The 3 Contractual Approaches to the Firm


The contractual approaches to the firm aims to define the most efficient form of organization given the context, particularly the informational context.

This article presents three different contractual approaches: transaction cost theory, incentive theory (or agency theory), and incomplete contract theory.

These contractual approaches differ in their analysis of agent behavior and interactions (limited or perfect rationality) and in their assumptions about the information agents have (perfect or imperfect information).

However, the unity between these works stems from a common conception of economic relationships: they are contractual relationships between free individuals. From this perspective, the firm is analyzed as a “nexus of contracts” between individuals.

Contract theory develops with the desire to overcome certain limitations of the neoclassical approach to the firm, without fundamentally challenging it.

Its objective is to offer a more realistic representation of the firm. Incentive theory remains the closest to standard theory, particularly because it does not break with the assumption of perfect rationality.

Transaction Cost Theory

R. COASE and the Question of the Existence of the Firm

In his 1937 article “The Nature of the Firm,” R. Coase was the first economist to ask why firms exist and, more broadly, the question of the nature of the firm.

For Coase, the firm is a mode of coordinating transactions that is an alternative to the market. If the market is not the only way to coordinate economic activity, it is because there are costs associated with using the price system – transaction costs.

These transaction costs correspond to the costs of searching for information, negotiating contracts, repeated contracting, etc. Williamson would later propose a precise definition of these costs.

The fundamental difference between market and firm is that in a market, coordination is done through the price system, while the firm offers administrative coordination that goes through authority and hierarchy. For Coase, it is authority that fundamentally characterizes a firm.

Once the existence of two alternative modes of coordination has been established, Coase questions why the two modes of coordination coexist.

In fact, if market coordination generates transaction costs, internalized management of transactions incurs coordination costs, particularly because the returns on managerial activities are decreasing.

Thus, as long as internal coordination costs are lower than transaction costs, coordination takes place within the firm; from the moment they become higher, market coordination is preferable.

WILLIAMSON and the Definition of Transaction Costs

Williamson’s work is explicitly an extension of that of R. Coase. This work will allow us to clarify the concept of transaction costs and specify certain key assumptions for understanding, in particular, in which cases the firm prevails as a mode of coordination, i.e., under what conditions the integration of an activity within the firm will be preferred to resorting to the market.

Williamson makes two assumptions about agent behavior:

  • Bounded rationality: Agents have limited cognitive abilities. When the environment is complex, they cannot consider all possible events and perfectly calculate the consequences of their decisions.
  • Opportunism of agents: This is a consequence of bounded rationality. As the contract cannot foresee all possible alternatives, an agent may be tempted to adopt opportunistic behavior to favor their interests to the detriment of others.

Bounded rationality and opportunism increase transaction costs, particularly contract design and control costs.

Williamson also makes assumptions about the characteristics of transactions:

  • Asset specificity: An asset is said to be specific if it requires specific investments. These are durable investments, made to carry out a particular transaction, which cannot be redeployed to other uses without cost.
  • Uncertainty: Uncertainty about the conditions for carrying out the transaction risks increasing its cost (uncertainty linked to exogenous disturbances to the transaction, for example).
  • Transaction frequency: The more a transaction is repeated, the more opportunities contractors have to be opportunistic, which increases transaction costs.

Given these characteristics of behavior and transactions, the challenge for Williamson is to find the most appropriate organizational form, in the sense that it limits transaction costs.

Thus, for Williamson, the firm is a particular contractual system, an “institutional arrangement” characterized by a hierarchical principle that allows company management to make decisions in the event of unforeseen events by contracts, and which limits the risks associated with opportunism.

Incentive Theory

This contractual approach is part of the new microeconomics and more precisely within the framework of information economics.

It differs from transaction cost theory insofar as it preserves the neoclassical assumption of perfect agent rationality.

It places the holding of information and its sharing between contractors at the heart of its analysis of the firm. At the origin of this approach are the analyses of Berle and Means (1932) and that of Michael Jensen and William Meckling (1976).

Incentive theory is based on the notion of an agency relationship. An agency relationship has two main characteristics: it is a delegation relationship and it assumes information asymmetry.

An agency relationship arises when one person hires another to carry out a mission that requires a delegation of power. The one who delegates is called the Principal, the one to whom the mission is entrusted is called the Agent.

The relationship between a shareholder and a manager is an agency relationship. The shareholder (the Principal) delegates decision-making to the manager (the Agent).

Delegation is associated with an imperfection of information which can be of two orders:

The Principal has only limited information on the characteristics of the Agent (a situation called adverse selection) and imperfectly observes their behavior (a situation called moral hazard). Information asymmetry therefore characterizes any relationship between the Principal and the Agent.

An agency relationship generates three types of costs, called agency costs:

Monitoring and incentive expenses (e.g., incentive systems) incurred by the Principal to guide the Agent’s behavior.

“Bonding” costs, borne by the Agent, i.e., the expenses they may have to incur to be able to guarantee that they will not implement certain actions that could harm the Principal, or to be able to compensate them if necessary.

The “residual loss”, which corresponds to the inevitable difference between the result of the Agent’s action for the Principal and what a behavior of effective maximization of the Principal’s well-being would have given.

In view of the hypotheses presented, incentive theory presents organizations as “nodes of contracts”, written and unwritten, between holders of factors of production and customers (for example the relationship between an employee and their employer).

Each contractual relationship is an agency relationship for which the optimal configuration must be found, i.e., the contractual rules that minimize agency costs.

This vision proposed by agency theory has several important implications regarding the representation of the firm (this representation is useful for the strategist’s work, especially since they are required to make long-term decisions).

The firm has no real existence (it is a “legal fiction”). It is not assimilated to an individual (the owner, the entrepreneur), as in the neoclassical approach. What is important to understand the firm are the characteristics of the different contractual relationships linking individuals.

Each factor in a firm is the property of an individual. The firm is a set of contracts relating to the way in which inputs are combined to create products, and to the way in which income is shared between the owners of these inputs.

There is no fundamental difference between firm and market, in particular contractual relations within the firm do not assume any relationship of authority. In this sense, the employment relationship is in no way specific, the employment contract is comparable to the commercial contract. The employment contract connects the owner of an input (the employee) and a applicant for this work (the employer).

  • Example of agency relationship:

The public limited company: This is characterized by an agency relationship between shareholders and managers (effectively characterized by a delegation of decision and information asymmetry in favor of the manager), likely to generate conflicts of interest.

This conflict arises from the fact that shareholders seek to maximize dividends and the manager seeks to maximize their salary, these two objectives being irreconcilable. But these divergences of interest can fade if the company sets up a suitable manager remuneration system, using a stock option system.

In this case, the interests of the shareholders and the managers converge further insofar as the latter now hold shares in their company.

Similarly, the analytical tools offered by agency theory are particularly well suited to analyzing the relationship between employer and employees within capitalist firms.

This relationship is an agency relationship, and certain procedures can limit the problems of controlling employee behavior.

For example, the employer can incentivize employees to work harder by linking their pay to their performance.

The initiative can also come from the employee who can provide information in order to make known the reality of their skills and their commitment.

For example, the employee has obtained a diploma which is not directly related to their job, but they thus demonstrate their ability to make an effort.

Naturally, all these procedures have a cost which can only be borne by a person rewarded by the allocation of part of the company’s profits.

Incomplete Contract Theory

This approach postulates the incompleteness of contracts. A contract is incomplete when it is not possible to foresee and therefore write what should happen in all possible cases.

Contractors cannot list all these cases, or even imagine them all. When an unforeseen circumstance occurs, there is room for a new negotiation to interpret or redefine the terms of the contract.

It is this renegotiation which is the central concept of incomplete contract models. Note that this assumption of incompleteness of contracts is also that made by Williamson, as soon as he postulates the limited rationality of agents.

What nevertheless distinguishes incomplete contract theory from transaction cost theory is the solutions offered to this incompleteness.

For Williamson, it is authority that gives its holder discretionary power, that is to say the power to make decisions in all situations not provided for by contract.

For incomplete contract theory, it is the allocation of property rights that gives the owner the right to dispose of the resource in case of uncertainty.

According to the theory of incomplete contracts (Hart and Moore, 1990), agents are unable to sign complete contracts due to imperfect information (information is symmetrical but agents lack information).

No one is actually able to verify ex post the actual state of certain variables characteristic of the relationships between contractors (in particular on investment in physical capital). It is the possession of the assets that will make it possible to exercise ex post control over them.

The question of the acquisition of assets refers to the problem of vertical integration with the associated questions: where to stop the expansion of the firm? What is its efficient size? The answer is to compare the costs and benefits of integration.

Incomplete contract theory is therefore not interested in contracts that bind the different members of a company, but in contracts between customers and suppliers. The firm is thus defined as the collection of non-human assets held by the individuals constituting the firm (equipment, capital, etc.).


In conclusion to the presentation of these three contractual approaches, we will retain that while transaction cost theory developed a lot in the 1980s and 90s, it now suffers from a lack of formalization which would allow it to explain certain hypotheses more clearly.

Likewise, the empirical tests carried out are sometimes disappointing, for example, the concept of specific assets is attractive, but econometric work has difficulty in defining a convincing measure of this specificity (cf. point 3. of this synthesis).

Incomplete contract theory is made up of a set of fairly heterogeneous models that can hardly be considered as a definitively constituted doctrine.

In addition, this theory has given rise to numerous formalizations, often complex, hence the difficulty in translating these theoretical models into testable hypotheses.

Incentive theory is undoubtedly the most developed contractual approach to the firm today, it has been enriched in recent years from both a theoretical and empirical point of view. This approach to the firm is part of the new industrial economics, initiated by J. Tirole in his 1988 book.

The fields of application of incentive theory are now numerous: insurance contracts, franchise contracts, employment contracts, etc.

The predominance of incentive theory over other contractual approaches is also explained by its degree of mathematical formalization, in a context where this has greatly increased in economic research for thirty years.

The definition of the optimal contract, in the sense of an incentive contract, in the presence of adverse selection or moral hazard, requires the development of constrained optimization models.

The Principal enters into a contractual relationship as soon as the contract allows them to maximize their profit, subject to the constraint that the Agent agrees to participate in the contract (participation constraint) and that they reveal their characteristics (revelation constraint in an adverse selection context) or be encouraged to make sufficient effort (incentive constraint in an adverse selection context).



Please enter your comment!
Please enter your name here