RESEARCH DOSSIER
Incomplete Contracts and Residual Control
Abstract. Incomplete contracts constrain the enforceable specification of future conduct, while residual rights of control allocate discretion over asset uses and decisions left open. This Research Dossier investigates how that allocation changes disagreement opportunities, bargaining outcomes, investment incentives, organizational adaptation and the quality of performance. It reconstructs the development of transaction-cost and property-rights economics, derives the investment incentive problem under explicit assumptions, examines implementation and renegotiation critiques, and integrates formal authority, relational enforcement and behavioral reference-point research. Applications encompass venture capital, US trucking, UK manufacturing, cross-border production, trade and public provision, with international commercial-contract and public-private partnership institutions examined on their own normative terms. The study distinguishes theoretical implications, reported empirical associations and experimental findings from original illustrative arithmetic. Its analytical conclusion is that control must be evaluated through the marginal consequences of feasible alternative decisions before and after investment, including the effects on people outside the contracting coalition. Governance improves efficiency when its benefits in investment, coordination and adaptation exceed the costs of displaced initiative, strategic deviation, rigidity and uncontractible quality. The work is an original literature-based synthesis, with access limitations and the absence of new empirical replication explicitly recorded.
Keywords: Incomplete contracts; residual rights of control; property rights; hold-up; transaction costs; firm boundaries; investment incentives; renegotiation; delegation; relational contracts; financial governance; international production; contract enforcement; public provision
The Economic Object of Incomplete Contracting
An economic contract is an institution for coordinating future actions and distributing claims among parties whose decisions are interdependent. The parties possess a technology of production, information about circumstances, resources that can be deployed in alternative ways, and a legally or socially supported ability to commit themselves. The agreement changes the feasible set of actions and transfers after the parties have entered their relationship. Its economic significance therefore extends across time: what happens when an obligation becomes due depends on what the parties previously promised, who controls the requisite assets, and what alternatives each party retains.
Contractual incompleteness is a limitation on the feasible specification and enforcement of obligations. In the canonical setting, economically relevant actions or states cannot all be made the basis of enforceable contingent promises at acceptable cost. A residual right of control is an entitlement to determine the use of an asset in contingencies left to the holder's discretion by specific contracts, law, and other binding arrangements. Grossman and Hart's distinction between specific and residual rights (1986) gives this otherwise broad idea a rigorous organization: the transfer of discretionary rights changes the incentives of both the acquiring and the relinquishing party.
The subject of this dossier is the relationship among incompleteness, adaptation, bargaining and investment. The study adopts the institutional scope of Economic Theory and the individual-choice and strategic mechanisms of Microeconomics. It treats financing, procurement, employment, ownership and international sourcing as analytically distinct applications. A general history of all institutions, a comprehensive exposition of principal-agent theory, and a survey of every legal remedy would obscure the mechanism that unifies these applications.
Four questions organize the investigation. Which economically relevant variables enter enforceable contracts? Who can act when those contracts leave a decision open? How does the distribution of such power change the return to earlier investment? What kinds of observation could establish that ownership and authority, rather than another institutional characteristic, produced the measured outcome? The answers require a temporal explanation of governance: commitments made before decisions, decisions taken before all relevant information arrives, subsequent adaptation, and consequences borne by different people.
The methodology is a critical synthesis of published theoretical and empirical research, complemented by transparent numerical examples and comparison of institutional mechanisms. The examples are pedagogical derivations under stated assumptions. The published findings are reported at the level supported by accessible records and abstracts; the work does not supply a new econometric estimate or claim that a particular legal rule applies worldwide. This distinction protects the identity of the research object and makes the argument reproducible within its evidentiary limits.
Genealogy of the Theory of the Firm
The economic explanation of the firm begins with the existence of multiple ways to coordinate production. Market exchange can induce separate owners to supply inputs at stated prices. Administrative arrangements can coordinate related activities through comparatively durable relationships. The two modes can accomplish the same physical production while requiring different processes to discover information, settle terms and respond to unforeseen demands. Coase's The Nature of the Firm (1937) established the foundational question of costly market coordination. A firm can be interpreted as a governance response to the practical costs of organizing exchange.
Williamson's governance of contractual relations (1979) places alternative institutional arrangements at the center of transaction-cost economics. An exchange may require investment whose value in the relationship exceeds its value in other uses. Where the terms of repeated adaptation are hard to agree or enforce, the governance of those changes becomes economically significant. A contract is situated within an arrangement for monitoring, adjustment, conflict and continuation. The explanatory task is to compare feasible arrangements rather than measure an actual arrangement against an idealized market.
Klein, Crawford, and Alchian's (1978) appropriable-rent analysis addresses integration in relation to the surplus exposed when productive assets become specific to an exchange. That surplus produces a potential dispute over the division of returns. Integration, long-term commitments and other safeguards represent different possible responses. An asset may be specialized because its shape fits a buyer's process, its location reduces transport cost, or accumulated knowledge is valuable principally with a particular organization. The fact that potential gains exist from coordination does not establish that a particular governance structure realizes them at the lowest cost.
The property-rights contribution narrows the mechanism further. The crucial question becomes how control of productive assets influences the payoff obtainable if the parties fail to agree on subsequent cooperation. That effect alters the expected return from a noncontractible investment undertaken earlier. In the original ownership analysis, integration is costly precisely because the acquisition of discretion by one party alters what another party can do. Integration must therefore be compared with nonintegration as a redistribution of incentives, rather than an automatic cure for conflict.
This genealogy supports a methodological separation. A study of adaptation costs may find that an administrative process settles recurring decisions with less delay. A property-rights study may find that the initial choice of owner changes what people invest before bargaining. A theory of relational contracts may find that a governance structure makes future cooperation more credible. These mechanisms can reinforce or offset one another in the same observed organization. To explain a particular case, the investigator must identify which constraint is actually binding, what time horizon matters and what alternative governance arrangement was available.
Contractibility, Information, and the Complete-Contract Benchmark
The complete-contract benchmark is most useful when specified as an analytical environment. Consider a finite set of states of the world, a set of feasible actions, and a schedule of transfers. If actions and states can be described and verified as required, and the parties can commit to the schedule, they can agree today on a desired allocation for each state tomorrow. Under suitable preferences and transfer possibilities, the benchmark implements a joint-surplus-maximizing allocation and then distributes its gains. Risk aversion, private information and limited wealth already complicate the benchmark even when the physical state is well defined.
Observability means that a participant can recognize a variable, such as the condition of a machine or the effort devoted to diagnosing its malfunction. Verifiability means that the variable can be established adequately for the chosen enforcement procedure. Contractibility additionally requires that a sufficiently determinate rule can be attached to the variable and enforced at tolerable cost. Two contractors may recognize careful work while anticipating that a court lacks the evidence or expertise to establish that care. Conversely, a publicly measured quantity may be formally verifiable while remaining a poor indicator of the quality the parties actually value.
Uncertainty and incompleteness are different dimensions. A large set of random outcomes does not itself preclude a contingent contract; insurance commonly operates with uncertain but defined events. An action in a known and predictable state can be noncontractible if it is difficult to assess. Unknown future technologies may impede specification, yet a contract can also grant a party authority to choose among alternatives as they emerge. These distinctions explain why the degree of uncertainty alone cannot establish the efficient ownership arrangement.
Moral hazard is another mechanism: privately chosen effort changes the probability distribution of a contractible outcome. The principal can choose incentives based on that outcome, possibly at a cost in risk bearing or distorted effort. In the property-rights setting, the gap concerns the feasible action or investment contract and the use of assets after disagreement. An arrangement can contain both features. For example, a technical supplier may receive a measurable delivery bonus and retain discretion over its specialized production process. The delivery bonus addresses an observable outcome; process control affects negotiation over modifications that cannot be specified.
A useful representation treats the set of feasible contracts as an institutional primitive rather than the same as the set of technologically feasible allocations. Improvements in metering and documentation expand the former set without necessarily changing the latter. By contrast, greater productive flexibility can change both sets. The separation matters for empirical inference: a change in organization after adoption of monitoring technology may arise because the technology makes promises enforceable, because it raises productivity, or because it changes the distribution of tasks. All three possibilities require distinct evidence.
Assets, Ownership, and Effective Rights
Ownership is economically important because it governs asset uses not determined by prior binding obligations. Its content is defined relative to the surrounding regime. A factory owner may be able to select suppliers and reconfigure production lines yet lack the authority to ignore a lease, a labor agreement or environmental regulation. Residual has a relational meaning: it denotes discretion remaining after applicable constraints. It does not describe an unlimited sovereignty over objects or people.
A physical asset confers productive possibilities. Machinery can be moved, used in another production process, leased or denied to a trading partner, subject to law and technical requirements. Ownership consequently affects access to these possibilities when cooperation breaks down. In contrast, an individual's knowledge, effort and labor cannot be transferred as ownership of a person. Where the usefulness of an asset depends on indispensable human skills, the value of discretionary asset rights may be limited by the worker's ability to withhold cooperation. This observation prevents a mechanical equation of all productive capital with property rights.
Hart and Moore's asset-control account (1990) distinguishes selective replacement within an integrated arrangement from ending dealings with an independent supplier as a whole. The example identifies control of a bundle of complementary possibilities: if the assets are accessible, it may be feasible to retain some productive relationships while replacing another participant. The legal and practical conditions for replacement are essential. Regulatory licensing, exclusive know-how, insolvency restrictions and employment protection may make nominal ownership a weak predictor of the practical outside option.
Cash-flow rights are entitlements to a defined portion of receipts, whereas voting rights determine specified collective decisions. A residual claimant captures a portion of income left after contractual obligations; a residual controller possesses discretion over uses left open. Neither concept necessarily implies the other. A lender can have contingent intervention power without ordinary equity ownership. A corporation can issue shares carrying different voting rights. A cooperative can distribute votes and surplus according to distinct formulas. Analytical categories must be assigned to the specific right and its activation condition.
Asset control can be narrowly partitioned. A party might own a machine while another has exclusive possession during a lease, a lender holds a security interest, and a regulator determines permissible operation. The question for incomplete-contract economics is who can implement or block a relevant alternative in the particular future state. Where rights are jointly held, a decision may require multiple approvals. That requirement can protect one side's contribution but also delay adaptation. A binary integration variable is therefore a useful simplification in an explicit model, not an exhaustive inventory of actual institutional rights.
Timing, Bargaining, and the Hold-Up Mechanism
Consider parties A and B who decide on a governance arrangement at time zero. At time one, A and B can undertake costly investments that affect the value of subsequent joint production. At time two, a trading opportunity becomes known; the parties can agree on an efficient use of their resources or fail to agree and exercise their alternatives. The choice of governance precedes the investments, and the investments precede bargaining. This ordering is foundational. If the parties can contract costlessly on the investments before incurring them, the incentive distortion considered here may be avoidable.
Let total attainable gross surplus from cooperation be V(a,b), with investments a and b. Let the value obtainable without agreement under governance g be d_A(a,b;g) and d_B(a,b;g). Suppose transfers are feasible, the parties have quasi-linear utility, information is symmetric at bargaining, and V is at least the sum of disagreement payoffs. Assume efficient bargaining with A receiving fraction θ of the incremental cooperative surplus, where 0 < θ < 1. Then the gross private payoffs are
Let investment costs be C_A(a) and C_B(b). A fixed ex ante transfer redistributes payoffs without affecting marginal investment incentives under quasi-linearity and unrestricted transfers. The first-order conditions for an interior Nash investment equilibrium are
In the efficient benchmark, with no omitted real costs or third-party effects, the marginal conditions are C_A'=V_a and C_B'=V_b. The difference between these conditions and the private conditions identifies the investment wedge. Derivatives of disagreement payoffs transmit control into investment incentives. The model does not require observed ex post bargaining breakdown: if reassignment and trade are efficient at time two, the distortion can arise entirely from anticipation at time one.
If the investment raises only joint value, with no effect on either disagreement payoff, A privately receives θ of the marginal cooperative benefit and B receives 1−θ of its own. The parties underinvest relative to the interior first best if marginal returns are positive and the relevant problems are well behaved. If control enables one party to obtain some investment return outside agreement, the incentive wedge can shrink. Where an investment instead improves the other party's alternative or creates strategic value without increasing joint surplus sufficiently, the private response can be weaker or excessively strong. Incomplete contracting therefore identifies a possible conflict between private and social marginal returns, not a theorem of universal underinvestment.
The contractual hold-up problem arises when a committed investment changes subsequent opportunities so that its financier cannot ensure receipt of the return anticipated before commitment. Threatening to withhold consent can shift surplus after the investment has become sunk. Such renegotiation may be lawful and efficient in the realized state. Calling it hold-up is an economic characterization of the incentive problem; it does not itself establish fraud, breach or bad faith. A study of governance must show how a counterfactual contractual or institutional arrangement would alter the investment payoff.
A Reproducible Ownership Comparison
The following numerical illustration isolates the investment mechanism without asserting an empirical magnitude. Suppose V(a,b)=6a+4b and each party's investment cost is one-half the square of its investment. Both investments are nonnegative, bargaining shares are equal, and ownership does not affect the feasible gross cooperative surplus. An owner can realize its own investment's return outside the intended cooperation: A's ownership gives d_A=4a and d_B=0; B's ownership gives d_A=0 and d_B=3b. For a neutral comparison assume both disagreement payoffs vanish. These assumptions ensure that incremental cooperative surplus is nonnegative for all feasible a and b.
The first-best investment satisfies a=6 and b=4, yielding net surplus 36+16−18−8=26. With zero disagreement payoffs each party internalizes half its own marginal joint return. Its investment is therefore a=3 and b=2, yielding net surplus 18+8−4.5−2=19.5. Under A ownership, A internalizes half of 6 plus half of 4, so a=5 and b=2. Net surplus is 30+8−12.5−2=23.5. Under B ownership, a=3 and b=3.5. Net surplus is 18+14−4.5−6.125=21.375.
| Arrangement | Investment A | Investment B | Net surplus |
|---|---|---|---|
| Complete investment contract | 6 | 4 | 26 |
| Zero disagreement payoffs | 3 | 2 | 19.5 |
| A owns | 5 | 2 | 23.5 |
| B owns | 3 | 3.5 | 21.375 |
The gain from A ownership over B ownership in this example is 2.125 units. This is a comparative result for an illustrative set of primitives, not a general empirical ranking of managerial or supplier ownership. The social criterion sums both parties' payoffs after subtracting the real investment costs. A transfer from A to B cancels in this sum. If the proposed transfer is infeasible because of credit constraints or limited enforceability, the privately implementable arrangement may differ from the maximum joint-surplus arrangement.
Consider sensitivity to the B owner fallback coefficient. Replace 3 by 4, its maximum consistent with the stipulated nonnegative cooperative increment for all b. Then b=4. the A contribution is 6×3−3²/2=13.5; the B contribution is 4×4−4²/2=8; the total is 21.5. With A's fallback coefficient unchanged at 4, A ownership remains preferable in this calibration. Reverse the investment technology by assigning cooperative marginal returns of 4 to A and 6 to B, fallback coefficients of 3 to A and 4 to B, and the ownership ranking reverses by symmetry. Which party's initiative matters is therefore an economic condition, not a label attached to one type of participant.
This stylized representation omits asset complementarities, dynamic wealth effects, uncertainty about bargaining shares and direct operating costs of integration. These are deliberate restrictions. The example serves to demonstrate what the residual-rights argument can establish and what it cannot. In particular, the beneficial owner is determined by the difference between the total value generated by alternative induced investment profiles. Comparing the distribution of bargaining proceeds alone would not perform that calculation.
Complementarity, Joint Control, and Human Capital
Ownership choices become richer when joint production requires complementary assets. If two machines are technologically useful together but weak individually, separate control can make each owner an indispensable consent holder. Both parties may acquire bargaining leverage from their veto positions. The configuration can also restrict the ability of either party to benefit from its own earlier improvement outside agreement. Concentration of asset control can increase the fallback value of the party receiving the combined assets while diminishing the autonomy of the other. The relevant sign depends on how investments change these feasible alternatives.
The property-rights approach to firm boundaries provides a language for comparing more elaborate allocations than a sale of a single asset. Its economic insight is that organization alters the ability to retain, replace or exclude participants from combinations of assets. The value of a veto can be understood as protection against changes that undermine one participant's earlier investment. The cost of a veto can be understood as a constraint on adapting the productive combination. Joint control may thus be reasonable when both participants make indispensable contributions and a unilateral switch would create a severe exposure, but the arrangement may also increase costly deadlock.
A related distinction concerns productive capital embodied in assets and capital embodied in people. A specialized engineer might be able to undertake additional investment in general technical knowledge regardless of the owner of the equipment. Firm-specific learning, however, may raise productivity principally when the engineer retains access to a given research program. If an owner can substitute another engineer after obtaining the investment's results, the original engineer's return may decline. If the engineer's tacit knowledge remains indispensable, ownership of the machine may be less consequential. Contractibility of knowledge transfer and the portability of capability shape the answer.
The consequences of organizational rights can extend beyond the direct contracting parties. Workers, minority investors and users of an essential service may make investments whose returns depend on decisions made by owners and managers. A bilateral surplus calculation omits these effects unless the affected individuals are represented by enforceable claims or explicit welfare weights. More efficient use of assets among the contracting parties can coexist with reduced opportunities for people outside the agreement. This is particularly important for social evaluation of mergers, public outsourcing and infrastructure concessions.
For a researcher, asset listings and corporate ownership records provide an initial map, not the full measure of control. Evidence on access, rights of refusal, licensing, replacement possibilities and practical dependence on workers is required to establish the predicted bargaining structure. Changes in joint or separate ownership can then be analyzed through the alternatives they create or remove, rather than by an unsupported assumption that legal concentration always implies operating autonomy.
Renegotiation and the Limits of Contingent Mechanisms
An incomplete agreement is frequently followed by additional bargaining. The contract can specify revision procedures, expert determination, options, price adjustments and termination rights. Such devices are valuable because they organize decisions that cannot be fixed in advance. Yet a procedure written before uncertainty is resolved may create incentives that disappear when the parties can jointly abandon it and replace it with a new arrangement. Hart and Moore's renegotiation analysis (1988) examines this limit to closing gaps through mechanisms of revision.
To see the distinction, suppose a rule assigns an inefficient outcome if the parties report different descriptions of a state. The possibility of that outcome can induce agreement on their reports when the rule is credible. After the reports are made, however, both parties may gain by replacing the prescribed inefficient outcome with an efficient agreement and sharing the incremental surplus. If renegotiation is allowed and predictable, the threatened outcome fails to provide the original deterrent in the anticipated form. The investment consequences therefore depend on the continuation game and on enforceable commitments about revision.
Renegotiation is not uniformly undesirable. When technology changes unexpectedly, or a customer discovers that a different output is more valuable, a prohibition on adjustment can impose real losses. A well-designed arrangement can use default positions and revision processes to permit efficient changes while limiting strategic appropriation. A rule assigning a party the right to accept or reject an adjustment can shape bargaining without prescribing every adjustment. Escalation procedures can reduce delay and encourage disclosure of information, although they do not automatically restore the initial commitment.
The exact mechanism-design assumptions matter. A contract can use messages, observable payments and decision options rather than attempting to describe all future physical circumstances. Risk preferences, opportunities for collusion, enforcement of message histories and the committed ability to impose a consequence determine what those mechanisms can accomplish. It is therefore inaccurate to infer from any contract's absence of a detailed contingency list that sophisticated contracting could not solve the corresponding investment problem.
The empirical implication is to examine contract amendments in their procedural context. A high amendment rate might reflect useful flexibility, opportunistic reallocation, poor initial design, new information or regulatory intervention. A low rate might reflect efficient specification or costly rigidity. A count of amendments without their content, counterfactuals and timing cannot identify hold-up. Relevant evidence would compare proposed adjustments, disagreement options and the investment commitments made before revision.
Foundations and the Unforeseen-Contingencies Debate
A central theoretical challenge concerns whether the impossibility of describing future states establishes a meaningful restriction on contracts. Maskin and Tirole's (1999) unforeseen-contingencies argument shows that difficulties in describing physical states need not prevent optimal contracting when participants can make appropriate probabilistic forecasts of future payoffs and can commit to the mechanism. Their abstract also identifies qualifications concerning renegotiation and risk aversion. The result calls for explicit analysis of what can be written, communicated, committed and enforced.
This challenge distinguishes two forms of knowledge. Parties might be unable to name a future technological device but able to anticipate a distribution of values from its possible uses. A contract can give parties rights to choose from future alternatives or make transfers depend on permitted reports. Where credible contingent mechanisms can rely on those informational resources, the absence of names for future objects is not sufficient to establish economically important incompleteness. A model must justify the restriction that prevents an alternative mechanism from implementing an efficient outcome.
Hart and Moore's Foundations of Incomplete Contracts (1999) directly responds to objections to the incomplete-contract framework. The relevant scientific lesson is to formulate a positive account of why the feasible agreement set is restricted. It would be inappropriate to treat the existence of this response as a universal refutation of implementation arguments. Its contribution belongs to the problem of establishing the precise conditions under which incomplete-contract modeling is defensible.
The present synthesis separates technological uncertainty, communication limits, verifiability, commitment to threatened outcomes and legal enforceability. They influence different stages of a contractual mechanism. A market participant may possess a good probabilistic description of future payoffs while lacking the ability to verify reports; an arbitrator may verify an outcome while the parties remain free to agree on new terms; a binding procedure may exist while its operation costs exceed the value of its informational refinement. Distinguishing these restrictions keeps the theory responsive to counterexamples.
The evidentiary status of the fundamental debate is primarily theoretical. A theorem can establish that a mechanism succeeds or fails under its assumptions. Observed contracts establish which mechanisms parties actually use, but their absence may reflect legal restrictions, cognitive burdens, equilibrium selection or their comparative costs. The empirical research task is consequently to connect a defensible theoretical restriction with measurable differences in institutional opportunity. That task is more demanding and more informative than describing all contracts as inevitably incomplete.
Formal Authority, Delegation, and Information
Asset control does not ensure that every decision within a firm is made by its legal owner. Formal authority identifies the party entitled to choose. Effective authority depends on who possesses the information needed to evaluate the alternatives and who can implement a decision in time. Managers may delegate decisions to people with specialized local knowledge, even where they retain power to intervene. The arrangement shapes how much those people invest in collecting information and developing proposals.
Aghion and Tirole's formal and real authority (1997) explains the tension between initiative and principal control through endogenous information. Delegation can improve the subordinate's incentive to identify projects when it increases the probability that a preferred proposal will be adopted. Retaining authority can protect the principal from a choice misaligned with its objectives. The value of delegation is jointly determined by the potential gain from better information and the consequences of disagreement over which project should be chosen.
A simple illustrative decision makes the trade-off concrete. Suppose two parties can choose between projects U and V. Project U yields 10 to the principal and 2 to the agent; V yields 6 to the principal and 8 to the agent. When only the agent has information sufficient to discover a feasible project, insisting on principal review can create costly delay. If the principal can establish that U is feasible immediately, delegation may instead yield a reduction of four in its payoff and an increase of six in the agent's. Joint surplus in this illustration favors V by two, but the principal's preference favors U. Those preferences and information acquisition conditions must enter the analysis independently. No empirical quantities are inferred from these numbers.
Authority inside the firm can be allocated across categories rather than handed over in its entirety. Engineers may choose how to meet a technical standard; managers can control budget; safety officers can forbid specified actions. The resulting allocation defines who supplies information and who can veto a decision. Veto and initiative are distinct forms of power. A party with a veto can stop a project but may have no right to initiate or implement a replacement.
This perspective explains why outsourcing and decentralization should not be treated as identical. An integrated firm can grant substantial local decision authority while retaining the productive assets. An independent supplier may have extremely restricted practical discretion under a detailed purchaser contract. The property-rights model concerns residual control over the assets and disagreement alternatives; the authority model adds the information and decision processes through which those rights are exercised. Empirical analysis should therefore distinguish legal ownership from the decision network and the degree of informative initiative.
Relational Contracts and Behavioral Adaptation
Many productive relationships rely on conduct that participants recognize without being able to enforce through a third party. Reputation, continuation and future gains can support promises that would otherwise be difficult to make credible. A relational contract is a self-enforcing agreement whose operation depends on the value of continuing cooperation relative to the gain from deviation. Its sustainability is affected by the allocation of residual rights because that allocation determines what a party can obtain without honoring the promise.
Baker, Gibbons, and Murphy's relational theory of the firm (2002) connects integration with the incentives to renege on informal agreements. Governance changes the deviation payoff as well as the opportunities for cooperation. That mechanism differs from the static investment problem: the central constraint is the ability to sustain an agreement across repeated interactions. Since residual control can change both compliance and deviation returns, the two mechanisms can interact. Ownership that promotes investment in a one-period calculation might make a valuable ongoing bonus promise difficult to sustain, or conversely.
The arithmetic of a simple repeated agreement clarifies the point. Suppose deviation yields a one-time extra gain D, and compliance preserves extra surplus s in every future period relative to the consequence of deviation. With constant discount factor δ, stationary payoffs, perpetual continuation and a credible punishment, a sufficient compliance condition for the relevant actor is
For example, D=8 and s=3 require δ at least 8/11, approximately 0.7273. If ownership changes D to 5 at the same continuation surplus, the threshold becomes 5/8, or 0.625. The numerical comparison is purely illustrative. Actual relationships face varying payoffs, limited information about deviation, renegotiation of punishments and uncertain horizons. These complications can invalidate the simple inequality as a complete account.
Another mechanism arises when parties interpret the original contract as a baseline for what they deserve. Hart and Moore's reference-point analysis (2008) examines how flexibility can permit efficient adjustment yet produce discontent about the chosen outcome. When a party perceives that its entitlement has not been respected, it may lower aspects of performance that cannot be compelled. The cost of contractual discretion may accordingly include reduced discretionary cooperation despite continued trade.
Fehr, Hart, and Zehnder's experimental evidence (2011) reports less performance shading under rigid contracts than under flexible ones in the studied laboratory conditions. This supports the existence of the proposed behavioral trade-off in that setting. External validity requires attention to the particular participants, institutional design and available actions; the finding does not by itself estimate an effect for a multinational supply agreement. The experimental and repeated-game mechanisms can coexist. Where participants cooperate for future gains, perceptions of entitlement may nevertheless influence their current willingness to deliver noncontractible quality.
A robust comparative analysis must therefore distinguish the enforceability of promises, the credibility of continued cooperation, and the behavioral response to discretionary choices. Flexibility can have an adaptation benefit, a strategic commitment cost and an entitlement-related cost. Their relative magnitudes can vary across institutions. Treating all three as the same unspecified transaction cost would eliminate useful empirical distinctions.
Financial Contracts and Contingent Control
Financial investment separates the provision of capital from the exercise of operating decisions. A founder may possess productive knowledge and initiative, while an outside investor finances expenditure and bears substantial risk. Because future opportunities and effort cannot all be verified, financing arrangements frequently establish decision rules for circumstances in which new finance, restructuring or replacement of management becomes relevant. The rights over these decisions influence behavior before those circumstances occur.
Kaplan and Strömberg's (2003) venture capital contract study examines actual contracts and distinguishes cash-flow, board, voting, liquidation and other control rights. The separation of these rights provides directly relevant institutional evidence: ownership percentages do not describe the complete governance arrangement. Rights may change with observable financial or operational performance and across financing rounds. A security carrying priority in liquidation may allocate losses differently from a security conferring authority over new investments.
An important temporal feature is the activation of control. While performance remains acceptable, a founder may retain authority over ordinary projects; when stated conditions are not satisfied, the investor may acquire additional rights. This arrangement can preserve the founder's incentive to develop the enterprise while protecting funds in adverse states. Its effectiveness depends on the trigger's verifiability, the threat of strategic manipulation and the feasible alternatives once authority switches. A contractual trigger attached to revenue, liquidity or another objective measure may be verifiable but imperfectly related to the value of continuing the enterprise.
An illustration clarifies the difference between incentives and risk distribution. Imagine that the holder of a contingent intervention right can improve liquidation proceeds from 30 to 40 after a state in which continued operations yield 35 to the sum of the contracting parties. Exercising the right may raise aggregate value by five relative to continuation under these assumptions. Change the continuation value to 50, with the liquidation value still 40, and immediate liquidation destroys ten. Actual parties may nevertheless favor different choices because their securities allocate the realized proceeds differently. The researcher must determine whether the controller captures the gain or loss accruing to all relevant parties and how the contract conditions intervention.
International legal arrangements influence the feasibility of contingent rights. Kaplan, Martel, and Strömberg's (2007) study of legal differences and experience asks how financial contracting depends on institutional environments and experience. The economic significance is that the same named security or voting provision may operate differently under distinct enforcement, corporate and insolvency rules. Cross-country comparisons must account for the composition and experience of investors, domestic institutions and selection into financing relationships. The existence of comparable instruments in several markets does not establish identical enforceability.
Corporate finance thus expands the property-rights question from ownership of tangible assets to control of the process by which new capital is committed and distressed assets are reorganized. It also shows why the economically efficient allocation might be state-contingent rather than a permanent assignment to one participant. Where states are verifiable, contingent control can be specified directly; for remaining decisions, the residual rights and practical alternatives determine bargaining power.
Measurement Technology and Governance in Production
Contractibility can change as measurement improves. A technology that records behavior may allow parties to monitor aspects of performance previously left to discretion. Its economic effects depend on whether the resulting records can be used to enforce a suitable agreement. A record may contain noise, measure a different objective or be costly to interpret. The relationship between monitoring and asset ownership must therefore be studied at the level of the tasks the technology makes observable and enforceable.
Baker and Hubbard's trucking ownership study (2004) provides an empirical approach to this question. The authors compare patterns of driver ownership under different haul conditions and the relationship between adoption of on-board computers and ownership. Their reported finding is that ownership by drivers is more common on longer hauls and declines with computer adoption, especially on longer hauls. This establishes a setting in which the ownership and measurement choices vary together in the manner investigated by contract theory.
The observation admits several possible causal pathways. If monitoring makes a driver's previously uncontractible actions enforceable through incentive agreements, the value of relying on driver ownership for incentive alignment may fall. Computers might also improve operational coordination, reduce maintenance cost or alter route selection. Firms may choose the technology precisely when they are reorganizing their operations. Establishing the magnitude of the ownership effect attributable specifically to contractibility requires information about the timing of adoption, the kinds of devices used, selection and competing changes in production.
Acemoglu, Aghion, Griffith, and Zilibotti's technology and integration study (2010) develops theoretical predictions and investigates them using UK manufacturing plant-level data. The research connects ownership with investment incentives and the technology intensity of productive relationships. The unit is materially different from the truck or driver in the preceding study; manufacturing integration may involve complementary capital, technological investments and larger organizations rather than a single mobile asset. A theory's usefulness lies in its capacity to specify how those differences enter the prediction.
These cases suggest an important measurement distinction for empirical research. A variable called technology can measure the introduction of monitoring, the complexity of capital, the productivity of a process or the intensity of research and development. These characteristics can influence governance in opposite directions. Monitoring may expand the feasible contract set, while more specific investment can increase the significance of residual control. Aggregating them into a single index risks identifying an average that does not correspond to either mechanism.
The design of a comparison should therefore classify the technological change and the previously uncontractible decision before observing the ownership result. This reduces the temptation to reinterpret every increase or decrease in integration as support for the same broad theory. Conditional predictions should be evaluated with evidence on the decision margin, the preexisting contractual restriction and the actual change in feasible alternatives.
International Production and Contract Enforcement
International exchange extends the analysis to relationships spanning legal regimes, national locations and production networks. A buyer may import an input from an affiliate or purchase the same input from an independent foreign supplier. These organizational alternatives distribute productive assets, information and decision rights differently. They also differ in fixed costs, taxation, transport, regulatory exposure and the composition of firms that can use them.
Antràs's (2003) firms and trade structure study associates the share of US imports within firms with the capital intensity of exporting industries and the capital-labor ratios of exporting countries in its historical data. The theoretical account joins a property-rights mechanism with international trade. Its empirical unit and denominator matter: an intrafirm share of US imports differs from the fraction of all production controlled by multinational corporations or the share of all contemporary world trade. The result is a historical association in specific data, not a current universal statistic.
Antràs and Helpman's global sourcing framework (2004) introduces heterogeneous producers choosing both supplier location and governance form. The decision reflects the trade-off between gains from the organizational arrangement and its fixed or variable costs. Firms with different productive capacities can select distinct forms within the same sector. An industry-level integration share can change as the composition of firms changes, even if each firm follows the same decision rule over time. Empirical tests consequently benefit from information on productivity, firm entry, sourcing and the actual location of ownership.
Contract enforcement can also shape comparative advantage without necessarily determining whether any single relationship should be integrated. Nunn's relationship-specificity and trade study (2007) constructs a sector-related measure from inputs requiring relationship-specific investments and combines it with trade and judicial-quality measures. The reported pattern links better contract enforcement with specialization in products dependent on specific inputs. The empirical interpretation requires distinction between legal quality, factor endowments, production capabilities and other country characteristics; the published association is not a direct observation of each bargaining disagreement.
These international applications highlight two potentially separate margins. Better enforcement can make arm's-length transactions more viable by expanding the feasible agreement set. It can also increase investment in complex production generally, including production within integrated firms. The direction of an observed integration-share change depends on which margin dominates, the underlying industrial composition and the alternatives available. Inference from aggregate national indicators without firm-level governance data can be misleading.
For cross-border research, the proper comparison distinguishes contract governing law, place of enforcement, owner jurisdiction, physical location of the asset, supplier residence and market of sale. These can differ in one relationship. A cross-country contract-enforcement index may proxy only a subset of relevant institutions. The property-rights explanation is strengthened when the analyst can identify the precise rights that remain credible in disagreement and the investments expected to respond to them.
Public Provision, Outsourcing, and Noncontractible Quality
Public services introduce objectives whose value may not be captured by the provider's contractual receipts. A buyer can specify measured output and pay for it, yet some consequences affect service users or the public in ways difficult to verify in a contract. In settings such as prisons, health services and infrastructure, quality encompasses dimensions that extend beyond the recorded quantity of provision. Ownership and control influence the provider's incentive to reduce cost and change the quality of service.
Hart, Shleifer, and Vishny's public provision analysis (1997) compares government employment with contracting out provision under incomplete contracts. Their theoretical application to prisons emphasizes stronger private incentives for both cost reduction and quality improvement, together with the risk of excessive reductions in noncontractible quality. This is a model-based institutional argument. It does not constitute a causal estimate of all private prisons, nor establish that public provision is always preferable.
A simple arithmetic example separates private and social returns. A provider can invest in a new process at real cost 4. The change saves 10 in operating expenditure, so if the provider retains the entire saving the net private gain is 6. If it reduces the value of uncontractible service quality by 9, the net social change is 10−4−9=−3. If quality loss is 1 instead, the net social change is 5. The example uses hypothetical units and omits any transfer effects. It shows why the efficient strength of cost-reduction incentives depends on the marginal quality consequence; it does not measure that consequence in a real public service.
The same arrangement might yield underinvestment in quality improvement when the provider captures a small fraction of the value created for users. Public management can also exhibit weak effort incentives and bureaucratic constraints. A comparative assessment should include the government's ability to observe and reward improvements, the contractor's residual rights, costs of procurement and transition, competitive entry, accountability and the costs of external effects. The governance alternative to contracting out should itself be described realistically.
Public-private partnership contracts can establish decision procedures for material changes during long projects. The World Bank's PPP renegotiation guidance discusses conditions for revising agreements, the importance of disclosing changes and the risks that substantial modifications create for procurement competition. Step-in and termination rights allocate discretion in specified circumstances and influence what each party can credibly threaten. Guidance is an institutional source and must be distinguished from a binding rule in any particular jurisdiction.
A fundamental tension arises where the state must continue providing an essential service even if the private provider fails. The public authority cannot credibly accept every threatened service interruption. This can change bargaining leverage, financing choices and expectations of future assistance. A regulator may specify intervention and compensation rules to preserve service while reducing incentives to shift avoidable costs onto the public. The appropriate evaluation concerns credible residual alternatives, the cost of failure and the effect on earlier bids and investment. Measuring only the nominal contract price would omit each of these dimensions.
Contract Law, Hardship, and Institutional Boundaries
Economic contractibility is embedded in law. Courts, arbitral tribunals, administrative bodies, contract managers and informal organizations provide different forms of enforcement and dispute resolution. Their capacity to establish relevant facts, deliver timely orders and impose remedies determines whether a contractual condition can operate as anticipated. Jurisdictions differ in the legal content of ownership and the limits on private agreements. Accordingly, a property-rights model must state which legal and practical entitlements it treats as feasible.
The UNIDROIT hardship provisions (2010 edition, articles 6.2.1–6.2.3) offer a specific international commercial-contract framework for exceptional changes in circumstances. The general obligation of performance coexists with a hardship category requiring a fundamental alteration of the contractual equilibrium and satisfaction of additional conditions. The framework provides a path for requesting renegotiation and, under specified circumstances, further adjudicative responses. Its relevance lies in the institutional organization of revision. It is not a universal statute and its effect depends on its applicable legal or contractual basis.
The legal category of hardship must be distinguished from economic hold-up. One concerns legally defined exceptional circumstances and possible responses; the other describes exposure of an investment's return to subsequent bargaining. A party can face an economic hold-up problem without satisfying legal conditions for hardship. An unexpected event can satisfy a legal category while leaving the original investment incentives largely unaffected. A careful cross-disciplinary account preserves the separate questions: what revision is legally authorized, and what anticipated payoff results from that authorization?
Contractual remedies influence outside options through their timing and coverage. A damages award may compensate a particular loss but arrive after a long delay; an order requiring performance can preserve access to an important asset but require monitoring; a termination provision may expand the set of alternatives while imposing a payment. The enforcement institution changes the value of alternatives and hence the bargaining problem even when its services are rarely invoked. Observations of litigated disputes are consequently selected from a much larger population of contractual interactions.
Suppose a supplier can redirect a machine to another customer, but its current purchaser holds enforceable exclusive-use rights for a specified period. During that period, the value of unilateral diversion as a disagreement option is constrained by the agreement. The relevant outside payoff must account for the enforceable obligation and the realistic cost of exercising an alternative. After the period ends, the same formal asset ownership can confer more discretion. The economic content of a residual right changes when specific contractual obligations expire or become activated.
The scientific conclusion is that institutions of enforcement must be represented inside the analysis of bargaining opportunities rather than added only after a model has been solved. International comparisons that use a generic enforcement score require a demonstrated connection to the specific rights or decisions examined. An evaluation of a current individual contract also requires its governing law and actual terms. The present dossier is an economic and institutional synthesis, not legal advice or an assertion of jurisdiction-specific enforceability.
Empirical Identification and Research Design
The major empirical challenge is that governance is chosen. Firms adopt contractual safeguards, acquire assets and delegate authority in response to expected benefits and costs. Observed integrated transactions therefore differ systematically from observed market transactions. A regression of performance on a binary integration indicator combines the causal effect of integration with preexisting differences between the relationships selected into each form.
Let G be an indicator of governance and Y an outcome such as investment or output quality. Potential outcomes Y(1) and Y(0) represent results under two well-specified arrangements. If each observation has a consistent treatment and there is no interference, the observed conditional difference can be written as
The second term is the selection component. A researcher who can identify only the left-hand side has not yet established the treatment effect. Moreover, the equation presupposes that the treatment has a coherent meaning across observations. Two firms classified as integrated might allocate very different residual rights, and treatment effects might depend on their suppliers' choices. Then the potential-outcome representation must be refined to reflect these institutional variations and interactions.
The study of actual VC agreements by Kaplan and Strömberg (2003) demonstrates the importance of measuring contract clauses rather than treating broad security categories as complete descriptions. The trucking evidence in Baker and Hubbard (2004) illustrates a distinct approach linking asset ownership to a technology relevant to monitoring. These are complementary forms of evidence. Clause description can establish that contractual dimensions exist and covary with observable features; a technology-based research design can investigate whether changed information or enforcement opportunities predict organization. Neither description alone guarantees an exogenous source of variation.
Possible stronger designs include discontinuities around binding regulatory thresholds, staggered reforms altering enforceability, externally imposed changes to measurement standards, and randomized variations in the allocation of decision rights where feasible and ethical. Each requires a defensible identification argument. A regulatory threshold can affect financing or firm size through other channels. A staggered reform can be anticipated or coincide with broader institutional changes. A monitoring intervention can directly improve production, violating an exclusion restriction if used as an instrument solely for contractibility. Randomization may identify the effect of offering an arrangement rather than that of actual take-up.
Research on ownership should report the unit of observation, right affected, outcome, observation horizon, comparison group, institutional constraints and source of identifying variation. Contract-text analysis should record which rights activate in which states. Investment measures should separate expenditure, asset purchase, research and human learning when feasible. Quality should be measured in terms that reflect the welfare-relevant product, rather than assuming costs are a sufficient statistic. Attrition and failed relationships require special attention because some consequences of hold-up never appear among surviving contracts.
The published empirical cases considered here support testing mechanisms at the level of concrete institutional margins. They do not provide a common, transferable numerical effect size. The dossier reports the directions and settings directly supported by accessible publication abstracts and does not pool coefficients, reconstruct proprietary data or claim full replication. Further research would require the underlying data and complete methodological materials before estimating or comparing causal magnitudes.
Original Analytical Synthesis and Counterexamples
The central synthesis concerns the temporal relationship between decision rights and specific investments. A contractual gap affects economic behavior when a future discretionary decision influences the expected private return to an action taken earlier. This relationship connects ownership with bargaining, bargaining with investment, and investment with feasible productive value. The chain is informative because each link can be checked separately. A gap in contractual language need not produce an investment distortion if no consequential decision remains discretionary. A discretionary decision need not affect investment when its payoff consequences are independent of that investment. An investment distortion need not justify integration if another arrangement addresses it at lower total cost.
The numerical example in this study isolates the connection. A fixed cooperative opportunity set produces different investment levels because changing control changes the derivative of a disagreement payoff. Efficient bargaining after investment is compatible with inefficient investment before bargaining. An organizational reform that reduces the probability of visible disputes can therefore fail to address the deeper incentive problem. Conversely, an arrangement generating occasional renegotiation can sustain high investment if the anticipated distribution of returns is sufficiently well aligned with the value created. Observed conflict and ex ante incentives are distinct economic outcomes.
The property-rights, authority, relational and behavioral accounts can be compared without collapsing them into a new theoretical system. The first emphasizes alternative asset uses and bargaining. The second emphasizes decisions conditioned on knowledge and initiative. The third emphasizes the future benefits supporting a self-enforcing promise. The fourth emphasizes expectations of entitlement and discretionary quality. In a single supplier relationship, a purchaser's acquisition of an asset can improve its expected investment returns, reduce the supplier's information-acquisition initiative, alter the profitability of reneging on a bonus, and change the interpretation of fair treatment. A scientific explanation identifies these mechanisms and their respective variables before offering a net organizational judgment.
Comparative efficiency also requires recognition of the boundary of the contracting coalition. Two investors may choose an arrangement that maximizes their joint surplus while externalizing costs to workers, consumers or taxpayers. The contract model's objective is a valid measure of social efficiency only if all relevant effects are included, rights are evaluated with appropriate weights, and transfers actually capable of compensating affected parties are available where assumed. In public services, the cost and quality example demonstrates the consequence of excluding third parties. In international production, differences in the enforcement of obligations can alter not only governance but also the selection of industries, firms and participants into markets.
Counterexamples sharpen the conclusion. With complete enforceable investment contracts and unrestricted transfers, the simple ownership wedge can disappear. With no feasible alternative use and fixed bargaining shares, transferring ownership may leave disagreement payoffs unchanged. With cooperative investments that produce negative externalities, greater private incentives can exacerbate socially excessive conduct. With sufficiently valuable repeated relationships, informal promises may support an outcome beyond the static one-shot arrangement. With a credible mechanism that implements the desired allocation through observable reports, difficulties in describing every physical state may have limited economic importance, as the implementation critique makes clear.
The resulting analytical procedure is to specify the uncontractible action or contingency, trace the practical default decision and disagreement payoff, derive the resulting marginal incentive, and compare feasible governance alternatives including their direct costs and effects on nonparties. Its application should identify observable institutional and behavioral evidence for each step. This procedure is an authorial organization of established mechanisms and a comparative interpretation of their implications. It introduces no newly named model, doctrine, or claim of priority.
Open scientific questions concern measurement of actual discretionary control when ownership is fragmented among shareholders, creditors, employees and regulatory institutions; comparative effects of monitoring technology when it also changes productivity; the empirical interaction between formal bargaining power and relational cooperation; and the valuation of quality that cannot be captured in a contract. Each question requires an identified contractual margin and an appropriately selected comparison, not extrapolation from a single industry. The most useful future evidence would establish how changes in specific enforceable rights affect the earlier actions of people whose productive contributions depend on those rights.
Conclusion
Contractual incompleteness makes governance an integral part of economic allocation. Contractual promises determine a portion of future action; the remaining discretion belongs to parties and institutions with enforceable rights and practical capacity to act. That discretion influences the division of gains when agreement must be revised and thereby the incentives to invest before future decisions are taken. The property-rights framework establishes a formal link between assignment of control and induced investment. Its explanatory force depends on the actual available contracts, the treatment of renegotiation, the complementarity of assets and people, and the response of disagreement payoffs to investment.
The study has distinguished the limited description of future contingencies from nonverifiability, the allocation of ownership from formal and effective authority, and private surplus from social welfare. These distinctions explain why an observed change in legal ownership may have a strong, weak or opposite incentive effect depending on the institutional environment. An arrangement can realize efficient subsequent exchange and still induce inefficient earlier investment. Control can also be allocated through contingent financial rights, cooperation sustained over time, and decision procedures that preserve flexibility while providing safeguards.
The applications provide differentiated evidence. Venture capital contracts demonstrate the separate allocation of financial and decision rights. Trucking and manufacturing studies connect production organization with precisely defined technological conditions. Research on trade relates contract enforcement and ownership patterns to the organization and specialization of international production. Experimental evidence supports a behavioral cost of flexibility within a particular laboratory design. Public provision theory shows why private cost reduction can conflict with quality that matters to the public. These results support the investigation of distinct mechanisms under explicit identification conditions; they do not produce a universal quantitative ranking of ownership systems.
The authorial contribution is the integration of these results around the temporal sequence from feasible commitment to anticipatory investment, subsequent discretion, bargaining, adaptation and the final distribution of consequences. The illustrative calculations demonstrate that a specific control assignment can improve or impair joint surplus depending on the induced investment profile, and that the same economic change can increase private returns while reducing social value when costs fall on third parties. The contribution is a comparative analytical synthesis of existing scientific work, not a claim of a new named theory.
The economically appropriate allocation of residual control is a conditional result of the feasible governance alternatives and their consequences. It must be determined by the value of investment and adaptation, the costs of enforcement and organization, the credibility of future cooperation, and the rights and welfare of affected persons. This conclusion defines a substantive program for research on organizations across markets, financial systems and public institutions.
References
Acemoglu, Daron, Philippe Aghion, Rachel Griffith, and Fabrizio Zilibotti. 2010. “Vertical Integration and Technology: Theory and Evidence.” Journal of the European Economic Association 8 (5): 989–1033. DOI: 10.1111/j.1542-4774.2010.tb00546.x. Publication or institutional record
Aghion, Philippe, and Jean Tirole. 1997. “Formal and Real Authority in Organizations.” Journal of Political Economy 105 (1): 1–29. DOI: 10.1086/262063. Publication or institutional record
Antràs, Pol, and Elhanan Helpman. 2004. “Global Sourcing.” Journal of Political Economy 112 (3): 552–580. DOI: 10.1086/383099. Publication or institutional record
Antràs, Pol. 2003. “Firms, Contracts, and Trade Structure.” Quarterly Journal of Economics 118 (4): 1375–1418. DOI: 10.1162/003355303322552829. Publication or institutional record
Baker, George P., and Thomas N. Hubbard. 2004. “Contractibility and Asset Ownership: On-Board Computers and Governance in U.S. Trucking.” Quarterly Journal of Economics 119 (4): 1443–1479. DOI: 10.1162/0033553042476152. Publication or institutional record
Baker, George, Robert Gibbons, and Kevin J. Murphy. 2002. “Relational Contracts and the Theory of the Firm.” Quarterly Journal of Economics 117 (1): 39–84. DOI: 10.1162/003355302753399445. Publication or institutional record
Coase, Ronald H. 1937. “The Nature of the Firm.” Economica 4 (16): 386–405. DOI: 10.1111/j.1468-0335.1937.tb00002.x. Publication or institutional record
Fehr, Ernst, Oliver Hart, and Christian Zehnder. 2011. “Contracts as Reference Points—Experimental Evidence.” American Economic Review 101 (2): 493–525. DOI: 10.1257/aer.101.2.493. Publication or institutional record
Grossman, Sanford J., and Oliver D. Hart. 1986. “The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration.” Journal of Political Economy 94 (4): 691–719. DOI: 10.1086/261404. Publication or institutional record
Hart, Oliver, and John Moore. 1988. “Incomplete Contracts and Renegotiation.” Econometrica 56 (4): 755–785. Publication or institutional record
Hart, Oliver, and John Moore. 1990. “Property Rights and the Nature of the Firm.” Journal of Political Economy 98 (6): 1119–1158. DOI: 10.1086/261729. Publication or institutional record
Hart, Oliver, and John Moore. 1999. “Foundations of Incomplete Contracts.” Review of Economic Studies 66 (1): 115–138. Publication or institutional record
Hart, Oliver, and John Moore. 2008. “Contracts as Reference Points.” Quarterly Journal of Economics 123 (1): 1–48. DOI: 10.1162/qjec.2008.123.1.1. Publication or institutional record
Hart, Oliver, Andrei Shleifer, and Robert W. Vishny. 1997. “The Proper Scope of Government: Theory and an Application to Prisons.” Quarterly Journal of Economics 112 (4): 1127–1161. DOI: 10.1162/003355300555448. Publication or institutional record
Kaplan, Steven N., and Per Strömberg. 2003. “Financial Contracting Theory Meets the Real World: An Empirical Analysis of Venture Capital Contracts.” Review of Economic Studies 70 (2): 281–315. DOI: 10.1111/1467-937X.00245. Publication or institutional record
Kaplan, Steven N., Frederic Martel, and Per Strömberg. 2007. “How Do Legal Differences and Experience Affect Financial Contracts?” Journal of Financial Intermediation 16 (3): 273–311. Publication or institutional record
Klein, Benjamin, Robert G. Crawford, and Armen A. Alchian. 1978. “Vertical Integration, Appropriable Rents, and the Competitive Contracting Process.” Journal of Law and Economics 21 (2): 297–326. DOI: 10.1086/466922. Publication or institutional record
Maskin, Eric, and Jean Tirole. 1999. “Unforeseen Contingencies and Incomplete Contracts.” Review of Economic Studies 66 (1): 83–114. DOI: 10.1111/1467-937X.00079. Publication or institutional record
Nunn, Nathan. 2007. “Relationship-Specificity, Incomplete Contracts, and the Pattern of Trade.” Quarterly Journal of Economics 122 (2): 569–600. DOI: 10.1162/qjec.122.2.569. Publication or institutional record
UNIDROIT. 2010. UNIDROIT Principles of International Commercial Contracts 2010. Chapter 6, Section 2, Articles 6.2.1–6.2.3, “Hardship.” Rome: International Institute for the Unification of Private Law. Publication or institutional record
Williamson, Oliver E. 1979. “Transaction-Cost Economics: The Governance of Contractual Relations.” Journal of Law and Economics 22 (2): 233–261. DOI: 10.1086/466942. Publication or institutional record
World Bank. n.d. “Renegotiation, Government Step-in Rights, Termination, and Dispute Resolution.” PPP Resource Center. Accessed November 17, 2026. Publication or institutional record