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(2)

 

The endowment effect surfaced in laboratory experiments (Thaler, 1992;

Korobkin, 1998). Experimenters intent on testing propositions about

bargaining typically gave half the subjects something of value (for example,

a lottery ticket, ballpoint pens, or a coffee mug) and the other half a sum of

money. One member of each group was paired with a member of the other

group. The pairs were then given an opportunity to exchange; the roles were

then re-assigned and the participants again had an opportunity to exchange.

This reversal of roles was done a number of times with the understanding

that only one of the attempted exchanges would actually be executed by the

experimenters. The subjects were given ample opportunity to learn the rules

of the game.

 

The purpose of the experiments was to test two propositions about

exchange suggested by rational choice theory. First was the proposition that

when there are no impediments to exchange, goods and services will move

to those who value them the most. Because there were no impediments in

the experiments, the tickets, ballpoint pens and coffee mugs should end up

in the hands of those who valued them the most. Sometimes that would be

the subject to whom the items had been originally given and sometimes it

would be to the person to whom cash had been given. Because the

investigators did not know beforehand what the tastes and preferences of the

subjects were, their prediction was that approximately half of the pairs

would engage in an exchange.

 

Second was a proposition about the prices at which the exchanges would

take place. Because of the role reversals and the repetition of the possible

exchanges, each subject found herself alternately in the role of seller and

buyer of the same object. The prediction of the experimenters was that the

prices asked by subjects in their role as sellers ought to be roughly the same

as the prices bid by them in their role as buyers.

 

The experiments confirmed neither of these propositions. First, far fewer

transactions took place than the theory predicted - approximately half those

anticipated. Second, the prices asked by those who were willing to sell and

those bid by those who were willing to buy were not in equilibrium. The

ratio of the median selling price and the median buying price was

approximately 2 to 1. These results were invariant to the objects being

exchanged and to other important factors.

 

9. Choice under Uncertainty

 

The rational choice theory of decision making under uncertainty posits that

decision makers attempt to maximize their expected utility by combining

three elements: their attitudes toward risk (risk neutrality, risk preferring, or,

the most commonly-assumed attitude, risk aversion); their stable,

well-ordered preferences for the possible outcomes; and estimates of the

likelihood of the various possible outcomes. But some recent experimental

results suggest that this is not an accurate description of how many people

make decisions about uncertain outcomes.

 

9.1 Preference Reversals

Consider the following choice under uncertainty. There are two gambles,

call them H and L. H entails a high probability of winning a small prize,

say, a 90 percent chance of winning $4. L entails a low probability of

winning a larger sum, say, a 10 percent chance of winning $40. When

presented with these alternatives, most people choose H. The subjects were

then asked to say for what price they would be willing to sell each gamble if

they owned it (as, say, a lottery ticket). Surprisingly, most subjects put a

higher price on L than on H (Lichtenstein and Slovic, 1971). This is

surprising because the expected value (the product of the probability of

winning and the value of winning) is almost identical in the examples given.

What is curious about this is that although when put to a choice between H

and L most people choose H, when asked to price the two gambles, most

people attribute a higher selling price to L than to H, which indicates that

they find L more valuable than H. The figures are dramatic. One scholar

reports that in a recent replication of the experiment that used the values

given above, 71 percent of the subjects preferred H but 67 percent priced L

above H (Thaler, 1992, p. 84). If one had predicted an outcome on the basis

of expected utility maximization, one would have confidently predicted that

these choices would have been consistent. That is, if H were preferred to L,

then the imputed selling price of H would have been higher than that of L

and vice versa. But that consistency is not at all what the experimenters have

found.

 

This curious phenomenon is called ‘preference reversal’. In so far as

there is a simple explanation for these reversals, it is that people apparently

use the payoffs rather than the expected values of gambles in pricing them.

What is troubling about this explanation for rational choice theory is that if

it is true, it can lead people to the sort of inconsistent and seemingly

irrational choices shown in the experiments. In the extreme, of course, one

can make sport of people who behave in this fashion by getting them to

make preposterous offers for very low probability gambles that have huge

monetary payoffs, a fact that may be evident to the organizers of

state-operated lotteries (Grether and Plott, 1979).

 

Three possible explanations have been given for preference reversals:

intransitive preferences, procedure invariance and violations of the

independence axiom. Let us take these possibilities up in turn.

 

As we have already seen, one of the common definitions of rationality in

economics is that preference orderings exhibit transitivity. It is easy to see

that the preference-reversal phenomenon might imply intransitive

preferences. A rational person ought to be roughly indifferent between the

imputed cash value of H and H itself. Similarly, a rational person ought to

be roughly indifferent between the imputed cash value of L and L itself. If,

therefore, one prefers H to L, then she ought, by transitivity, to prefer the

cash value of H to the cash value of L. But preference reversal means that

when H is preferred to L, the cash value of L is preferred to the cash value of

H. It turns out that this pattern of preferences is intransitive only if

something called procedural invariance does not hold.

 

‘Procedural invariance’ refers to a result’s being invariant to the

particular procedure designed to measure it. And most scientific

investigation presumes procedural invariance to hold. The distance from

Berlin to Munich should be the same whether we start our measurement in

Berlin and go south or in Munich and go north. In the context of choice

under uncertainty, the phrase refers to the invariance of preference rankings

when the investigator uses different means of eliciting the subject’s

preferences. It is standard in modern economics to say that A is preferred to

B if A is selected when both A and B are available or if the subject has a

higher reservation price for A than for B. That is, we can determine the

preference ranking by two different procedures: either presenting the subject

with the choice and seeing which she chooses or by asking the subject which

good has the higher reservation price per unit. It is almost never stated as an

axiom in microeconomics (but probably should be) that these different

procedures must yield the same result. The notion is that the preference

ranking of A and B is (or ought to be) independent of the procedure by

which the investigator determines that ranking. As a result, the preference

equivalence of the cash value of H and H itself and that of the cash value of

L and L itself is the result of an assumption of procedural invariance.

 

The third possibility is that the subjects violate the independence axiom

of expected-utility theory. That axiom says, in essence, that if you prefer X to

Y, then you should also prefer the chance to win X with probability p to the

chance to win Y with probability p. This seems as straightforwardly

appealing as does the axiom of transitivity, but it turns out that the

independence axiom is sometimes violated in decision making under

uncertainty (Machina, 1990). The preference-reversal phenomenon would

clearly be a violation of this axiom.

 

9.2 Intertemporal Choice

Decision making over uncertain outcomes frequently involves choosing

between a current and a future outcome or between two future outcomes.

There is a standard, rational-choice-based theory of this allocation of

resources over time, but there is now experimental evidence that contradicts

this theory. People seem not to be fully aware of the special problems and

opportunities that the passage of time raises. As a result, they frequently

make decisions about the allocation of resources over time that seem to be

difficult to square with rational choice theory. Take the example of paying

income taxes. Many taxpayers routinely have too much income tax withheld

during the year so that they can receive a refund from the Internal Revenue

Service after filing their tax returns in the Spring of the following year. This

over-withholding constitutes an interest-free loan to the federal government.

The taxpayers who currently do this would be better off (according to

rational choice theory) if they were to reduce the amount withheld so that at

the end of the year they neither owed money to nor were owed money by the

IRS.

 

At the other extreme are examples of absurdly high discount rates. For

example, people routinely ignore the warnings of dermatologists that

over-exposure to the sun can cause skin cancer later in life, apparently

preferring the current benefits of a suntan. But they may pay attention if the

dermatologist tells them that the sun may cause large pores or blackheads in

the near future. Most homeowners do not have nearly enough insulation in

their attics and walls, even though the cost of installing more would lead to

significant savings on energy use within one year. Nor do they buy more

expensive energy-efficient appliances, even though the energy-use savings

will more than make up for the increased purchase price within a year.

Economists have calculated that the purchase of the lower-priced, less

efficient appliances implies a discount rate of between 45 percent and 130

percent at low energy costs and between 120 percent and 300 percent at

higher energy costs. Either set of discount rates is absurdly high.

 

Why do people make such anomalous decisions where intertemporal

choice is involved? One of the most robust findings in the experimental

literature is that discount rates decline sharply with the length of time that

the subject must wait for her reward and with the size of the reward. These

experimental results are not consistent with rational choice theory, which

holds that discount rates should generally equal the market rate of interest,

that the discount rates should be constant (that is, invariant to the period of

time considered) and certainly invariant with respect to the amount of

money involved. The difficulty posed by the declining discount rate as the

date of the reward recedes further into the future is that it implies an

anomalous preference reversal. This is because the individual’s preference as

between, say, Project A and Project B could initially be in favor of A

(because it is, let us assume, the nearer in time) and then switch to B, all

other things remaining equal, if the time at which B will be realized is

brought forward (but is still realized after A is realized). If discount rates are

constant, this sort of switching, all other things equal, cannot occur

(Loewenstein and Elster, 1992).

 

The effect of the size of the reward on the discount rate is as strong as

the effect of time delay. The general problem is that people perceive the

difference between $100 today and $150 in a year as greater than the

difference between $10 today and $15 in a year. As a result, many people are

willing to wait for the extra $50 in the first instance but not for the extra $5

in the second instance. Rational choice theory cannot explain this robust

experimental result. Shefrin and Thaler have proposed that the explanation

lies in how people take mental account of small and large windfalls. They

hypothesize that small windfall gains are put into a mental account that

allows for immediate consumption, while large windfall amounts are put

into a separate mental account for which there is a much lower propensity

for immediate consumption. Thus, the opportunity cost of waiting for a

small windfall may be perceived to be foregone consumption. But the

opportunity cost of waiting for a large windfall will be foregone interest or

investment. If foregone consumption is more difficult to resist than foregone

interest or investment, that would explain the observed effect of the size of

the award causing a decline in the discount rate (Shefrin and Thaler, 1988).

 

D. Implications of the Criticisms of Rational Choice Theory for the

Economic Analysis of Law

 

As we have seen, law and economics has premised much of its scholarship

on rational choice theory. Therefore, the implications of the literature

critical of that theory for law and economics are profound. In this part I want

to focus on four of those implications - on the relationship between

transactions costs and the law, on the choice between mandatory and default

rules in the law, on the best means of dealing with risky decisions by

consumers and on some issues in tort law.

 

10. The Coase Theorem and Criticisms of Rational Choice Theory

 

The most famous piece of scholarship in law and economics is ‘The Problem

of Social Cost’ by Professor Ronald Coase (1960). The broad inquiry to

which that article is addressed is this: when may society rely upon

bargaining to achieve the efficient use of resources and when may it not?

That inquiry then leads to a discussion of how the law should be structured

so as to encourage efficient resource use in those circumstances in which it

is inappropriate to rely upon bargaining. The Coase Theorem says that when

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

zero), the efficient use of resources will result, regardless of the assignment

of property rights. Appropriate legal policy depends on being able to identify

impediments to exchange and to specify correctives when those impediments

are significant. Law and economics scholarship has concentrated on search,

bargaining and enforcement costs as the elements of transaction costs and

has sought to identify the objective characteristics of transactions (for

example, the number of people involved, whether the transaction is for a

fungible or a unique item and so on) that cause these three elements of the

costs of exchange to be high.

 

The literature reported in Section 8 has two important implications for

the standard view of the Coase Theorem. First, the reported results on

cooperation and fairness suggest that people are far more ready to cooperate

and that they have a much stronger sense of what is an equitable outcome

than rational choice theory predicts. These conclusions point in two very

different directions on the Coase Theorem. On the one hand, the broad

willingness to cooperate (as revealed in the public goods experiments)

suggests that voluntary exchange may be able to achieve an efficient

allocation in a broader range of circumstances than those of zero transaction

costs and, further, that the need to intervene in private decision making to

enhance efficiency, even when transaction costs are positive and significant,

may be less than previously thought. For example, if people appear to be

more willing to contribute to the provision of public goods than rational

choice theory predicts, then there may be less need for the compulsory public

subsidization of those goods or the level of subsidization can be less

extensive. There are implications, too, regarding the need for or the most

appropriate structure of environmental regulations - for example, people

may be more willing to bestow the external benefits of

environmentally-conscious activity than previously supposed.

 

On the other hand, the finding of the experimental literature that people

appear to be extremely sensitive to the equitable distribution of resources

suggests that more intervention in private decision making may be justifiable

than previously thought. This is because the experiments suggest that people

may be so sensitive to fairness issues that they would rather not cooperate

than cooperate on terms that they consider to be excessively one-sided. This

is a cause for the failure of bargaining that has not heretofore been given

much weight. Even when transaction costs are very low, some otherwise

efficient exchanges will not take place because some of the participants do

not like the proposed division of the cooperative surplus. The experimental

findings provide an efficiency justification for legal intervention in private

decision making in order to prevent over-reaching by one of the parties that

might forestall an otherwise efficiency-enhancing exchange.

 

The second major implication for the Coase Theorem of the criticisms

reported in Section 8 arises from the endowment effect (or status quo bias).

Recall that that effect suggested two anomalies in bargaining behavior: first,

when transaction costs were very low, people were far more reluctant to

transact than rational choice theory predicted and, second, subjects typically

demanded twice as much to sell something they owned as they were willing

to pay in order to acquire it. The troubling implication of those findings is

that there may be cases in which there is no such thing as a uniquely

efficient assignment of rights. Where society initially assigns an entitlement

is where it is likely to remain; we should be far less sanguine about

entitlements moving to their highest-valued use, even when transaction costs

are zero, than we have been heretofore. (Indeed, status quo bias makes the

notion of ‘highest valuing use’ less clear.)

 

11. Default and Mandatory Rules and the Criticisms of Rational Choice

Theory

 

If one assumes, as does law and economics, that the law can increase the

efficient use of resources by creating rules of conduct that correct for market

failures, two issues that must be resolved are, first, the specification of a rule

or standard and, second, whether that rule or standard is mandatory or may

be waived by those affected. One of the areas of the law in which this issue

has been central is corporation law. There the debate has been between those

who favor non-waivable mandatory rules of corporate conduct and those

who favor allowing corporations to opt out of some rules. Consider, for

example, insider trading. Everyone admits that there are potential

inefficiencies from allowing insider trading, although there are

disagreements about the extent and likelihood of these inefficiencies. Most

commentators, therefore, agree that there ought to be some legal regulation

of the practice. However, there is disagreement about whether this regulation

should take the form of a prohibition or merely a default rule from which

those corporations that so choose might wish to opt out. Those who favor

making the prohibition waivable argue that some corporations might wish to

offer their managers partial compensation in the form of allowing them to

trade on the basis of the inside information that they acquire in the course of

working for the corporation. If that method of compensation is more

efficient than the alternatives, then, the argument goes, those corporations

and their managers ought to be allowed to opt out of the default rule. The

other side argues that both private and public difficulties in policing the

behavior of managers make the realization of those efficiencies illusory.

Thus, they argue, the prohibition of insider trading should be non-waivable.

 

How are the findings of Part C relevant to this issue? Status-quo bias

suggests that people will not make changes away from a default position

unless the expected benefits from so doing substantially exceed the expected

costs. That is, the default position has a strong anchoring effect. With

respect to insider trading rules, the presence of status-quo bias might

indicate that even if the prohibition on insider trading was waivable, very

few corporations would take advantage of that waivability.

 

There is another relevant implication of status-quo bias. If most people

are reluctant to leave the status quo, whatever that is, then the law ought to

establish the starting position (that is, establish the status quo) at an efficient

point. In the case of insider trading that might suggest that the appropriate

starting point is a prohibition of the practice, not the freedom to engage in

the practice unless one’s employer has forbidden it. This sort of

consideration no doubt has other applications in the law well beyond

corporation law. For instance, it may say something about whether society

should make the status quo one in which addictive drugs are legal or one in

which they are illegal but one may (explicitly or implicitly) opt out of this

illegality. (For the implications of status quo bias for a broad range of

contract issues, see Korobkin, 1998.)

 

12. Risk Regulation and the Criticisms of Rational Choice Theory

 

The regulation of risk is a topic upon which there is very large and growing

literature and about which there is a surprising lack of consensus. Many are

convinced that the panoply of regulations dealing with risky behavior is not

well conceived and the criticisms of rational choice under uncertainty

contribute to an understanding of this position. Recall that, broadly

speaking, people do not seem to do a very good job of appraising risky

outcomes. For example, they tend to overestimate the value of

low-probability, high-payoff gambles. And because of status-quo bias, they

prefer a known, high risk to an unknown, low risk. These imperfections in

the way people deal with risk may motivate them to demand legislative

regulation of risk that reflects their own, not entirely coherent, views. For

example, on an average day in the United States 30 people are killed on the

job, 56 are killed in accidents in the home, 133 die in automobile accidents

and 4,000 die from cancer. Of those who die each day from cancer, 30

percent of those deaths are attributable to tobacco; 4 percent are attributable

to cancers arising from occupational hazards; 1 percent to medical

treatment; and 2 percent to air and water pollution (Breyer, 1993). All other

things equal, these figures would suggest two predictions about current

regulations designed to minimize the harms from risky activities: first, that a

large amount of effort ought to be directed at reducing the risk of cancer and

second, that a large portion of the cancer-reducing effort ought to be directed

at tobacco-related cancers. Neither prediction is correct. Rather, the risk

regulation of the United States government has a willy-nilly aspect with

little rational regard for the value of the good it might be doing. For

instance, there is no single implicit value of a life saved that is used by the

federal government in regulating risk. Rather, the government’s regulations

imply that the value of a life saved ranges from $10,000 to $1 billion

(Viscusi, 1992).

 

Just bringing coherence to risk regulation would be a substantial

improvement in the efficient allocation of governmental resources. But the

experiments on decision making under uncertainty described in Section 8

suggest an important new way of looking at the regulation of risk.

Heretofore, much government risk regulation has been premised on the

belief that individuals make errors in dealing with risk because they do not

have correct information. If they had that information, they would make the

appropriate maximizing decision. Thus, the government’s role ought

principally to be to disseminate accurate information to assist individuals

and organizations in their decision making. But the material on

intertemporal choice in Part C suggests that even if they had the appropriate

information, some people would not make the right decision about risky

activities.

 

How these insights should translate into a reform of risk regulation is a

very broad issue. Here I want only to suggest that they might lead to

principled justifications for far more paternalistic policies than those that

rational choice theory typically recommends. For instance, where rational

choice theory might suggest that the comparison of the costs and benefits of

wearing motorcycle helmets ought to be left to motorcyclists so long as they

are well-informed about the true costs and benefits, the findings about

mistakes in intertemporal choice and in the assessment of risk imply that

motorcyclists will always underestimate the benefits of wearing helmets and

that, therefore, the best regulation for minimizing head injuries among

motorcyclists may be one mandating helmet-wearing. These are significant

differences in policy and we must wait on further empirical work to clarify

the extent of the cognitive errors and the various policy choices before us.

 

13. Tort Law and the Criticisms of Rational Choice Theory

 

Finally, I come to the issue of whether the anomalies discussed in Part C

affect the economic analysis of tort law. The connection between those

anomalies and the economic analysis turns, I think, on this central issue: the

economic analysis perceives that potential victims and potential injurers are

capable of understanding and acting rationally in response to the

implications of the tort liability system for their choices about which

activities to pursue, how and when to pursue those activities, how cautious

they should be, how much they should spend on warnings of danger to

others and the like. If those whose behavior we seek to affect by imposing

tort liability do not have the cognitive abilities to understand and act in

accord with the law’s desires, then we should not be surprised to learn that

the tort liability system is not achieving its desired efficiency ends. For

example, if decision makers make systematic errors when faced with

uncertain outcomes or if they are systematically overconfident about their

abilities to avoid an accident or injury, then they may behave in ways that

are contrary to those anticipated by rational choice theory. The next sections

seek to clarify how these imperfections might influence several issues in the

economic analysis of tort liability.

 

13.1 The Choice Between Statutory Regulation and the Risk-Utility Test

There are two important points to be made about the efficiency of negligence

and of strict liability. First, within negligence there are two very different

means of determining whether someone had complied with a legal duty of

care. In one set of circumstances compliance is determined by comparing the

victim’s or injurers actions with a clear rule - for example, a speed limit or

manufacturing standard proposed by an administrative agency (or possibly

by some respected private standard-setting group). This sort of negligence

(negligence per se) is relatively easy for the court to determine and easy for

potential injurers and victims to perceive and to follow. No sophisticated

calculations are required and, therefore, the demands on the cognitive

abilities of the potential victim and injurer are not large.

 

The other, more common form of negligence delegates to potential

injurers and potential victims the determination of the appropriate amount of

care to take. There is no hard-and-fast rule specifying the suitable amount of

precaution; rather, each potential victim and potential injurer calculates

what is appropriate in the understanding that, in the event of an accident, a

court may check those calculations to see if they have been done reasonably.

This standard of due care is frequently determined according to a

‘risk-utility test’ or the Hand Test. The court assumes that the parties who

may injure or be injured compare the costs of precaution with the benefits of

taking precaution (the reduction in the probability and severity of an

accident) and that they will take all cost-justified precaution - that is,

precaution that confers greater (expected) benefits than it costs.

 

The cognitive demands of the risk-utility test are substantial. In order to

comply with the legal duties imposed by negligence, potential injurers and

victims must independently form an estimate of the probability of an

accident’s occurring as a function of the amount of precaution they take and

an estimate of the size of the accident losses that will result from various

levels of precaution. An implication of the experiments noted in Part C

above is that potential injurers and victims may make systematic errors in

these calculations.

 

The potential shortcomings of human decision making with regard to

risk may be the key to understanding when it is socially efficient to use the

rule-like negligence per se and the due-care standard. Put simply, if one

believes that those likely to be involved in a particular kind of accident are

prone to cognitive errors and limitations, then the more appropriate method

of achieving social efficiency (in the sense of minimizing the social costs of

accidents) would be to state rules with which it is relatively easy to comply.

 

13.2 Strict Liability and Negligence

Recall that the law and economics literature has identified one principal

factor that should figure in the efficient choice between negligence and strict

liability: whether precaution is unilateral or bilateral. The presence of

cognitive errors and limitations in the ability to perceive and act rationally

upon risk complicates this distinction between unilateral and bilateral

precaution. It is not unlikely that situations arise in which both the potential

injurer and the potential victim could have taken precaution that would

reduce the expected social costs of accidents but in which one of the two

parties was very much less likely to have experience with the sort of

calculations of risk and expectation that the economic theory supposes that

both parties have. That is, a cognitive limitation in dealing with uncertain

outcomes may be an independent factor in determining whether precaution

was unilateral or bilateral and, therefore, in choosing between negligence

and strict liability.

 

Consider, for example, product-related accidents. Suppose that we were

free to decide for the first time which form of liability to use in those

accidents and suppose further that we intend to use the economic theory

exclusively in reaching our decision. Which form of liability - negligence or

strict liability - ought we to use in order to minimize the social cost of

product-related accidents? We might conclude that precaution is bilateral:

producers can reduce the likelihood and severity of accidents by taking care

in the design and manufacture of their goods and by warning consumers of

any non-obvious dangers; consumers can reduce the expected social costs of

accidents by taking care in the use of the product, by following the

manufacturer’s instructions, by using the product in a manner that it was

intended that it be used and so on. But suppose that we make one more

assumption - namely, that producers have much greater facility in making

decisions about uncertain outcomes than do consumer because consumers

are prone to the sort of miscalculations that we noted in Part C. We might

now doubt that precaution is truly bilateral. Perfectly rational consumers

might be able to calculate the appropriate level of risk and the expected level

of accident costs, given different levels of precaution, but these are not, by

assumption, perfectly rational consumers. They will make errors;

importantly, they will make more and more costly errors than will

producers. If so, then a situation that assumed the affected parties to be

rational and that precaution was bilateral becomes one in which only one of

the parties is reliably rational and there is, therefore, unilateral precaution.

This makes out an argument for treating at least some product-related

accidents under the strict liability rule. To put the point more generally, I am

suggesting that the recognition that there may be cognitive limitations

among potential victims and injurers should alter the search for the

appropriate liability standard from one in which the law looks for the

least-cost avoider to one in which the law looks for the least-cost decision

maker or least irrational party.

 

14. Conclusion

 

We have seen how important rational choice theory is to law and economics.

But we have also seen that there is an increasing body of experimental work

that questions some of the assumptions of that theory. We must amend the

rational-choice model, but precisely how we should amend the model is not

yet clear. I want to conclude with a cautionary statement about the crucial

questions that must be addressed in undertaking these emendations in

rational choice theory and in drawing conclusions about law on the basis of

these emendations.

 

Some may mistakenly think that we are put to a stark choice between, on

the one hand, rational choice theory and, on the other hand, the extreme

position that no coherent theory of human decision making is possible. That

is a dangerous illusion. A synthesis is possible and is, I believe, coming. But

it is not yet here and until it is, we must remain uncomfortably in the middle

- somewhat skeptical about rational choice theory but not so skeptical that

we abandon that theory. To see the dangers of moving too far, too fast in the

application of Part C’s findings, consider the experiments that suggest that

cooperation in the provision of a public good is much more likely than

rational choice theory predicts. This is, so far, merely suggestive; it is not a

complete guide to behavior. Therefore, no one could responsibly use these

experiments as a warrant for cutting the public subsidies for basic research

and public television or for laxer enforcement of the intellectual property

laws. Before we make policy pronouncements on the basis of these

anomalies, we need to know much more. The implication of some of the

experimenters is that their findings apply to all decision makers in all

circumstances. But that seems highly unlikely. Surely there are important

differences among circumstances and among people. There may be some

people who always obey the predictions of rational choice theory; there may

be some circumstances in which no one obeys those predictions. And there

may be more subtle differences. For example, are there systematic

differences in the dispensation to cooperate by age and gender? Are there

objective circumstances about the manner in which the cooperation is

solicited (for example, how long the people have known each other and

whether they are allowed to communicate) that lead to a greater likelihood

of cooperation? How robust is the finding that repeated playing leads to a

diminution of the propensity to cooperate? These and many more questions

need to be addressed.

 

Some day, perhaps soon, we shall have a complete account of human

decision making than that provided by rational choice theory. And when we

do, that account will greatly enhance our understanding of the law and our

ability to draft the law for desirable ends.

 

Acknowledgements

 

The author would like to thank an anonymous referee for helpful comments.

 

Bibliography on Rational Choice Theory in Law and Economics (0710)

Andreoni, James (1988), ‘Why Free Ride? Strategies and Learning in Public Goods Experiments’,

38 Journal of Public Economics, 291-305.

Becker, Gary (1962), ‘Irrational Behavior and Economic Theory’, 70 Journal of Political

Economy, 1-13.

Becker, Gary (1968) ‘Crime and Punishment: an Economic Approach’, 76 Journal of Political

Economy, 169-217.

Breyer, Stephen (1993), Breaking the Vicious Circle, Cambridge, MA, Harvard University Press.

Coase, Ronald A. (1960), ‘The Problem of Social Cost’, 3 Journal of Law and Economics, 1-28.

Cooter, Robert D. and Ulen, Thomas S. (1997), Law and Economics, (2nd edn), Reading, MA,

Addison-Wesley-Longman.

Dixit, Avinash and Nalebuff, Barry (1991), Thinking Strategically, New York, W.W. Norton &

Co.

Grether, David and Plott, Charles (1979), ‘Economic Theory of Choice and the Preference Reversal

Phenomenon’, 69 American Economic Review, 623-638.

Guth, Werner, Schmittberger, Rolf and Schwarze, Bernd (1982), ‘An Experimental Analysis of

Ultimatum Bargaining’, 3 Journal of Economic Behavior and Organization, 367- 383.

Kahneman, Daniel, Knetsch, Jack and Thaler, Richard (1986), ‘Fairness as a Constraint on Profit

Seeking: Entitlements in the Market’, 76 American Economic Review, 728-741.

Korobkin, Russell (1994), ‘Policymaking and the Offer/Ask Price Gap: Toward a Theory of

Efficient Entitlement Allocation’, 46 Stanford Law Review, 663-708.

Korobkin, Russell (1998), ‘The Status Quo Bias and Contract Default Rules’, 83 Cornell Law

Review, 608-687.

Lichtenstein, Sarah and Slovic, Paul (1971), ‘Reversals of Preference between Bids and Choices in

Gambling Decisions’, 89 Journal of Experimental Psychology, 46-61.

Liebenstein, Harvey (1950), ‘Bandwagon, Snob and Veblen Effects in the Theory of Consumers’

Demand’, 64 Quarterly Journal of Economics, 183-201.

Loewenstein, George and Elster, John (eds) (1992), Choice Over Time, New York, Russell Sage

Foundation.

Machina, Mark (1990), ‘Expected Utility Hypothesis’, in John Eatwell, Milgate, Murray and

Newman, Peter (eds), The New Palgrave: Utility and Probability, London, Macmillan

Publishing.

Marwell, Gerald and Ames, Ruth (1981), ‘Economists Free Ride, Does Anyone Else?’, 15 Journal

of Public Economics, 295-310.

Nozick, Robert (1993), The Nature of Rationality, Princeton, Princeton University Press.

Plous, Scott (1993), The Psychology of Judgment and Decision Making, New York,

McGraw-Hill.

Posner, Richard A. (1998), Economic Analysis of Law, (4th edn), Boston, MA, Aspen Publishing

Co.

Samuelson, William and Zeckhauser, Richard (1988), ‘Status Quo Bias in Decision Making’, 1

Journal of Risk and Uncertainty, 7-17.

Shefrin, Hersh and Thaler, Richard (1988), ‘The Behavioral Life-cycle Hypothesis’, 26 Economic

Inquiry, 609-643.

Thaler, Richard (1992), The Winner’s Curse, New York, W.W. Norton & Co.

Ulen, Thomas S. (1994), ‘Rational Choice Theory and the Economic Analysis of Law’, 19 Law and

Social Inquiry, 487-522.

Ulen, Thomas S. (1998), ‘The Growing Pains of Behavioral Law and Economics’, Vanderbilt Law

Review (forthcoming).

Viscusi, W. Kip (1992), Fatal Tradeoffs: Public and Private Responsibilities for Risk, New York,

Oxford University Press.

Wilson, James Q. and Abrahamse, Alan (1992), ‘Does Crime Pay?’, 9 Justice Quarterly, 359-377.

 


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 16:11:49浏览次数:2463
发表评论
名号:
内容:
验证: 6528
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