On NPR's Morning Edition Steve Inskeep talked to Bengt Holmstrom of MIT, who was one of the winners of the 2016 Nobel Prize in economics.
Showing posts with label Holmstrom. Show all posts
Showing posts with label Holmstrom. Show all posts
Saturday, 31 December 2016
Monday, 12 December 2016
Nobel Prize lectures in economic sciences for 2016
From Nobelprize.org:
Oliver Hart: Incomplete Contracts and Control
Oliver Hart delivered his Prize Lecture on 8 December 2016 at the Aula Magna, Stockholm University.
Bengt Holmström: Pay for Performance and Beyond
Bengt Holmström delivered his Prize Lecture on 8 December 2016 at the Aula Magna, Stockholm University.
Oliver Hart: Incomplete Contracts and Control
Oliver Hart delivered his Prize Lecture on 8 December 2016 at the Aula Magna, Stockholm University.
Bengt Holmström: Pay for Performance and Beyond
Bengt Holmström delivered his Prize Lecture on 8 December 2016 at the Aula Magna, Stockholm University.
Monday, 24 October 2016
Kevin Bryan on Bengt Holmstrom and the black box of the firm
Kevin Bryan writes at VoxEU.org on the key contributions of Bengt Holmstrom to the theory of contracts and its application to the theory of the firm. Bryan opens by saying,
So real people where into incentives contracts long before economists started formal investigation of them!!!
The British economist Edwin Chadwick was also thinking about about such issues in the mid-1800s. Chadwick pondered the question of incentives for the transportation of prisoners from the UK to Australia. Chadwick noted:
Bryan goes on:
Holmström’s contribution lies most centrally in the area of formal contract design. Imagine that you want someone – an employee, a child, a subordinate division, an aid contractor or, more generally, an ‘agent’ – to perform a task. How should you induce them to do this?Perhaps a real world example of such a contract may help illustrate that is happening here. Below are details of the contract for surgeons on the ships that transported colonists to the province of Canterbury in New Zealand in the 1850s.
If the task is ‘simple’ – meaning that the agent’s effort and knowledge about how to perform the task most efficiently is known and observable – you can simply pay a wage, cutting off payment if effort is not being exerted. When only the outcome of work can be observed, if there is no uncertainty in how effort is transformed into outcomes, knowing the outcome is equivalent to knowing effort, and hence optimal effort can be achieved via a bonus payment made on the basis of outcomes.
All straightforward so far. The trickier situations, which Holmström and his co-authors have analysed at great length, are when neither effort nor outcomes are directly observable.
Consider paying a surgeon. You want to reward the doctor for competent, safe work.
The British economist Edwin Chadwick was also thinking about about such issues in the mid-1800s. Chadwick pondered the question of incentives for the transportation of prisoners from the UK to Australia. Chadwick noted:
[I]n the first instance, a capitation payment was made on embarkation, and this resulted in the loss of half the convects put on board ; by degrees that loss was reduced to one-third ; but when, under the auspices of a new colonial administration, the system was altered to a capitation payment for all the convicts that were landed at their destination, the contrast was very striking indeed, and tho owners of the vessels carried surgeons, and the best means were devised for landing the largest possible number at the port for which they were bound.But as Byran notes that Holmstrom was aware that in general, in the modern case, it is very difficult to observe perfectly what the surgeon is doing at all times, and basing pay on outcomes has a number of problems:
Bryan continues,Holmström wrote the canonical paper on each of these topics. His 1979 paper shows that any information that reduces the uncertainty about what an agent actually did should feature in a contract, since by reducing uncertainty, you reduce the risk premium needed to incentivize the agent to accept the contract.
- First, the patient outcome depends on the effort of not just one surgeon, but on others in the operating room and prep table. Team incentives must be provided.
- Second, the doctor has many ways to shift the balance of effort between reducing costs to the hospital, increasing patient comfort, increasing the quality of the medical outcome, and mentoring young assistant surgeons. So paying on the basis of one or two tasks may distort effort away from other harder-to-measure tasks. There is a multitasking problem.
- Third, the number of medical mistakes, or the cost of surgery, that a hospital ought to expect from a competent surgeon depends on changes in training and technology that are hard to know, and hence a contract may want to adjust payments for its surgeons on the performance of surgeons elsewhere. Contracts ought to take advantage of relevant information when it is informative about the task being incentivised.
- Fourth, since surgeons will dislike risk in their salary, the fact that some negative patient outcomes are just bad luck means that you will need to pay the surgeon very high bonuses to overcome their risk aversion. When outcome measures involve uncertainty, optimal contracts will weigh ‘high-powered’ bonuses against ‘low-powered’ insurance against risk.
- Fifth, the surgeon can be incentivised either by payments today or by keeping their job tomorrow, and worse, these career concerns may cause the surgeon to waste the hospital’s money on tasks that matter to the surgeon’s career beyond the hospital.
It might seem strange that contracts in many cases do not satisfy this ‘informativeness principle’. For example, CEO bonuses are often not indexed to the performance of firms in the same industry. If oil prices rise, essentially all oil firms will be very profitable, and this is true whether or not a particular CEO is a good one. Bertrand and Mullainathan (2001) argue that this is because many firms with diverse shareholders are poorly governed.
Much of Holmström’s work in the 1980s and 1990s tried to square the gap between theory and empirics by finding justifications for the simplicity of many real world contracts that can be rationally justified.A brief discussion of a simple version of Holmstrom's 1982 paper, due to Kim C. Border, and its relationship to Alchian and Demsetz's work is given in chapter 4 of The Theory of the Firm: An overview of the economic mainstream.
Written jointly with Paul Milgrom, the famous ‘multitasking’ paper published in 1991 notes that contracts shift incentives across different tasks in addition to serving as risk-sharing mechanisms and as methods for inducing effort. Since bonuses on task A will cause agents to shift effort away from hard-to-measure task B, it may be optimal to avoid strong incentives at all (just pay teachers a salary rather than a bonus based only on test performance) or to split job tasks (pay bonuses to teacher A who is told to focus only on mathematics test scores, and pay salary to teacher B who is meant to serve as a mentor).
That outcomes are generated by teams also motivates simpler contracts. Holmström’s 1982 article on incentives in teams points out that if both my effort and yours is required to produce a good outcome, then the marginal product of our efforts are both equal to the entire value of what is produced, hence there is not enough output to pay each of us our marginal product. What can be done?
Alchian and Demsetz had noticed this problem in 1972, arguing that firms exist to monitor the effort of individuals working in teams. With perfect knowledge of who does what, you can simply pay the workers a wage sufficient to make the optimal effort, then collect the residual as profit.
Holmström notes that the monitoring isn’t the important bit; rather, even shareholder-controlled firms where shareholders do no monitoring at all are useful. The reason is that shareholders can be residual claimants for profit, and hence there is no need to distribute profit fully to members of the team.
Bryan goes on:
Free-riding can therefore be eliminated by simply paying team members a wage of X if the team outcome is optimal, and zero otherwise. Even a slight bit of shirking by a single agent drops their payment precipitously (which is impossible if all profits generated by the team are shared by the team), so the agents will not shirk. Of course, when there is uncertainty about how team effort transforms into outcomes, this harsh penalty will not work, and hence incentive problems may require team sizes to be smaller than that which is first-best efficient.And when thinking about agency costs and innovation Bryan writes,
A third justification for simple contracts is career concerns: agents work hard today to try to signal to the market that they are high-quality, and do so even if they are paid a fixed wage. This argument had been made less formally by 2013 Nobel laureate Eugene Fama, but Holmström in a 1982 working paper (finally published in 1999) showed that this concern about the market only completely mitigates moral hazard if outcomes within a firm are fully observable to the market, or the future is not discounted at all, or there is no uncertainty about agent’s abilities. Indeed, career concerns can make effort provision worse; for example, agents may take actions to signal quality to the market that are negative for their current firm.
A final explanation for simple contracts comes from Holmström’s 1987 paper with Milgrom. They argue that simple ‘linear’ contracts, with a wage and a bonus based linearly on output, are more ‘robust’ methods of solving moral hazard because they are less susceptible to manipulation by agents when the environment is not perfectly known. [...]
These ideas are reasonably intuitive, but the way Holmström answered them is not. Think about how an economist before the 1970s, like Adam Smith in his famous discussion of the inefficiency of sharecropping, might have dealt with these problems. These economists had few tools to deal with asymmetric information, so although economists like George Stigler (1961) analysed the economic value of information, the question of how to elicit information useful to a contract could not be discussed in any systematic way.
These economists would also have been burdened by the fact that the number of contracts one could write are infinite. So beyond saying that under a contract of type X does not equate marginal cost to marginal revenue, the question of which ‘second-best’ contract is optimal is extraordinarily difficult to answer in the absence of beautiful tricks like the revelation principle, partially developed by Holmström himself.
Holmström’s work is brilliant in how it clarifies many puzzles that are tricky to understand without thinking about incentives within a firm. For example, why would a risk-neutral firm not work enough on high-variance moonshot-type R&D projects? This is a question Holmström asks in his 1989 paper. Four reasons:
- First, in Holmström and Milgrom’s 1987 linear contracts paper, optimal risk-sharing leads to more distortion by agents the riskier the project being incentivised, so firms may choose lower expected value projects even if they themselves are risk-neutral.
- Second, firms build reputation in capital markets just as workers do with career concerns, and high-variance output projects are more costly in terms of the future value of that reputation when the interest rate on capital is lower (for example, when firms are large and old).
- Third, when R&D workers can potentially pursue many different projects, multitasking suggests that workers should be given small and very specific tasks so as to lessen the potential for bonus payments to shift worker effort across projects. Smaller firms with fewer resources may naturally have limits on the types of research a worker could pursue, which surprisingly makes it easier to provide strong incentives for research effort on the remaining possible projects.
- Fourth, multitasking suggests that agent’s tasks should be limited, and that high-variance tasks should be assigned to the same agent, which provides a role for decentralising research into large firms providing incremental, safe research, and small firms performing high-variance research. A deep understanding of how these types of internal incentives aggregate into explanations for why firms appear the way they do can best be achieved by a thorough reading of Holmström and Milgrom’s beautiful 1987 paper, “The Firm as an Incentive System”.
Wednesday, 12 October 2016
Hart and/or Holmstrom and the theory of the firm.
One part of the contribution of Hart and Holmstrom that has not been emphasised in other coverage of the Nobel award is their huge contribution to the theory of the firm, both separately and together. I wish to briefly cover this area. In doing so I draw on material from The Theory of the Firm: An Overview of the Economic Mainstream. See also Walker (2015).
One grouping in the literature on the firm is referred to as the ‘firm as a solution to moral hazard in teams approach’. This is an area where Holmstrom has been active. In a paper, Holmstrom (1982), he looks looks at the incentive problems of teams and identifies possible solutions. Holmstrom assumes that the members of the team each take actions which are unobservable to the monitor but the overall result of the combined actions is observable. What Holmstrom shows is that it is only under very restrictive assumptions that the monitor can ensure that efficient effort levels will be provided by each team member. The monitor can do this by designing a sophisticated incentive scheme. But Holmstrom shows that given unobservable effort levels, the requirements for the incentive scheme to be a Nash equilibrium, to balance the budget and to be Pareto optimal, cannot be met together. More specifically, a budget-balancing incentive scheme cannot reconcile the Nash equilibrium requirement and Pareto optimality because each team member will equalise the costs and benefits of extra effort; that is, if the team revenue is increased by the efforts of a single member, then that member should receive that revenue to ensure that they are properly motivated. But as the monitor only knows that team revenue has increased and not the effort levels of each individual member, then all members of the team would have to each receive the extra revenue to ensure that the hardworking member is rewarded for his or her efforts. But this will, obviously,violate the balanced budget condition. This suggests that there is an advantage, in terms of incentives, in the team not having to balance their budget.
Another group in the literature is the ‘firms as an incentive system view’. This approach to the theory of the firm was developed by Holmstrom and Milgrom (1991, 1994), Holmstrom and Tirole (1991) and Holmstrom (1999). In their view the firm is characterised by a number of factors: (1) the employees do not own the non-human assets of the firm; (2) the employees are subject to a ‘low-powered incentive scheme’; and (3) the employer has authority over the employee.
Holmstrom and Milgrom (1991) make two observations. First, they note that there are a number of ways that an employee can spend their time, many of which can be of value to an employer. But if these multiple activities compete for the worker’s attention, then the incentives offered for each of the activities must be comparable. Otherwise, the employee will put most effort into those things that are most well compensated and put less effort into the others activities. The second observation relates to the provision of strong incentives to a risk-averse employee. Providing strong financial incentives is costly because it loads extra risk into the worker’s pay. In addition, the cost is greater the more difficult it is to measure performance. This means that, other things being equal, tasks where performance is hard to measure should not be given as intense incentives as ones that are more accurately observed. But having low-powered incentives means that the employer needs to be able to exercise authority over the use of the employee’s time, since the employee will not have the proper incentives to be productive.
This logic suggests that, conversely, an independent contractor should face the opposite combination of instruments. The choice between having an employee or using an independent contractor depends on the ability of the principal to measure each dimension of the agent’s contribution. Thus, in the Holmstrom and Milgrom approach, measurability of performance is one important determinant of the boundaries of the firm. In addition their approach incorporates the importance of the allocation of property rights to the physical assets in determining incentives via determination of bargaining positions as is the case with the Williamson and Grossman-Hart-Moore approaches.
Hart enters our story with the 'firm as an ownership unit' approach to the firm. This approach is more commonly called the property rights theory or incomplete contracts theory of the firm. Early contributions to this approach include Grossman and Hart (1986,1987), Hart and Moore (1990) and Hart (1995).
The central idea in the property rights approach is that as contracts are incomplete the allocation of control rights affects the incentives that people face, and thus their behaviour and the allocation of resources.This theory defines ownership of an asset as the possession of the residual control rights
over that asset. A firm is defined as a collection of jointly-owned (non-human) assets. This means, for example, that the distinction between an independent contractor and an employee turns on who owns the non-human assets with which the agent works. An independent contractor owns his own 'tools' while an employee does not.
But, how and why does ownership matter? The answer is that in a world of incomplete contracts, ownership (i.e. having residual control rights) can serve as a source of power. Given that incomplete contracts contain gaps (or ambiguities) the question arises of who gets to make decisions in these non-contracted for situations? For the property rights theory, it is the owner. This matters since if there are two separate firms, A and B say, then the management of each firm can make decisions for their firm in the uncontracted for situations. If, on the other hand, A was to take over B then A’s management could make decisions for both A and B in the uncontracted for cases. To see the implications of this imagine that B supplies A with an input for A’s production process. If A and B are separate firms then B’s management could threaten to withdraw both its assets and its own labour if the firms cannot, say, agree on the terms for an increase in the supply of the input. If A owns B then B can only threaten to withhold its labour. The latter threat is normally weaker than the former. Such differences in power will effect the distribution of surplus generated by the relationship between A and B. If the firms are separate, then A may have to pay a lot to induce B to supply the increased level of inputs whereas if A owns B, then it can enforce the supply at a much lower cost since B’s management has a reduced threat, and thus, bargaining power.
Determining the boundaries of the firm requires us to balance the advantages of integration against its disadvantages. The benefit of integration is that the acquiring firm’s, A above, incentives to make relationship-specific investments is stronger because it now has greater residual control rights and thus can command a larger share of the ex post surplus created by such investments.The disadvantage of integration is that the incentives of the acquired firm, B, to make relationship-specific investments is reduced since they now have fewer residual controls rights and thus are able to capture less of the ex post surplus that their investments creates. To put this in employee/independent contractor terms, the optimal size of a firm trades off the fact that hiring an employee means hiring someone who lacks optimal incentives since they risk being held up by the firm because they can be fired, and thereby separated from the assets they need to be productive, versus using an independent contractor who could hold-up the firm by threatening to quit the relationship and taking his assets with him.
An implication of this is that if a non-contractible, specific to a particular set of assets, investment is undertaken then a non-owner risks being held-up by the owner. Thus,the property rights theory would say that whoever makes the most important, non-contractible,asset-specific investment should be the owner of the asset.and a larger share of the ex post surplus created by such investments.
Another approach in which Hart is a major contributor is the 'the reference point approach' to the firm.
Hart (2008) presents a simple reference points model and applies it to the theory of the firm. What is presented below is a slightly modified version of Hart's model. I hope that you can tell the difference between the zero's and the theta's below. Hart assumes that a seller, S, can provide a good, costing 10, to a buyer, B, who is willing to pay 20. Let us assume that we are talking about a public lecture on some aspect of microeconomics which B is organising and which B wants S to give. A successful lecture is worth 20 to B and it costs 10 for S to give the lecture.
At this stage Hart ignores the fact that B could engage other economists or that S could give lectures elsewhere. While trade could proceed smoothly, it is also possible that it will not. We will assume that B and S each have some discretion over the ‘quality’ of performance. For example, S could give a witty, lively, entertaining lecture or a very boring one. B on the other hand could treat S well, give her a nice dinner and pay quickly, or treat her badly.
In the language of Hart and Moore (2008) each party is able to provide basic (perfunctory) or exemplary (consummate) performance. It is further assumed that only the basic (perfunctory) level of performance can be legally enforced: exemplary (consummate) performance is entirely discretionary. It is assumed that each party is more or less indifferent between providing each level of performance − exemplary performance costs only a little more than basic or may even be slightly more pleasurable − and will provide exemplary performance if they feel they are being 'well treated' but not if they feel they are being 'badly treated'. Cutting back on exemplary performance is called ‘shading’. Such behaviour cannot be observed or punished by an outsider. Shading hurts the other party.
Hart emphasises that each party will feel 'well treated' if they receive what they think they are entitled to; that a contract between the parties is a reference point for perceived entitlements; and that should there be no reference point, then entitlements can diverge, wildly in some cases.
To return to the example above. First, we will add a time line. The time line tells us that B and S will write a contract some months before the lecture is given, at date 0, rather than at the last minute, date 1. One reason for this is that each party has more options earlier on. In fact it is assumed that there is a competitive market for sellers, at date 0.
Assume, further, that although B and S sign a contract at date 0, they leave the question of how much B will pay S open until the night before the lecture, date 1. This may seem a bad idea, and later it will be shown that it is. If no price is specified, then any p between 10 and 20 is possible. What might each party feel entitled to?
Hart and Moore (2008) take the view that entitlements can diverge. S may feel that the whole success of the talk will be due to her giving it and thus she feels entitled to p = 20. On the other hand B may have a somewhat different view of S’s abilities and likely contribution and thus feel that S is worth much less, say, p = 10.
Even though they disagree as to what p should be, they are rational enough to arrive at a compromise, say p = 15. According to Hart and Moore (2008) each party will feel short-changed and therefore aggrieved. Since B is aggrieved by 5, (15-10), B shades to the point where S’s payoff falls by 5θ, where θ is the constant of proportionality. And since S is also aggrieved by 5, (20-15), S shades to the point where the payoff for B falls by 5θ.
The end result of this is that if S and B leave the determination of the price until the night before the lecture, there will be a deadweight loss of 10θ due to the shading activities of each party. This reduces the value of the relationship between S and B from 10 to 10−10θ = 10(1−θ).
Next Hart asks the question: Can anything be done to avoid this deadweight loss? His answer is yes. But first note an answer that doesn’t do the job. Ex post Coasian bargaining at date 1 doesn’t work. The reason is that shading is not contractible and thus an agreement not to shade is not enforceable. Or to put this another way, if B offers to pay S more to reduce her shading, say B offers to pay p = 16 to S rather than 15, then this will indeed reduce S’s shading, from 5θ to 4θ, since S will now feel less aggrieved, but it will also increase B’s shading from 5θ to 6θ, since he now feels more aggrieved. Total deadweight loss does not change, it remains at 10θ. However there is a simple solution; the parties just put the price in the contract at date 0. Since it has been assumed that the market for lectures is competitive at date 0, B will be able to hire S for a price p = 10. With this price specified in the contract, there is nothing for B and S to disagree about at date 1. The fact that B and S may disagree about the contribution that S makes to the success of the lecture no longer matters. B and S have agreed that B will pay S 10, and neither B nor S will be disappointed or aggrieved when that happens. Importantly, agreeing in advance, at date 0, to a payment of 10 eliminates ex post argument and aggrievement, and thus both parties will be willing to provide exemplary performance. Here we have the first best being achieved and zero deadweight losses as a result. This does raise an obvious question: What changes between dates 0 and 1? Why does a date 0 contract that fixes p avoid aggrievement, whereas a date 1 contract that fixes p does not? The crucial point here is the role of the ex ante market at date 0. This market gives an objective measure of what B and S bring to the relationship. Given the assumption of a competitive date 0 market, there are many sellers willing to supply at p = 10 and thus S accepts that she cannot expect to receive more than 10, while B understands that he can’t expect to pay less, as no one would be willing to give the lecture for less. Thus, neither party is aggrieved by p = 10. This gives us a model of the contractual relationship between B and S, but, as Hart explains, we need one further ingredient to create a theory of the firm.
Now let us add a little more realism by assuming that not all of the details of the lecture can be anticipated at date 0. To keep things simple we will assume that two different lectures can be given.
Lecture 1 - say, a theory of the firm lecture - is the same as above, with a value of 20 and costs of 10. Lecture 2 - say, a microeconometrics lecture - yields value of 14 and costs of 8. Note that lecture 1 is more efficient in that it generates a greater surplus - 10 v's 6. Assume that the lectures can not be specified in the date 0 contract, since thinking about econometrics is sooooo boring that no one can stay awake long enough to write the contract! At date 1, however, the choice between them becomes clear.
Now we have to compare two organisational forms: an employment contract and an independent contractor. First, let B and S fix the price of the good at date 0, at say 10, and let B and S agree at date 0 that S will be an independent contractor. This is, in other words, a market exchange between two separate economics agents. Independent contractor means here that S gets to pick which lecture to give.
Hart then asks, What will S do? Given that the price has been fixed by the date 0 contract, S will pick lecture 2, since it is cheaper for her. But note this is inefficient. B will be aggrieved because S didn’t choose lecture 1, which B feels entitled to; B is short-changed by 6 (20-14), and he will therefore shade enough to reduce S’s payoff by 6θ. Total surplus in this case will be 6 − 6θ.
The second organisation form to be considered is an employment contract. B and S agree at date 0 that S is an employee of B. This we take to mean that S will work for B at a fixed wage, again assume 10. B, being the employer, has the right to decide on which lecture is to be given. As the wage is fixed B will choose lecture 1, as this gives him the greater value. This is efficient. S will be aggrieved since lecture 2 wan’t chosen, but S’s aggrievement is only 2. This induces S to shade by enough to reduce B’s payoff by 2θ. Total surplus is therefore 10 − 2θ.
Under the conditions specified, the employment contract is better. This is true for two related reasons. First, the lecture matters more to B than to S. B will lose 20 − 14 = 6 if his favoured lecture is not chosen while S only loses 10 − 8 = 2 if her favoured lecture is not chosen. This means it is efficient for B to choose the lecture. Second, and related, S’s aggrievement is low since she doesn’t care very much.
Hart now changes the numbers. Keep lecture 1 as it is, but change lecture 2 so while it still yields 14, it now costs only 2. Lecture 2 is now the more efficient (12 v’s 10). Under employment, lecture 1 will be chosen, yielding a total surplus of 10−8θ. If S is an independent contractor, lecture 2 will be chosen resulting in a total surplus of 12 − 6θ.
What this suggests is that employment is good if the lecture matters more to B than to S, while independent contracting is good if the lecture matters more to S than to B.
Hart goes on to say,
The basic ingredients of the reference point approach to contracts have been applied to the theory of the firm more fully in papers by Hart and Moore (2007), Hart (2009) and Hart and Holmstrom (2010) - yes they get together!. For more on these papers see Walker (2013).
Refs.:
One grouping in the literature on the firm is referred to as the ‘firm as a solution to moral hazard in teams approach’. This is an area where Holmstrom has been active. In a paper, Holmstrom (1982), he looks looks at the incentive problems of teams and identifies possible solutions. Holmstrom assumes that the members of the team each take actions which are unobservable to the monitor but the overall result of the combined actions is observable. What Holmstrom shows is that it is only under very restrictive assumptions that the monitor can ensure that efficient effort levels will be provided by each team member. The monitor can do this by designing a sophisticated incentive scheme. But Holmstrom shows that given unobservable effort levels, the requirements for the incentive scheme to be a Nash equilibrium, to balance the budget and to be Pareto optimal, cannot be met together. More specifically, a budget-balancing incentive scheme cannot reconcile the Nash equilibrium requirement and Pareto optimality because each team member will equalise the costs and benefits of extra effort; that is, if the team revenue is increased by the efforts of a single member, then that member should receive that revenue to ensure that they are properly motivated. But as the monitor only knows that team revenue has increased and not the effort levels of each individual member, then all members of the team would have to each receive the extra revenue to ensure that the hardworking member is rewarded for his or her efforts. But this will, obviously,violate the balanced budget condition. This suggests that there is an advantage, in terms of incentives, in the team not having to balance their budget.
Another group in the literature is the ‘firms as an incentive system view’. This approach to the theory of the firm was developed by Holmstrom and Milgrom (1991, 1994), Holmstrom and Tirole (1991) and Holmstrom (1999). In their view the firm is characterised by a number of factors: (1) the employees do not own the non-human assets of the firm; (2) the employees are subject to a ‘low-powered incentive scheme’; and (3) the employer has authority over the employee.
Holmstrom and Milgrom (1991) make two observations. First, they note that there are a number of ways that an employee can spend their time, many of which can be of value to an employer. But if these multiple activities compete for the worker’s attention, then the incentives offered for each of the activities must be comparable. Otherwise, the employee will put most effort into those things that are most well compensated and put less effort into the others activities. The second observation relates to the provision of strong incentives to a risk-averse employee. Providing strong financial incentives is costly because it loads extra risk into the worker’s pay. In addition, the cost is greater the more difficult it is to measure performance. This means that, other things being equal, tasks where performance is hard to measure should not be given as intense incentives as ones that are more accurately observed. But having low-powered incentives means that the employer needs to be able to exercise authority over the use of the employee’s time, since the employee will not have the proper incentives to be productive.
This logic suggests that, conversely, an independent contractor should face the opposite combination of instruments. The choice between having an employee or using an independent contractor depends on the ability of the principal to measure each dimension of the agent’s contribution. Thus, in the Holmstrom and Milgrom approach, measurability of performance is one important determinant of the boundaries of the firm. In addition their approach incorporates the importance of the allocation of property rights to the physical assets in determining incentives via determination of bargaining positions as is the case with the Williamson and Grossman-Hart-Moore approaches.
Hart enters our story with the 'firm as an ownership unit' approach to the firm. This approach is more commonly called the property rights theory or incomplete contracts theory of the firm. Early contributions to this approach include Grossman and Hart (1986,1987), Hart and Moore (1990) and Hart (1995).
The central idea in the property rights approach is that as contracts are incomplete the allocation of control rights affects the incentives that people face, and thus their behaviour and the allocation of resources.This theory defines ownership of an asset as the possession of the residual control rights
over that asset. A firm is defined as a collection of jointly-owned (non-human) assets. This means, for example, that the distinction between an independent contractor and an employee turns on who owns the non-human assets with which the agent works. An independent contractor owns his own 'tools' while an employee does not.
But, how and why does ownership matter? The answer is that in a world of incomplete contracts, ownership (i.e. having residual control rights) can serve as a source of power. Given that incomplete contracts contain gaps (or ambiguities) the question arises of who gets to make decisions in these non-contracted for situations? For the property rights theory, it is the owner. This matters since if there are two separate firms, A and B say, then the management of each firm can make decisions for their firm in the uncontracted for situations. If, on the other hand, A was to take over B then A’s management could make decisions for both A and B in the uncontracted for cases. To see the implications of this imagine that B supplies A with an input for A’s production process. If A and B are separate firms then B’s management could threaten to withdraw both its assets and its own labour if the firms cannot, say, agree on the terms for an increase in the supply of the input. If A owns B then B can only threaten to withhold its labour. The latter threat is normally weaker than the former. Such differences in power will effect the distribution of surplus generated by the relationship between A and B. If the firms are separate, then A may have to pay a lot to induce B to supply the increased level of inputs whereas if A owns B, then it can enforce the supply at a much lower cost since B’s management has a reduced threat, and thus, bargaining power.
Determining the boundaries of the firm requires us to balance the advantages of integration against its disadvantages. The benefit of integration is that the acquiring firm’s, A above, incentives to make relationship-specific investments is stronger because it now has greater residual control rights and thus can command a larger share of the ex post surplus created by such investments.The disadvantage of integration is that the incentives of the acquired firm, B, to make relationship-specific investments is reduced since they now have fewer residual controls rights and thus are able to capture less of the ex post surplus that their investments creates. To put this in employee/independent contractor terms, the optimal size of a firm trades off the fact that hiring an employee means hiring someone who lacks optimal incentives since they risk being held up by the firm because they can be fired, and thereby separated from the assets they need to be productive, versus using an independent contractor who could hold-up the firm by threatening to quit the relationship and taking his assets with him.
An implication of this is that if a non-contractible, specific to a particular set of assets, investment is undertaken then a non-owner risks being held-up by the owner. Thus,the property rights theory would say that whoever makes the most important, non-contractible,asset-specific investment should be the owner of the asset.and a larger share of the ex post surplus created by such investments.
Another approach in which Hart is a major contributor is the 'the reference point approach' to the firm.
Hart (2008) presents a simple reference points model and applies it to the theory of the firm. What is presented below is a slightly modified version of Hart's model. I hope that you can tell the difference between the zero's and the theta's below. Hart assumes that a seller, S, can provide a good, costing 10, to a buyer, B, who is willing to pay 20. Let us assume that we are talking about a public lecture on some aspect of microeconomics which B is organising and which B wants S to give. A successful lecture is worth 20 to B and it costs 10 for S to give the lecture.
At this stage Hart ignores the fact that B could engage other economists or that S could give lectures elsewhere. While trade could proceed smoothly, it is also possible that it will not. We will assume that B and S each have some discretion over the ‘quality’ of performance. For example, S could give a witty, lively, entertaining lecture or a very boring one. B on the other hand could treat S well, give her a nice dinner and pay quickly, or treat her badly.
In the language of Hart and Moore (2008) each party is able to provide basic (perfunctory) or exemplary (consummate) performance. It is further assumed that only the basic (perfunctory) level of performance can be legally enforced: exemplary (consummate) performance is entirely discretionary. It is assumed that each party is more or less indifferent between providing each level of performance − exemplary performance costs only a little more than basic or may even be slightly more pleasurable − and will provide exemplary performance if they feel they are being 'well treated' but not if they feel they are being 'badly treated'. Cutting back on exemplary performance is called ‘shading’. Such behaviour cannot be observed or punished by an outsider. Shading hurts the other party.
Hart emphasises that each party will feel 'well treated' if they receive what they think they are entitled to; that a contract between the parties is a reference point for perceived entitlements; and that should there be no reference point, then entitlements can diverge, wildly in some cases.
To return to the example above. First, we will add a time line. The time line tells us that B and S will write a contract some months before the lecture is given, at date 0, rather than at the last minute, date 1. One reason for this is that each party has more options earlier on. In fact it is assumed that there is a competitive market for sellers, at date 0.
Assume, further, that although B and S sign a contract at date 0, they leave the question of how much B will pay S open until the night before the lecture, date 1. This may seem a bad idea, and later it will be shown that it is. If no price is specified, then any p between 10 and 20 is possible. What might each party feel entitled to?
Hart and Moore (2008) take the view that entitlements can diverge. S may feel that the whole success of the talk will be due to her giving it and thus she feels entitled to p = 20. On the other hand B may have a somewhat different view of S’s abilities and likely contribution and thus feel that S is worth much less, say, p = 10.
Even though they disagree as to what p should be, they are rational enough to arrive at a compromise, say p = 15. According to Hart and Moore (2008) each party will feel short-changed and therefore aggrieved. Since B is aggrieved by 5, (15-10), B shades to the point where S’s payoff falls by 5θ, where θ is the constant of proportionality. And since S is also aggrieved by 5, (20-15), S shades to the point where the payoff for B falls by 5θ.
The end result of this is that if S and B leave the determination of the price until the night before the lecture, there will be a deadweight loss of 10θ due to the shading activities of each party. This reduces the value of the relationship between S and B from 10 to 10−10θ = 10(1−θ).
Next Hart asks the question: Can anything be done to avoid this deadweight loss? His answer is yes. But first note an answer that doesn’t do the job. Ex post Coasian bargaining at date 1 doesn’t work. The reason is that shading is not contractible and thus an agreement not to shade is not enforceable. Or to put this another way, if B offers to pay S more to reduce her shading, say B offers to pay p = 16 to S rather than 15, then this will indeed reduce S’s shading, from 5θ to 4θ, since S will now feel less aggrieved, but it will also increase B’s shading from 5θ to 6θ, since he now feels more aggrieved. Total deadweight loss does not change, it remains at 10θ. However there is a simple solution; the parties just put the price in the contract at date 0. Since it has been assumed that the market for lectures is competitive at date 0, B will be able to hire S for a price p = 10. With this price specified in the contract, there is nothing for B and S to disagree about at date 1. The fact that B and S may disagree about the contribution that S makes to the success of the lecture no longer matters. B and S have agreed that B will pay S 10, and neither B nor S will be disappointed or aggrieved when that happens. Importantly, agreeing in advance, at date 0, to a payment of 10 eliminates ex post argument and aggrievement, and thus both parties will be willing to provide exemplary performance. Here we have the first best being achieved and zero deadweight losses as a result. This does raise an obvious question: What changes between dates 0 and 1? Why does a date 0 contract that fixes p avoid aggrievement, whereas a date 1 contract that fixes p does not? The crucial point here is the role of the ex ante market at date 0. This market gives an objective measure of what B and S bring to the relationship. Given the assumption of a competitive date 0 market, there are many sellers willing to supply at p = 10 and thus S accepts that she cannot expect to receive more than 10, while B understands that he can’t expect to pay less, as no one would be willing to give the lecture for less. Thus, neither party is aggrieved by p = 10. This gives us a model of the contractual relationship between B and S, but, as Hart explains, we need one further ingredient to create a theory of the firm.
Now let us add a little more realism by assuming that not all of the details of the lecture can be anticipated at date 0. To keep things simple we will assume that two different lectures can be given.
Lecture 1 - say, a theory of the firm lecture - is the same as above, with a value of 20 and costs of 10. Lecture 2 - say, a microeconometrics lecture - yields value of 14 and costs of 8. Note that lecture 1 is more efficient in that it generates a greater surplus - 10 v's 6. Assume that the lectures can not be specified in the date 0 contract, since thinking about econometrics is sooooo boring that no one can stay awake long enough to write the contract! At date 1, however, the choice between them becomes clear.
Now we have to compare two organisational forms: an employment contract and an independent contractor. First, let B and S fix the price of the good at date 0, at say 10, and let B and S agree at date 0 that S will be an independent contractor. This is, in other words, a market exchange between two separate economics agents. Independent contractor means here that S gets to pick which lecture to give.
Hart then asks, What will S do? Given that the price has been fixed by the date 0 contract, S will pick lecture 2, since it is cheaper for her. But note this is inefficient. B will be aggrieved because S didn’t choose lecture 1, which B feels entitled to; B is short-changed by 6 (20-14), and he will therefore shade enough to reduce S’s payoff by 6θ. Total surplus in this case will be 6 − 6θ.
The second organisation form to be considered is an employment contract. B and S agree at date 0 that S is an employee of B. This we take to mean that S will work for B at a fixed wage, again assume 10. B, being the employer, has the right to decide on which lecture is to be given. As the wage is fixed B will choose lecture 1, as this gives him the greater value. This is efficient. S will be aggrieved since lecture 2 wan’t chosen, but S’s aggrievement is only 2. This induces S to shade by enough to reduce B’s payoff by 2θ. Total surplus is therefore 10 − 2θ.
Under the conditions specified, the employment contract is better. This is true for two related reasons. First, the lecture matters more to B than to S. B will lose 20 − 14 = 6 if his favoured lecture is not chosen while S only loses 10 − 8 = 2 if her favoured lecture is not chosen. This means it is efficient for B to choose the lecture. Second, and related, S’s aggrievement is low since she doesn’t care very much.
Hart now changes the numbers. Keep lecture 1 as it is, but change lecture 2 so while it still yields 14, it now costs only 2. Lecture 2 is now the more efficient (12 v’s 10). Under employment, lecture 1 will be chosen, yielding a total surplus of 10−8θ. If S is an independent contractor, lecture 2 will be chosen resulting in a total surplus of 12 − 6θ.
What this suggests is that employment is good if the lecture matters more to B than to S, while independent contracting is good if the lecture matters more to S than to B.
Hart goes on to say,
“[o]ne point worth emphasizing is that in neither of the above examples is the following contract optimal: to leave the choice of price and method until date 1, i.e. to rely on unconstrained Coasian bargaining. This would always yield the efficient method, but the aggrievement costs would be high. In [Table not shown] the parties would agree on method 1; however, since there are 10 dollars of surplus to argue over, shading costs equal 10θ: net surplus = 10(1 − θ), which is less than that obtained under the employment contract. In [Table 171.1] there are 12 dollars of surplus to argue over and net surplus = 12(1 − θ), which is less than that obtained under independent contracting.” (Hart 2008: 409).Clearly the examples above are toy ones, but Hart argues they contain the basic ingredients of a theory of the firm in that they consider the choice between carrying out a transaction in the market, using an independent contractor, and 'inside the firm', via an employment contract. This was the trade-off at the heart of Coase (1937).
The basic ingredients of the reference point approach to contracts have been applied to the theory of the firm more fully in papers by Hart and Moore (2007), Hart (2009) and Hart and Holmstrom (2010) - yes they get together!. For more on these papers see Walker (2013).
Refs.:
- Coase, Ronald Harry (1937). ‘The Nature of the Firm’, Economica, n.s. 4(16) November: 386–405.
- 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.
- Grossman, Sanford J. and Oliver D. Hart (1987). ‘Vertical Integration and the Distribution of Property Rights’. In Assaf Razin and Efraim Sadka (eds.), Economic Policy in Theory and Practice (504–48), London: Macmillan Press.
- Hart, Oliver D. (1995). Firms, Contracts, and Financial Structure, Oxford: Oxford University Press.
- Hart, Oliver D. (2008). ‘Economica Coase Lecture: Reference Points and the Theory of the Firm’, Economica, 75(299) August: 404–11.
- Hart, Oliver D. (2009). ‘Hold-up, Asset Ownership, and Reference Points’, Quarterly Journal of Economics, 124(1) February: 267–300.
- Hart, Oliver D. and Bengt Holmström (2010). ‘A Theory of Firm Scope’, Quarterly Journal of Economics, 125(2) May: 483–513.
- Hart, Oliver D. and John Moore (1990). ‘Property Rights and the Nature of the Firm’, Journal of Political Economy, 98(6): 1119–58.
- Hart, Oliver D. and John Moore (2007). ‘Incomplete Contracts and Ownership: Some New Thoughts, American Economic Review, 97(2) May: 182–6.
- Hart, Oliver D. and John Moore (2008). ‘Contracts as Reference Points’ Quarterly Journal of Economics, 123(1) February: 1–48.
- Holmstrom, Bengt (1982). ‘Moral Hazard in Teams’, Bell Journal of Economics, 13(2) Autumn: 324–40.
- Holmstom, Bengt (1999). ‘The Firm as a Subeconomy’, Journal of Law, Economics, and Organization, 15(1) April: 74-102
- Holmstrom, Bengt and Paul Milgrom (1991). ‘Multitask Principal-Agent Analyses: Incentive Contracts, Asset Ownership, and Job Design’, Journal of Law, Economics, & Organization, 7(Special Issue): 24–52.
- Holmstrom, Bengt and Paul Milgrom (1994). ‘The Firm as an Incentive System’, American Economic Review, 84(4) September: 972–91.
- Holmstrom, Bengt and Jean Tirole (1991). ‘Transfer Pricing and Organizational Form’, Journal of Law, Economics, and Organization, 7(2) Fall: 201-28.
- Walker, Paul (2013). ‘The ‘Reference Point’ Approach to the Theory of the Firm: An Introduction’, Journal of Economic Surveys, 27(4) September: 670–95.
- Walker, Paul (2015). 'Contracts, Entrepreneurs, Market Creation and Judgement: The Contemporary Mainstream Theory of the Firm in Perspective'. Journal of Economic Surveys, v29 no2, April: 317-38
Monday, 10 October 2016
2016 Nobel Prize in economics
Press Release: The long and the short of contractsPopular Science Background
Modern economies are held together by innumerable contracts. The new theoretical tools created by Hart and Holmström are valuable to the understanding of real-life contracts and institutions, as well as potential pitfalls in contract design.
Society’s many contractual relationships include those between shareholders and top executive management, an insurance company and car owners, or a public authority and its suppliers. As such relationships typically entail conflicts of interest, contracts must be properly designed to ensure that the parties take mutually beneficial decisions. This year’s laureates have developed contract theory, a comprehensive framework for analysing many diverse issues in contractual design, like performance-based pay for top executives, deductibles and co-pays in insurance, and the privatisation of public-sector activities.
In the late 1970s, Bengt Holmström demonstrated how a principal (e.g., a company’s shareholders) should design an optimal contract for an agent (the company’s CEO), whose action is partly unobserved by the principal. Holmström’s informativeness principle stated precisely how this contract should link the agent’s pay to performance-relevant information. Using the basic principal-agent model, he showed how the optimal contract carefully weighs risks against incentives. In later work, Holmström generalised these results to more realistic settings, namely: when employees are not only rewarded with pay, but also with potential promotion; when agents expend effort on many tasks, while principals observe only some dimensions of performance; and when individual members of a team can free-ride on the efforts of others.
In the mid-1980s, Oliver Hart made fundamental contributions to a new branch of contract theory that deals with the important case of incomplete contracts. Because it is impossible for a contract to specify every eventuality, this branch of the theory spells out optimal allocations of control rights: which party to the contract should be entitled to make decisions in which circumstances? Hart’s findings on incomplete contracts have shed new light on the ownership and control of businesses and have had a vast impact on several fields of economics, as well as political science and law. His research provides us with new theoretical tools for studying questions such as which kinds of companies should merge, the proper mix of debt and equity financing, and when institutions such as schools or prisons ought to be privately or publicly owned.
Through their initial contributions, Hart and Holmström launched contract theory as a fertile field of basic research. Over the last few decades, they have also explored many of its applications. Their analysis of optimal contractual arrangements lays an intellectual foundation for designing policies and institutions in many areas, from bankruptcy legislation to political constitutions.
Scientific Background.
A great award, contract theory is important in so many areas of economics now. For me, of course, its importance is in the theory of the firm. Both the work of Hart and Holmstrom has played a major role in the modern theory of the firm. In addition to the Ronald Coase (Nobel 1991) and Oliver Williamson (Nobel 2009) these guys are two of the biggest players in the field. Holmstrom's work has been in the area of the moral hazard in teams approach to the firm while Hart was one of the major developers of the incomplete contracts (or property rights) and reference point approaches to the firm.
Subscribe to:
Posts (Atom)