首页 | 农村土地改革 | 法经济学茶座 | 法经济学论坛
     网站首页≯ 法律经济学理论≯
RATIONAL CHOICE THEORY IN LAW AND ECONOMICS(1)

Thomas S. Ulen

Alumni Distinguished Professor of Law, College of Law, University of

Illinois at Urbana-Champaign and Professor, University of Illinois Institute

of Government and Public Affairs

© Copyright 1999 Thomas S. Ulen

Abstract

 

The great appeal of law and economics has been its use of a coherent theory

of human decision making (rational choice theory) to examine legal rules

and institutions. While the innovations and accomplishments of that theory

in the analysis of the law have been many and important, there has been a

great deal of dissatisfaction among more traditional legal scholars with the

rational-choice foundation of law and economics. This chapter, first,

explains rational choice theory and its importance in the economic analysis

of law; second, summarizes some of the literature from economics, cognitive

psychology, and other disciplines that have been critical of rational choice

theory; and, third, speculates on the impact of those criticisms on the

economic analysis of law.

JEL classification: K00

Keywords: Rationality, Bargaining, Human Decision Making,

Methodological Criticism

 

1. Introduction

 

When law and economics was a new field in the legal curriculum and just

becoming a regular part of academic legal discourse, the use of

microeconomic theory to discuss traditional legal topics aroused interest but

also suspicion and hostility. Prominent among the reasons for this suspicion

and hostility was the feeling that the economist’s account of human decision

making - rational choice theory - was so deeply flawed that conclusions

derived from that account ought to be taken with a very large grain of salt, if

not rejected outright. To take one example, the economic theory of the

decision to commit a crime asserted that the potential criminal evaluated the

expected costs and expected benefits of the criminal act and committed the

crime only if the expected benefits exceeded the expected costs.

 

Many traditional legal scholars, judges and practitioners to whom such

examples of law and economics were given felt that the root of their

unhappiness with the conclusions of the new discipline lay with the

economist’s contention that all decisions (like that to commit a crime) are

the result of rational deliberation. The rational utility- or profit-maximizers

of microeconomic theory seemed to bear very little correlation to the

flesh-and-blood human beings with whom the law dealt. Therefore, to the

extent that law and economics used rational choice theory as its principal

theory of human decision making, the field had a difficult time in

convincing traditional legal scholars that it should be taken seriously.

 

In this chapter I first describe the rational choice model of decision

making and then give some examples of the use of that theory in law and

economics. Next I describe some criticisms of rational choice theory that

have been made principally by cognitive and social psychologists. And

finally, I speculate on the implications of these criticisms for the economic

analysis of law.

 

A. Rational Choice Theory

 

Rational choice theory is at the heart of modern economic theory and in the

disciplines contiguous to economics, such as some parts of political science,

decision theory, sociology, history and law, that have adopted the theory as

their model of decision making. In this section I define rational choice, show

how it is used in economics and describe its use in other disciplines and

suggest why traditional scholars in those other disciplines find problems

with rational choice theory.

 

2. Definitions of Rational Choice

 

There is no widely accepted definition of rational choice theory, but there are

two important senses in which the term is used. The first is an informal

sense: choice is said to be rational when it is deliberative and consistent. The

decision maker has thought about what he or she will do and can give a

reasoned justification for the choice. And taking choices over time or

focusing on their choices about particular things, such as food or class

choices in college, one expects rationality to lead to consistent (and

relatively stable) choices. That is, one expects that there will be no wild and

inexplicable swings in the objects of their choices and that the means chosen

to effectuate the goals of the decision maker will be reasonably well-suited to

the attainment of those goals (Nozick, 1993)

 

Like many informal definitions this one is highly imprecise. Indeed,

because almost all action would seem to be deliberative and consistent, this

informal definition does not seem to allow us to distinguish rational from

irrational action. Everything confirms the definition and nothing refutes it.

 

The second sense in which the profession uses “rational choice” is more

formal: consumers have transitive preferences and seek to maximize the

utility that they derive from those preferences, subject to various constraints.

Transitive preferences are those for which, if some good or bundle of goods

denoted A is preferred to another good or bundle of goods denoted B and B

is preferred to a third good or bundle of goods denoted C, then it must be the

case that A is preferred to C. By contrast, if it were the case that A were

preferred to B, B were preferred to C and C were preferred to A, we would

find that distinctly odd - indeed, irrational. Similarly unobjectionable is the

assumption that the decision maker seeks to maximize utility subject to

various constraints (such as those imposed by income, time, cognitive

resources and the like). Most economists find this more formal sense of

rational choice to be so obvious that they never doubt it and are puzzled by

those who do. (For a discussion of other formal conditions on rational

choice, see Plous, 1993, pp. 80-82.)

 

However obvious the formal sense of rational choice may be to

economists and to those in other disciplines who have adopted it as their

model of human decision making, the formal sense has not been without its

critics. Two such criticisms are worth noting here. First, some have said that

the formal notion of rational choice is as tautological as the informal sense.

That is, there is no, or almost no, behavior that refutes the formal sense of

rationality. All behavior may be said to be directed at utility maximization

(who would ever do otherwise?) and all preferences can be said to be

transitive. For instance, one might explain many instances of seemingly

intransitive preferences as being the result of a change in preferences over

time. Second, one can show some inconsistencies or puzzles in the notion of

transitive preferences. Suppose that we have asked a subject how he feels

about a teaspoonful of sugar. Now we add a grain of sugar to that spoonful.

If the subject likes sugar, he should presumably prefer the augmented

spoonful to the original on the theory that more is better. (If he does not like

sugar, then he should prefer the original spoonful.) However, it is likely to

be the case that the subject cannot distinguish the spoonful with one more

grain of sugar from the original spoonful. If so, he may say that he is

indifferent between the two. (This confusion between “more is better” and

indifference is itself puzzling, but set that puzzle to one side.) If we continue

to add grains of sugar to the original spoonful and ask the subject each time

we do so how he compares the augmented with the previous spoonful, he

will probably continue to say that he cannot distinguish and is, therefore,

indifferent. But ultimately the grains of sugar will add up to something

substantially greater than the original spoonful. So, even though the subject

will have contended that each successive spoonful was just as good as the

original so that, by transitivity, the final heaping spoonful should be

indifferent to the first, the subject is almost certain to prefer the heaping

spoonful to the original teaspoonful.

 

3. The Uses of Rational Choice Theory in Economics

 

These problems notwithstanding, economists have found rational choice

theory to be a very useful model for forming hypotheses about market

behavior. There are five principal reasons for this. First, the theory allows

economists to make predictions about economic behavior and, by and large,

those predictions are borne out by the empirical evidence. For example,

rational choice theory predicts and empirical work confirms, that when the

wage rate rises, all other things held equal, the supply of labor increases and

the demand for labor decreases; when the price of alcohol rises, relative to

that of other goods and services, the quantity demanded goes down (if not by

much); when the price of a good or service rises, again, relative to that of

other goods and services, productive effort tends to shift into the supply of

that good or service; and when the price of an input rises relative to that of

its substitutes, producers tend to use less of that input and relatively more of

the substitutes. These sorts of results are so widespread, so familiar to

professional economists and so central to the tenability of modern

microeconomic theory that it is not surprising that rational choice theory

forms such an important part of the canon of modern microeconomics.

 

Second, whenever there are seeming deviations from the predictions of

price theory, economists can usually explain those deviations without having

to assume that the decision makers involved are irrational. If the deviations

are a matter of degree (for example, an increase in the tax rate on incomes

above $250,000 was followed by a much smaller increase in government

revenues than predicted), there are a large number of hypotheses that can

explain this deviation that are well short of questioning the rationality of the

parties involved. To take just one such hypothesis, there may have been

means by which those with incomes over $250,000 could shelter income

from the tax authorities that had not been worth pursuing until the tax rate

increased. Alternatively, one might argue that the deviation from the

prediction of rational choice theory is a statistical fluke, due to some oddity

of the data set, that the seeming anomaly is the result of the decision

makers’ not having had the appropriate information to reach the result

predicted by the theory, that there was some theretofore unnoticed market

failure, such as monopoly or monopsony, external costs or benefits, public

goods, or informational asymmetry, that accounts for the discrepancy

between the theory’s predictions and the observed behavior. These statistical

and structural problems are real and common, so that economists are not

themselves being irrational in clinging fiercely to rational choice theory in

the face of seeming anomalies in the theory’s predictions.

 

Third, an economist may explain behavior that seems anomalous to

rational choice theory by appeal to a slight emendation of the theory.

Suppose, by way of example, that someone demonstrates that for a particular

good, when the price rises, all other things held equal, the quantity

demanded increases. Is this sufficient evidence on which to abandon the

rational choice theory? Typically not. For instance, in the case of a good

whose quantity demanded rises rather than falls when the price of the good

increases, one might propose a new phenomenon called a “snob effect”,

which arises when consumers take an increased price for a good as a sign of

its desirability, not as a sign to switch to cheaper alternatives. As a

consequence, the demand curve for a good subject to a snob effect may slope

upward, to indicate that an increase in the price of that good leads to an

increase in the quantity demanded (Liebenstein, 1950).

 

Fourth, there is a strong presumption among economists in the

evolutionary fitness of rational behavior - at least in the economic realm.

That is, rational consumers will prosper, while irrational consumers will

squander their resources and, perhaps, become money pumps for rational

calculators. More importantly, rational profit-maximizing businesses will

dominate those businesses that do not operate according to a rational plan.

 

Fifth and finally, Professor Gary Becker has shown that even if there

were consumers who behaved irrationality, in the sense of having

intransitive preferences, the standard predictions of price theory (such as

that an increase in the relative price of a good will lead to a decline in the

quantity of that good demanded) would still hold (Becker, 1962). Since that

article appeared, there have been more formal demonstrations that the

conclusions of price theory and of welfare economics are not much affected

by the presence of even a large number of consumers with intransitive

preferences. Therefore, while irrationality might still be an issue with

respect to the behavior of certain individuals, it is not an issue with respect

to aggregate behavior in markets and may, as a result, be ignored.

 

The point of all this is to suggest that there are plausible reasons why

economists cling tenaciously to rational choice theory. The theory is

extremely useful and powerful. Its predictions are frequently accurate and a

valuable guide to the formulation of public policy. And it is facile enough to

explain phenomena that seem anomalous without necessitating an

abandonment of the theory.

 

4. The Use of Rational Choice Theory in Other Disciplines

 

Criticisms abound when scholars seek to use rational choice theory to

describe non-market behavior, as has been the case in such oppressed

disciplines as demography, history, biology, political science, international

relations and law. None of these disciplines involves explicit market choices,

but all of them have been revolutionized by the importation and use of

rational choice theory.

 

Why has rational choice theory been so attractive to some scholars in

these contiguous disciplines? The principal reason is that the theory is the

most complete and coherent account of human decision making in the social

sciences. Moreover, the acknowledged success of economics in the public

policy arena in the past fifty years, which may be attributed in part to its

grounding in a coherent theory of rational behavior, may have inspired

emulation of the modeling aspect of economics by other disciplines in the

hope that this will lead to academic and policy successes similar to those of

economics.

 

But, as I noted, there have been sharp criticisms of rational choice theory

in the study of non-market behavior. Why should this be the case? One

possible answer is that traditional scholars are threatened by rational choice

theory: it is an unfamiliar technique, wielded principally by young scholars

and clearly threatens the academic standing of those who use traditional

methods. But there is more to the objections than mere self-interest. Put in a

light most favorable to the objectors, their query may be put this way:

Rational choice theory may be fine for the consideration of explicit market

decisions - such as which car to buy, whether to lease the car or to purchase

it with a loan, which job to take and what terms and conditions to accept and

how to invest one’s savings. These are, after all, quantifiable decisions. They

all involve money and that currency allows comparison among different

economic courses of action. But what reason is there to believe that

non-market decisions - such as whom to marry or how many children to

have or how to care for each of them or whether to trust one’s ally in foreign

affairs and so on - are made according to the same calculations?’ Put

someone succinctly, the question is ‘Why is the rational choice model

suitable to market behavior but not to non-market behavior?’ The question is

a serious one and deserves an answer.

 

I can think of three factors that might make the rational choice model a

better general model for market choices than for non-market choices. First,

market choices are frequent and routine. Even if people make mistakes when

they make their first market choices, they have an opportunity to learn

through repeated transactions. Moreover, in those instances in which market

transactions are rare in an individual’s life - as, for example, the purchase of

a house, there are many people who have made these purchases, so that

there is the possibility of learning from others about the pitfalls of those rare

transactions. Nonetheless, the general point bears making that market

choices are more problematic for individuals, the rarer they are. Related to

this matter is the fact that many non-market choices are so infrequent that

people do not have repeated opportunities to learn and to make corrections.

Love and marriage are examples. Even though one might consult others for

their experiences with these infrequent decisions, each individual’s

circumstances with respect to many of these non-market choices are so

highly particularized that the others’ experiences may not be an appropriate

guide to one’s own best course of action.

 

Second, as already noted, market choices are mediated through a

common medium - money - that makes commensurability easier. We do not

have problems of ‘comparing apples and oranges’ in many market

transactions because the choices almost always involve the purchaser’s

giving up money. Because the purchaser knows or could know the market

price of other goods and services or can compute an opportunity cost, he or

she can make a fairly accurate estimate of the comparative worth of very

different courses of action, such as whether to purchase or lease a new car or

whether to spend another year in school or get a job. By contrast,

non-market choices usually do not involved a common measuring rod like

money. Therefore, making comparisons across non-market alternatives or

between a market and a non-market alternative may be very difficult. How

does one compare the profound experience of parenthood with the cost of an

exotic vacation?

 

Third, there are problems of transparency in non-market choices. Market

choices involve relatively straightforward comparisons, save when they are

complex and reserved for specialists, as in some complicated options

valuations. There is, frequently, a single best (an optimal) decision. But

many non-market choices are simply difficult to understand and have a

variety of suitable outcomes. Consider the decision of whether or not to

invite a friend to travel a long distance with one. There is no ‘correct’

answer to the question and there are lots of nuanced meanings, including

misunderstandings, that may be read into the question and its answer.

 

Taken together, these issues of frequency, commensurability and

transparency may suggest why rational choice theory is widely accepted as

an explanation of market choices but has difficulties in acceptance as a

model of non-market choices (Ulen, 1998).

 

B. The Application of Rational Choice Theory in the Law

 

The most important, but not the only, characteristic of law and economics is

its use of rational choice theory to examine legal decisions. In this section I

describe the general reasons why rational choice theory may be appropriate

for the description and prediction of legal decision making and give

examples of the use of the theory in the analysis of private law rules of

contract and tort law and in the analysis of criminal law.

 

5. Why Rational Choice Theory is an Appropriate Model of Legal

Decision Making

 

Rational choice theory is one of three distinguishing characteristics of law

and economics. In light of the general criticism of the applicability of

rational choice theory to non-market choices given above, one is entitled to

ask why rational choice is appropriate for the discussion of legal matters,

most of which are non-market choices.

 

The answer is that many legal decisions are indeed market-like choices.

They may be said to be so on the ground that legal rules create implicit

prices on different behaviors and that legal decision makers conform their

behavior to those prices in much the same way as they conform their market

behavior to the relative prices there. For example, the law imposes a

monetary sanction (called ‘compensatory money damages’) on those who

unjustifiably interfere with another’s property, breach a contract, or

accidentally injure another person or his property. These money amounts

may be taken to be the ‘prices’ of engaging in certain kinds of behavior,

such as a failure to take due care or to perform a contractual obligation.

Presumably, rational decision makers will compare those legal prices with

those of the alternatives and will comply with the law’s duties (that is, not

interfere with another’s property without their permission, perform a

contractual obligation, or take due care) if the price for doing so is greater

than the price of not doing so. For example, if the benefit of breaching a

contract is $10,000 and the money damages that the breacher can anticipate

paying to the innocent party are $5,000, then there is likely to be breach of

contract (Cooter and Ulen, 1997; Posner, 1998). It is the central innovation

of law and economics to have recognized that many legal decisions have this

market-choice-like quality and that, therefore, rational choice theory is an

appropriate model of much legal decision making.

 

6. Private Law Examples of Rational Choice Theory

 

I shall here give only a few broad examples of rational choice theory in

private law, rather than an exhaustive survey. One particular omission

deserves mention. I shall have nothing to say about the Coase Theorem, the

most famous example of the economic analysis of law and a superb example

of rational choice theory in private law decision making, on the grounds that

this Encyclopedia covers that theorem extensively elsewhere. Here I merely

note that the bargaining behavior that the Coase Theorem posits will occur

in the absence of transaction costs precisely because the parties are rational

calculators in the manner assumed by rational choice theory.

Notwithstanding the fact that I have not discussed the Coase Theorem, in

Part C below I shall describe some criticisms of the assumptions of that

theorem and then in Part D I shall some implications for the theorem of

those criticisms. Here I shall give examples of rational choice theory as it

informs the economic analysis of contract law and tort law.

 

6.1 Contract Law

In an economic analysis of contract law, the place to begin is with the

question, Why do rational parties need the law’s help in concluding

consensual agreements? One might well argue that in the absence of

transaction costs, parties seeking to conclude agreements would not need

help from the law. They would costlessly conclude mutually beneficial

terms. It follows that contact law aids parties to conclude agreements when

transaction costs are positive.

 

What gives rise to positive transaction costs in contracting? There are

two general sorts of reasons. First, there may be problems in the

environment in which the parties negotiate and these problems can lead to

inefficiencies. For example, there could be third-party effects and in the

absence of legal intervention the contracting parties are not likely to pay

attention to those external effects. Additionally, one of the contracting

parties could be a monopolist and could, therefore, put the other party in a

situation in which ‘consent’ would be meaningless. The law can correct for

this social cost by insisting that parties reach roughly competitive terms in

their agreement.

 

The second general source of transaction costs in contracting are

problems that individual contractors may have. For example, some parties

may have unstable or intransitive preferences because, say, they are very

young, insane, or suffering from Alzheimer’s disease. When people have

unstable or intransitive preferences, there is no guarantee that they are in a

position to gauge the benefit of bargaining and cannot, therefore, form

mutually beneficial agreements. Predictably, contract law does not enforce

agreements in which one of the parties has unstable preferences.

 

The transaction costs arising from factors of the contractual environment

and individual transactors may be so high as to preclude contracting or to

make it take place on inefficient terms. As a general corrective, contract law

can present a set of pre-determined contract terms that take account of these

transaction costs and save parties the costs of specifying these terms each

time they negotiate to enter an agreement.

 

6.2 Tort Law

According to economic analysis, the tort liability system seeks to minimize

the sum of prevention, accident and administrative costs. Potential injurers

and potential victims are rational calculators who compare the expected

costs and benefits of various states of the world (such as those arising from

taking different kinds or amounts of precaution) and, given their tastes,

maximize their utility subject to several constraints. By assumption, the

transaction costs between potential injurers and victims are so high that they

cannot form a contractual agreement regarding their obligations in the event

of an accident. That being so, the potential injurer has virtually no incentive

to take the expected costs of his failure to take adequate precaution into

account. As a result, there are too many or too severe accidents and potential

victims may inefficiently seek to protect themselves from uncompensated

injury. Economic analysis of tort law focuses on using legal rules to induce

the (rational) potential injurer to internalize these costs of failing to take

adequate care. Specifically, by holding out the possibility that the potential

injurer will be deemed liable for failure to take due care and, if liable, will

have to pay the victim’s damages arising from the accident, tort law induces

the rational potential injurer to take the social-cost-minimizing level of care.

 

As an example of the rational-choice aspect of this analysis, consider the

economic view of negligence versus strict liability. To be extremely terse

about a complex matter, economic analysis suggests that some form of

negligence is efficient when precaution is bilateral and that strict liability is

efficient when precaution is unilateral. The intriguing novelty in this view is

the implication that the negligence standard addresses itself to both potential

victims and potential injurers in order to induce both of them to take care.

 

Imagine the calculations that a rational person will make when faced

with some form of the negligence standard for determining liability. Assume

that this person does not know whether she will be injured or will injure; for

example, she could be an automobile driver. She knows that under the

negligence standard the injurer who complies with the legal duty of care will

not be held liable for the victim’s injuries. Therefore, if she were to be an

injurer, the best thing for her to do would be to comply with the legal duty of

care. That action will minimize her expected liability and, being rational,

she decides to comply with the legal duty of care. But suppose that she is the

victim in an automobile accident. In that case she will almost certainly be

injured by someone who has complied with the legal duty of care. (Why?

Because every potential injurer is, like her, rational and will have recognized

that his expected liability is zero if he complies with the legal duty of care.

Being rational, he will seek to minimize his expected liability by taking the

appropriate amount of care.) Having reasoned that she will be injured by a

rational injurer who will not be found liable, she recognizes that if there is

an accident in which she is the victim, she will have to bear her own

accident losses. She must, therefore, take action so as to minimize those

expected accident costs by taking the optimal amount of care (whose

marginal cost equals its marginal benefit - the expected reduction in

expected accident costs). Thus, negligence induces optimal care by both

potential injurers and potential victims.

 

7. Public Law: The Decision to Commit a Crime

 

As a final example of rational choice theory as applied to legal decision

making, consider the well-known Becker (1968) model of the decision to

commit a crime. Becker hypothesized that criminals are rational calculators

and that, therefore, they made their decisions about compliance with

criminal law on the basis of a comparison of the expected costs and benefits

of criminal and legal activity. The expected costs of crime result from

multiplying the probabilities of the activity’s being detected and of the

perpetrator’s being apprehended and convicted by the monetary value of the

legal sanction and the value of any non-pecuniary losses he might suffer,

such as a loss in reputation from being branded a criminal. The expected

benefits of the crime result from multiplying the probability of success times

the monetary and non-pecuniary benefits of the particular crime. These latter

include both the value of the goods or the amount of money resulting

directly from crime and such intangible but potentially valuable outcomes as

being known in one’s community as a law-breaker. According to the Becker

model, the rational criminal will commit the crime if these expected costs

are less than the expected benefits and will refrain from crime if the reverse

is true. (For a critique of the Becker model’s predictions, using the criticisms

of Part C of this entry, see Wilson and Abrahamse, 1992).

 

C. Criticisms of Rational Choice Theory

 

Recent scholarship by some cognitive psychologists and by economists

familiar with the cognitive psychological literature describes experimental

results that are difficult to reconcile with rational choice theory. The

experiments have questioned implications of that theory with regard to at

least four different areas. First, subjects in carefully-designed experiments

seem to reject mutually beneficial exchanges when they believe that the

proposed division of the cooperative surplus violates widely-accepted norms

of fairness. Rational choice theory predicts that this will not happen. Second,

subjects in another series of experiments in which there are several stages of

bargaining involved do not devise rational strategies. Third, most decision

makers have cognitive limitations that cause systematic deviations in their

behavior away from that predicted by the theory of rational choice. For

instance, those engaged in a common-value auction fall prey to the

‘winner’s curse’; and people cling to the status quo, even though an

alternative likely has much greater value. Fourth, experiments have shown

that people do not make decisions about uncertain outcomes in the way that

the theory of rational choice predicts.

 

I shall briefly summarize some of these results in this section before

turning in Part D to a discussion of the important implications of this

literature for the rational-choice-based economic analysis of the law.

 

8. Rational Bargaining

 

Rational choice theory makes two broad claims about bargaining. One is that

whenever there is a cooperative surplus greater than the transactions costs of

splitting that surplus, parties will find a means of dividing the surplus. The

second is that there are certain situations in which people will not fully

participate in bargaining behavior, such as in the provision of and payment

for public goods. Experimental evidence questions both of these claims.

People apparently willingly cooperate in circumstances in which rational

choice theory predicts that they will not cooperate and they frequently do not

bargain in circumstances in which the theory predicts that they will.

 

8.1 Cooperation in the Production of Public Goods

Rational choice theory predicts that for public goods - that is, goods that

exhibit non-rivalrous consumption and for which the costs to private

profit-maximizing suppliers of excluding non-paying beneficiaries are

prohibitively high - rationally self-interested, utility-maximizing consumers

will not pay for the units of a public good from which they benefit. They

will, in the classic phrase, ‘free ride’, that is, consume the public good

without paying for it.

 

However, a series of experiments reveals that people do willingly and

voluntarily pay for their share of public goods (Thaler, 1992; Ulen, 1994).

The experiments are variations on the following general set of rules. A

group of people, usually college students, are brought together and each is

given the same sum of money. They are told that they can invest some, none,

or all of that money in something called a ‘group exchange’. (The decision

to invest in the group exchange is a secret one. That is, one does not know

whether or not the other players have contributed. All one knows is that they

have all been given the same amount of money and are all subject to the

same rules.) The group is also told that the game operator will multiply the

total sum invested by the group by a number that is larger than one but

smaller than the number of people in the group and will then divide the

resulting sum equally among all of the group members, whether they have

invested in the group exchange or not. These rules make the group exchange

into a public good. Presumably, the temptation on rational actors will be to

contribute nothing to the group exchange and then benefit with an equal

share of the sum generated by the game operator.

 

To see how this works, suppose that there are five people in the group

and that each of them is given $5. If no one contributes anything to the

group exchange, then there is nothing for the game operator to multiply and

nothing, therefore, for the group to divide. But suppose that only one person

contributes nothing and the other four people in our example contribute their

entire $5 to the group exchange. Further, suppose that the group operator

doubles the resulting $20 to $40 and then distributes that sum equally

among all five players. Each, therefore, receives $8. The incremental return

to the four players who contributed $5 is $3, but that of the player who

contributed nothing is $8. This logic should be clear to all the subjects, so

that none of them should contribute to the group exchange; all of them

should seek to free ride. Thus, the prediction of the theory of rational choice

is that no one will invest in the group exchange.

 

In laboratory experiments of this game, the predictions of rational choice

theory are not borne out. Although not everyone contributes to the group

exchange, a substantial number do. On average, subjects in the experiments

contributed between 40 and 60 percent of their initial sum to the public

good. When experimenters vary the conditions of the game - by, for

example, increasing the number of times the game is played, giving the

players some prior experience with the game, or increasing the size of the

stakes - the general outcome is the same: contributions to the public good are

well above what the theory of rational choice would predict. The only

exception to the 40-60 percent contribution rate was when the subjects were

graduate students in economics at the University of Wisconsin. The

contribution rate for that group was only 20 percent (Marwell and Ames,

1981).

 

One variation of the experiment is particularly interesting: that of the

players repeating the game several times. Rational choice theory would

predict that with repeated plays the rate of contribution to the group

exchange would decline (perhaps because the players would come to

understand the disadvantages of contribution and the advantages of free

riding). And that is what the experimenters found. (There is some

controversy about whether the decline is rapid or gradual, but there is

agreement that there is decline.) Were the reasons for the decline those

given by rational choice theory? Plausible as the theory’s conjecture sounds,

it is not supported by the experimental results. The 40-60 percent

cooperation rate of the earlier experiment is found to hold on the first trial of

the game even for experienced players - that is, even for those players who

have participated in other multiple-play public-goods experiments in which

the contribution rate fell with repeated plays. Andreoni confirmed this

surprising result in the following way. He assembled a group to play the

public-goods experiment and announced the usual rules of the game and,

further, that the game would be played for ten trials. He found, as expected,

that the contribution rate declined over the course of those trials. At the end

of the ten trials, he announced that the same players would play the game for

an additional ten trials. When the game was re-started for the second run of

ten trials, the participation rate rose back to the 40-60 percent range before

declining again. (Andreoni, 1988)

 

These experimental results present a puzzle for rational choice theory:

why do people cooperate when there appears to be a rational basis for not

cooperating? One possibility is that people start any given interaction from

the presumption that it is better to cooperate than not; they continue to

cooperate until the evidence shows this to be ill-advised; and then they quit

cooperating.

 

8.2 Rational Bargaining over a Cooperative Surplus

Rational choice theory offers no prediction about the particular proportions

in which voluntary traders will divide a cooperative surplus; it merely

suggests that if such a cooperative surplus exists and, very importantly, if

there are no serious impediments to exchange (that is, no transaction costs),

traders will find a way to divide that cooperative surplus so that both of them

are better off than they would have been if they had not traded. The theory

provides a complete explanation for exchanges that do take place and those

that do not: if a voluntary exchange takes place, then there must have been a

cooperative surplus to be divided and the impediments to exchange must

have been trivial; if an exchange does not take place, then there was either

no cooperative surplus to be divided (that is, the minimum price for which

the seller was prepared to sell was greater than the maximum price the buyer

was prepared to pay) or the costs of concluding an exchange were greater

than the cooperative surplus. Rational choice theory offers no other reasons

for a failure to exchange. Clearly, what would be troubling for rational

choice theory would be exchanges that failed to materialize even though

there was a cooperative surplus to be divided and there were no impediments

to exchange.

 

Experimenters have probed these possibilities in a very wide-ranging

series of experiments regarding the ‘ultimatum bargaining game’. (Guth,

Schmittberger and Schwarze, 1982). The game works as follows. There are

two participants, call them Player 1 and Player 2. They do not know one

another and are not allowed to communicate. The object is to divide a fixed

sum of money, say, $20. Player 1 makes an offer to divide the sum; Player 2

then either accepts the division, in which case the players receive the actual

division proposed by Player 1, or rejects it, in which case they each receive

nothing. Thus, if Player 1 proposes that they each receive $10 and Player 2

accepts that proposal, that is what they actually receive. If Player 1 proposes

that he receive $19 and Player 2 receive $1 and if Player 2 accepts that, that

is what they each receive; if Player 2 rejects that division, they each receive

nothing.

 

The prediction of rational choice theory is that Player 1 will recognize

that the best thing for her to do is to propose a one-sided division of the

fixed sum in her favor. This is because Player 2 will then be in the position

of accepting whatever Player 1 proposes or getting nothing and the clearly

rational thing for Player 2 to do is to accept something rather than nothing.

 

The experimental results do not confirm the prediction of rational choice

theory. Those in the position to make the initial proposed division generally

do not propose a one-sided division in their favor. Rather, in a wide-ranging

number of experiments over many years and in many different countries, the

modal (that is, most common) proposal is for a 50-50 split and the mean

proposal has been for a 37-73 split. Nor was the prediction for the sheepish

acquiescence of Players 2 to the proposed division confirmed. Most of them

accepted the split (presumably because the modal proposal was an even

split), but, interestingly, almost 25 percent of the proposals were rejected

(with the most one-sided proposals being almost uniformly rejected)

(Kahneman, Knetsch and Thaler, 1986).

 

There have been numerous variations on these basic versions of the

ultimatum game. In every instance, no matter how complex the

experimenters make the game, the results offer little support for rational

choice theory’s account of how people do or ought to behave.

 

8.3 The Endowment Effect or Status Quo Bias

Recall that rational choice theory predicts that in the absence of transaction

costs and in the presence of a cooperative surplus, there will be an exchange.

One of the most important discoveries in the experimental literature is an

effect that suggests that bargains will not necessarily take place under the

ideal conditions posited by rational choice theory. The reason is the presence

of what is called an ‘endowment effect’ or ‘status quo bias’. Thaler defines

that effect as ‘the fact that people often demand much more to give up an

object than they would be willing to pay to acquire it’. (Thaler, 1992, p. 63).

The closely related ‘status quo bias’ may be defined as a general preference

for the current state of holdings over any alternative (Korobkin, 1994;

Samuelson and Zeckhauser, 1988).

 

 


中国法律经济学网登载此文出于学术研究之目的,绝不意味着中国法律经济学网赞同其观点或证实其描述。以上内容仅供研究者学习与交流,无意侵犯版权。如有侵犯您的利益,请告知。我们将尽快删除。

加入日期:2007/1/6 16:11:32浏览次数:2837
发表评论
名号:
内容:
验证: 3338
法律经济学网
联系站长: 柯华庆 lawgame@263.net 京ICP备09028584号
北京市昌平区中国政法大学法学院(102249)
本网站由卡卡鱼网提供技术支持 网站总访问量:1996356