Title: An Introduction to Law and Economics: The Coase Theorem and Behavioral Economics
1An Introduction to Law and Economics The Coase
Theorem and Behavioral Economics
- Thomas S. Ulen
- University of Illinois College of Law
- University of Colorado Law School
- June 11, 2008
2Introduction
- Why all the fuss?
- A practical and useful innovation or one that is
confined to the academy? - A stalking horse for a particular political
ideology? - A (scientific) method of evaluating the impact of
law?
3Assumptions and premises
- People respond to incentives.
- Law is a method of ordering society to further
social goals. - Law creates incentives for people to behave in
certain ways. - Law can help people by facilitating their
achievement of their legitimate goals. - Economics provides both theoretical and empirical
techniques for examining laws likely and actual
effects in the world.
4Assumptions and premises 2
- So, to discourage a particular activity, law
should increase the price for engaging in that
activity. - E.g., by increasing the sanction for engaging in
undesirable behavior. - Or by increasing monitoring and enforcement of
the undesirable activity. - If legal sanctions increase or become more
likely, people will consume less of the
sanction-triggering activity.
5Coase Theorem
- Ronald A. Coase, The Problem of Social Cost, 3
J.L. Econ. 1 (1960). - If transaction costs are zero, bargaining will
lead to an efficient allocation of resources,
regardless of the law. - Law is only necessary to induce efficient
behavior when transaction costs are positive. - An implication assign legal entitlements to the
party who would have ended up with the
entitlement (that is, to the person who values it
the most).
6The Economics of Contract Law
- Law should seek to reduce the costs of concluding
and enforcing consensual agreements. - Law does so by providing a set of default and
mandatory rules that contracting parties can use
as a template for concluding an agreement. - Default rules.
- Mandatory rules.
7Contract Law 2
- Why do private parties need help in forming,
relying upon, and completing consensual
agreements? - Time-intensive transactionsi.e., those that take
time to complete. - Coordination, commitment, and cooperation.
- Risk allocation.
- Information exchange.
8The Economics of Tort Law
- Minimize the social costs of accidents.
- Precaution costs.
- Accident losses.
- Administrative costs.
- Continue to assume that potential victims and
injurers are rational decisionmakers.
9Economics of Tort Law II
- Tort law holds out the prospect of liability for
accident losses so as to - create incentives for parties to choose levels of
precaution and activities in which to participate
in order to minimize their liability, - And thereby to minimize the social costs of
accidents.
10Precaution costs
- How does a rational potential injurer decide how
much precaution to purchase? - Assume provisionally that the injurer will be
liable for the victims losses if there is an
accident. - But recognize that most parties are acting
behind a veil of ignorancethey do not know if
they will be a victim or an injurer. - A rational potential injurer takes all
cost-justified precautioni.e., precaution for
which the cost of the last unit of precaution
taken is just equal to the benefit provided by
that precaution.
11Precaution costs 2
- The expected benefit of a unit of precaution
equals the probability of an accidents occurring
times the anticipated accident losses. - Suppose that one more unit of care will reduce
the probability of an accidents occurring by
0.005 and that if an accident occurs, given that
amount of precaution, the losses are likely to be
100,000. - The expected benefit of that unit of precaution
is (0.005) x 100,000 500. - So, if that unit of precaution costs less than
500, society would like a potential injurer to
purchase the precaution because the cost is less
than the expected benefit.
12Additional topics in the economics of tort
liability
- Different tort liability standards.
- The relationship between administrative agency
safety regulation and tort liability. - Should regulatory compliance be a defense in a
private tort action? - Who are these rational people youre talking
about? - If injurers and victims are not fully rational,
then a situation that might seem to be one of
bilateral precaution may be, instead, one of
unilateral precaution.
13The Economics of Crime and Punishment
- Recall the assumption of rational decisionmakers.
- Gary Becker, Nobel Laureate in Economics, 1992.
14Crime and Punishment 2
- Beckers rational-choice theory of the decision
to commit a crime - Criminal compares the expected costs and expected
benefits of a crime. - Expected costs include the probability of
detection, arrest, and conviction times the value
of the sanction imposed. - Expected benefits include the monetary value of
the crime plus any non-monetary satisfaction the
criminal receives.
15Crime and Punishment 3
- A rational potential criminal commits the crime
if EB gt EC and refrains if EB lt EC. - Society can reduce crime by raising the expected
costs of crime. - Raise the probability of detection, arrest, and
conviction or, - Increase the criminal sanction or,
- Do both.
16Recent U.S. Statistics on Crime
- Since the mid- and late 1980s a decline in
non-violent crime. There is now less auto theft
in the U.S. than in much of Western Europe. - Property crimes down 30 percent in the 1990s.
- Since 1991 a precipitous decline in violent
crime, with homicide at the lowest level since
the 1930s. - Homicide rates down 40 percent in the 1990s.
17Why the Decline in Crime?
- Deterrence works.
- Increasing incarceration rates.
- More police and improved policing strategies.
- Decline in crack cocaine.
- A robust economy.
- Increased victim precaution.
- Alarms and security procedures.
- Faster and more effective trauma treatment.
- The legalization of abortion?
18Donohue Levitt, The Impact of Legalized
Abortion on Crime
- 116 Q. J. Econ. 379 (2001).
- Donohue and Levitt attribute half of the decline
in crime since 1991 to the legalization of
abortion in 1973. - Roe v. Wade led to a significant increase in the
number of abortions. (1.6 million per year by
1980 1 abortion per 2 live births.) Therefore,
relatively fewer 18 year-olds in the population
beginning in 1991. - All the other factors together account for the
other half of the decline.
19Donohue Levitt 2
- Whats the evidence for legalized abortions
effect on crime? - (1) Broad consistency with the prevailing
patternnamely, most crime is committed by 18-24
year-old males because of legalized abortion,
there are fewer 18 year-olds exactly 18 years
after Roe, and thats when the downturn in crime
began. - (2) Five states legalized abortion in 1970
(before Roe v. Wade), and they experienced a
decline in crime before the rest of the country
did.
20Donohue Levitt 3
- (3) Higher rates of abortion in a state in
the late 1970s and early 1980s are strongly
linked to lower crime in that state for the
period from 1985 to 1997. - (4) There is no relationship between abortion
rates in the mid-1970s and crime changes between
1972 and 1985. - (5) Almost all of the decline in crime in the
1990s can be attributed to reduction in crime
among the cohorts born after abortion
legislation there is little change in crime
among older cohorts over the last 30 years.
21Donohue Levitt 4
- The other hypotheses are unlikely to explain the
drop in crime in the 1990s. - The greater use of imprisonment, more police, and
changes in police strategies have been going on
for a long time. Its unlikely that they could
cause a sudden and sharp drop in crime just in
the 1990s. And the drop occurred in places, such
as Los Angeles, where there was no particular
improvement in the police force. - Similarly for the decline in the crack cocaine
trade. That was largely a phenomenon of major
urban areas. But the crime drop occurred not
just in major urban areas but everywhere. - The robust economy has been with us since the
early 1980s, not just in the 1990s. And,
moreover, there is a relatively weak correlation
between macroeconomic activity and crime levels.
Indeed, there is some evidence that much crime is
anti-cyclical, increasing when the economy is
doing well and declining when it is doing poorly.
22Donohue Levitt 5
- Donohue and Levitt identify two components that
make up the total effect that legalized abortion
had on crime - The cohort size effect.
- When the cohort reaches the late teensthe prime
years for committing crimes, there are fewer of
them and, therefore, less crime. - The cohort quality effect.
- Children born after abortion legalization may
on average have lower subsequent rates of
criminality. - (1) Women who have abortions are those most
at risk to give birth to children who would
engage in criminal activity. Teenagers,
unmarried women, and the economically
disadvantaged. - (2) Women may use abortion to optimize the
timing of childbearing. Through abortion women
may delay childbearing till later if their
current conditions are suboptimal. Children tend
to be born into better environments.
23Donohue Levitt 6
- Of the half of the drop in crime in the 1990s
that Donohue Levitt attribute to the
legalization of abortion, - about half of that total effect is attributable
to the cohort size effect and - about half to the cohort quality effect.
24New developments in law and economics
- Empirical research.
- Behavioral law and economics.
- The impact of law and economics on the legal
academy.
25Empirical law and economics 2
- Robert Ellickson, Of Coase and Cattle, 38 Stan.
L. Rev. 623 (1986), and Order Without Law (1991).
- Natural experiment damage done to property by
unsupervised cattle in Shasta County, California.
- In part of the county the owners of cattle were
responsible for damage done by their unsupervised
cattle. - In the other half of the county, owners were not
responsible.
26Ellickson and social norms
- Should there be a difference between the two
halves of the county in terms of the number of
cattle and other indicators of efficiency? - Not necessarily, if the Coase Theorem is true.
- There was no difference in behavior between the
two halves, even though the liability rules were
different. Why?
27Ellickson 2
- Not because neighbors were bargaining to the most
efficient result, regardless of the law. - Rather, because neighbors were not paying
attention to the law. - They sought to conform their behavior to the
prevailing social norms, not to the law.
28Behavioral law and economics
- Recall the close connection between rational
choice theory and traditional law and economics.
- Social and cognitive psychologists have found
some systematic deviations from the predictions
of RCT. - Taking these deviations into account in analyzing
law leads to changes in the economic analysis of
that flowed from RCT.
29Behavioral law and economics 2
- Consider four examples
- Endowment effect / status quo bias.
- The ultimatum bargaining game experiments.
- Loss aversion.
- Difficulties with probabilistic reasoning.
- In each case Ill seek to seek to show how these
empirical findings necessitate our amending some
settled conclusions of traditional law and
economics.
30Endowment effect / status quo bias
- People seem to place a very high value on the
things they have and the way things are. - Systematic difference between the
willingness-to-pay (WTP) price to acquire
something one doesnt have and the
willingness-to-accept (WTA) price to give up that
same thing if one already possesses it.
31Status quo bias 2
- WTA 2 WTP.
- Not experience-related.
- Applies to pens, coffee mugs, and other trivially
valuable items. - Implication
- Far more difficult to change the way things are
than one might anticipate. - Change may not just be a matter of transaction
costs.
32Status quo bias 3
33The ultimatum bargaining game.
- Two parties are to split 20.
- This is a pure cooperative surplus.
- The players are not allowed to talk or meet.
- Player 1 makes an offer for division of the 20.
- Player 2 can accept the offer, in which case they
each get the proposed division, or reject the
offer, in which case neither player receives
anything.
34Results of the game
- The game has been played in over 100 countries
with thousands of players of all ages and
socio-economic circumstances. - The modal result is a 50-50 split.
- An unexpected finding (although not necessarily
inconsistent with rational choice theory) is that
if Player 1 proposes a split that give him or her
more than 70 percent of the surplus, Player 2
almost always rejects the offer. - People have a strong sense of what is fair.
- Interestingly, if Player 1 is selected on some
seemingly meritorious criterion, Player 2 will
tolerate Player 1s receiving more of the surplus
than if Player 1 is selected randomly.
35Implications
- Perhaps we need not worry overly about how
parties divide up cooperative surpluses. They
seem to do it equitably. - But we need, perhaps, to pay attention to the
fact that overreaching can cause otherwise
mutually beneficial transactions to fail. - The Normative Hobbes Theorem.
36Loss aversion
- The standard social science theory of
decisionmaking under uncertainty is that of
subjective expected utility (SEU). - Individuals are thought to maximize expected
utility rather than expected value. - The difference arises from attitudes toward risk
risk-neutrality, risk-preferring, and
risk-aversion.
37Loss aversion 2
- Kahneman and Tversky found that most people are
risk-averse with respect to gains but
risk-seeking with respect to losses. - Option A 50 with certainty
- Option B 100 with probability 0.5 or 0 with
probability 0.5 - Same expected value.
- Most people prefer Option A.
38Loss aversion 3
- Option C -50 with certainty
- Option D -100 with probability 0.5 or 0 with
probability 0.5 - Same expected value.
- Most people prefer Option D.
- Implication In civil actions, defendants may be
less likely to settle than plaintiffs. - The standard law-and-economics theory of
litigation versus settlement is that trial almost
always results from mistaken and inconsistent
estimates of the likelihood of prevailing at
trial.
39Difficulties with probabilistic reasoning
- Many legal situations imply that decisionmakers
make probabilistic calculations. - Rational criminals are thought to compare
expected costs and expected benefits of crime. - Tortfeasors and victims are imagined to compare
precaution costs with expected liability. - Jurors engage in Bayesian updating in determining
liability in civil trials and guilt in criminal
trials.
40Probabilistic reasoning 2
- The Monty Hall (or Three Door) Problem
- You are a guest on Lets Make a Deal and are
invited to play the final prize game. - There are three doors on stage, each marked with
a number1, 2, or 3. - Behind one door is 60,000 in cash. Behind the
other two doors are goats. - Monty invites you to choose a door.
41Probabilistic reasoning 3
- Suppose that you choose Door 1.
- Monty turns to his assistant and asks her to open
Door 3, revealing a goat. Doors 1 and 2 remain
closed. - Now Monty says to you, Youve chosen Door 1.
That door and Door 2 remain closed. Would you
like to remain with Door 1 or would you like to
switch to Door 2? - What should you do?
42Probabilistic reasoning 4
- You should switch!
- Your chances of winning if you switch are 2/3.
43Why should you switch?
44Implications
- Decisionmakers may not do as able a job in
estimating probabilities as RCT assumes. - For example, they may overestimate their
abilities to avoid an accident. Therefore, there
may be too many accidents. - A possible corrective is to take safety decisions
away from individuals and place them on
manufacturers. Auto safety.