Abstract
What distinguishes a market economy is that a producer produces goods not directly for his own consumption but for markets through which he sells outputs and buys in inputs. Accordingly, his income and hence his utility do not just depend upon how much he has produced from given inputs but also upon how much he can charge for his outputs and how much he has paid for the inputs. What concerns him is the dot of the price vector and product vector (therefore the net return) rather than the product vector itself.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Huang (1973) distinguishes between risks associated with decision making and exogenous or natural risks beyond the party’s control. To my understanding, Huang’s distinction coincides with Knight’s (1921) classical interpretation of the distinction between risk and uncertainty. Roughly speaking, “exogenous or natural risk” in Huang is “risk” in Knight; “risk associated with decision making” in Huang is “uncertainty” in Knight. In the present thesis, for simplicity, we call the first type risk “natural risk” and the second “business risk”.
- 2.
Robinson Crusoe was very disappointed when he found the boat which he had spent four years in building couldn’t be moved into the water. However, if he were a businessman, he would have gone bankrupt rather than just be disappointed. Karl Marx referred to the market as a “thrilling jump”, failure in which would destroy not just the product but also its producer himself.
- 3.
For more discussion about entrepreneurial qualities, see Casson (1982), Chap. 2.
- 4.
Yang and Ng (1995) assume that individuals are ex ante identical in their marketing ability. In their model, the firm exists because of economy of specialization and transaction costs.
- 5.
It should be noted that although we borrow this term from Alchian and Demsetz (1972), what we emphasize here is that the marginal contribution of producing (marketing) member’s effort depends on marketing (producing) member’s effort, rather than worker A’s contribution depends on worker B’s effort.
- 6.
In Alchian and Demsetz (1972), only the incentive problem has been identified. In their model, although the monitor claims the residual, the returns to monitoring are certain for given monitoring effort. Here, following Knight (1921), we relate the incentive problem to the risk problem.
- 7.
We use “incentive costs” instead of “agency costs”. In the thesis, the agency costs are defined as the sum of risk costs and incentive costs.
- 8.
The prisoners’ dilemma problem among the producing members (workers) in monitoring may enable a single marketing member to have advantages in competing for principalship. There are two reasons why we do not emphasize this argument. First, our proposition must hold even if the firm consists of only one producing member and one marketing member. Second, this argument implies that principalship held by the marketing member is “deepening” as the number of producing members increases. We have no evidence to support this prediction.
- 9.
An alternative is first to show why capitalists should hold the principalship, given that the marketing party is entitled to monitor the producing party, and then to demonstrate why the marketing party rather than the producing party should be the monitor. We prefer our approach because it is more logical as well as historical.
- 10.
The main advantage of this so-called parameterized distribution formulation is to allow us to capture Frank Knight’s uncertainty which says that business risks are dominated by actions.
- 11.
In a simple agency model, it is always possible for the agent to take full responsibility for his own actions by letting him be the residual claimant, because the distribution function of the outcome is only conditional on the agent’s actions. The problem is that such a full responsibility system cannot be efficient if the agent is strictly risk-averse (regardless of the principal’s risk attitude).
- 12.
Here we follow Frank Knight who argued that “With human nature as we know it would be impraticable or very unusual for one man to guarantee to another a definite result of the latter’s actions without being given power to direct his work. And on the other hand the second party would not place himself under the direction of the first without such a guarantee.” (pp. 270).
- 13.
We will discuss how acquisition of authority of monitoring is constrained by acceptability for the other later.
- 14.
The assumption of linearity is made only for simplicity.
- 15.
The agent may play a monitoring-anti-monitoring game with the principal. For instance, if the principal checks the agent n times randomly a day, the agent may make use of an “spy-hole” so that he needs to work only when the principal is coming. Then the shirking time will certainly fall as n increases. (In the jargon of games, the agent’s strategic space for shirking shrinks as the principal’s monitoring effort increases.).
- 16.
Although positive effects of monitoring on work effort are quite intuitive and widely observed, theoretical views are far from unanimous. In Putterman and Skillman (1988), it is argued that the positive incentive effect of monitoring depends critically on the compensation scheme employed, the risk preferences of the agents, and the informational content of increased monitoring. In particular, they show that: when monitoring is understood to produce a noisy signal of working effort with a first- or second-order stochastic dominance condition, the positive effect cannot be guaranteed in general under either the share payment scheme or the wage payment scheme, and moreover under some reasonable assumptions of the risk-preferences, the effect is actually negative; on the other hand, the positive effect is most easily guaranteed (under both compensation schemes) if monitoring produces an accurate signal of working effort with a probability which depends on the level of monitoring intensity.
- 17.
Is such an assumption really acceptable in the sense of fairness? Obviously it would be irrational for j to accept monitoring by i which makes him worse-off. The problem is: why cannot j do better? The answer is the authority of monitoring is equally open to j with a symmetric compensation rule: j is free to monitor i as long as his monitoring would not make i worse-off. Under such a symmetric treatment, the condition seems fair.
- 18.
When \(w_j=w_j^s\), i would choose too much monitoring effort since the externality of monitoring cost is not fully internalized.
- 19.
We will discuss this point later. Briefly, the residual share effect implies that the principal can never fully internalize the benefit from his effort unless he is the only residual claimant.
- 20.
It is interesting to note that under Assumption 4, i’s monitoring actually improves j’s welfare (unless j takes no residual share) compared to the non-monitoring equilibrium (an equilibrium when monitoring is technically impossible). The arguments in Sect. 2.3 will show that Assumption 4 has implicitly incorporated a bargaining procedure into the contract.
- 21.
I believe this is one of the major reasons for the observation that the entrepreneur (or manager) have higher expected income than the worker.
- 22.
The number can be replaced by parameters. Here the numbers are chosen so as to make the expressions simple.
- 23.
One argument is that the optimal contract requires that the risk-neutral member becomes the principal and the risk-averse becomes an agent. But the most of literature simply assumes that the principal is risk-neutral.
- 24.
In Itoh (1991), imperfect substitution between own effort and helping effort is essential for team-work to be optimal in designing an incentive scheme.
- 25.
To avoid complexity of notation, in the following analysis except Sect. 2.5, we use Y instead of EY to denote the expected return.
- 26.
Strictly speaking, Leontief technology (\(\gamma =\infty \)) is pure teamwork.
- 27.
- 28.
Note an increase in the degree of teamwork has symmetric effects on both member’s incentive functions, while an increase in alpha has asymmetric effects.
- 29.
The objective function is derived from (2.9). \(\beta Y^s\) enters the function because of the effect on \(M^{\prime }\)s status quo of \( P^{\prime }\)s self-incentive effort.
- 30.
The argument cannot be extended to the case when the monitor takes no residual share at all. The reason is that in that case there is no channel through which the ‘monitor’ can capture any surplus generated by monitoring, unless the contract specifies the monitoring effort. But this case need not trouble us because the working member has already obtained a full incentive to work.
- 31.
In both Holmstrom and McAfee and McMillan, the principal is not a member of team. Under this assumption, Holmstrom argues that the principal’s primary role is to break the budget-balancing constraint so as to create group incentive to work, while McAfee and McMillan argue that the role of monitoring is to discipline the monitor himself instead of the team. Holmstrom has clearly realized that “it is important that the principal not provide any (unobservable) productive inputs or else a free-rider problem remains.” (pp. 328) We say our argument is coincident with theirs only in spirit because they do not explicitly model the choice of monitoring effort.
- 32.
It is widely known that the incentive problem depends on the degree to which the net surplus of a decision can be internalized by the decision-maker. A full internalization does not necessarily require a full residual claim; but if it does, the residual share is important.
- 33.
A mathematical problem with the mutual-monitoring is that the two compensation rules cannot be simultaneously binding. One way to deal with this problem is to assume the two members play a non-cooperative monitoring game; that is, each member chooses his own monitoring effort subject to the compensation rule for the other. Alternatively we can assume that the two members play a cooperative monitoring game; that is, they maximize the joint output net of the total effort costs. The reader can check to confirm the two methods give the same solution.
- 34.
The argument may not hold when both members are risk-averse. See Sect. 2.5.
- 35.
We use \(\gamma =0\) only for comparison. In fact, in the case of \(\gamma =0\), no firm exists.
- 36.
It is still possible that \(\beta _\Pi \) does not belong to the mutual-monitoring regime.
- 37.
When should a professor treat her research assistant as a co-author or only acknowledge him in a footnote? Observation is that when the research requires assistance from brains, the research assistant appears as a co-author; on the other hand, when the research requires assistance mainly from hands (collecting and calculating data), the research assistant will be “gratefully acknowledged”.
- 38.
The exponential form of utility function is one under which the value maximization principle (or the certain equivalent approach) is perfectly satisfactory. See Holmstrom and Milgrom (1991). For more discussion of this approach, see Milgrom and Roberts (1992), Chap. 7.
- 39.
This reflects a difference of mode of thinking between our model and the standard principal-agent model. In the standard principal-agent model, the “agent” shares the residual because the principal cannot observes his actions, while in our model, the “agent” shares the residual because the “principal” is too risk-averse.
- 40.
Another possibility is that an increase in work effort increase both the mean and the variance subject to the condition that the first-order stochastic dominance still holds.
- 41.
Like in case 1, mutual-monitoring may be efficient if the marketing member is sufficiently risk-averse.
- 42.
We choose this particular form for simplifying the calculation. \(V^0\) is chosen such that V will not become negative for the relevant range of effort. Note V is convex in \(a_M\).
- 43.
It should be pointed out that although our formal propositions are derived under some special technical assumptions about production function, preferences and monitoring technology, the basic arguments should hold under more general assumptions, because what is important is the extent to which external costs of a action are internalized, rather than concrete parameters.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2018 Truth and Wisdom Press and Springer Science+Business Media Singapore
About this chapter
Cite this chapter
Zhang, W. (2018). Marketing, Producing, Monitoring and the Assignment of Principalship. In: The Origin of the Capitalist Firm. Springer, Singapore. https://doi.org/10.1007/978-981-10-0221-2_2
Download citation
DOI: https://doi.org/10.1007/978-981-10-0221-2_2
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-0220-5
Online ISBN: 978-981-10-0221-2
eBook Packages: Business and ManagementBusiness and Management (R0)