EC476 Contracts and Organizations, Part III: Lecture 4 Leonardo - - PowerPoint PPT Presentation
EC476 Contracts and Organizations, Part III: Lecture 4 Leonardo - - PowerPoint PPT Presentation
EC476 Contracts and Organizations, Part III: Lecture 4 Leonardo Felli 32L.G.06 2 February 2015 Benjamin Cardozo 1921 Law never is , but is always about to be. It is realized only when embodied in a judgment, and in being realized,
Benjamin Cardozo — 1921
“Law never is, but is always about to
- be. It is realized only when embodied
in a judgment, and in being realized,
- expires. There are no such things as
rules or principles: there are only isolated dooms. [...] [...] No doubt the ideal system, if it were attainable, would be a code at
- nce so flexible and so minute, as to
supply in advance for every conceivable situation the just and fitting rule. But life is too complex to bring the attainment of this ideal within the compass of human powers”.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 2 / 45
Common Law — 1066-1189
The English system of Common Law, originally pure Case Law was established during the 11th
- r 12th centuries.
In particular it was spread at the beginning of the reign of William the Conqueror, 1066. It was fully established by the end of the reign
- f Henry II, 1189.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 3 / 45
Justinian — AD 529
537 years earlier — Justinian had quite a different idea of what the Law ought to be. The “Corpus Juris Civilis” in English translation adds up to 5,818 pages. Table VIII — Law VIII — Where a road runs in a straight line, it shall be eight feet, and where it curves, it shall be sixteen feet in width.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 4 / 45
Code Napol´ eon — 1804
◮ “By virtue of its simplicity, my one code
was the source of more good in France than all the laws which preceded me.” (Bas-relief in Les Invalides, Paris.)
◮ The Code Napol´
eon (renamed the Civil Code) was promulgated in 1804.
◮ The Code Napoleon has served as the
model for the codes of law of more than twenty nations.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 5 / 45
Law of the Italian Republic — AD 2000
If the birth takes place during a railway trip, the declaration must be rendered to the railroad officer responsible for the train, who will draw a transcript of verbal declarations, as prescribed for birth certificates. Said railroad officer will hand over the transcript to the head of the railroad station where the train next stops. The head of such station will transmit the documents to the local registrar’s office to be appropriately recorded.
This is Article 40 of the regulations for registrar’s offices, issued as Decree Number 393 of November 3rd 2000 of the President of the Republic of Italy. Regulations issued to ensure the streamlining of procedures, as prescribed by Article 2, comma 12,
- f Law Number 15 of May 1997 of the Republic of Italy.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 6 / 45
Motivation
◮ There is a large empirical literature that shows that legal
- rigin matters (La Porta, Lopez-de-Silanes, Shleifer, and
Vishny 1997,1998, 1999, 2002, La Porta, Lopez-de-Silanes and Shleifer 2005, etc.) known as Law and Finance.
◮ Identify a stylized model of both Case Law and Statute Law. ◮ Compare the two systems and provide conditions under which
- ne legal system dominates the other.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 7 / 45
Stylized differences between legal systems
◮ Statute Law is a legislator-made Law: judges apply the
statute.
◮ In the Case Law, there is (implicit) delegation of the
law-making power to judges.
◮ In Case-law we have the rule of precedent (stare decisis) that
commits future judges to the decisions of past judges, but not in Statute Law.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 8 / 45
A Big Caveat
◮ Can we reinterpret this comparison in terms of Civil Law vs.
Common Law?
◮ To some extent, clearly yes (at least historically) ... but this is
controversial (Hatfield, 2006, Whitman 2006).
◮ Common Law countries have accumulated a lot of statutes
through time.
◮ In Civil Law judges have some leeway as well and precedents
play some role (especially in Germany).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 9 / 45
The Question
◮ Is the pragmatism of Case Law simply always superior to the
rigidity of Statute Law?
◮ Are there universes in which Statute Law is instead superior? ◮ Or is it the case that the ascent of Case Law is like a scientific
discovery?
◮ It just was not known before the 11th or 12th century, and
those who discovered it and started using it became unambiguously better off.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 10 / 45
Case Law vs Statute Law: A First Answer
◮ Premise: “[...] life is too complex [...]” Laws are bound to be
Incomplete (impossible to fully describe/conceive ex-ante).
◮ Leaving things to the Courts is a way to delay the decisions to
when more detailed information is available: ex-post.
◮ What are the Courts’ objectives? They are “ruling over
- thers” so we should be sure they have “social welfare” in
mind.
◮ If the Courts’ motives are “right” then the Case Law regime
- ught to be superior.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 11 / 45
Punch-Line
◮ In some cases Statute Law is instead superior. ◮ This is because the Case Law advantage — ex-post decisions
— may also be a disadvantage.
◮ Case Law Courts may suffer from time inconsistency (present
bias) due to the timing of their decisions.
◮ In a dynamic framework this time-inconsistency problem is
unavoidable.
◮ When cases are sufficiently homogeneous, the draw-backs of
incompleteness (Statute Law) are outweighed by the costs of time-inconsistency (Case Law).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 12 / 45
Laws
◮ With “fully contingent” (complete) Laws, “[...] a code at
- nce so flexible and so minute, as to supply in advance for
every conceivabe situation the just and fitting rule [...]” the Statute Law is clearly the ideal regime.
◮ “[...] But life is too complex [...]” hence laws are not flexible,
at least not as much as Courts decisions that occur after the fact and hence can depend on the details of the realized contingency.
◮ Statutes are incomplete laws: fix rules that do not depend on
the realized contingencies.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 13 / 45
Time Inconsistent Courts: Examples
◮ Patent policy: wide breadth is optimal ex ante to give
incentives, but inefficient ex post.
◮ Investor Protection: ex ante borrowers are in favor of strong
investor protection to have cheap credit, but not ex post.
◮ Disclosure: Court’s decision may induce the parties to disclose
their private information at the ex-ante stage at the cost of preventing profitable ex-post trades.
◮ Three strikes you are out: long Sentences deters crime ex
- ante. Ex post, the deterrent effect does not play any role.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 14 / 45
Preliminaries
◮ Infinite-period-lived Courts and sets of one-period-lived
parties.
◮ Courts can either take a weak decision (W) or a tough
decision (T ).
◮ Parties take private decisions based on the Court’s expected
ruling (in equilibrium, perfect foresight).
◮ Two states of natures : the complex state C (with prob. ρ)
and the simple state S (with prob. 1 − ρ).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 15 / 45
Simple Environment S, probability (1 − ρ)
◮ Π denotes Court’s ex-ante payoffs (social welfare),
Π denotes Court’s ex-post payoffs.
◮ In environment S it is optimal for the Court to choose W
ex-ante: Π(S, T ) < Π(S, W)
◮ In environment S it is also optimal for the Court to choose W
ex post:
- Π(S, T )
<
- Π(S, W)
◮ Clearly in this environment there exists no time inconsistency.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 16 / 45
Complex Environment C, probability ρ
◮ In environment C it is optimal for the Court to commit
ex-ante to choose T Π(C, T ) > Π(C, W)
◮ However, ex post, after private decisions have been made, it is
- ptimal for the Court to choose W
- Π(C, T )
<
- Π(C, W)
◮ This is the source of the Court’s time inconsistency. ◮ The present-bias temptation is therefore
- Π(C, W) −
Π(C, T ) > 0.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 17 / 45
Setup
◮ At the beginning of each period Nature:
◮ determines whether the environment is C or S with probability
ρ, 1 − ρ,
◮ in the Case Law regime draws ℓS and ℓC (more below)
◮ The common discount factor of the Legislator and the Court
is δ < 1 (chance that the same case occurs next period).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 18 / 45
Case Law: Preliminaries
◮ Court can pick a different ruling in the two environments (S
- r C) unless it is constrained by precedents (stare decisis).
◮ Everyone (Parties and Court) knows whether precedents are
binding.
◮ Precedents are a quadruple J = (τS, ωS, τC, ωC) satisfying
the restrictions τS < ωS and τC < ωC. Let ω0
E = 0, τ 0 E = 1,
E ∈ {S, C}.
◮ At the beginning of each period a random variable is realized:
ℓS or ℓC, uniformly distributed over [0, 1].
◮ The, say, C Court is free to choose if ℓC ∈ (τC, ωC). It is
bound to W if ℓC ∈ [ωC, 1]. It is bound to T if ℓC ∈ [0, τC].
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 19 / 45
Case Law: Precedents
τC ωC 1 constrained to T constrained to W discretion
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 20 / 45
Case Law: Precedents (cont)
◮ In each period — Case Law Court takes precedents J t as
given and is either bound or not bound by them.
◮ The ruling of today’s (period t) Case Law Court, if it has
discretion, affects the body of precedents for tomorrow’s (period t + 1) Case Law Court: J t+1.
◮ Each Case Law Court — if it has discretion — can also
choose the breadth of its ruling bt ∈ {0, 1}.
◮ Narrow Ruling bt = 0 ⇒ Small effect on precedents. ◮ Broad Ruling bt = 1 ⇒ Larger effects on precedents. Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 21 / 45
Benchmark Case: No Rule of Precedents
Assume first no rule of precedents: J t+1 does not depend on (J t, bt). Case Law Court’s problem if it has discretion: max
Rt∈{T ,W}(1 − δ)
Π(Et, Rt) + δ
- Zt+1(J t+1, σ, ρ)
- Lemma
In the absence of the rule of precedents the Case Law court chooses Rt = W for every E ∈ {S, C} and every t. Case Law does not evolve: the Case Law Court is completely myopic.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 22 / 45
Assumptions on precedents technology J t+1(J t, bt)
Assumption
◮ Case Law Court always has discretion:
ωt
S−τ t S > 0, ωt C−τ t C > 0 ⇒ ωt+1 S
−τ t+1
S
> 0, ωt+1
C
−τ t+1
C
> 0.
◮ Case Law Courts can choose rulings that have no effect on
- precedents. Zero breadth: bt = 0 ⇒ J t = J t+1. Restrict
the holding to the facts of the case/ Summary order. “Our consideration is limited to the present circumstances, for the problem of equal protection in election processes generally presents many complexities.” Bush v. Gore (00-949). US Supreme Court Per Curiam
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 23 / 45
Case Law: Judge’s Problem
Case Law Court’s problem if it has discretion: max
Rt∈{T ,W},bt∈[0,1](1 − δ)
Π(Et, Rt) + δ
- Zt+1(J t+1(J t, bt), σ, ρ)
- Lemma
Expected welfare in Case Law is weakly monotonically increasing. Case Law evolves for the better (thanks to zero breadth) (Posner 1973, Rubin 1977).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 24 / 45
Evolving Case Law
Proposition
Consider any realized path of the Case Law equilibrium
- environment. Then there exists an integer m such that:
Along any such path the Court will rule T in environment C when it has discretion for a total of at most m times. After a finite number of periods the Case Law Court when in C reverts to take the weak decision W whenever it has discretion: Case Law eventually stops evolving.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 25 / 45
Case Law Evolution: Intuition
◮ Under Case Law, the Court wants to take the time
inconsistent decision (W) if it looks only at today’s case.
◮ If it looks forward, via precedents, it wants to take the “right”
decision (T in environment C).
◮ The effect it can have via precedents is marginal. ◮ What matters is the change in the body of precedents. ◮ The present-bias temptation is equal to the difference
- Π(C, W) −
Π(C, T ) > 0 (does not changes through time).
◮ Therefore, when the marginal effect becomes ”small” (DRS
and monotonicity), the Case Law will stop “evolving.”.
◮ At some point all Case Law Courts with discretion choose the
time-inconsistent decision W and zero breadth.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 26 / 45
Well Behaved Precedent Technology
Assumption
◮ The precedent technology—the cutoffs τE(τ, ωR, 1) and
ωE(τ, ω, R, 1)—are monotonic (continuous) in τ and ω, respectively, in both states E ∈ {S, C};
◮ The precedent technology satisfies “no cross-state effects,”
e.g. at t if the state is C then τ t
S = τ t+1 S
and ωt
S = ωt+1 S
;
◮ The precedent technology satisfies “reversibility,” e.g. if at t a
Court takes decision (T , b = 1) and at t + 1 decision (W, b = 1) then τ t
E = τ t+2 E
, ωt
E = ωt+2 E
;
◮ The cutoff τC(τ, ω, R, 1) satisfies DRS in τ and “independent
impact” from ω;
◮ The cutoffs τ t S and ωt C have “no interior asymptotes.”
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 27 / 45
How Does the Equilibrium Look like?
◮ We construct a Markov Perfect equilibrium (with state
variable J t).
◮ Consider first state S:
◮ the Court takes decision W whether it has discretion or not,
the breath b is indeterminate.
◮ The steady-state Court’s decision is W.
◮ Consider now state C:
◮ notice first that the Court’s decision (W, bt = 1) is strictly
dominated.
◮ We can focus on the two Court’s decisions:
(T , bt = 1) (W, bt = 0).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 28 / 45
How Does the Equilibrium Look like? (cont)
◮ There exist a unique threshold ¯
τ such that the Court is indifferent between the two decisions above: (1 − δ)[ Π(W) − Π(T )] = ρ δ[τC(¯ τ, T , 1) − ¯ τ] [Π(T ) − Π(W)]
◮ When (τ t C > ¯
τ), precedents do not move, (W, bt = 0), the system is in steady state.
◮ When (τ t C < ¯
τ), precedents move (they improve) towards a steady state, (T , bt = 1).
◮ The Markov perfect equilibrium cannot be in pure strategies.
If there exist a last period in which the Court chooses T and b = 1 procrastination is a profitable deviation for the Court.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 29 / 45
How Does the Equilibrium Look like? (cont)
τC 1 τ 0
C
s ❯ ❄
- τ
¯ τ steady state discretion
❃
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 30 / 45
How Does the Equilibrium Look like? (cont)
Proposition (Equilibrium Characterization)
There exists a Markov perfect equilibrium in mixed strategies in which the Court is indifferent at any t such that τ t
C < ¯
τ. In this case the Court randomizes with probability pt between (T , bt = 1) (evolution of precedents) and (W, bt = 0) (no evolution). At any t such that τ t
C > ¯
τ the Court chooses (W, bt = 0)(steady state). We interpret randomizing Court as a situation in which the body of precedents is ambiguous (incentives to transact and go to Court).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 31 / 45
Case Law: Welfare
ρ 1
✻ ✲
WCL WCL ¯ ρ W
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 32 / 45
Statute Law: The Problem
◮ By incompleteness of laws, Statute Law can select a single
ruling regardless of environment: either T or W.
◮ The optimal Statute Law solves:
max
R∈{T ,W} (1 − δ) ∞
- t=0
δt [(1 − ρ)Π(S, R) + ρΠ(C, R)]
◮ or
max
R∈{T ,W} [(1 − ρ)Π(S, R) + ρΠ(C, R)]
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 33 / 45
Statute Law: Characterization
Proposition
Choose W if S is more likely: ρ < ρ∗
SL.
Choose T if C is more likely: ρ > ρ∗
SL.
Where ρ∗
SL is such that:
(1 − ρ∗
SL) Π(S, W) + ρ∗ SL Π(C, W) =
= (1 − ρ∗
SL) Π(S, T ) + ρ∗ SL Π(C, T )
Denote the maximized welfare WSL(ρ).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 34 / 45
Statute Law: Welfare
ρ 1 ρ∗
SL
✻ ✲
WSL W
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 35 / 45
Statute Law vs. Case Law
Proposition (Statute Law Dominance)
Statute Law may dominate Case Law: There exists a ρ∗
CL ∈ (0, 1)
such that for every ρ ∈ (ρ∗
CL, 1] we have that
WSL(ρ) > WCL(J 0, ρ).
Proposition (Case Law Dominance)
Case Law may dominates Statute Law in a neighborhood of ρ∗
SL,
(for δ high enough): WSL(ρ∗
SL) < WCL(J 0, ρ∗ SL).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 36 / 45
Welfare under Statute Law and Case Law.
ρ 1
✻ ✲
WCL WCL ¯ ρ W
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 37 / 45
Welfare under Statute Law and Case Law.
ρ 1 ρ∗
SL
✻ ✲
¯ ρ W WSL = WCL WSL WCL ρ∗
CL
✗
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 37 / 45
Welfare under Statute Law and Case Law.
ρ 1 ρ∗
SL
✻ ✲
¯ ρ W WSL = WCL WSL WCL ρ∗
CL
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 37 / 45
Statute Law Dominance: Intuition
◮ The dominance of Statute Law is due to the lack of evolution
- f the Case Law.
◮ Eventually, Courts that have discretion take the present-biased
decision (W).
◮ This stops the evolution of case law implying that the parties’
welfare under case law is bounded away from the first best welfare.
◮ Clearly, time consistency problems have bite when the
environment is C. This happens a fraction ρ of the times.
◮ When ρ is close to one the “lack of flexibility” of Statute Law
matters less: Statute Law dominates.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 38 / 45
Case Law Dominance: Intuition
◮ Recall when ρ = ρ∗ SL the Statute Law legislators are indifferent
between W or T .
◮ The Parties’ expected welfare is
WSL(ρ∗
SL) = ρ∗ SL Π(C, W) + (1 − ρ∗ SL)Π(S, W) ◮ Consider state S: Common Law and Statute Law welfare
coincides.
◮ Consider state C: let
τ be the steady state cutoff, τ ≥ ¯ τ. The Parties’ steady state expected welfare is: WCL(J 0, ρ∗
SL) = ρ∗ SL [
τ Π(C, T )+(1− τ)Π(C, W)]+(1−ρ∗
SL)Π(S, W) ◮ Clearly: [
τ Π(C, T ) + (1 − τ)Π(C, W)] > Π(C, W)
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 39 / 45
Some Evidence...
◮ Niblett, Posner, and Shleifer (2010) investigate the
application of the ”Economic Loss Rule” (ELR) in 465 appellate decisions over 35 years.
◮ Economic Loss Rule implies that one cannot sue in tort for
“economic loss” unless that loss is accompanied by personal injury or property damage (with standard exceptions, fraud).
◮ ”Tort law is a superfluous and inapt tool for resolving purely
commercial disputes. We have a body of law designed for such disputes. It is called contract law.” (902 F.2d 573, 574 , 7th Cir. 1990).
◮ ELR encourages parties to solve their potential problems
through contracts. It is widely recognized as the (ex-ante) efficient rule.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 40 / 45
Some Evidence... (cont.)
◮ However, ex-post the judge may have sympathy for wronged
plaintiffs (for ex. because warranty has expired) and make exceptions to the ELR.
◮ The initial twenty years after the Seely decision are best
described as years of growing acceptance of the ELR.
◮ In the final decade of the sample, however, courts moved away
from strict application of the ELR and invoked more and more idiosyncratic exceptions.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 41 / 45
Percentage of Tort wins across states
.2 .4 .6 .8 1 % of cases 1970 1980 1990 2000 2010 Year California Florida Illinois New York Ohio
Figure 5.3: Tortwin over time in the five states with the highest caseloads
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 42 / 45
Justinian
The Appian Way was built approximately in 300 BC and was fully replaced 2,084 years later in AD 1784. A “slow changing” environment: a very high (or very low) ρ above. The Statute Law regime is likely to be superior.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 43 / 45
Cardozo
◮ Edison’s “improved” stock ticker was introduced in 1871, the
last mechanical stock ticker was introduced in 1960, the wireless device in 1996.
◮ A fast changing environment: intermediate values, ρ = ρ∗ SL.
(La Porta et al, Law and Finance — but Rajan and Zingales 2003 and Lamoreaux and Rosenthal 2005.).
◮ Statute Law is likely to be inferior because of its lack of
flexibility.
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 44 / 45
Cardozo: pessimism
◮ Cardozo was also pessimistic
“[...] there are only isolated dooms.”
◮ In our framework, this statement can be
reinterpreted as: Case Law is “doomed” to become mature, and after maturing to succumb to the time-inconsistency problem.
◮ Case Law improves through time (zero
breadth), but it does not reach full
- ptimality (Posner, 1973, 1990, 2003,
Rubin 1977, Gennaioli and Shleifer 2005).
Leonardo Felli (LSE) EC476 Contracts and Organizations, Part III 2 February 2015 45 / 45