Introduction
The theoretical basis of economics is based on perfections like perfect competition, perfect markets, zero transaction cost, perfect access to all the information, rational decisions, etc. This indicates an ideal situation.
However, the real world is full of imperfections. There are imperfect markets and people have incomplete information, there is a cost associated with each transaction and not all decisions are rational. Focusing on the economics of information, perfect information entirely removes uncertainty in decision-making.
Perfect information means that anything that may have a bearing on a buyer’s or seller’s decision-making process is known in advance. For instance, in a perfectly competitive market for used cars, buyers would have full information about the car and pay directly in relation to the expected quality of the used car.
But, in reality, information is less than perfect and whatever imperfect information is accessible tends to reduce but not remove the uncertainty in decisions. In this example, sellers would be better informed than buyers about the quality of the car.
This missing or imperfect piece of information may lead to market failure because the party with better or more information would have a competitive advantage.
In order to have access to information, people need to incur search and transaction costs which prove to be costly. The problems encountered due to asymmetric information could be understood in the Principal-agent framework. Also, there are many mechanisms which society can use to cope with such information asymmetries.
Market with Incomplete Information
In today’s world, all economic activities are taking place in an uncertain environment. There is risk and uncertainty involved in stock markets, exchange rate, profit rate, business environment etc.
The main reason for this uncertainty is access to limited and imperfect information with the economic decision-makers at the time of taking decisions. The extent of uncertainty, in turn, depends on the reliability of available information.
Information asymmetry occurs when one party in a transaction has less information than the other party. This affects the decisions and outcomes on the individual as well as aggregate level.
Information asymmetries may be due to hidden characteristics (Adverse selection) or due to hidden actions (Principal-agent problem). In the case of the former, one party to the transaction has better access to crucial information, which it uses to its private advantage, thereby harming the uninformed side of the market.
For instance, due to inefficiencies brought about by asymmetric information in insurance contracts, the resulting market outcome is distorted.
Similarly, in the case of used car markets, sellers are better informed about their car’s quality, and they decide whether or not to offer their cars for sale. At median prices, which uninformed buyers decide to pay (because of imperfect information about the car) only lower-quality cars or lemons dominate the market. This results in a rising proportion of lemons for sale.
In the case of hidden actions, due to a lack of perfect monitoring or due to divergence in the utilities of contracting parties, one party acts differently. This unobservable behaviour causes the problem of moral hazard. For example, when a person buys health insurance, he starts to take lesser care of his health as he knows that the insurance company will bear the costs in case he falls ill.
Similarly, in the Principal-agent problem, the principal’s and agent’s objective functions are divergent, and when an agent works on behalf of the principal, he tends to serve his own interests more by pursuing hidden actions.
In totality, all such instances result in inefficient outcomes in the market, where one party, which holds the private information, manipulate the situation to its own advantage.
This problem of asymmetric information can be analyzed in the principal-agent framework and the probable solutions can be extended by creating some form of incentives and contracts so that the altered behaviour of economic agents can be taken care of.
Search and Transaction Cost
Even though economic models presume perfect markets but imperfections are indispensable in the real world. Whenever information asymmetries prevail in the market, this leads to market failure such that an efficient outcome is not reached. To overcome such asymmetry of information, people tend to acquire the requisite information. This entails various types of costs, which are enormous at times.
All the relevant information which is required in any exchange is not easily and readily available. A person needs to search for information. All those costs which are incurred in searching for the requisite information and consulting the agents with whom the transaction takes place are called as search and transaction costs.
For instance, buyers and sellers of financial assets incur costs for getting access to insider information and thereby reduce uncertainty. The fees that are paid to mediators such as brokers are an important constituent of information costs associated with financial markets.
Similarly, there are search costs which a buyer incurs in determining which good is available in the market at the lowest price. Apart from this, there are bargaining costs associated with the transactions, which require negotiation between the parties in order to arrive at a reasonable agreement. These costs are negligible in some market transactions (e.g. buying newspapers) but huge in others (e.g. used car markets).
Apart from search and bargaining costs, an economic transaction under an incomplete information framework requires certain litigation and enforcement costs too. This is incurred to ensure that the terms of a contract are adhered to. This restrains people from deviating from the terms of the contract. Litigation costs are the most obvious indicator of enforcement costs in economic transactions.
Principal-Agent Framework
The Principal-agent framework facilitates the understanding of the problems that are encountered by economic decision-makers when faced with asymmetric information. This is due to the existence of an information gap between the contracting parties. The principal-agent problem is a specific game-theoretic description of a situation.
In this problem, the principal wants to delegate a task to the agent to undertake some action which is costly to the agent. The welfare of the principal depends on what the agent does. The utility function of an agent is different from that of the principal.
There exists a contract between the principal and the agent. This contract is self-enforcing. But there exists an information gap between the principal and the agent, and such a gap has implications for the decision of the contract they sign. The principal can act more effectively through the agent rather than directly because that would be a costly affair for the principal.
The behaviour of the agent is such that he is willing to work hard as long as his net utility from working is at least as much as he can earn from the next best opportunity. Hard work gives disutility to the agent, so he prefers not to work hard. This will result in lower work which adversely affects the principal’s utility.
As the welfare of the principal depends upon the performance of the agent, so he must construct incentive schemes in the form of bonuses or non-monetary to induce the agent to behave at least partially according to the principal’s interests.
This problem basically focuses on the designing of an incentive scheme. As the actions are unobservable, the incentives have to be created in such a way that the interests of agents get aligned with that of the principal.
The principal-agent problem is encountered in a wide variety of contexts, including Owners and managers of a firm, Manager-worker, etc. Taking up the case of the principal-agent relationship between owner and manager and also between manager and worker.
The managers, when working as agents for the owners, possess more information. For owners to acquire such information, a huge monitoring and research cost has to be incurred. The managers know that it is unlikely that the owners will involve themselves in such an expensive activity, so due to this asymmetry of information, managers behave discretionally, thereby promoting their own interests rather than that of the owners.
Similarly, when workers work as agents of managers (principal here), workers too will serve their interests by not working hard. Managers cannot fully scrutinize the actions of the worker, who tends to shirk. The objective of the manager is to produce maximum output through the efforts of workers by designing an appropriate incentive scheme for workers so that he doesn’t shirk.
This maximisation problem can be stated as:
maxx [x – s(x)]
In the above problem, x is the output which the principal aims to maximize. From the set of feasible actions A, an agent chooses actions a and b. If we assume that there is no uncertainty, then the output will be fully determined by the actions of the agent. The output-effort relationship is written as x=x(a), the cost of action a is written as c(a) and the incentive payment from the principal to the agent as s(x).
The utility function of the principal is given by output (x) minus the incentive payment s(x), i.e., x-s(x).
The utility function of the agent, on the other hand, is given by the difference between incentive payment and the cost of action, [s(x) – c(a)]
The utility function of the agent turns out to be:
U [s(x) – (x)]
The utility of the principal is maximised subject to the constraint [s(x) – c(a)], which is imposed by the optimising behaviour of the agent.
The constraints that the agent faces can be one of the following two types:
1. Participation Constraint
It is possible that the agent may have another opportunity available, imparting a certain level of utility. Let us call it as reservation level of utility. For this agent to be willing to work with the principal, the offered utility must be equal to at least this reservation utility. This constraint is also called as individual rationality constraint.
2. Incentive Compatibility Constraint
Under this, an agent chooses the best action for himself, given the incentives offered by the principal. Whatever is chosen by the agent is compatible with all the other options available, and the selected contract seeks to maximise the utility of the agent.
To determine the optimal incentive scheme, we assume that the principal has full information about the actions and costs of the agent while designing the incentive scheme for the agent.
Output produced is a function of action taken by the agent. Let us say the agent takes action ‘a’. The principal wants the agent to put in best effort ‘b’ in contrast to the available action ‘a’. So, the optimal incentive turns out to be:
maxb,s(.) x(b) – s(x(b))
such that s(x(b)) – c(b) ≥ ū …….[Participation constraint]
and s(x(b)) – c(b) ≥ s(x(a)) – c(a) [Incentive compatibility constraint]
for all a in A
For an optimal incentive design, the principal chooses the agent’s best effort ‘b’. Considering the participation constraint only along with the objective function, the principal wants s(x) to be as small as possible. The principal chooses ‘b’ in order to maximise
(x(b)) – c(b) ≥ ū
So, the first-order condition is x’(b) = c’ (b), and the solution can be b*.
If we consider the incentive compatibility constraint along with the objective function, then the principal should choose a function s(.) such that it is in the interest of the agent to choose b*, given s(.). This means that s(x*) – c(b*) ≥ s(x(a)) – c(a) for all a in A. This is the second-order condition.
An instance of the principal-agent problem can be taken from the behaviour of rating agencies which work as agents to the companies (the principals) that hire them to set their credit rating in the market.
Lower ratings by such agencies raise the cost of borrowing in the market for the firm. Thus, it is in the interest of the company to design an incentive plan so that the agency highly rates the company.
An important instance of principal-agent is found when a person takes his car to be serviced by the mechanic. A mechanic has more information about the car and has an incentive to charge discretionally. In this case, the principal, i.e. the person, must offer some incentive to the mechanic in order to reduce his ability to charge at discretion.
This principal-agent problem is not a situation or person-specific phenomenon, rather it is applicable under varying situations, and this exists not only in private corporate enterprises but also in public enterprises.
Coping with Imperfect Information
The above discussion of information asymmetries in the principal-agent framework and the resulting problems raise the need to search for possible solutions in order to reduce, if not eliminate, the problem of asymmetric information.
In the case of different types of problems like hidden actions, hidden information, etc., a different mechanism is used to deal with such asymmetries.
1. Incentive Structure or Information Revealing System
Whenever market failure (imperfect information) occurs due to hidden actions of one of the contracting parties, the establishment of an appropriate incentive structure can help in aligning the behaviour of the agent with that of the principal.
Similarly, when there exists hidden information, and the party which has access to the information uses this to its advantage, setting up of relevant information revealing system should be there. This system can either be legally enforcing or self-enforcing. Developing and improving our legislation and intensifying law enforcement can bring down the problem of adverse selection in financial securities, insurance markets, etc.
2. Signaling
This acts as a proxy measure to communicate information about unobservable characteristics of the good or service traded. This is generally seen as an attempt by the informed side of the market to communicate valuable information to the unknown party.
For example, in the market for lemons, the seller is better informed about the quality of the car and can signal his superior product from lemons by giving a warranty to the buyers for the quality of the car. This acts as a signal for the buyer to pay a higher price for the plum (good-quality car) rather than paying the average price.
Similarly, in the insurance market, those on the lower side of risk will be less willing to pay high premiums. Signalling, thus, is a useful tool to lessen the problem of asymmetric information. Warranties, discounts, guarantees, and educational levels all these are the means of signalling used to eliminate information asymmetries.
3. Screening
This process is used in order to arrive at the best conclusion in the presence of asymmetric information. When faced with a choice to engage the most efficient worker in the company, the recruiter seeks to screen all the profiles and set a screen or basis on observable characteristics of the workers.
The asymmetric information is actually due to certain hidden characteristics of the workers. The observable characteristic can be educational levels, health history, lifestyle, physical fitness, etc., the selection of which, in turn, depends upon the objective function.
4. Self-Selection in Insurance Market
When deciding the premiums to be charged to different risk groups, an insurance company uses the weighted average probability of illness for the whole group. This is done because the insurance company can not distinguish the high-risk group from the lower-risk. This weighted average leads to the drop-out of low-risk individuals, and the company serves only the high-risk group.
To surpass this problem, the insurance company gets the low-risk people to reveal themselves by offering coinsurance or some deductibility scheme. This process is called as self-selection, and this helps the insurance company to infer the risk features of the individuals better.
Efficiency Wage Model
One of the applications of the principal-agent framework and the prevalent asymmetric information is the efficiency wage model. This model seeks to explain the reason for the presence of unemployment.
Here, the participation constraint of employees is slack due to informational issues or other limited liability restraints. This says that unemployment arises because wages need to be above the market-clearing equilibrium level in order to give incentives to workers.
In order to boost the unobservable behaviour of workers and to incentivize them to work, unemployment and high wages coexist. This induces workers to keep on putting in effort and doing well, thereby serving the interests of the principal as well.
Assuming a situation where there is no unemployment and workers get paid the market-clearing wage rate, workers will have no incentive to put in effort. This is because there is no penalty involved with shirking from work. Whenever shirking leads to loss of job for a worker, he instantaneously gets rehired. So, in order to make workers not shirk, firms pay above-market wages. This creates its own penalty for shirking.
Let us say that a worker gets w* (the market clearing wage rate). If he shirks, gets caught and is fired, then he can again search for some job with the same wage rate w*. Now, suppose the principal offers a wage above this w*, i.e. we.
Then, it would mean that a worker can be induced to be productive when the difference between this shirking wage w* and non-shirking wage we are significant. The selected wage where there occurs no shirking is called as efficiency wage we.