecto enorme. De manera clara, Hart empleaba lateoría de los contratos incompletos para explicar una cantidad tremenda de observaciones sobre la estructura interna de las empresas, cómo estas se financiaban o el procedimiento de bancarrota. El momento, además, no podía ser más propicio. En una clase detrás de otra del último año de carrera hablábamos de la empresa y de su gestión pero no eran más que ideas deshilvanadas, carentes de estructura teórica y sin soporte empírico. Hart, en cambio, me enseñaba que uno puede pensar en cómo funcionan las empresas de manera sistemática y abría, aunque él no lo hiciera de manera extensa en el libro, un programa de investigación empírica.
Luego, al llegar al doctorado en Minnesota, Tom Holmes empleó el libro casi completo para la secuencia de organización industrial de segundo año, una secuencia en la que aprendí muchísimo y donde releyendo el libro de Hart me di cuenta de mil cosas que solo dos años antes se me habían escapado. Realmente el primer año de doctorado en una universidad como Minnesota le cambia a uno la cabeza de manera profunda.
No he trabajado académicamente ni en organización industrial ni en teoría de contratos incompletos, no por falta de ganas, mas por falta de tiempo (aunque ahora justo estoy intentando acabar un papel con Pat Bajari sobre el tema). En los próximos días, otros colaboradores de este blog que conocen la literatura mejor que yo escribirán entradas más elaboradas que la mía (incluyendo una descripción del trabajo de Bengt R. Holmström, que también ha tenido aportaciones fundamentales), pero quería aprovechar esta entrada para resaltar varias ideas.
Primero, con la anécdota de mi lectura del libro de Hart señalar como toda una generación de economistas ha crecido bebiendo de Hart y Holmström. Como me decía esta mañana un amigo ¿pero no tenían ya el nobel?
Segundo, que el prestigio del premio se fundamenta en la cuidadosa selección de premiados. Cuando uno mira a los nombres de los últimos años, desde Hansen a Tirole, pasando por Roth o Sims (y tantos otros que me dejo), es casi imposible pensar en que es factible hacerlo mejor. Excepto Myrdal en 1974 (al que le concedieron el premio porque les daba apuro dárselo a Hayek en solitario, pero así, entre nosotros, no se lo merecía), el comité una y otra vez ha demostrado un gusto excepcional. Aquí hay un episodio de Freakonomics que explica muy bien cómo se concede el premio: si algo queda demostrado en el mismo es la profesionalidad del comité.
Como decía antes, tiempo habrá de escribir con más calma, pero mientras tanto, una lista básica de enlaces:
La explicación de la academia sueca, como todos los años muy bien escrita. Si yo tuviese que dar clase sobre este tema mañana, esta sería mi “lectura” para los estudiantes.
Sobre el impacto de los contratos incompletos en economía: aquí.
Scientific Background on the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2016
OLIVER HART AND BENGT HOLMSTRÖM: CONTRACT THEORY
The Committee for the Prize in Economic Sciences in Memory of Alfred Nobel
An eternal obstacle to human cooperation is that people have di§erent interests.
In modern societies, conáicts of interests are often mitigated ñif not completely
resolved ñby contractual arrangements. Well-designed contracts provide incentives
for the contracting parties to exploit the prospective gains from cooperation.
For example, labor contracts include pay and promotion conditions that
are designed to retain and motivate employees; insurance contracts combine the
sharing of risk with deductibles and co-payments to encourage clients to exercise
caution; credit contracts specify payments and decision rights aimed at
protecting the lender, while encouraging sound decisions by borrowers.
The idea that incentives must be aligned to exploit the gains from cooperation
has a long history within economics. In the 1700ís, Adam Smith argued
that sharecropping contracts do not give tenants su¢ cient incentives to improve
the land. In the 1930ís, Chester Barnard considered how employees could be
incentivized to contribute e§ort within large organizations.1 This yearís laureates
have approached these old ideas using theoretical models that have given us
new insights into the nature of optimal contracts. The models have also allowed
researchers to sharpen existing arguments and pursue them to their logical conclusions.
As a result, contract theory has made major strides during the last few
decades. Today, incentive problems are almost universally seen through its lens.
The theory has had a major impact on organizational economics and corporate
Önance, and it has deeply ináuenced other Öelds such as industrial organization,
labor economics, public economics, political science, and law.
A classic contracting problem has the following structure. A principal engages
an agent to take certain actions on the principalís behalf. However, the
principal cannot directly observe the agentís actions, which creates a problem
of moral hazard: the agent may take actions that increase his own payo§ but
reduce the overall surplus of the relationship. To be speciÖc, suppose the principal
is the main shareholder of a company and the agent is the companyís
manager. As Adam Smith noted, the separation of ownership and control in a
company might cause the manager to make decisions contrary to the interests
of shareholders.
To alleviate this moral-hazard problem, the principal may o§er a compensation
package which ties the managerís income to some (observable and veriÖable)
performance measure. We refer to this as paying for performance. The companyís
proÖt or stock-market value are frequently used performance measures,
but they have well-known drawbacks. They may depend largely on factors beyond
the managerís control, so that the manager would be rewarded for luck.
Intuitively, it would be desirable to Ölter out as much of the luck component as
possible, perhaps by measuring the Örmís performance relative to other Örms in
the same industry. But any performance measure is likely to be imprecise and
noisy, so in the end the optimal compensation schedule must trade o§ incentiveprovision
against risk-sharing.
To go beyond these vague intuitions requires a formal analysis. Some formal
results were obtained in 1975 by James Mirrlees, the 1996 Economics Laureate.
In 1979, Bengt Holmstrˆm provided a formalization which would prove to have
a lasting impact. In addition to characterizing the optimal trade-o§ between
incentives and risk-sharing, Holmstrˆmís article contained a fundamental result
on optimal performance measures, namely the informativeness principle. A second
generation of moral hazard models, developed in the 1980s by Holmstrˆm,
sometimes by himself and sometimes with coauthors ñin particular Paul Milgrom
ñintroduced dynamic moral hazard, multi-tasking and other key issues.
Personnel economics, and organizational economics more broadly, have been
strongly ináuenced by this line of work. We discuss the pay-for-performance
approach in more detail in Section
Paying for performance requires both the ability to write su¢ ciently detailed
contracts ex ante, as well as the ability to measure and verify performance ex
post. These requirements are sometimes hard to satisfy. Suppose, for example,
that the agent is a researcher whose delegated task is to develop a new technology
for the principalís company. Due to the uncertainties inherent in the R&D
process, it may be impossible to specify ex ante exactly what the innovation
should be. Moreover, neither the quality of the new technology nor its impact
on the principalís proÖt may be veriÖable ex post. Since performance-based contracts
may not be of much use in this kind of situation, an alternative approach
is needed. The incomplete contracting approach, pioneered by Oliver Hart and
his collaborators Sanford Grossman and John Moore, emphasizes the allocation
of decision rights.
Decision rights are often determined by property rights ñi.e., by ownership.
In the R&D example, one possibility is that the agent is employed with a Öxed
salary by the principal. In this case, the agent has no ex post bargaining power:
the principal owns any innovation and can use it freely. Another possibility is
that the researcher independently owns any innovation that he develops....
Aplications
Privatization and Public versus Private Ownership The property-rights
framework helps us understand the costs and beneÖts of privatization and public
versus private ownership (Schmidt, 1996, and Hart, 2003). In a completecontracting
model, public or private ownership does not matter, since the relationship
between the government and a Örm can be fully resolved in a detailed
incentive contract. By contrast, when contracts are incomplete, the identity of
the owner becomes highly relevant, because the owner retains the residual rights
of control. In the context of our illustrative model, let P be the government who
desires the production of some public service, and let A be a manager in charge
of producing this service. As usual, the ownership of assets will a§ect bargaining
power and incentives. In the context of privatization, the government cares
about production e¢ ciency as well as quality of service. But the quality of
welfare services, say, may be hard to specify in a contract, while the private
contractor has an incentive to produce at lowest cost.34
In an ináuential article, Hart, Shleifer, and Vishny (1997) study privatization
by combining the incomplete-contracts approach with the multi-tasking model
of Holmstrˆm and Milgrom (1991). Let us illustrate a special case of their
approach in our simple model. Suppose an innovation would reduce both costs
and beneÖts. The asset owner has the right to implement the innovation or to
prevent its implementation. As before, actions are not contractible......
Corporate Finance The most important application of incomplete-contracts
theory to date is in the Öeld of corporate Önance. Traditionally, Önancial contracts
were seen as devices for furnishing investment capital and sharing risk.
With agency theory, the emphasis changed. Economists came to see what many
practitioners and legal scholars had seen all along, namely that a major purpose
of Önancial contracting is to ensure that entrepreneurs and managers act in the
interests of investors. Perhaps this is the greatest moral-hazard problem of all,
at least in the business sector......
Finance and Governance of Entrepreneurial Firms In a paper originally
circulated in 1989, Oliver Hart and John Moore address several of these issues.37
The analysis turns on the concept of foreclosure rights and the role that they
play in forcing entrepreneurs to repay their investors. More speciÖcally, Hart
and Moore consider an entrepreneur who may divert project returns, but cannot
divert the Örmís productive assets. Under some additional assumptions, the
optimal (incomplete) contract has the following features
Evidence from Venture Capital Financing Kaplan and Strˆmberg (2003,
2004) report evidence from a large number of venture capital (VC) contracts.
This is a real-world setting that quite closely corresponds to the environment
of the Önancial contracting theories, where entrepreneurs who raise money for
start-ups are the agents, and VC investors who invest in start-up Örms are the
principals. They Önd that VC Önancial contracts separately allocate cash-áow
rights as well as di§erent control rights ñincluding board rights, voting rights,
and liquidation rights ñbetween investors and entrepreneurs. The allocation of
these rights is frequently state-dependent, being contingent on Önancial as well
as non-Önancial measures of performance. Both cash-áow and control rights are
allocated such that investors obtain full control of poorly performing Örms. But
entrepreneurs retain more control rights in better performing Örms, and their
cash-áow rights increase with Örm performance, often in a non-linear fashion.
Moreover, the allocation of rights responds to the perceived risk investors
see in the venture, as well as the likelihood that investors will have to intervene
in the future, for example by replacing management. Kaplan and Strˆmberg
argue that these contracts can best be understood through a combination of
incomplete-contracts theory and classic contract theory. Importantly, the separate
allocation of cash-áow and control rights cannot be fully understood outside
an incomplete-contracting framework.
Finance and Governance of Widely Held Firms Hart and Moore (1995)
shift the focus from entrepreneurs to publicly held companies. In this work,
managers (or company boards) take decisions that a§ect many dispersed claim
holders. The managers, whose stake is usually small, may want to expand
investment beyond the level that maximizes returns....
Bankruptcy Reform and Investor Rights Through its emphasis on which
stakeholders have the right to do what under what circumstances, the framework
of incomplete contracts allow us to sensibly talk about more or less e¢ cient procedures
for resolving conáicts of interest between entrepreneurs and investors,
and thus about bankruptcy legislation. Building on his foundational contributions,
Hart has studied empirically the ine¢ ciencies of debt enforcement around
the world (Djankov et al. 2008). He has also o§ered a number of suggestions
for bankruptcy reform, as well as reform of Önancial institutions more broadly
(e.g., Aghion et al., 1992; Hart et al., 1997; Hart and Zingales, 2011). More
generally, the incomplete-contracting framework is now central to the analysis
of corporate governance (see e.g. Shleifer and Vishny, 1997), and there is now
considerable empirical work showing the importance of property rights and investor
protection for Önancial market development (see, e.g., La Porta et al.,
1997, 1998).42
Other Extensions and Applications The basic idea of assigning property
rights in order to provide incentives has been elaborated in several other directions,
of which we will mention a few. Bolton and Whinston (1993) analyze
vertical integration in the presence of multiple potential trading partners.43
Grossman and Helpman (2002) develop a general-equilibrium model in which
there is a non-trivial interaction between the integration decisions of di§erent
Örms in the same industry. Antras (2003) and Antras and Helpman (2004)
consider international outsourcing, a setting in which it is especially di¢ cult
to enforce complete contracts, owing to diverging national rules and possible
discrimination of foreign claim holders.
Besides the issue of asset ownership, the incomplete-contracts approach has
been used to study the allocation of decision rights within the Örm. This work
concerns the optimal delegation of authority (Aghion and Tirole, 1997), as well
as the separation of ownership and control (Burkart et al., 1997). In the highly
cited Aghion and Tirole (1997) model, delegating tasks increases the initiative
of the agent at the cost of less principal control. One implication of this model
is that delegation increases the agentís incentive to collect and use soft decisionrelevant
information, which cannot be credibly transmitted up in the hierarchy
in a less delegated structure...
Other Contributions
Before his focus shifted to contract theory, Oliver Hart contributed to a number
of di§erent areas, producing one or two highly ináuential articles on a topic
before moving on to di§erent territory. Early in his career, he worked on general
equilibrium with incomplete markets. Hart (1974) considered the existence
of incomplete-markets equilibrium and showed that in general, existence is not
guaranteed. This inspired research on the conditions under which, generically
at least, an equilibrium would exist. Hart (1975) considered the optimality
properties of incomplete-markets equilibrium, and proved another result of lasting
importance: making the market more complete, but not fully complete, by
opening new markets, does not necessarily lead to Pareto improvement. Hartís
work on monopolistic competition (in particular, Hart 1979 and 1982) inspired
a signiÖcant literature on the role of imperfect competition in product markets
as an explanation for Keynesian unemployment. Among Hartís early work on
contract theory, we may highlight two contributions. The Örst is Grossman and
Hartís (1983b) work on adverse selection in labor markets, which shows how
underemployment can result from an optimal contract between a Örm and its
workers. The second, and highly ináuential contribution, is Hart (1983)
In this article, Hart asked if competition in the product market can alleviate managerial
incentive problems: that is, does more competition make it harder for
managers to deviate from proÖt maximization? He found an a¢ rmative answer
under the assumption that Örms are exposed to common shocks. However, in
general this turns out to be a very di¢ cult question to answer, and this is still
an active research area more than 30 years after Hartís pioneering article.
Bengt Holmstrˆm is best known for his work on contract theory and the
theory of incentives, much of which we have already discussed. In recent years,
he has studied issues relating to Önancial market liquidity and its relevance to
Önancial regulation. Holmstrˆm and Tirole (1997, 1998, 2000, 2001, 2011) use
a model of managerial wealth constraints to investigate a number of important
issues in corporate Önance, including the impact of wealth shocks on the banking
system, banking regulation, the role of public liquidity provision for Örms, and
how liquidity a§ects asset prices and expected returns. Recently, Holmstrom has
also explored the trade-o§ between Önancial market transparency and market
liquidity (e.g. Dang et al., 2015).
Final Remarks
Contract theory identiÖes a variety of obstacles to cooperation and suggests
which contracts are best suited to overcome them. The theory is both positive
and normative: it o§ers coherent explanations for the contracts that are
commonly written, as well as a method for Önding appropriate contractual solutions
to new problems. It generates precise hypotheses that can be confronted
with empirical data and lays an intellectual foundation for the design of various
policies and institutions, from bankruptcy legislation to political constitutions.
Scientific Background on the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2016
OLIVER HART AND BENGT HOLMSTRÖM: CONTRACT THEORY
https://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2016/advanced-economicsciences2016.pdf
http://marginalrevolution.com/marginalrevolution/2016/10/bengt-holmstrom-nobel-laureate.html
http://marginalrevolution.com/marginalrevolution/2016/10/performance-pay-nobel.html
https://www.amazon.com/Impact-Incomplete-Contracts-Economics-ebook/dp/B0190CYHYI/ref=sr_1_1?ie=UTF8&qid=1476112782&sr=8-1&keywords=incomplete+contracts
http://nadaesgratis.es/admin/oliver-hart-y-bengt-holmstrom
Un año más el Premio Sveriges RiskBank en memoria de Alfred Nobel ha recaído en dos figuras clave de la revolución de la Teoría Microeconómica que tuvo lugar entre el final de los años 70 y el principio de los años 90.Bengt Holmstrom y Oliver Hart, juntamente con sus colaboradores Milgrom, Grossman y Moore y con los anteriormente laureados Myerson, Maskin, Roth y Tirole y otros posibles futuros ganadores como Kreps, Wilson y Rubinstein, desarrollaron los métodos y las ideas que han dado forma a un corpus de conocimiento que los economistas ahora utilizan en los más variados campos, desde la regulación y la economía de la competencia al desarrollo económico y las finanzas corporativas.
Holmstrom y Hart han realizado contribuciones esenciales al estudio del diseño de contratos en situaciones en las que una de las partes tiene la capacidad o el deber de actuar en beneficio de la otra (contratos de agencia). Algunos ejemplos clásicos son el trabajador que produce para su empleador, el directivo que representa los intereses de sus accionistas o el político que toma decisiones para su electorado. El trabajo de Holmstrom se centra en aquellas situaciones en las que las partes tienen la capacidad de firmar contratos completos, describiendo los pagos a realizar en cada contingencia futura mientras que Hart se especializó en situaciones en las que los contratos son incompletos y una de las partes posee el derecho a decidir en caso de que hubiera una contigencia imprevista.
En su artículo ‘Moral Hazard and Observability’ (1979), Holmstrom analiza la relación entre un trabajador, que es averso al riesgo, y su empleador, cuando este último no puede observar directamente el esfuerzo del primero pero sí tiene acceso a varias medidas de su productividad. Por ejemplo, el propietario de un concesionario de coches no puede observar el esfuerzo que ponen sus vendedores pero sí sabe cuántos coches venden, a qué precio, etc. El problema fundamental es el conflicto entre el riesgo que sufre el trabajador si su compensación es variable y sus incentivos a esforzarse si su compensación es fija (como dice Jim en Huckleberry Finn: “What’s the use you learning to do right when it’s troublesome to do right and ain’t no trouble to do wrong, and the wages is just the same”)
Holmstrom muestra que los vendedores deben ser retribuídos de acuerdo a sus resultados (performance pay) pero el contrato solo debe incluir aquellas variables que el trabajador pueda influir directamente y cada variable debe pesar de acuerdo a su nivel de correlación con el esfuerzo (informativeness principle). Como ninguna de las medidas es perfecta, el trabajador estará sometido a un riesgo residual que reduce su bienestar, y requiere que su pago esperado sea mayor, lo que reduce los beneficios marginales del esfuerzo para el empleador y aconseja que el trabajador se esfuerze menos que en el óptimo social.
Hay quien asegura que tales respuestas son ‘obvias’ (véanse los comentarios a la noticia del galardón en diversos medios). Y en cierto modo lo son. Pero lo que desde luego no es obvio es poder formular este tipo de preguntas de manera general de modo que estemos seguros que no existe ninguna otra fórmula para proveer incentivos que resulte en una situación mejor. Piense el lector las múltiples variedades de contrato existentes y las múltiples características que los definen. El verdadero avance es el método que Holmstrom ofrece para contestar estas preguntas y desarrollar predicciones contrastables empíricamente. Desde entonces, son numerosísimos los estudios que han mostrado la relevancia de este modelo para entender el diseño de incentivos en el mundo real (para muestra, Marcos Vera Hernández nos relataba en esta entrada su trabajo con médicos en India).
En su trabajo posterior Holmstrom desarrolló este método en múltiples direcciones, incluyendo el diseño óptimo de incentivos para equipos de trabajadores, para trabajadores que pueden dedicar su tiempo a múltiples tareas (junto con Milgrom) o el diseño óptimo de la relación entre un principal que delega su decisión a un agente con incentivos sesgados (mi favorito).
El punto de partida de Oliver Hart (en colaboración con Sanford Grossman (1984) y Moore (1990)) es que en entornos llenos de incertidumbre, los contratos son necesariamente incompletos, pues las partes no pueden describir todas las contigencias futuras que pueden afectar a su relación contractual. Piense el lector en la relación entre una fábrica siderúrgica y una mina de carbón. La incertidumbre sobre los precios futuros del carbón y el acero, los avances tecnológicos, la política energética y medioambiental afectan los retornos de las inversiones pero son dificilmente predecibles a priori. En muchas ocasiones, esta limitación será irrelevante, pues las partes pueden renegociar el contrato cuando esa incertidumbre se disipe.
La idea fundamental de Grossman y Hart es que la posibilidad de renegociación reduce los incentivos de las partes para realizar inversiones en la relación contractual, pues los retornos brutos de esta inversión serán renegociados. Por ejemplo, si la mina de carbón construye una vía férrea que una la mina con la fábrica, y el precio se renegocia, una parte de la ganancia puede pasar a manos de la fábrica. ¿Qué parte retrendrá la mina? Pues dependerá, en gran medida, de quien posee la propiedad de la vía y que usos alternativos se le puedan dar. Es decir, los derechos de propiedad determinan los resultados siempre que los contratos sean incompletos y las partes deban invertir en la relación. Dependiendo de qué parte deba realizar más inversiones, la propiedad debe pertenecer a una de las partes o si ambas deben contribuir a partes iguales, existen grandes beneficios de integrarse en una sóla empresa. Nace así una teoría de integración y desintegración de la empresa basada en la existencia de activos específicos (la vía férrea) y los contratos incompletos.
El trabajo de Hart, Grossman y Moore ha constituido una verdadera revolución en el estudio de la naturaleza de las empresas. Cientos de artículos han contrastado sus predicciones empíricamente y el resultado es, en general, positivo. El artículo ha sido enormemente influyente en campos tan dispares como las finanzas corporativas y el comercio internacional (aquí destaca sobre manera el trabajo del antiguo colaborador de este blog Pol Antrás sobre el fenómeno conocido como offshoring).
Permítanme terminar con una breve anécdota que refleja el valor de la formalización de los argumentos económicos y el uso de las matemáticas en las ciencias sociales. En 1980Eugene Fama (Premio Nobel en 2013) publicó un provocativo artículo en el que argüía que las empresas no necesitaban utilizar incentivos explícitos para motivar a sus trabajadores pues estos sabían que su futuro profesional dependía en gran medida de su desempeño actual. Según Fama, los incentivos de carrera son un sustituto perfecto del mecanismo de precios y los incentivos explícitos solo son necesarios en entornos de corto plazo o en trabajos donde el talento individual cuente poco. Holmstrom recogió el guante y dos años después publicó un artículo en el que ofrecía un modelo formal de un mercado de trabajo dinámico en el que las empresas aprenden sobre la productividad intrínseca de los trabajadores mediante la observación de su rendimiento pasado (Holmstrom 1982,reimpreso en 1999). La conclusión es que si bien los incentivos de carrera inducen a los trabajadores a esforzarse más, las asignación de recursos es ineficiente. Los incentivos implícitos, por tanto, no son un susittuto perfecto de los contratos explícitos como tampoco la intuición es un sustituto perfecto de la modelización.
http://nadaesgratis.es/admin/un-nobel-para-la-teoria-de-contratos-hart-y-holmstrom
No hay comentarios:
Publicar un comentario