ArticlePDF Available

Abstract and Figures

This Article explores the economic nature of law and courts as an explanation for the world’s endemic court congestion problem. The economic theory of goods and services is used to demonstrate that law has a dual nature—coercion and compliance—and that law as coercion is actually a club good that requires a complementary good to be useful, courts. But because courts are private goods in nature, the bundled product will behave as a private good. However, the unrestricted implementation of access-to-justice policies with the objective of increasing the people’s access to courts will transform the bundled product into a common pool resource. The counterintuitive result of this transformation is that granting unrestricted access to justice might actually prevent people from accessing their rights—the tragedy of the judiciary. Two policy implications are explored: The importance of legal certainty for the tragedy mitigation, and the potentially adverse selection problem resulting from court congestion.
Content may be subject to copyright.
ARTICLE
The Tragedy of the Judiciary: An Inquiry into
the Economic Nature of Law and Courts
Ivo Teixeira Gico Jr*
(Accepted 05 August 2019)
Abstract
This Article explores the economic nature of law and courts as an explanation for the worlds endemic
court congestion problem. The economic theory of goods and services is used to demonstrate that law
has a dual naturecoercion and complianceand that law as coercion is actually a club good that requires
a complementary good to be useful, courts. But because courts are private goods in nature, the bundled
product will behave as a private good. However, the unrestricted implementation of access-to-justice pol-
icies with the objective of increasing the peoples access to courts will transform the bundled product into a
common pool resource. The counterintuitive result of this transformation is that granting unrestricted
access to justice might actually prevent people from accessing their rightsthe tragedy of the judiciary.
Two policy implications are explored: The importance of legal certainty for the tragedy mitigation, and the
potentially adverse selection problem resulting from court congestion.
Keywords: Court congestion; legal theory; nature of law; nature of courts; tragedy of the judiciary
A. Introduction
For who would bear the whips and scorns of time,
Thoppressors wrong, the proud mans contumely,
The pangs of despised love, the laws delay
...
When he himself might his quietus make
With a bare bodkin?1
*Professor Ivo Teixeira Gico, Jr. teaches at the Law Department of Centro de Ensino Unificado de BrasíliaUniCeuB,
Brazil (e-mail: gico@ghdadvogados.com.br). Prof. Gico Jr. does research in Contracts, Antitrust, Regulation, Civil
Procedure, Legal Theory and Law & Economics. His current project is The Judiciary and the Rule of Law.The author thanks
Bernardo Mueller, Luciana Luk-Tai Yeung, Bruno Salama, Danielle Cristina Lanius, Tom Ginsburg, William Hubbard,
Carolina Sales Cabral Arlota, John Leubsdorf, Mariana Mota Prado, the participants at the Research Workshop at the
Law School of Fundação Getúlio Vargas Rio, Scholars Colloquium at the Summer Institute in Law and Economics at
the University of Chicago of 2018 and the German Law & Economics Association Annual Meeting of 2019, and the blind
review referees for helpful comments and suggestions.
© The Author(s), 2020. Published by Cambridge University Press on behalf of the German Law Journal. This is an Open Access article,
distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unre-
stricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
1W
ILLIAM
S
HAKESPEARE
,H
AMLET
act 3, sc. 1.
German Law Journal (2020), 21, pp. 644673
doi:10.1017/glj.2020.34
The judiciary is an institutional technology developed over thousands of years of human exper-
imentation for one sole purpose: To solve disputes by enforcing the rules.2,3A well-functioning
judiciary is essential to the development of any nation. From a public perspective, the existence
of an impartial dispute settlement mechanism allows political groups to reach ex ante compro-
mise solutionsthat is, lawsthat, if violated, will be duly imposed ex post by this external
mechanismenforcement. From a private perspective, not only does the judiciary protect
citizensrights against possible violations by the government itselfchecks and balances
and others, but it also allows individuals to cooperate with one another to achieve their private
goals through the generation of credible commitments called contracts. The institutional func-
tion of the judiciary, however, requires that any right violation be corrected in a timely manner.
In this context, court congestion is a socioeconomic problem that reduces the effectiveness of
the judiciary as a mechanism for fostering cooperation, development,4and in the long run, the
coordination value of law itself.
Almost anywhere in the world, it is commonplace to say that the judiciary is in crisis.5In many
countries, the judiciary is considered slow, inefficient, and expensive.6Numerous reforms have
been implemented to try to expedite courts,7and many more are currently underway.8But
the results have thus far not been quite satisfactory and it is fair to say that the caseload and court
2The question: Who makes the rules?from a political science approach or, more traditionally in the legal literature: What
are the sources of law?is a different matter.
3There is some debate over whether the adjudicatory system also serves other functions, such as social controllocally
enforcing legal rules formulated by a central governmentor rule creation. See, e.g.,M
ARTIN
S
HAPIRO
,C
OURTS
:
AC
OMPARATIVE AND
P
OLITICAL
A
NALYSIS
(1981). It is clear to me that courts do exercise other functions in some countries,
but the only universal trait that is present in every single adjudicatory system is the dispute resolutionone. The other functions
are contingencies. From a political science perspective, the social control function is a feature of the legal system as a whole
of which the courts are solely an elementand the idea of social control is embedded in the condition enforcing the rules.
As for the legislative function, even in common law systems, the rules created by courtsprecedentsare actually only
byproducts of their dispute resolution activities, and the legislative function is normally restricted by statutes and requires
the emergence of a dispute. Ordinarily, apart from constitutional review, precedents cannot overrule statutes. Even in the
exceptional case in which a court may have suo moto jurisdiction, as in Pakistan, the suo moto jurisdiction is still limited
to the dispute in question and the court cannot issue legislationmeaning rule-makingwithout a dispute. See generally
Asher A. Qazi, Suo Motu:Choosing Not to Legislate,in P
OLITICS AND
J
URISPRUDENCE OF THE
C
HAUDHRY
C
OURT
(2015).
4For example, there is some evidence that court delays increase crime, as some criminals are either sensitive to the dis-
counting process of punishment, are aware of the probability of prescription, or both. See generally Lucia Dalla Pellegrina,
Court Delays and Crime Deterrence: An Application to Crimes Against Property in Italy,26E
UR
. J.L. & E
CON
. 267 (2008).
5See generally M
ARIA
D
AKOLIAS
,C
OURT
P
ERFORMANCE AROUND THE
W
ORLD
:AC
OMPARATIVE
P
ERSPECTIVE
(1999); see
also H
ÉCTOR
F
IX
-F
IERRO
,C
OURTS
,J
USTICE AND
E
FFICIENCY
:AS
OCIO
-L
EGAL
S
TUDY OF
E
CONOMIC
R
ATIONALITY IN
A
DJUDICATION
(2003); see also C.H. V
AN
R
HEE
&F
U
Y
ULIN
,C
IVIL
L
ITIGATION IN
C
HINA AND
E
UROPE
:E
SSAYS ON THE
R
OLE OF THE
J
UDGE AND THE
P
ARTIES
(2014).
6A report from the Justice Studies Center of the Americas concluded that the current civil system prevalent in Latin
America and most of Continental Europe created a procedure that is too slow, formal and bureaucratic, incorporated in
a judicial expedient as the center of the procedure and of the decisions [and this caused] other problems, like the excessive
delay of cases and their lack of transparency.See C
AROLINA
V
ILLADIEGO
B
URBANO
,B
ASES
G
ENERALES PARA UNA
R
EFORMA A
LA
J
USTICIA
C
IVIL EN
A
MÉRICA
L
ATINA Y EL
C
ARIBE
4 (2009).
7The oldest judicial reform I have heard about was conducted by the Sumerian King Shulgi of Ur, son of King Ur-Namma,
between 209447 B.C.E. See M
ARTHA
T
OBI
R
OTH
,L
AW
C
OLLECTIONS FROM
M
ESOPOTAMIA AND
A
SIA
M
INOR
13 (1997). See
generally P
IOTR
M
ICHALOWSKI
& C.B.F. W
ALKER
,AN
EW
S
UMERIAN
L
AW
C
ODE
(1989). We have been trying to improve
upon that ever since.
8One can easily find continuous reforms of the judicial system around the world, normally for the same reasons. For exam-
ple, eighteen countries from the European Union adopted some form of procedural reform in 2017, while nineteen others were
negotiating further reforms. See E
UROPEAN
C
OMMISSION
,T
HE
2018 EU J
USTICE
S
COREBOARD
4 (2018). One can identify
recent reforms in countries such as Brazilwhich just adopted a new civil procedure codeor reports such as Scot.
Govt, Making Justice Work: Courts Reform (Scotland) Bill A Consultation Paper (2013), https://www2.gov.scot/
Resource/0041/00415373.pdf, and Benjamin Perrin & Richard Audas, Report Card on the Criminal Justice System #2,
M
ACDONALD
-L
AURIER
I
NST
., (2016), https://www.macdonaldlaurier.ca/report-card-criminal-justice-system/ (last visited
Apr. 8, 2020) (Evaluating Canadas justice deficit).
German Law Journal 645
congestion may be rising in many jurisdictions. Numerous reasons have already been offered
to explain court congestion: Lack of resources,9complex procedures,10 lawyersincentives,11
judgesincentives,12 poor management,13 culture,14 bad laws,15 and an excessive number of laws,16
among others. Efforts have been made to first try to identify the relevance of each of these possible
explanations and then to solve them.17 Nevertheless, until now, the idea that the very nature of law
and the adjudicatory system may contribute to the congestion problem has not been fully
explored. This very idea is what I plan to explore in this Article.
The knowledge of the economic nature of law and courts is an important step toward fully
understanding the behavior of adjudicative systems around the world and may be a substantial
step toward a positive economic theory of law. Until now, the literature has generally conceived
law as a public good. In this Article I would like to diverge from this literature to show that law has
a dual naturecoercion and complianceand that law as coercion is actually a club good that
requires a complementary good to be useful, courts. But because courts are private goods in
nature, the bundled product will behave as a private good. However, the unrestricted implemen-
tation of access-to-justice policies with the objective of increasing the peoples access to courts will
transform the bundled product into a common pool resource. The counterintuitive result of this
transformation is that granting unrestricted access to the judicial system might actually prevent
people from accessing their rights.
It is no coincidence that, in parallel with the increase in the demand for public adjudicative
services, many countries have also been adopting reforms that have substantially decreased the
costs of utilizing the judiciary to solve all types of disputes, including procedures simplification,
creation of small-claims courts, and implementation of free public defendersschemes, among
others. Nonetheless, access to public adjudicating services continues to be limited in practice
and court congestion remains an endemic problem. Unexpectedly, the inquiry conducted in this
Article predicts that this condition is the natural result of those policies due to the economic
nature of law, courts and the resulting bundled product.
After this brief introduction, Section B.I initiates the discussion by exploring the economic
nature of law using the traditional economic theory of goods and services to recognize the duality
of laws nature. Section B.II discusses that law and courts are complementary goodsa bundled
productwhile Section B.III shows the private nature of courts. Once the economic nature of law
and courts is established, Section B.III.1 explains that access-to-justice policies will transform the
economic nature of courts into a common-pool resource, and that such a transformation will
result in its overusehence, utter court congestionwhat I call the tragedy of the judiciary.
Two policy implications are then explored: Section Cexplores the relationship between the
9For example, the Task Force on Administration of Justice stated, The causes of delay are manifold: lack of resources,
inefficient management and an increasing number of cases to be decided.See O
FFICE OF
J
USTICE
P
ROGRAMS
,N
AT
L
C
RIM
.J
UST
.R
EFERENCE
S
ERV
., Task Force Report: The Courts, at 80 (1967) [hereinafter O
FFICE OF
J
USTICE
P
ROGRAMS
].
10See generally S
IMEON
D
JANKOV ET AL
., C
OURTS
453517 (2003).
11See generally D
ANIELA
M
ARCHESI
,L
ITIGANTI
,
AVVOCATI E MAGISTRATE
D
IRITTO ED ECONOMIA DEL PROCESSO CIVILE
(2003).
12See generally Deborah Beim, Tom S. Clark & John W. Patty, Why Do Courts Delay?, 5 J.L. & C
TS
. 199 (2017).
13See generally E
DGARDO
B
USCAGLIA
&M
ARIA
D
AKOLIAS
,J
UDICIAL
R
EFORM IN
L
ATIN
A
MERICAN
C
OURTS
:T
HE
E
XPERIENCE
IN
A
RGENTINA AND
E
CUADOR
(1996).
14See generally Tom Ginsburg & Glenn Hoetker, The Unreluctant Litigant? An Empirical Analysis of Japans Turn to
Litigation,35J.L
EGAL
S
TUD
. 31 (2006).
15See Thomas F. Hargis, The Laws Delay, 140 N
ORTH
A
M
.R
EV
. 309 (1885) (stating, more than 100 years ago, that
[t]he principal source of the laws delay is the laws defects, originating immediately in the venality, neglect, or incapacity
of legislators); see also Lode Vereeck & Manuela Mühl, An Economic Theory of Court Delay,10E
UR
. J.L. & E
CON
. 243 (2000)
(discussing the quality of laws as an explanation for court delay).
16See generally Giuseppe Di Vita, Production of Laws and Delays in Court Decisions,30I
NT
L
R
EV
.L.&E
CON
. 276 (2010).
17See generally R
ICHARD
E. M
ESSICK
,J
UDICIAL
R
EFORM AND
E
CONOMIC
D
EVELOPMENT
:AS
URVEY OF THE
I
SSUES
(1999).
646 Ivo Teixeira Gico
tragedy of the judiciary and legal uncertainty, and Section Dexplores the potential adverse selec-
tion problem resulting from court congestion. Brief conclusions follow in Section E.
B. The Economic Nature of Law and Courts
I. The Economic Nature of Law
Because lawyers and economists frequently use the same words with different meanings, it is use-
ful to have some conceptual uniformity. For a long time in the economic literature, the basic dis-
cussion about public versus private goods has been an endeavor about distinguishing between
what should be the province of markets and what should be the province of governments. If some-
thing were called a private good, it should be provided by markets. If something else were called a
public good, it should be provided by governments.18 Yet, our comprehension of the attributes of
goods or services has evolved toward encompassing more than merely the rivalry attribute iden-
tified by Samuelson 19 to also include a second attribute called excludability.20
In this modern approach, access excludability refers to the possibility for the possessor of a
certain good or provider of a certain service to exclude others, at low costs, from enjoying the
good or service provided. On the one hand, a computer is an exclusive good because its possessor
can prevent others from using it at low costs. On the other hand, national security is a non-
exclusive good because it is not feasible to exclude other people from enjoying it once it has been
provided. It is widely accepted that excludability is a requirement for the market provision of any
good or service, because it allows for the provider to deny access to the prospective consumer
unless both parties can reach mutually agreeable terms. It is the excludability feature that imposes
upon one party the obligation to come up with something of interest for the other party so that the
exchange may voluntarily occur; in other words, it is the excludability feature that makes the
creation of a market possible.
Alternatively, rivalry occurs when consumption by one person either substantially prevents the
same good from being consumed by another or otherwise significantly diminishes its utility. An
apple and a glass of water are rival goods because their consumption by one individual prevents
another from fully enjoying them. On the one hand, a good that is not eligible for consumption
by someone else would be considered a rival good. On the other hand, if a good may be enjoyed by
more than one person without substantially reducing the possibility of that goods enjoyment by
another, then it would be considered a non-rival good. National security and weather forecasts are
examples of non-rival goods, because the consumption of these services by one individual does not
preclude others from enjoying the same benefits, but rather they remain available for use by others
in substantially undiminished quantity and quality.
Although we normally talk about these two characteristics in binary terms to make the discus-
sion easier, in fact, they vary significantly in degree from one extreme to the other in accordance
with the good or service and the actual context. In any case, excludability and rivalry are two
independent characteristics that can be combined to create four different types of goods:
(1) Public goods are, simultaneously, non-rival and inclusive (non-exclusive);21 (ii) private goods
are opposed to public goods as they are rival and exclusive; (3) common pool resources (CPRs)
18See, e.g.,R
ICHARD
A. M
USGRAVE
,T
HE
T
HEORY OF
P
UBLIC
F
INANCE
:AS
TUDY IN
P
UBLIC
E
CONOMY
(1959).
19See Paul A. Samuelson, The Pure Theory of Public Expenditure,36R
EV
.E
CON
.&S
TAT
. 387, 38788 (1954).
20See James M. Buchanan, An Economic Theory of Clubs,32E
CONOMICA
1 (1965); see also Vincent Ostrom & Elinor
Ostrom, Public Goods and Public Choices,in P
OLYCENTRICITY AND
L
OCAL
P
UBLIC
E
CONOMIES
:R
EADINGS FROM THE
W
ORKSHOP IN
P
OLITICAL
T
HEORY AND
P
OLICY
A
NALYSIS
,75103 (Michael D. McGinnis ed., 1999).
21It is important to notice that, although it is commonly accepted that public goods are both non-rival and non-excludable,
some technical books still define a public good by Samuelsons traditionconsidering only the rivalry attribute. See, e.g.,
A
NDREU
M
AS
-C
OLLEL ET AL
., M
ICROECONOMIC
T
HEORY
350 (1995); see also S
TIGLITZ
&R
OSENGARD
,E
CONOMICS OF THE
P
UBLIC
S
ECTOR
10203 (2015) (stating that public goods are characterized by non-rival consumption and non-excludability,
only to state in the following passage that public goods can either be excludable or non-excludable).
German Law Journal 647
share the non-exclusivity of public goods, but their consumption substantially diminishes utility
for other users, rendering them rivals in use, as are private goods; and (4) club or toll goods are
exclusive as are private goods, but are non-rivals as are public goods.22 For clarity purposes, we can
summarize this classification as demonstrated in Figure 1.
Now the traditional discussion regarding the provision of goods by governments or markets
becomes even more nuanced. It is relatively clear that private goods can be efficiently provided by
markets. Efficiently,in this context, means only thatin a perfectly competitive marketthis
kind of good would be provided for all those who would be willing to pay the costs for the pro-
vision of the good. It is also clear that market mechanisms can be used to provide club goods, as
the provider may easily demand a price for access to the service or good.23 The good may still be
enjoyed collectively, although only by those who actually paid the toll, unless the provider decided
to offer the good for free and bear the costs of provision alone.
Having equalized terms and concepts, we can move on to explain how the application of the
economic theory of private and public goods may illuminate our search for the economic nature of
law and courts and assist our understanding of legal phenomena.
The question regarding what the law is, or rather, the definition of law, is a recurring issue in
legal philosophy that has spurred all kinds of answers.24 It is not the purpose of this Article to
tackle the issue from a philosophical perspective. Instead, this Article aims to use the public
and private good theory to try and better understand both the economic nature of what we nor-
mally call lawand the economic nature of accompanying adjudicative systems, in order to
advance toward a positive economic theory of law.
The majority of the discussion on the economic nature of law is primarily, in effect, just an
argument within the broader debate about whether an anarchist society is viable,25 or instead
whether a government should be created and called upon to impose order. For those who argue
Figure 1. Economic Types of Goods.
22See Buchanan, supra note 20. Another way of approaching the issue is to say that pure public goods have no upper bound
to club size, while club goodsimpure public goodsdo have an optimal club size.
23One can say that charging for a non-rival good may be inefficient because this action results in underconsumptionprice
higher than the marginal cost. Yet not charging for a non-rival good will lead to no offer or not enough offerundersupply. If
its provision is funded with a tax, then you will be charging for it anyway, just not allocating the burden on the actual users of
the service, as not all taxpayers may enjoy the service.
24See, e.g.,J
OHN
A
USTIN
,T
HE
P
ROVINCE OF
J
URISPRUDENCE
D
ETERMINED
(1995); Oliver Wendell Holmes Jr., The Path of
the Law,10H
ARV
.L.R
EV
. 457 (1897); H
ERBERT
L. A. H
ART
,T
HE
C
ONCEPT OF
L
AW
(Oxford University Press 2d ed. 2012)
(1961); H
ANS
K
ELSEN
,P
URE
T
HEORY OF
L
AW
(Max Knight trans., The Lawbook Exchange 2d ed. 2005) (1967); R
ONALD
D
WORKIN
,T
AKING
R
IGHTS
S
ERIOUSLY
(1977); N
EIL
M
AC
C
ORMICK
,L
EGAL
R
EASONING AND
L
EGAL
T
HEORY
(1978); J
OSEPH
R
AZ
,T
HE
C
ONCEPT OF A
L
EGAL
S
YSTEM
:A
N
I
NTRODUCTION TO THE
T
HEORY OF
L
EGAL
S
YSTEM
(2d ed. 1980); R
OBERT
A
LEXY
,C
ONCEITO E
V
ALIDADE DO
D
IREITO
(2005); F
REDERICK
S
CHAUER
,T
HE
F
ORCE OF
L
AW
(2015). In any case, it is rea-
sonable to say that the inquiry into the nature of law has essentially been an inquiry into the three questions posed by Hart:
(1) How does law differ from and how is it related to orders backed by threats?; (2) how does legal obligation differ from, and
how is it related to, moral obligation?; and (3) what are rules and to what extent is law an affair of rules? I intend to answer
none of these questions here.
25See, e.g., David Friedmann, Private Creation and Enforcement of Law: A Historical Case, 8 J.L. S
TUD
. 399 (1979); B
RUCE
L.
B
ENSON
,T
HE
E
NTERPRISE OF
L
AW
:J
USTICE
W
ITHOUT THE
S
TATE
(Independent Institute 2d ed. 2011) (1990); Tyler Cowen,
Law as a Public Good: The Economics of Anarchy,8E
CON
.&P
HIL
. 249 (1992).
648 Ivo Teixeira Gico
in favor of the public provision of lawmeaning its creation and enforcement by government
law is considered a public good. For example, Buchanan used the public and private good theory
with Samuelsons traditional approach to affirm that, since law is a kind of heterodox public
goodan externalityit would not be provided by markets.26
Expressly mentioning Buchanans work, Landes and Posner27 also argued that law is a public
good, but their argument to support this claim entailed that precedentsnot necessarily law
could be used by one person without substantially diminishing its utility as a precedent to another
person. We should observe here that the argument is based solely upon the precedent non-rivalry
attribute. In a later work, Landes and Posner28 also argued that a precedent is a public good,
although this time the argument entailed that precedents created in a private litigation may also
be useful in the future for those not taking part in the original dispute that created the precedent.
One can observe that the argument now claims that precedent creates a positive externality.29
Interestingly enough, Landes and Posner argued that if the precedent was useful only for the
original litigating parties, then the precedent would actually be considered a private good.30
The same kind of argument is presented by authors worldwide, such as Carvalho, who stated that
something is a public good if it generates a positive externality, or more directly that the law is a
public good.31 Regardless of the reasoning, the idea of law as a public good seems to be quite
commonly accepted.
Along with this tradition, although slightly diverging, I would like to argue that law can, in
fact, act both as a public and a club good. In its simplest conception, law is nothing but
information regarding how adjudicative systems are likely to behave when presented with a
disputecoercionas well as information regarding how people are likely to behave in
generalcompliance. As I will demonstrate shortly, just as light may behave either as a particle
or a waveknown as the wave-particle duality32law behaves as a club good when we are
26See J
AMES
M. B
UCHANAN
,T
HE
L
IMITS OF
L
IBERTY
:B
ETWEEN
A
NARCHY AND
L
EVIATHAN
99 (Liberty Fund 2000) (1975)
(In somewhat more technical language, lawof the sort analyzed here qualifies as a pure collective-consumption or public
good, and one for which independent adjustment yields corner solutions for each person. No person will provide, by his own
restricted behavior, the benefits of law-abiding to others.).
27The authors were trying to use the capital investment theory to assess the behavior of precedents, and they defined legal
capital as the set of judicial decisions available in a given period. Because a precedent can be used by one person without
diminishing its utility as a precedent for another person, they argued that legal capital was a public good in nature. See
William M. Landes & Richard A. Posner, Legal Precedent: A Theoretical and Empirical Analysis, 19 J.L. & E
CON
. 264,
264 (1976). I have shown that the same model could be used to include statutes and also be applied to civil law systems.
See Ivo Teixeira Gico, Jr., O Capital Jurídico e o Ciclo da Litigância,9R
EVISTA
D
IREITO
GV 435 (2013).
28See William M. Landes & Richard A. Posner, Adjudication as a Private Good, 8 J. L
EGAL
S
TUD
. 235, 248 (1979); see also
Verreck & Mühl, supra note 15, at 244 ([V]erdicts are categorized as collective goodsbecause the exclusion of citizens from
that information is undesirable . . . .).
29A close argument is made by Shavell in Steven Shavell, Suit, Settlement, and Trial: A Theoretical Analysis under
Alternative Methods for the Allocation of Legal Costs, 11 J.L. S
TUD
. 71 (1982) ([L]itigation will be worthwhile to society
as long as its deterrent and compensatory value exceeds total legal costs plus public administrative expenses . . . .).
30See Landes & Posner, supra note 28, at 261. Notice that this argument is incorrect because a public good does not become
a private good only because it is not useful for someone else, nor do precedents become rival only because they are not useful
for others. A video from my latest musical composition freely available on the Internet is a public goodnon-rival and non-
exclusiveeven if I am the only person on Earth that actually thinks it is useful.
31See C
RISTIANO
C
ARVALHO
,T
EORIA DA
D
ECISÃO
T
RIBUTÁRIA
146, 283 (2018).
32See A
LBERT
E
INSTEIN
&L
EOPOLD
I
NFELD
,T
HE
E
VOLUTION OF
P
HYSICS
:T
HE
G
ROWTH OF
I
DEAS FROM THE
E
ARLY
C
ONCEPTS TO
R
ELATIVITY AND
Q
UANTA
278 (1938):
But what is light really? Is it a wave or a shower of photons? . . . There seems no likelihood of forming a consistent
description of the phenomena of light by a choice of only one of the two possible languages. It seems as though we
must use sometimes the one theory and sometimes the other, while at times we may use either. We are faced with a
new kind of difficulty. We have two contradictory pictures of reality; separately neither of them fully explains the
phenomena of light, but together they do!
German Law Journal 649
discussing coercion, and it behaves as a public good when we are dealing with spontaneous com-
pliancethe law duality.
In the first, more traditional perspective of law as coercion,33 I acknowledge that when someone
goes to a lawyer and inquires about the law, he is actually asking one of several questions: How
courts will behave if a given dispute arises, how the government expects him or his business to
behave, or otherwise, how the government may employ its powerful apparatus to make him
behavehence, enforcing the law. This is why it is possible to say that the first nature of law
is to inform in which direction enforcement will go, and the utility that law provides is the knowl-
edge of this directionand how to channel or avoid it.34
From this perspective, it is easier to perceive that, when authors declare that law is a public
good, they are normally dictating it is a public good in the same way that any information is
a public good.35 The fact that I know how courts will behave and then use this information to
conduct my daily business affairs does not preclude your knowing of the same information,
nor does it affect your conducting of your business accordingly. This analysis fails, however,
to grasp that what is being discussed here is not actually the economic nature of law itself.
Rather, what is being discussed is that law is non-rival.36 When we say that the consumption
of law by one person does not preclude its consumption by another, we are only saying that this
knowledge can be consumed by both you and me. We are not saying anything about its exclud-
ability, and I would like to argue that law as coercion is, in reality, highly excludable.
Legal theorists are constantly involved with the question of jurisdiction, referring to either a
governments general power to exercise authority over all persons and things within its territory or
a courts power to decide a case or issue a decree.37 The key word here is powerthe simple fact
that the enforcement of legal rules requires jurisdiction means that, at times, a government or a
court is legally powerless to enforce its rules. In this case, even though the parties may know which
law would theoretically be applied de jure, the lack of jurisdiction means that the considered law
will not be enforced; hence, the law is inapplicable de facto. Consider an international arbitration
award that is to be enforced in another country that does not recognize part of the applied sub-
stantial law used to solve the dispute. As a matter of fact, although you know what the applicable
law is, its enforcement is effectively denied. This hypothetical situation is just one example of how
law can be excluded from the consumption of an individual. I will, however, offer two alternatively
interesting historical examples to make the point even clearer.
Originally, when Rome was just a city-state, Romans named the law that applied to themselves
ius civilis (civil law), and this law applied exclusively to the citizens of Rome. Thus, although most
people dwelling within Roman territory knew about ius civilis, those who were not Romans were
effectively excluded from its application and protection; hence, they were excluded from the con-
sumption of ius civilis.38 With the growth of the Roman empire, Romans gradually introduced
legal rules also applicable to those who were not Roman citizens, which became known as ius
gentium: The body of legal institutions and principles common to all Roman subjects regardless
33For a modern discussion, see S
CHAUER
,supra note 24.
34For an example of how this works, see Robert H. Mnookin & Lewis Kornhauser, Bargaining in the Shadow of the Law:
The Case of Divorce,88Y
ALE
L.J. 950 (1979).
35Note that information can be rival, if it is in fact a secret and the sharing of the information will destroy its strategic value,
like in know-how cases. In this scenario, the use of the information by one agent reduces the information utility for another;
hence, it is rival.
36Of course, if you define a public good as anything that is non-rival regardless of excludability, as was done previously, then
law is a public good.
37See Jurisdiction,B
LACK
S
L
AW
D
ICTIONARY
855 (7th ed. 1999).
38See P
EDRO
M
ARTINS
,H
ISTÓRIA
G
ERAL DO
D
IREITO
R
OMANO
,P
ENINSULAR E
P
ORTUGUÊS
11419 (1906); L
UÍS
C
ABRAL DE
M
ONCADA
,E
LEMENTOS DE
H
ISTÓRIA DO
D
IREITO
R
OMANO
328 (1924); A
NDREW
M. R
IGGSBY
,R
OMAN
L
AW AND THE
L
EGAL
W
ORLD OF THE
R
OMANS
21528 (2010); G
EORGE
M
OUSOURAKIS
,R
OMAN
L
AW AND THE
O
RIGINS OF THE
C
IVIL
L
AW
T
RADITION
Chapter 2 (2015).
650 Ivo Teixeira Gico
of their origins.39 In this sense, although both ius civilis and ius gentium are non-rival in nature
and would be considered public goods by the authors discussed before,40 the former was actually
excluded from consumption by non-Romans, while the latter was consumed freely.
Another clear example of the excludability of law is the history of commercial law in Brazil.
Prior to the enactment of the 2002 Civil Code,41 Brazilian citizens were divided into two different
groups: The regular citizens, whose business and actions were considered civil in nature and were
duly regulated by the 1916 Civil Code,42 and the merchants, who normally explored commerce
and industry and, as a result, were regulated by a different set of rules as established by the 1850
Commercial Code.43 Note that the action would often be strictly the same in all relevant aspects,
such as the sale of a good, but if the seller was a merchant, the Commercial Code would be applied
to the transaction. Alternatively, if the seller was not a merchant, the Civil Code would be applied.
As in the ius civilis/ius gentium dichotomy, here, the traditional authors44 would say the law
embedded in the Civil and Commercial Codes was a public good because its application to
one subject would not prevent its application to another subject; but again, this reflects only
its rivalry dimension. In fact, the Commercial Code brought many special provisionssuch as,
asking for bankruptcy,45 issuing a public power of attorney without a notary,46 and executing
specific contracts47that regular citizens could not enjoy. As a result, non-merchants were
barred from consuming the law embedded in the Commercial Codeillustrating
excludabilityalthough, in principle, the consumption by one agent would not preclude another
from enjoying itillustrating non-rivalry.
In this sense, although we all agree that law is non-rival in natureas is information in
generalit is quite clear that law as coercion is highly excludable; ergo, law is not a public good
in nature, but rather it is a club good.48 One may ask: What are the implications of recognizing
that law is a club good if it is non-rival anyway? The simplest answer to this question is that,
since law as coercion is excludable, the provider or supplier of the adjudicative service may
withhold services from those that refuse to pay the costs of providing such services. As a result
of the excludability of law as coercion, there can be a market for law even in the absence of gov-
ernment.49 In a more technical way, law as coercion is not vulnerable to the free-rider problem
because it is excludable. Yet, the issue here is not that the market for law would be vulnerable to
39See supra note 38.
40See supra notes 27, 28, and 30.
41See Lei No. 10.406, de 10 de Janeiro de 2002 (Braz.), C
ÓDIGO
C
IVIL
[C.C.].
42See Lei No. 3.071, de 10 de Janeiro de 1916 (Braz.), C
ÓDIGO
C
IVIL
[C.C.].
43See Lei No. 556, de 25 de Junho de 1850 (Braz.), C
ÓDIGO
C
OMERCIAL
[C.C
OM
.].
44See supra note 38.
45See Lei No. 556, de 25 de Janeiro de 1850, C
ÓDIGO
C
OMERCIAL
[C.C
OM
.] at art. 797841.
46See id. at art. 21.
47See, e.g.,id. at art. 14064.
48See B
UCHANAN
,supra note 26, at 100 (arguing that law could be excluded from those outside a coalition, even though he
called it a public good due exclusively to its non-rivalry attribute).
49The greatest example of law without government is the lex mercatoria, the body of commercial law constructed and
employed by merchants in medieval and early modern Europe that was enforced by merchant courts along the main trade
routes. For a discussion of the decentralized production of law, see generally Robert D. Cooter, Structural Adjudication and the
New Law Merchant: A Model of Decentralized Law,14I
NT
L
R
EV
.L.&E
CON
. 215 (1994). Though there is a high probability
that this market would be biased in favor of plaintiffs, because plaintiffs choose the court, as discussed generally in Daniel
Klerman, Jurisdictional Competition and the Evolution of the Common Law,74U.C
HI
.L.R
EV
. 1179 (2007) and Daniel
Klerman & Greg Reilly, Forum Selling,89S.C
AL
.L.R
EV
. 241 (2016). If this plaintiff bias in courts could be avoided though
legislation, giving judges no leeway for interpretation, the possibility of competition would then shift to competition between
legislatures. This a common phenomenon, as countriesor even internal statestry to attract companies to incorporate in
their jurisdictions so as to increase tax revenueand business for the local bar. The author thanks Prof. John Leubsdorf for
calling attention to these issues.
German Law Journal 651
the free-rider problem and law would therefore not be provided by a market, as Buchanan50
argued, but that law is strongly characterized by a positive externalityundersupply by the
market.
When I say that law is knowledge of how courts or government agencies will behave
enforcementwe can separate two aspects of this assumption: One being for the consumer of
this information, and the other being for the employer of this informationthe adjudicator.51
On the one hand, the subjects under the law know how to behave and how to predict the behavior
of others simply by possessing the knowledge of what the law isthe applicable rules. This knowl-
edge will allow these subjects to make choices and coordinate their behavior. On the other hand,
the adjudicators will know how to solve a dispute simply by being aware of what the law isthe
applicable rulesand then applying those rules to the facts before themknown as syllogism. In
both cases, law is information regarding what rule is most likely to be enforced in a given case.52
This system creates a feedback loop in which the more a rule is applied to the same set of cases, the
more the rules value is reinforced and spontaneous or enforced behavior will fit into the expected
rule of conduct, constituting a positive feedback. This is the demand-side economies of scale of the
rule of law, or its network effect.53 The greater number of people who know about what the law is
and the more they can sustain a rational expectation that these are the rules that will be enforced in
the emergence of a dispute, the more people are likely to spontaneously follow ithence, the more
valuable this knowledge is.54
The network effect associated with law reinforces the importance of statutes created by parlia-
mentswhich are nothing more than publicly provided, specialized legislative systemsand may
put into check the highly debated efficiency hypothesis of the common law system.55 In any case,
the assertion that the production of law creates a positive externality or that law as enforcement is
highly characterized by network effects does not alter its economic nature, meaning that it is non-
rival and highly excludableergo, a club good.
The focus on law as knowledge of how the adjudicative system will behave when presented with
a dispute explains its value in social contexts where enforcement plays a major role. This explan-
ation, however, does not elaborate upon why we witness so much compliance with legal rules even
in contexts in which enforcement is less likely. To understand this phenomenon, one must rec-
ognize that law also serves as a focal point for how people are likely to behave and, in doing so,
induces collective behavior toward certain paths that will generate self-sustaining equilibria,
creating compliance without enforcement.
50See B
UCHANAN
,supra note 26.
51In a previous work, I have shown that, in order to comprehend the behavior of the judicial system, one must simulta-
neously take into consideration the incentive structure of both parties and judges. See generally Ivo Teixeira Gico Jr.,
A Tragédia do Judiciário, 267 R
EVISTA DE
D
IREITO
A
DMINISTRATIVO
163 (2014).
52The legal system does not need to be perfect; it can be probabilistic, although it will work more effectively the more
predictable it is.
53Notice that, under the rule of law, a judge is a consumer of law as much as are the parties, while a court is as much a
consumer of law as is a provider. In the common law system, the rule-making function of courts is self-evident, while in the
civil law system, it is somewhat disguised, but still existent. See generally Ivo Teixeira Gico Jr., Hermenêutica das Escolhas e a
Função Legislativa do Judiciário,15R
EVISTA DE
D
IREITO
E
MPRESARIAL
R
DEMP
55 (2018).
54This rational expectation of how the law will be applied is an essential element of the rule of law and of what I call the
litigation cycle.See infra Section C, Figure 4. For a discussion about how this positive feedback feature is expected to create a
homeostasis behavior in the judicial system, see generally Gico Jr., supra note 27. For a discussion about the consequences of
not having the judiciary following the rule of law as well as the breaking of this positive feedback loopwhat I call judicial
anarchismsee generally Ivo Teixeira Gico Jr., Anarquismo Judicial e Segurança Jurídica,5R
EVISTA
B
RASILEIRA DE
P
OLÍTICAS
P
ÚBLICAS
479 (2015).
55See generally Ronald H. Coase, The Problem of Social Cost, 3 J.L. & E
CON
. 1 (1960); Isaac Ehrlich & Richard A. Posner, An
Economic Analysis of Legal Rulemaking, 3 J. L
EGAL
S
TUD
. 257 (1974); Richard A. Posner, What Do Judges and Justices
Maximize? (The Same Thing Everybody Else Does),3S
UP
.C
T
.E
CON
.R
EV
. 1 (1993); Francesco Parisi, The Efficiency of
the Common Law Hypothesis,in T
HE
E
NCYCLOPEDIA OF
P
UBLIC
C
HOICE
51922 (2004). For a somewhat dissenting opinion
on the efficiency of the common law, see generally Cooter, supra note 49.
652 Ivo Teixeira Gico
In this second, less traditional perspective of law, I refer to the fact that people at times comply
with legal rules even in the absence of a credible threat of coercionin the absence of enforce-
ment. Social scientists have struggled to explain why this is the case, and some argue that, when
agents perceive legal order to be legitimate, they will follow the rules naturally or their behavior
will be restricted by social beliefs.56 More recent authors have tried to explain the same phenome-
non by employing game theory to show that social behavior can emerge naturally from human
interaction, and the existence of law itself provides focal points57 that will attract collective behav-
ior and generate behavioral equilibria.58 In accordance with this more modern approach, it is rea-
sonable to say the second nature of law involves informing how people normally behave, and
because most people will likely behave in such a way, there is ultimately no gain in behaving differ-
ently; hence, compliance becomes a self-sustaining equilibrium. In this case, the utility that law
provides is not the enforcement of a certain rule, but the knowledge of this social direction, and
how to behave accordingly.59
It is important to stress that a Nash equilibrium is defined as a stable state of a system involving
the interaction of different agents in which no participant can gain by a unilateral change of strat-
egy, if the strategies of the others remain unaltered. In other words, a social state is considered a
Nash equilibrium when no individual gains from changing his own behavior unless someone else
also changes her behavior, and because there is no individual gain in changing behavior, that social
stateequilibriumis stable.
One may observe how the network effect of law is even more relevant when dealing with law
as compliance. As the greater number of people behave in a certain way, the higher the
probability becomes that others will adjust their behavior accordingly; hence, the prominent
behavior role as a focal point is stronger and that particular equilibrium is more likely to result.
I may even claim that law as compliance becomes an equilibrium only when the network effect
is so powerful that it transforms general non-compliance behavior into irrational behavior,
meaning the payoff for non-compliance is generally lower and the behavior becomes self-
sustainable.
In any case, if law as coercion is clearly a club goodnon-rival and excludablethe same can-
not be said about law as compliance. Law as compliance is also non-rival because your knowledge
of how most people behave within a given society does not preclude me from knowing the same
and enjoying the benefits resulting from this knowledge. Yet, since law as compliance does not
require enforcement, it is pure information and, once available, it cannot be withdrawn from the
56Compare M
AX
W
EBER
,E
CONOMIA E
S
OCIEDADE
:F
UNDAMENTOS DA
S
OCIOLOGIA
C
OMPREENSIVA
(4th ed. 1999) (1920)
(explaining how legitimacy may lead to compliance as agents perceive the rulecustomary or legalas the correct thing to
do), with E
MILE
D
URKHEIM
,A
S
R
EGRAS DO
M
ÉTODO
S
OCIOLÓGICO
(Martins Editora 1982) (1895) (explaining how collectivity
could create beliefs and how those beliefs would shape and restrict behavior).
57In game-theoretic parlance, a focal pointor Schelling pointis a salient equilibrium in a multi-equilibria game that
attracts players because it is salient, meaning that this focal point is more commonly a practical solution of the game than are
the other equilibria, even if the former is not a dominant strategy. See generally T
HOMAS
C. S
CHELLING
,T
HE
S
TRATEGY OF
C
ONFLICT
(Reprint ed. 1981) (1960).
58See A
VNER
G
REIF
,I
NSTITUTIONS AND THE
P
ATH TO THE
M
ODERN
E
CONOMY
:L
ESSONS FROM
M
EDIEVAL
T
RADE
(2006)
(using game theory to show that rather than assuming that individuals follow behavior, social behavior can be generated
endogenously, although limited to behavior that are self-enforcing equilibria); see also R
ICHARD
H. M
C
A
DAMS
,T
HE
E
XPRESSIVE
P
OWERS OF
L
AW
:T
HEORIES AND
L
IMITS
(2017) (explaining how law can produce coordinated behavior even
in the absence of a credible threat of coercion simply by providing a focal point that will attract behavior).
59It is very interesting to imagine how both functions of law correct divergent behavior. Clearly, enforcement would include
punishment with prison, fines, and additional costschanging the tradeoff of divergent behaviorwhile non-compliance
would lead to disconnection, and moral and other social costs, such as stigmas or accidents. Nonetheless, in the end, both
are expected to complementarily lead to convergent behavior; otherwise, the system may collapse. For a discussion about how
to reach cooperation where legal enforcement fails, see generally A
VINASH
K. D
IXIT
,L
AWLESSNESS AND
E
CONOMICS
:
A
LTERNATIVE
M
ODES OF
G
OVERNANCE
(2007). For a discussion on the importance of coercion that expands its importance
to this second conception of law, see generally S
CHAUER
,supra note 24.
German Law Journal 653
public; hence, law as compliance is non-excludable. As a result, I reason that law as compliance is a
public good in nature because it is non-rival and non-excludable.
Considering that here I am really interested in the relationship between law and courtsthe
bundled productand, debatably, law as compliance may not require enforcementmeaning,
law as compliance may not require an adjudicative system to workI shall focus on law as coer-
cion henceforth. By analyzing the economic nature of adjudicative systems and their relationship
with law as enforcement, it is my objective to explain interesting aspects of judicial systems
at large.
II. The Complementarity of Law and Courts
In the previous section, I demonstrated that law as coercion is non-rival though excludable from
consumption at low costsexcludabilityand hence, it is a club good. Now, I would like to
explain how this insight reveals something very important about the relationship between law
and adjudicative systems: Law and courts are complementary goods.
When we understand law as coercion, it becomes clear that we are referring to law as the set of
rules60 that are applied by an adjudicative system to solve disputes within a given jurisdiction61
and consequently shape collective behavior. In this sense, the informational utility of law resides
precisely on the indication of how the adjudicative system will behave or how force will be
employed if a given dispute is brought before it.62 This informational role may, in fact, be enjoyed
only if there is an actual enforcement mechanism and if this mechanism is able to enforce its
decision. In other words, law is useful only if there is an effective adjudicative system to support
it, just as an intravenous medicine is useful only when accompanied by a syringe. From an eco-
nomic perspective, these statements are equivalent to saying that law and adjudicative systems are
complements. In fact, I would like to argue here that law and adjudicative systems are perfect
complements, meaning they must be consumed together in order to be considered useful. They
are a bundled product.
On the one hand, an adjudicative system that follows no rulecourt without lawwould be
no different from an ergodic stochastic system. Regardless of the case brought before it, there
would exist a random chance that the claimant would succeed and, at any dispute, the chance
of prevailing would be independent of the initial state of the system. In this scenario, there would
be no need for a judge or any alternative decision mechanism other than a computer with a ran-
dom seed63 to settle the dispute. The dispute resolution service would be provided almost instan-
taneously, but the result would be a complete social waste, as agents would be unable to coordinate
60Here, the discussion regarding whether or not principles are part of the legal system is immaterial because they will con-
vert to rules once appliedas shown in Gico Jr., Tragédia do Judiciário,supra note 51or they will not be able to guide
behavior, resulting in judicial anarchyas discussed in Gico Jr., Hermenêutica das Escolhas,supra note 53.
61Jurisdiction is used here as the ability or power to enforce a decision, as previously mentioned.
62This argument was previously offered by Holmes Jr. in the U.S., when he stated, The object of our study, then, is pre-
diction, the prediction of the incidence of the public force through the instrumentality of the courts.See Wendell Holmes Jr.,
supra note 24, at 61. In France, a close remark was made by Planiol, who stated:
When the Code was new the law schools had their opinions which they taught and often each professor had his
own; on its side, jurisprudence sought somewhat at random the solution of questions that were given it to resolve.
Thus there was formed in our law faculties a doctrine, often original but rarely followed in practice. Today juris-
prudence is fixed in many points. What good is there in still teaching, as though they were formulas of actual
French law, theories that are neither written in the law nor admitted in jurisprudence? . . . Let us take jurisprudence
for what it is, a customary law of recent creation and let us teach its solutions, subject to saying what we think of it.
See 3M
ARCEL
P
LANIOL
,T
RAITÉ
É
LÉMENTAIRE DE
D
ROIT
C
IVIL
at X (1904).
63One may experience despair if she realizes that random numbers created by computers are not really random, as they use
some kind of algorithms to create those numbers. Real randomness is only achieved by measuring phenomena outside the
system itself.
654 Ivo Teixeira Gico
behavior ex anteone cannot know what decision will be takenand the result would be useless
for coordinating behavior ex posthow this case was decided is irrelevant for similar future cases.
In other words, the whole point of having a legal system is to coordinate behavior by establishing a
rational expectation of othersbehavior, including the governments.64 If the dispute resolution
mechanism is ergodic and stochastic, this basic function of law is denied, and one may claim
not only that there is no law at all, but also that courts are useless. That is why I stated at the
beginning of this Article that the judiciary is an institutional technology developed through thou-
sands of years of human experimentation with one purpose only: To solve disputes by enforcing
the rules. If rules are not enforced, the sole purpose of law and courts is denied.
On the other hand, a legal system without an enforcement mechanismlaw without court
would be ineffective for coordinating behavior when people disagree about the law. First, if two
interacting agents ex ante agree that their course of action is the best option for their own interest,
there is no need for law to establish what conduct should be followed.65 There is no dispute. The
agreed-upon behavior is a dominant strategy, and therefore cooperation will result. This is also
true if the agents are already in a Nash equilibrium, even though, ex post, they may not agree that
this situation is the best option for one of them. Since no agent can improve his or her results by
unilaterally changing his or her own behavior, the equilibrium is self-sustainable, and no addi-
tional enforcement is required.
Law is useful when there exists more than one interesting conduct possible either preceding or
following the interaction. If the law precludes one of the parties from engaging in ex post divergent
behavior, the other party may trust ex ante that the agreement will follow through, and cooperation
is rendered viable. In order to do so, however, law must be able to coercively preclude such divergent
behavior; hence, enforcement is required. One may clearly argue that the more frequentlythose rules
of conduct are internalized, the less enforcement will be required, but unless one can construct a
society of perfect human beings, some level of enforcementhowever lightis still required.
Second, from a social perspective, law is also relevant when, although the engaged parties
fully agree with the conduct in question, their cooperation produces negative effects over third
partiesknown as externalities. In this scenario, assuming that the transaction costs are
considerable,66 it is in the publics interest that law is able to prevent, to limit, or to require
adequate compensation for such behavior; because the agents involved may be unwilling to com-
ply, again, enforcement will be required.
Third, people at times may disagree among themselves regarding what was actually agreed
upon, what the legal rule is, or whether there exists a legal rule at all. In these cases, a dispute
mechanism is required to resolve ambiguities and fulfill legal gaps, which is normally the other
ancillary function of the adjudicative system.67 Still, even though we normally consider ambiguity
reduction and gap-filling functions as attributed to the courts in most, if not all, legal systems
around the world, these roles are not really a required function of the adjudicative systems.
For example, in 1603, the Portuguese Philippines Ordinations allowed for courts to resolve law
ambiguities and fill the gaps using many techniques, including subsidiarily applying the Roman
Corpus Iuris Civilis, or the canonic law.68 Due to excessive freedom in interpretation, however, the
Law of Good Reason of 1769 withdrew most of such functions from courts and allocated them
back to the Portuguese Crown.69 Another similar example is the Brazilian Imperial Constitution of
64For a further discussion about the role of the judiciary in enabling ex ante cooperation, see Gico Jr., O Capital Jurídico,
supra note 27.
65Remember that, for the purposes of this discussion, law as compliance was excluded; I am not considering the self-
sustaining equilibria.
66See generally Coase, supra note 55.
67Note that these auxiliary functions are not distinct functions from the dispute resolution function, but a necessary step
towards solving disputes when there is an ambiguity or a gap.
68See Gico Jr., Tragédia do Judiciário,supra note 51, at 60.
69See Gico Jr., Tragédia do Judiciário,supra note 51, at 61.
German Law Journal 655
1824, which separated the role of fact-finding for juries and the role of law adjudication for courts
and restricted the rule-making and legal interpretation roles for the General Assembly.70 In
any case, even if the gap-filling role is attributed to courts, enforcement will be required to make
such judge-created rules effective, and even if the ambiguity-solving task is attributed to courts,
enforcement will be required to make the chosen interpretation effective. In this sense, choice
hermeneutics71 also requires enforcement.
As a result, whatever technique courts adopt to solve a disputemere adjudication, ambiguity
reduction, or gap-fillingthe fact is that the result most likely will become law, and enforcement
will be necessary if spontaneous compliance does not come through. In any case, law will be useful
only if a workable adjudicative mechanism is available; hence, law and courts are complementary
goodsa bundled product.
III. The Economic Nature of Courts
Once it has been established that law as coercion is a club good and its excludability is deeply
related to its complementarity with courts, the inquiry into the economic nature of the adjudi-
cative system should help us understand some aspects of the legal systems dynamic. This will lead
to the surprising conclusion that unrestricted access-to-justice policies may, instead of granting
rights, actually result in their de facto denial.
Similar to law, any adjudicative system is excludable by nature. Consider an arbitration court
and the requirement that, prior to the commencement of the arbitration, the complainant or both
parties pay part of the filing fee or the arbitrators fees. Without payment, the arbitration remains
uninitiated: Until consumers pay for the service, the adjudicative service provider will not provide
the service. The same can be said about public courts. In many jurisdictions, for every filed case,
there must be the payment of an initial filing fee, otherwise the case will not move forward. The
filing fee may fully cover the costs of providing the service or it may be subsidized, but in both
cases, without payment, the adjudicative service will not be provided. The idea here is not to dis-
cuss whether or not the filing fee actually exists or whether the service is subsidized or not, but to
make clear that such a fee could easily be implemented to regulate access to the service; ergo,
adjudicative services are excludable at very low costs.
To make the point even clearer, the excludability of the adjudicative service need not occur in
the form of a filing fee nor expressly as a price. Access to the services can be rationed using other
mechanisms, such as quotas or cost-benefit analyses. For example, since the Constitutional
Amendment No. 45 of 2004 was implemented, the Brazilian Federal Supreme Court will hear
constitutional cases through extraordinary appeal only if the Court deems the issue at hand to
be of general repercussion.72 Even if a case truly involves a potential constitutional violation
but the issue is restricted to the parties, the Brazilian Supreme Court may refuse to hear it.
One can observe here that the general repercussion legal standard is actually an inquiry into
the positive externality nature of the issue at hand. If the issue brought before the Supreme
Court generates a positive externality, meaning it is useful for other actual or potential parties,
then the scarce resources of the Supreme Court can be allocated toward solving it. If not, then
the case is dismissed and, for all purposes, the litigants are denied the adjudicative services of
70See Gico Jr., Tragédia do Judiciário,supra note 51, at 62.
71I have called before choice hermeneuticsthe essential legislative role that judges play in civil law countries when they
reduce ambiguities in statutes or contracts in dispute settlements and, in doing so, establish the legal rule that will prevail,
excluding all other possible interpretations, although they are all compatible with the legal texthence, potential rulesor
when they fill the gaps. See generally Gico Jr., Tragédia do Judiciário,supra note 51.
72Paragraph 1 of Article 1,035 of the Brazilian Civil Procedure Code establishes: For the purposes of general repercussion,
it will be considered the presence or absence of relevant questions from the economic, political, social or legal point of view
that go beyond the subjective interests in the case.See Lei No. 13.105, de 16 de Março de 2015, C
ÓDIGO DE
P
ROCESSO
C
IVIL
[C.P.C.] at art. 1,035, § 10(Braz.).
656 Ivo Teixeira Gico
the Supreme Court. Here, the adjudicative service is rationed without the use of a price mechanism
and rather employs an implicit cost-benefit analysis called general repercussion.
The same kind of service-denial discretionary power was granted to the Supreme Court of the
United States by the Judiciary Act of 1925 and the Supreme Court Case Selections Act of 1988.
According to these statutes, most potential consumers of law are prevented from receiving adju-
dicative services from the Supreme Court as a matter of right.73 Just as in the general repercussion
case in Brazil, a party who desires that the American Supreme Court review a decision of a federal
or state court must first file a petition for writ of certiorari, and the Supreme Court possesses the
discretionary power to decide whether or not to grant certiorari. In fact, most cases are denied
certiorari. Once again, access to the judicial service is rationed at low costs without the use of
a price mechanism, and hence, courts are excludable.
The next question is whether or not adjudicative services are rival. When discussing the law, we
have observed that, because law is information, it is non-rival. The knowledge of the law by one
person does not preclude another from enjoying the same benefits and conducting her business
accordingly. The same can be said about judges when applying the law. The fact that a judge
applied a statute or a precedent in a case does not preclude another judge from applying the same
legal rule. In fact, in both cases, the application of law by one agent only reinforces the utility of
the law for others, its network effect. The case for the adjudicative systems is different, as they are
normally rival in nature, although the discussion is somewhat subtler.
If we understand that the judiciary possesses high rivalry, then we must conclude that it is a
private goodrivalry excludability. If we understand that it is non-rival, then, similar to law,
it is a club goodnon-rivalry excludability. I believe the judiciary can be both, depending
upon the context. Consider a single judge called to solve a single dispute. Depending upon the
complexity of the case, the judge could simultaneously hear another case without substantially
slowing down the analysis of the first. Remember that a case is a series of actions, not all of
them simultaneous or exclusively executed by the judge; hence, there are many opportunities
for the judge to perform other parallel acts while he or she waits for the other parties to perform
their actions. In this scenario, at least initially, adjudicative services are non-rivaltherefore,
the judiciary would be a club good, just as law is a club good. In contrast, if simultaneously
analyzing two or more cases would lead to excessive complexity or if the number of cases
is greater than available supply, then adjudicative services are rival in use and the judiciary
would start behaving similarly to a private good. In this sense, it is reasonable to say that,
at first, the judiciary is a club good if the installed capacity is superior to the perceived demand.
As quantity demanded increases, it begins acting more like a private good, becoming rival and
prone to congestion if its use is not rationed somehow. In other words, the judiciary is a highly
congestible good (see Figure 2).
At this point, it should be clear that, even at an earlier stage when the supply is greater than the
perceived demand, the judiciary is a club goodeasy excludabilitynever a public goodhard
excludability. Nevertheless, when it begins to befall rivalry due to a mismatch between supply and
demand, the judiciary will gradually begin behaving as a private good because each additional case
filed will be added to the case log and will slow down the adjudicative service that is to be provided
in other cases; henceforth, each additional case filed will decrease the utility of employing the
judiciary for everyone. In this sense, although consumption is not entirely rivalrous initially,74
each additional filed case imposes a negative externality on the other users of the judicial system.
If no other rationing mechanism is installed, that negative externality will materialize in the form
of queues, which, combined with the shorter time available to judges to consider each case, will
73Judiciary Act of 1925, Pub. L. No. 68-415, 43 Stat. 936 (codified as amended in scattered sections of 28 U.S.C.); see also
Supreme Court Case Selections Act of 1988, Pub. L. No. 100-352, 102 Stat. 662.
74Note that, as opposed to the traditional private good case, here we do not have a proportional relationship between the
amount consumed by one person and the amount available for consumption by others.
German Law Journal 657
result in increasing degradation of the quantity and quality of the services provided. For our
present purposes, this utility-reduction effect means that the judiciary will begin behaving as a
rival service.75 The fact that the judiciary is congested in so many places worldwide only corrob-
orates the idea that courts are a rival resource.
1. The Tragedy of the Judiciary
Now that I have shown that courts behave like private goods while law as coercion is a club good, it
is relatively easy to understand that the unrestricted access to adjudicative services will lead to the
natural formation of queues to ration the supply. Any additional use of the adjudicative service
would prevent or substantially diminish the utility of the service for other users, just as the inclu-
sion of an additional car diminishes the utility of congested roads for other drivers.
Since courts are private goods, the simplest solution for the court congestion problem would be
to exclude users from the services through means such as general repercussion or certiorari civil
procedure toolsnon-price rationing mechanisms. Another solution may be to let the free market
work and the price mechanism regulate demand in the same way that it regulates demand for
arbitration courts. The competition among users would raise the prices of the adjudicative ser-
vices, and the resulting price increase would ration the scarce resources, naturally excluding users
that attribute less value to the service. Thus, an equilibrium between supply and demand would be
reached. As a result, there would be a smaller number of actions filed and those claims whose
litigants assigned the lowest value, according to their willingness to pay, would not be judged
or would be served with a much longer waiting time.76
Of course, since the market only captures the preferences of those who are capable and willing
to express their preferences through paymentreferred to as revealed preferencesthe price
mechanism would solve the congestion problem at the expense of the most vulnerable element
of societythe poorwho would be excluded from enjoying the service even though they might
actually value the service more than its costs. In addition, because law and courts are complemen-
tary goods, the exclusion of some users from access to the adjudicative services necessarily implies
they will also be excluded from enjoying the benefits of law itself. In other words, by excluding
some citizens from accessing the courts, we would also be excluding them from accessing their
rights. This is unacceptable. In this sense, from an economic perspective, adjudicative services may
be considered merit goods, meaning they are so important for human development that every
Figure 2. The Judiciary as a Private Good.
75For a discussion using economic theory of clubs to identify the optimum size of the club that could be used to investigate
the optimum size of the judiciary and the case filing fees, which is beyond the scope of this work, see D
ENNIS
C. M
UELLER
,
P
UBLIC
C
HOICE
III 183 (3d ed. 2003) and R
ICHARD
C
ORNES
&T
ODD
S
ANDLER
,T
HE
T
HEORY OF
E
XTERNALITIES
,P
UBLIC
G
OODS
,
AND
C
LUB
G
OODS
347 (Cambridge University Press 1999) (1986).
76As an example, see Cost Calculator,I
NT
L
C
HAMBER OF
C
OM
., https://iccwbo.org/dispute-resolution-services/arbitration/
costs-and-payments/cost-calculator/ (last visited Apr. 8, 2020). There, one can estimate the unsubsidized costs of arbitration
through a private adjudicative system, and the user can opt to pay more if she wants more arbitrators looking at her case
establishing a lower probability of erroror if she wants a quicker decisiongranting an expedited procedure.
658 Ivo Teixeira Gico
individual should have access to them on the basis of some concept of need rather than ability or
willingness to pay.
The idea that the judiciary is a merit good has taken over the world, as most countries have
decided that adjudicative services are a fundamental right and, therefore, governments should
grant free access to all in equal terms. According to the United Nations:
Access to justice is a basic principle of the rule of law. In the absence of access to justice,
people are unable to have their voice heard, exercise their rights, challenge discrimination
or hold decision-makers accountable. The Declaration of the High-level Meeting on the
Rule of Law emphasizes the right of equal access to justice for all, including members of
vulnerable groups, and reaffirmed the commitment of Member States to taking all necessary
steps to provide fair, transparent, effective, non-discriminatory and accountable services that
promote access to justice for all....United Nations activities in support of Member States
efforts to ensure access to justice are a core component of the work in the area of rule
of law. 77
In reality, this idea has been around for quite a while. The twentieth century witnessed an increas-
ing preoccupation with access to justice. The first wave was concerned with legal aid for the poor
and started in the West in the 1920s in Germany and England.78 The second wave was concerned
with the representation of diffuse interests and efforts to address the problem of representing
groups and collectives other than the poor, which started in the late 1960s in the U.S. 79 The third
wave went beyond advocacy by focusing not only on legal aid and the defense of diffuse interests,
but also on all kinds of institutional barriers to a more effective access to justice.80 These waves of
reform, aimed toward increasing access to justice, keep coming from time to time.81
In the 1970s, Mauro Cappelletti directed a research project funded by the Ford Foundation and
the Italian National Council of Research (CNR), called the Florence Access-to-Justice Project.82
Cappelletti and Garth stated:
The right of effective access is increasingly recognized as being of paramount importance
among the new individual and social rights since the possession of rights is meaningless with-
out mechanisms for their effective vindication. Effective access to justice can thus be seen as
the most basic requirement the most basic human right”–of a modern, egalitarian legal
system which purports to guarantee, and not merely proclaim, the legal rights of all.83
In other words, as I propose here using economic theory, Cappelletti and Garth recognized that
law and adjudicative services are perfect complementary goods. One is useful only in the presence
77See Access to Justice,U
NITED
N
ATIONS
,https://www.un.org/ruleoflaw/thematic-areas/access-to-justice-and-rule-of-law-
institutions/access-to-justice/ (last visited Apr. 8, 2020).
78See Mauro Cappelletti & Bryant G. Garth, Access to Justice: The Newest Wave in the Worldwide Movement to Make Rights
Effective,27B
UFFALO
L. R
EV
. 197 (1978).
79Id. at 209. For an earlier discussion about not rationing access to justice, see generally Geoffrey C. Hazard, Jr., Rationing
Justice, 8 J.L. & E
CON
. 1 (1965).
80See Cappelletti & Garth, supra note 78, at 22227.
81For a somewhat broader and more modern report on access to justice waves, see generally C
HRISTINE
P
ARKER
,J
UST
L
AWYERS
:R
EGULATION AND
A
CCESS TO
J
USTICE
(1999).
82See Cappelletti & Garth, supra note 78 (noting that the project results were published in a four-volume series). The vol-
umes, published by Sijthoff (Leyden and Boston) and Giuffrè (Milan) under the general editorship of Mr. Cappelletti, are:
Volume I. Access to Justice: A World Survey (edited by Messrs. Cappelletti and Garth); Volume II. Access to Justice: Studies of
Promising Institutions (edited by Mr. Cappelletti and Mr. John Weisner); VolumeIII. Access to Justice: Emerging Perspectives
and Issues (edited by Messrs. Cappelletti and Garth); and Volume IV. Patterns in Conflict Management: Essays in the
Ethnography of Law. Access to Justice in an Anthropological Perspective (edited by Professor Klaus-Friedrich Koch).
83See Cappelletti & Garth, supra note 78 at 18485.
German Law Journal 659
of the other. In other words, law as coercion and courts are a bundled product. The problem is that
the behavior of the bundled product will be determined by the limitations of the combined good.
No reasonable person would deny that access to justice is a fundamental requirement for
both human development and the implementation of the rule of law. Nevertheless, what this kind
of initiative fails to understand is that, if public policy subsidizes access to the adjudicative services
or creates a legal right to unrestricted, free access to the judiciarywhich is a private good in
naturein practice, this public policy changes the economic nature of courts by making it legally
hard to exclude users. Yet, because the judiciarys rivalry attribute is inevitable, if the judiciary is
turned into a free access system, then the result is the transformation of the judiciary into a
common pool resource (see Figure 3).
As a result of unrestricted free access policies, the judiciary becomes a common pool resource
system vulnerable to overexploitation and prone to court congestion.84 The more governments
around the world implement policies to incentivize people to use the judiciary, the more courts
will become congested,85 just as would occur with any other overexploited common resource. The
resulting court congestion substantially increases both the amount of time required to solve a dis-
pute and the probability of low quality or simply incorrect results, as judges become overwhelmed
by growing caseloads while simultaneously receiving demands for increasingly expeditious
answers. Within the chaotic world of overexploitation of the judicial system, it is not unreasonable
to expect that judges would focus as much effort as possible on dispute resolution to reduce the
case log and invest as little effort as possible on legal certainty,86 the latter of which is a positive
externality that only marginally reduces the case log for any specific court.
In sum, once unrestricted access-to-justice policies are in place, it becomes relatively straight-
forward that the judiciary will become non-excludable and more users will access the system.
However, since adjudicative systems are rival in nature, the expected result will be the overexploi-
tation of the system leading to the substantial degradation of the services provided.87 We all want
everyone to have access to adjudicative services, because being able to vindicate ones rights is an
essential feature of being able to exercise them. Nevertheless, once unrestricted access to courts is
granted, the system itself will be overloaded with cases, and although access to the system will be
Figure 3. The Judiciary as a Common Pool
Resource.
84See generally E
LINOR
O
STROM ET AL
., R
ULES
,G
AMES
&C
OMMON
-P
OOL
R
ESOURCES
(1994).
85Considering, of course, that scarce resources prevents the unlimited expansion of the judiciary, This line of reasoning is
applicable when the supply of adjudicative services is lower than total demandmaking courts rival. If a society is capable of
and decides to invest enough resources so that supply is always higher than demand for adjudicative services, then the judi-
ciary will not become rival and there will be no congestion, although there is always a congestion equilibrium given a certain
level of services provided. See infra Section C.
86Here, I am assuming that there are costs associated with discovering and applying the law, in general, just like in acquiring
and processing any kind of information; otherwise, this may not happen. Regarding the coordination cost of judges withinthe
judiciary when judges have a common goalthat is, same utility functionsee Ivo Teixeira Gico Jr., Anarquismo Judicial e
Teoria dos Times,4E
CON
.A
NALYSIS
L. R
EV
. 269 (2013). For a discussion about the coordination costs when they may have
different goalsthat is, different utility functionssee generally Gico Jr., Hermenêutica das Escolhas,supra note 53.
87There is some empirical evidence that, depending on the context, an increase in access to justice might even generate
regressive effects. See Anthony Niblett & Albert H. Yoon, Unintended Consequences: The Regressive Effects of Increased Access
to Courts,14J.E
MPIRICAL
L. S
TUD
. 1, 5 (2017). The author would like to thank Prof. Mariana Mota Prado for calling attention
to this study.
660 Ivo Teixeira Gico
granted, the adjudicative service itself will not be provided in a timely manner, or it will be pro-
vided, but at a much lower quality; this result is exactly what I call the tragedy of the judiciary.88
The more people use the judiciary, the more congested it becomes and the less useful it will be
for the next user, as its ability to provide public adjudicating services will be hindered. The prob-
lem is focusing on access to the systemthe judiciarywhen focus should be placed on the pos-
sibility of using and enjoying its fruitsrightswhich are, to a large extent, what people actually
want. By focusing only on rights awareness and encouraging the unrestricted use of the judiciary
without recognizing that it is already overloaded with cases, public policy is accelerating and
encouraging the overexploitation of the judicial system, which is already dysfunctional in many
jurisdictions. By recognizing the common pool resource nature of the judiciary, we can better
understand what is happening within courts all across the world in which free access policies
are increasing pari passu with court congestion.
It may be no coincidence that some authors have identified an unprecedented growth in case-
load, resulting in court congestion during the 1960sthe same period in which the biggest wave
of access-to-justice policies occurred. Posner showed that, although court caseload has been
steadily increasing in the Federal Courts of the United States, this growth was modest and easily
accommodated by the three-tiered system created in 1891 until the 1960s, when the caseload
exploded.89 The Task Force on Administration of Justice reached the same conclusions,90 and
Landes reinforced the argument, stating, [i]t is widely recognized that the courts are burdened
with a larger volume of cases than they can efficiently handle.91
One may notice the irony and clear contradiction of Cappellettis and others92 defense of free
access to justice while simultaneously defending that court congestion is actually inaccessible
justice.93 Similar to modern defenders of unrestricted access to justice, Cappelletti simply could
not perceive that these two goals are plainly contradictory. The objectives of greater access and
greater speed are, to a large extent, incompatible. If the number of cases widely exceeds the judi-
ciarys capacity for analysis and prosecution, each case will take more and more time to be solved,
and the analysis of each case will be increasingly more superficial because judges will not have
enough time to carefully go through each one. This condition diminishes the quality of decisions
and, at the limit, undermines the reason for having a judiciary.
On top of this, another issue that seems to escape the perception of some social scientists is that,
while the production of law by courts creates a positive externality94 and a positive network effect,
88This term is a direct reference to the phenomenon of overexploitation of common pool resources, when access to the
resource is not controlled or rationed, normally known as the tragedy of commons.See generally Garrett Hardin, The
Tragedy of the Commons, 162 S
CIENCE
1243 (1968).
89See R
ICHARD
A. P
OSNER
,T
HE
F
EDERAL
C
OURTS
:C
HALLENGE AND
R
EFORM
5386 (1999).
90See generally O
FFICE OF
J
USTICE
P
ROGRAMS
,supra note 9.
91See William M. Landes, An Economic Analysis of the Courts, 14 J.L. & E
CON
. 61, 74 (1971). Along the same lines, Clemenz
and Gugler determined that [t]he annual number of civil litigation cases taken to Austrian courts rose from 544,000 in 1960
to 951,000 in 1995. At the same time the population remained almost constant. Similar trends can be observed in other
countries as well.See Gerhard Clemenz & Klaus Gugler, Macroeconomic Development and Civil Litigation,9E
UR
. J.L. &
E
CON
. 215 (2000).
92For a modern defense of unrestricted access to justice, see generally Martin D. Beier, Economics Awry: Using Access Fees
for Caseload Diversion, 138 U. P
A
.L.R
EV
. 1175 (1990).
93See Cappelletti & Garth, supra note 78, at 190:
The effect of this delay, especially given the prevailing rates of inflation, can be devastating; it increases the parties
costs and puts great pressure on the economically weak to abandon their claims or settle for much less than that to
which they are entitled. As the European Convention for the Protection of Human Rights and Fundamental
Freedoms, in Article 6, paragraph 1, explicitly recognizes, justice that is not available within a reasonable time
is, for many people, inaccessible justice.
94For a discussion about the optimum level of litigation in a society, see generally Steven Shavell, The Social Versus the
Private Incentive to Bring Suit in a Costly Legal System, 11 J.L. S
TUD
. 333 (1982). See also Shavell, Suit, Settlement, and
Trial,supra note 29.
German Law Journal 661
the judiciary is a rival resource and its overexploitation creates congestion, which is a negative
externality. In other words, we can consider the precedents or the jurisprudence constante or
the ständige Rechtsprechung produced by courts, as a byproduct of their dispute resolution activ-
ities, a positive externality, because others may be able to enjoy the benefits of the created prec-
edent or case law without having contributed toward the cost of producing it.95 Nonetheless, the
filing of a case itself will slow down the provision of adjudicative services as a whole for everyone
else, resulting in an offsetting negative externality. Since most cases do not produce a precedent or
jurisprudence constante or ständige Rechtsprechung, while all filed cases contribute to court con-
gestion, I believe the negative externality of court congestion by far offsets any value resulting from
the positive externality of precedent or case law production. This conclusion renders moot any
economic argument for the provision of subsidies for litigation for the purpose of increasing prec-
edent or case law production.
The tragedy of the judiciary leads us to a puzzling conundrum: People should be able to vin-
dicate their rights without substantial barriers, but unrestricted access to the adjudicative system
for each person undermines access to justice for everyone else. It should be possible to create some
governance mechanisms that prevent the overexploitation of the judiciary,96 although the recog-
nition of the economic nature of courts leads to the conclusion that some kind of rationing may be
necessary. This requires some strategic thinking about how to better structure courts and the judi-
cial system, which may ultimately lead to some policy implications.
For illustrative purposes, I can think of a few examples of policies focused on mitigating the
tragedy of the judiciary, such as higher sham litigation penalties when parties abuse the adjudi-
cative system (bad faith litigation) and increased incentives for and possibilities of collective
actions to improve the provision of the services with reduction of traffic. Just as public transpor-
tation is a more efficient solution for traffic jams than is the building of more roads, collective
actions are more useful in reducing court congestion than is the building of more courts. By
the same token, the number of collective actions available against government bodies, including
class actions about tax and social security purposes, should be expanded, as a great number of
cases are filed against governments. Nevertheless, there is one special issue that I would like to
explore further: The importance of legal certainty.
C. Legal Certainty and the Tragedy of the Judiciary
In the law and economics tradition on litigation inaugurated by Friedman,97 the initial premise is
that the agent choosing to litigate makes a rational choice. A lawsuit for the litigantwhether the
plaintiff or the defendantis a rational decision in which the costs and benefits expected from
using this adjudicative mechanism are weighed. With or without lawyersassistanceas in small-
claims courtseven if intuitively, the litigants try to estimate the probability of success as well as
the costs associated with reaching a settlement or continuing to litigate. Both parties know they
incur in a margin of error along this estimate.
The outcome of these individual estimates will largely determine the partiesability to settle. As
in any voluntary exchange, there is room for a bargain when the maximum proposal accepted by
the rational defendant exceeds the minimum proposal acceptable to the rational plaintiff. The
rational plaintiff will settle when the expected return of the claim is equal to or close to the amount
offered by the defendantassuming risk neutrality. The rational defendant will enter the
95For simplicity purposes, here, I am ignoring that the judiciary is actually publicly funded by tax money from all members
of society rather than fees paid by litigants.
96For a general discussion of governance mechanisms, see E
LINOR
O
STROM
,G
OVERNING THE
C
OMMONS
:T
HE
E
VOLUTION
OF
I
NSTITUTIONS FOR
C
OLLECTIVE
A
CTION
(James E. Alt & Douglass C. North eds., 1990).
97See Alan E. Friedman, An Analysis of Settlement,22S
TAN
.L.R
EV
. 67 (1969); see also Landes, supra note 91; John P. Gould,
The Economics of Legal Conflicts, 2 J.L. S
TUD
. 279 (1973); Richard A. Posner, An Economic Approach to Legal Procedure and
Judicial Administration, 2 J.L. S
TUD
. 399 (1973); Landes & Posner, Legal Precedent,supra note 27.
662 Ivo Teixeira Gico
settlement when the expected value of the action is greater than or near the value that the rational
plaintiff is willing to acceptalso moving away from complexities related to risk aversion. It is
assumed that the joint private cost of litigation is higher than the joint private cost of reaching a
settlement. In such a scenario, the settlement will not occur only when the partiesestimates do
not coincide minimally, and thus there is no perceived cooperative surplus to be divided. In other
words, the logic of a settlement is similar to the logic of a voluntary exchange, that is, a contract:
It will occur only if the parties perceive a cooperative surplus to be distributed.
Thus, on the one hand let πrepresent the plaintiff in a potential litigation and let us denote, by
Uπ, the expected benefit of that litigation at an expected cost, Cπ
for example, initial fees, attor-
ney fees, and expert fees. On the other hand, let Δrepresent the defendant of this potential lit-
igation, and let UΔrepresent her return from that litigation, which is most often negativeit is a
costbut not necessarily. The costs incurred by Δto use the judicial system as a dispute settle-
ment mechanism are denoted by CΔ. Consequently, in the event of a litigation, the maximum
return for the plaintiff is UπCπand UΔCΔfor the defendant.
If the parties were able to predict exactly what the outcome of the trial would be, that is in
a world in which there were maximum legal certaintyperfect information and a perfect
judiciarythere would exist no asymmetry of information between the parties and the issue
would be a purely distributive one. In addition, the loss for the defendant would be equal to
the gain for the plaintiff and vice versa UπUΔ, and the plaintiff and the defendant could
then maximize joint welfare by making an out-of-court settlement and dividing the cost saved
from the court dispute CπCΔ, minus the cost of entering into a settlement. Along this sim-
plified approach in which the parties know exactly what the outcome of the judiciary would be, the
tendency is to make an out-of-court settlement as it is unreasonable to waste resources by
litigating.98
Considering the Brazilian experience as an example, this first model may seem unrealistic
because the number of settlements in filed cases seem to be much lower than the number of cases
that go to trial following the investigation procedure. In 2018, it was estimated that only 11.5% of
filed cases in Brazil ended in settlement.99 Nevertheless, in the U.S. legal system, the judgment rate
is substantially lower, with only 10% of cases being initiated100that is, more than 90% of cases
end in settlements prior to final judgment. As an example, in New York, 98% of bodily injury cases
resulting from negligence end in settlement,101 while cases regarding medical errors end in settle-
ment 50% of the time even before reaching courts and about 40% are closed during the inves-
tigationthat is, before the final judgmentwhile less than 10% actually go to trial.102
Still, the theory must be able to explain why some cases do not end in settlements, be it the
larger or the minor portion of the universe. Therefore, we must insert the idea of risk into this
evaluation. The judiciary does not generate perfect information regarding how each case will be
decided, the parties are not able to interpret these signs perfectly, and there also exists private
information between the partiescommonly referred to as asymmetry of information.
Moreover, there is evidence that parties and lawyers are consistently optimistic about the outcome
of future judgments,103 and lawyers may not fully disclose the facts to their clients, as they gain
98See George L. Priest & Benjamin Klein, The Selection of Disputes for Litigation, 13 J.L. S
TUD
. 20 (1984) (first proposing the
idea that only hard cases are tried while easy ones are settled, therefore creating the tendency toward a 50 percent plaintiff
victoriesamong litigated cases). The Priest-Klein hypothesis was generalized by Yoon-Ho Alex Lee & Daniel Klerman, The
Priest-Klein Hypothesis: Proofs and Generality,48I
NT
.R
EV
.L.&E
CON
. 59 (2016).
99See C
ONSELHO
N
ACIONAL DE
J
USTIÇA
, Justiça em Números 142 (2019).
100See George L. Priest, Measuring Legal Change, 3 J.L. E
CON
.&O
RG
. 531 (1987); see also Oren Bar-Gill, The Evolution and
Persistence of Optimism in Litigation, 22 J.L. E
CON
.&O
RG
. 490 (2006).
101See H. L
AURENCE
R
OSS
,S
ETTLED
O
UT OF
C
OURT
:T
HE
S
OCIAL
P
ROCESS OF
I
NSURANCE
C
LAIMS
A
DJUSTMENT
4 (1980).
102See Patricia Munch Danzon & Lee A. Lillard, Settlement Out of Court: The Disposition of Medical Malpractice Claims,12
J.L. S
TUD
. 365 (1983).
103See S
TEVEN
S
HAVELL
,F
OUNDATIONS OF
E
CONOMIC
A
NALYSIS OF
L
AW
405 (2004).
German Law Journal 663
more if there is a disputea principal-agent problem. I will interpret this noise as a riskthat is, a
measurable amount of probability, as opposed to uncertainty, in which case the risk is not meas-
urable. In this case, we would have:
VRπPπUπCπAπ(1)
VRΔPΔUΔCΔAΔ(2)
where Pπand PΔare the subjective probabilities attributed by the plaintiff and the defendant,
respectively, to the success event of the plaintiff; VRπis the reserve value of the plaintiffthat
is, the minimum that he must receive to accept an agreement; VRΔis the defendants reserve
valuethat is, the maximum she is willing to offer for settlement; and Aπand AΔare, respectively,
the costs of entering into an agreement for the plaintiff and the defendant. As one may observe, in
this model the litigation costs borne by the parties affect their propensity to litigate rather than
settle: The higher the costs, the lower the litigation rate, and the lower the cost, the higher the
litigation rate. Perceptibly, access-to-justice policies decrease those costs.
Since the litigants are rational agents, the condition of litigation104 states that VRπ>VRΔ, or:
PπUπCπAπ>PΔUΔCΔAΔ(3)
assuming UπUD, the inequation can be rewritten as:
Pr Litigation fPπPΔ
U>CπCΔ
AπAΔ
 (4)
This simple model has some direct and important implications. First, all other things being equal,
the greater the utility of the good in dispute (U), the greater the likelihood of litigation. Second, the
probability of litigation is a growing function of the expectations gapthat is, the distance
between the assessment of the probability of success by the plaintiff and the defendant
PπPΔ
. Third, the less costly a settlement AπAΔ
and the more expensive it is to litigate
CπCΔ
, the greater the probability becomes that a settlement will be reached. Thus, the
cheaper the litigation, the greater the probability of litigation.
Clearly, changing the assumptions of the model can substantially alter this scenario. For exam-
ple, if the parties are risk-averse, the increase in Ucan proportionally increase more the number of
settlements, as the parties would value a more certain return than the expected value of the lit-
igation. But if the damage to one of the parties has already occurred and the action is interpreted as
a way of reducing or eliminating such loss, aversion to loss105 may work in reverse and thus
encourage litigation. While the concavity of the value curve in relation to gains generates risk
aversion, the convexity of the value of losses can generate risky behavior and thus result in more
litigation. Here, the way in which society or the lawyer poses the question to the potential litigant
the framingcan make a significant difference. Alternatively, a repeated litigantfor example,
the government or a telecommunication companymay value the byproduct of the litigation for
future casesprecedent or jurisprudence constantemore than the outcome in any single
casedispute resolution. In such a scenario, the value of litigation for one of the parties may
be higher than the amount actually discussed in the particular case (UπUΔ), encouraging
the litigation according to Inequation 4. The same may occur if the litigant wants to build a
reputation of being a hard negotiator and, as a result, is willing to invest in litigation.
Exploring the basic model, legal uncertainty may lead to an increase in the distance between the
plaintiffs subjective probability attributed to his own success (Pπ) and the probability attributed to
the same event by the defendant PΔ. An increase in the expectations gap increases the expected
return on litigation and, all other things being equal, the amount of litigation. If there are adequate
104See Landes, supra note 91, at 6669; see also R
ICHARD
A. P
OSNER
,E
CONOMIC
A
NALYSIS OF
L
AW
599 (2007); Gico Jr.,
Tragédia do Judiciário,supra note 51.
105See generally Daniel Kahneman & Amos Tversky, Choices, Values, and Frames,39A
M
.P
SYCHOL
. 342 (1984).
664 Ivo Teixeira Gico
incentives for judges to invest in legal certainty, the expansion of the number of cases would cause
precedents or jurisprudence constante to be formed, which, for a time, would inform the plaintiff
and defendant at a low cost as to which set of rules would be applied by the judiciary in similar
cases in the future, therefore converging the subjective probabilities Pπ!PΔ. At the limit,
assuming the absence of private information, PπPΔ.
As discussed in a previous work,106 the law of a given society is the result of its experiences and
values over time. The larger the number of legal rules accrued by a society and applied by the
judiciary, the greater the number of situations in which agents can foresee the probable outcome
of a particular conflict if taken to the judiciary; therefore, it is easier to allocate risk or enter into an
out-of-court settlement. This legal certainty allows for long-term planning, better allocation of
risks, discouragement of opportunistic behavior, andultimatelycooperation.
Nevertheless, the behavior inducement effect presupposes that the legal rule is previously estab-
lished and that it is known to or can easily be discovered by all interested parties. This is not always
the case.Suppose there is a shockthat causes the predictability of legalrules in a particularfield of law to
fall. This may result from a Supreme Court changeof position due to a new composition, a change in
legislationsuch as the introduction of a new civil codeor unexpected changes in socioeconomic
conditionssuch as the 2020 pandemic crisis. Any of these shocks can immediately render obsolete
part of the stock of existing legal rules that will lead to legal uncertainty. Assuming the legislaturedoes
not address the issue in a timely manner, this uncertainty will increase the number of cases brought
before the judiciary, as the partieswill find it more difficult to foresee what the legal rule applicable to
their specific case would be, and, therefore,what the expected value of a particular settlement would be.
The result will be a temporary increase in litigation that will force the judiciary to reevaluate the issue
until it settles on a new rule or reaffirms the old one, thus eliminating legal uncertainty. Once legal
certainty is reestablished, litigation in that particular area of law will decrease, and a new litigation
equilibrium will be reached until the next shock occurs, when the cyclewill then resume. This simple
dynamic between litigation and legal certainty can be easily illustrated107 (see Figure 4).
From a tragedy of the judiciary perspective, policymakers and judges should be working relent-
lessly to achieve as much legal certainty as possible and, when a shock is inevitable, reduce the
expansion phase and decrease the contraction phase of the litigation as much as possible by
quickly re-establishing the applicable legal rule. The litigation cycle is inevitable as a whole,
but a comprehension of the economic nature of courts and the impact of legal uncertainty on
access to justice should be more than enough to convince legal scholars that legal certainty is
the most important feature of any legal system.108
Notwithstanding this, if the legal rule applied to each case varies along with the preferences of
each judge109 or the judiciary is too noisy to create and apply a single legal rule in a timely man-
ner,110 and the distribution of cases is random, then the plaintiff and the defendant may experi-
ence great difficulty in estimating Pπand PΔ, which may subsequently increase the expectations
gap. At the limit, such an estimate may be impossible, and the parties would be faced with
uncertainty.
106See generally Gico Jr., O Capital Jurídico,supra note 27.
107Note that the cycle only illustrates the conceptual behavior of the legal system over a single legal issue. One should not
imply a necessary relationship of causality at each step, as, for example, shocks are exogenous. In other words, litigation
decreases or legal certainty does not create a shock.
108Compare Anthony DAmato, Legal Uncertainty,71C
ALIF
.L.R
EV
. 1 (1983) (arguing that legal systems move inexorably
toward greater legal uncertainty and not the other way around), with Giuseppe Dari-Mattiacci & Bruno Deffains, Uncertainty
of Law and the Legal Process, 163 J. I
NST
.&T
HEORETICAL
E
CON
. 627 (2007) (agreeing that legal uncertainty can reduce
litigation).
109For a discussion about judicial anarchy and judicial activism, see generally Gico Jr., Hermenêutica das Escolhas,supra
note 53.
110For a discussion about how the judiciary may have difficulties creating and following precedents, or jurisprudence con-
stante, see generally Gico Jr., Anarquismo Judicial,supra note 86.
German Law Journal 665
The difficulty of estimating the probability of success of a demand may trigger a cognitive limi-
tation known as a bias of unrealistic optimism or even comparative optimism.111 The bias of opti-
mism is a cognitive limitation already identified in the literature in various contexts, according to
which the human brain is programmed to be optimisticthat is, on average, people believe they
are exposed to minor risks of experiencing negative events than are others. Because they are opti-
mistic, people can, for example, invest less in prevention by utilizing fewer contraceptive meth-
ods,112 not wear seatbelts, or drive in excess of the speed limit.113 In the present case, an optimistic
plaintiff and an optimistic defendant may overestimate their respective chances of success, there-
fore diminishing or simply eliminating the possibility of reaching a settlement. The more difficult
it is to estimate these probabilities, the greater the chance becomes that the optimism bias will be
more relevant.
Legal uncertainty, however, affects not only the partiesability to estimate their success prob-
abilities, but also their ability to estimate what is legally available as a result of the litigation itself
that is, U. Faced with the absence of clear legal rules, potential plaintiffsoptimists or notmay
initiate actions to reduce interest on a consumer loan even if such interest is in line with the mar-
ket, request that a judge compel the other party to negotiate a discount on the school monthly fee,
request $500,000 from a hospital for alleged moral damages resulting from supposedly rough
medical care even though the treatment was effective, and so on. Without a clear parameter
as to what is or is not a valid claim,114 the plaintiffs imagination and optimism are the limit
for what can be claimed in courts. Entrepreneurial lawyers will certainly play a significant
role here.
Figure 4. The Litigation Cycle.
Source: adapted from Gico Jr. supra, note 27.
111See, e.g., David A. Armor & Shelley E. Taylor, Situated Optimism: Specific Outcome Expectancies and Self-Regulation,
30 A
DVANCES IN
E
XPERIMENTAL
S
OC
.P
SYCHOL
. 309 (1998).
112See generally J. M. Burger & L. Burns, The Illusion of Unique Invulnerability and the Use of Effective Contraception,
14 P
ERSONALITY
&S
OC
.P
SYCHOL
.B
ULL
. 264 (1988).
113See generally Frank P. McKenna et al., Factors Underlying Illusory Self-Assessment of Driving Skill in Males and Females,
23 A
CCIDENT
A
NALYSIS
&P
REVENTION
45 (1991).
114Notice that juseconomists would claim without the clear definition of property rights,even though in many of these
cases we are not discussing property rights. I believe a more precise term would be without the clear definition of the titularity
of the right.
666 Ivo Teixeira Gico
In this sense, the existence of legal uncertainty leads to a reduction in the number of settlements
and an increase in the number of tried cases. According to the litigation-cycle logic, this litigation
expansion will be contained only when the judiciary itself starts applying the same rules to the
same cases and thus clearly sends signals to the parties regarding its prospective behavior.
Without internal coherence, legal uncertainty could lead to an increase in the use of the judiciary
and, in effect, its overuse. Old issues will be repeatedly litigated through a vicious circle that will
create a negative feedback that, on its turn, will increase legal uncertainty and result in the inevi-
table (over)use of the judiciary.
The problem of overexploitation of the judicial system due to legal uncertainty is even more
likely when we consider that, as a rule, public policies of free access to the judicial system, such as
legal aid, public defenders, creation of special courts, and expected decrease in the average value of
attorneys fees act on the right side of Inequation 4that is, they decrease the difference:
CπCΔ
AπAΔ. As litigation becomes cheaper, the quantity demanded of public
adjudicative services increases and the inclusive public policies, all other things being equal, only
contribute to the tragedy of the judiciary.
These public policies focus on reducing the private costs of using the judicial system in order to
expand access, increasing the likelihood of litigation at first. Following the litigation cycle logic,
this initial increase in litigation would only be reflected in a greater protection of communitys
rights as a whole if the additional litigation turned into legal certainty and, consequently,
generated a retraction of litigation at a later stage resulting from the increased spontaneous com-
pliance (self-composition). Otherwise, such policies reinforce and subsidize the free access to the
resource system, but end up generating shortages of judicial services due to the excess of demand
(congestion).
As discussed, the bundled product law-court is a scarce rival resource in that the more it is used,
the more difficult it is for others to use it. However, when an individual litigant decides to take his
case to court, he considers only his private costs and benefits. The litigant does not compute the
social cost of his litigation, which includes the amount of time that other more or less important
cases will have to wait until his case is decided. Just as the cow owner in the tragedy of the
commons metaphor has the incentive to drive as many cattle as possible to pasture, litigants have
the incentive to litigate as long as their expected private marginal benefits are greater than their
expected private marginal costs. Their individual contribution to court congestion is externalized
to society.
The combination of legal uncertainty with subsidized costs to litigate generates incentives for
the parties to litigate too much, demanding public adjudicative services above the installed
capacity of the judiciary. This excess demand generates queues and lower-quality judicial deci-
sions. Litigants who do not fully bear the social cost of litigation generate queues and end up
paying with their time. The result then becomes the endemic judicial difficulty of settling issues
within a reasonable period of time, which has become known as the judicial crisisworldwide.
Considering that adjudicative services are merit goods, without the proper investment in legal
certainty, the expansion of litigation is followed neitherbyaperiodofretractionnorlegalcer-
tainty. Therefore, underinvestment in uniformization of the law contributes directly to the
overexploitation of the judicial system, and its congestion equals non-access to adjudicative
services.
Assuming there are limited resources that a given society can allocate to provide adjudicative
servicesthat is, a supply solution to the congestion problem is limitedthen, we need a
demand-side solution. Nonetheless, since adjudicative services are merit goods, most societies will
be unwilling to restrict access to justice, effectively limiting the strategies available from the
demand side too. In this sense, alongside collective actions, legal certainty can be seen as the best
tool any jurisdiction has to decrease court congestion without unduly interfering with the adju-
dication of rights. In sum, under those conditions, legal certainty is the best solution or at least the
best mitigating strategy for the tragedy of the judiciary.
German Law Journal 667
D. The Adverse Selection Problem: The Other Face of the Tragedy of the Judiciary
Court delay is only one more obvious aspect of the tragedy of the judiciary. There is another aspect
that can be even more serious and pernicious from the perspective of social justice as well as the
economic function of the judiciary as a guarantor of political and private bargaining: The adverse
selection problem resulting from court congestion and the transformation of the judiciary into a
mockery of the law. On the one hand, the idea here is that once the time element is introduced in
the analysis, the biggest the court delay, the lower the present value of a claim to the plaintiff. On
the other hand, the biggest the court delay, the lower the present value of the obligation to the
defendant is.115 In this scenario, the decreasing value of a right will prevent marginal plaintiffs
from using the adjudicative system, while, at the same time, it will encourage obligations holders
to litigate only to benefit from the diminishing effects of court delay.
In Section Cabove, I analyzed the incentives structure for settling among plaintiff and defend-
ant as well as the impact of underinvestment in legal certainty on this incentive structure. As
explained, the result was an increase in the incentive to litigate, which leads to overuse of the
judiciary and the so-called tragedy of the judiciary. At that time, however, I did not consider
the mere fact that the time gap between the beginning and the end of a demand would single-
handedly reduce the present value of a possible action and therefore reduce the expected return
of litigation. In other words, the parties are not indifferent to judicial delays when deciding
whether or not to enter into a settlement; the longer a filed case takes to be judged, the lower
its present value becomes to those vindicating their rights. The chilling effect of judicial delay
was first noticed by Posner116 and by Cappelletti and Garth,117 and was later developed by
Priest,118 the last of whom I will follow here.
In order to take into consideration the time value of money, we can modify the previously
discussed model to include a discount rate to U, making the present value of the lawsuit dependent
upon the magnitude of court delay and the time value of money, which will be represented by an
interest or discount rate. Adapting Inequation 4, we have:
Pr Litigation with gapfPπPΔU
1r
t>CπCΔAπAΔ(5)
where ris a discount rate per period and tis the number of periods between the filing of the lawsuit
and its judgment. Here, the time gap between the right violation and the lawsuit filing date is
discarded for the sake of simplicity.
Inequation 5shows that the resolution periodthat is, the time required for the adjudicative
system to settle the issueper se is capable of reducing the present value of a lawsuithence, the
probability of a dispute arising. Ceteris paribus, there is an inverse relationship between the res-
olution period and the probability of litigation. Court congestionor judicial delayincreases
the probability of settlement, while judicial celerity increases the probability of litigation. This
non-intuitive, inverse relationship suggests the existence of a dynamic interrelationship between
the resolution period and the litigation rate.
The faster the judicial system, the greater the present value of a lawsuit, and therefore the
greater the probability of a lawsuit being filed. Still, the more actions are filed, the greater the
burden on the judiciary becomes, which forces it to slow down. This subsequently increase in
the resolution periodan increase in tdecreases both the present value of lawsuits and the
115Another way of saying this is that, disconsidering legal interest rates, if applicable, the value of the claim or its utility is
decreasing in time for the plaintiff and increasing for the defendantthat is, the bigger the court delay, the better for the
defendant and the worse for the plaintiff.
116See Posner, supra note 104, at 448.
117See Cappelletti & Garth, supra note 78, at 190.
118See George L. Priest, Private Litigants and the Court Congestion Problem, 69 B.U. L. R
EV
. 53339 (1989).
668 Ivo Teixeira Gico
future demand for adjudicatory services. Thus, according to theory, changes in the resolution
period should create compensatory effects on the number of lawsuits filed and vice versa. This
relationship suggests that, given an installed capacity to provide judicial services, there must exist
an equilibrium between litigation and judicial delay in each jurisdiction: A congestion equilibrium.
Just as legal uncertainty leads to litigation expansion, which should lead to law uniformity that
leads to legal certainty, followed by litigation retractionthe litigation cyclethe litigation level
should float above and below the congestion equilibrium level as the resolution period changes.
On the one hand, the delay reductionresolution celeritycauses marginal litigators to initiate
litigation since the present value of lawsuits U
1r
tmay increase and cause the congestion to reach
another equilibrium level. On the other hand, if the judiciary is too slow, the present value of
lawsuits is reduced and marginal litigators either settle or refrain from litigating, which reduces
judicial congestion and reestablishes the equilibrium level.
The incorporation of the time element into the condition of litigation is a simple way to illus-
trate how judicial delay may destroy the social value of the judiciary and, by extension, the social
value of the law. Any group that is interested in the expansion of rights and their effectiveness
should consider judicial delay their worst enemy. It does not matter if a certain group is de jure
entitled to a certain right: If the judicial system is unduly slowtis large enoughthe present
value of this right tends to be zero, which essentially means they are de facto not entitled to the
referred right. The longer it takes to vindicate a right, the lower the present value of that right
becomes. At the limit, an arbitrarily slow judicial system destroys the value of the right itself.
This inverse relationship between court congestion and litigation volume can explain, for
example, the mystery of the Brazilian small-claims courts and their almost immediate congestion
after creation. Brazilian small-claims courts were allegedly created to reduce the long-
acknowledged court congestion problem. The traditional scholarship that motivated the creation
of the small-claims courts focused only on diminishing access and procedural costs, as can be
exemplified by Figueira Jr. and Lopes:
[T]he legal and judicial crisis that we have experienced in recent times, in the search for
results that facilitate the concretization and the effectiveness of civil procedure through
the rapid and efficient solution of intersubjective, collective or diffuse conflicts . . . The
new system liberates what was called contained litigiousness because it not only amplifies
the access to tribunals, but it also offers a much faster flow of lawsuits due to its simplified
procedure, that follows the criterion of orality, simplicity, informality, procedural economy
and celerity.119
In Brazil, small-claims courts are entirely free to usethey charge no filing feesthere is no need
to be represented by a lawyerup to a certain thresholdthe loser incurs in no legal fee in cases
of defeat, and the procedure is highly simplified. In other words, free access to the judicial system
is granted as much as possible. The objective was to grant free access to adjudicative services and
reduce court congestion. Obviously, as the tragedy of the judiciary and the congestion equilibrium
theory predict, within a very short period of time, the small-claims courts were as congested and
slow as were the regular courts.120 The court congestion problem has now endured for years, as
recognized by the National Conseil of Justice (CNJ): Created in 1995 to quickly solve the citizens
lawsuits, the Smalls-Claims Courts . . . are suffering from the same disease that has long bothered
119J
OEL
D
IAS
F
IGUEIRA
J
UNIOR
&M
AURÍCIO
A
NTONIO
R
IBEIRO
L
OPES
,C
OMENTÁRIOS À
L
EI DOS
J
UIZADOS
E
SPECIAIS
C
ÍVEIS
E
C
RIMINAIS
31, 35 (2d ed. 1997).
120See generally Conselho da Justiça Federal, Centro de Estudos Judiciários, Diagn´ostico da Estrutura e Funcionamento dos
Juizados Especiais (2003); see also Ministério da Justiça, Secretaria de Reforma do Judiciário, Juizados Especiais Cíveis: Estudo
(2005) (both recognizing that the small-claims courts were overloaded and as congested as regular courts).
German Law Journal 669
the Judiciary, the accumulation of unresolved cases and the consequent increase in congestion.121
In other words, because free access to small-claims courts is granted, queues ration access and a
congestion equilibrium is reached.
Using the theoretical framework I just laid out, one would not be surprised by the counterin-
tuitive result. A public service whose price is fixed at zero and is structured to be faster than its
competitorthe regular courtsobviously attracted a great number of litigants who gradually
elevated the congestion rate of the small-claims courts until the new congestion equilibrium point
was reached. The small-claims court that started as a club good quickly began behaving like a
common-pool resource. This behavior is no different from what we observe when the creation
of an accessory roadway first decongests the main road, but the reduction of the driving costs
brings back to the streets the marginal drivers, increasing demand for both roads until they
are both congested again. In both cases, a new congestion equilibrium is reached.
The proposition that each judicial system faces a congestion equilibrium point does not mean
that it is impossible to change that balance. It only suggests that, when considering any reform to
address the court congestion problem, the compensatory effect that will arise by the decrease in
the resolution period must also be considered. There is always a compensatory effect.
In addition, the effect of any reform individually considered will depend, to a large extent, upon
the institutional context of the judicial system in which it is implemented. The final outcome will
depend upon the degree of dispersion of the judgment result, U, the divergence of estimates of
success, PπPΔ
, the discount rate, r, the resolution period, t, the costs to litigate, CπCΔ
,
and the costs of settlement, AπAΔ
. Inequation 5illustrates how the effects of each of these
relevant dimensions of the condition of litigation can be estimated. For example, what is the
impact of hiring more judges on the number of disputes? Ceteris paribus, an increase in the num-
ber of judges reduces the resolution period of each case (t). But the reduction of tmeans the
expected return on litigation increases as the present value of the litigation, U
1r
t, increases.
This means that the volume of decisions will initially increase and the stock of cases may even
decrease, but as a compensatory effect kicks in, new cases become interesting and tend to decrease
these gains until a new congestion equilibrium is reached, as occurred with Brazilian small-claims
courts.122 Alternatively, this effect may be completely different in a given jurisdiction if, for exam-
ple, the interest rate, r, changes from year to year. The magnitude of the interaction among each of
these dimensions depends upon the characteristics of each judicial system and is ultimately an
empirical question.
Studying thirty-six European countries over the period of 200612, Bielen, Peeters, Marneffe,
and Vereeck123 found that the congestion equilibrium hypothesis holds true for highly litigious
countries with a moderate-to-high judicial independence. For those countries, the results provide
compelling evidence for the existence of an inverse relationship between litigation rates and court
backlogs. This suggests that a reduction in court congestion induces additional disputants to go to
court. As a result, an increase in the supply of adjudicative services is likely to be partly ineffective
in reducing court delay, as it creates a compensatory additional demand. Consequently, backlog
levels return to the equilibrium level.
Most interestingly, though, the authors found no evidence for an inverse relationship between
litigation rates and court backlogs for lightly litigious countries. Using the theory developed here,
one possible explanation for their result is that, in lightly litigious countries, the judiciary is still
behaving like a club good rather than a common resource, and those countries likely possess a
121See G
ILSON
L
UIZ
E
UZÉBIO
, DPJ
TRAÇARÁ DIAGNÓSTICO DOS
J
UIZADOS
E
SPECIAIS
(2012).
122Here, I am excluding the potential result of higher legal uncertainty, as an increase in the number of judges will also
increase the coordination costs among themif they are cooperating, (see Gico Jr., Anarquismo Judicial,supra note 86)and
the monitoring costs of tribunalsif the judges are within a principal-agent context, (see Gico Jr., Hermenêutica das Escolhas,
supra note 53). In either case, legal rules dispersion is likely to increase, which will subsequently increase the expectations gap.
123See Samantha Bielen et al., Backlogs and Litigation Rates: Testing Congestion Equilibrium Across European Judiciaries,53
I
NT
L
R
EV
.L.&E
CON
. 9 (2018).
670 Ivo Teixeira Gico
high level of legal certainty and a fully working litigation cycle, which ultimately induce sponta-
neous compliance. As the threat of enforcement is highly credible because the judiciary is not
congested and there exists legal certainty, the level of spontaneous cooperation is higher; hence,
the use of the adjudicative system is lower. In a way, these countries appear to have somehow
managed to maintain the supply of adjudicative services at a level higher than the quantity
demanded.
In this sense, as I have argued before, any policy aimed toward tackling the tragedy of the judi-
ciary should take into consideration the specific context of the target country. Exclusively hiring
more judges to try and expedite courts in highly litigious countries is unlikely to work, just as is the
case in the example presented above from Brazil, as the compensatory effect is likely to take hold
and the congestion equilibrium is likely to resume. Reinforcing this conclusion, Bielen, Peeters,
Marneffe, and Vereeck agree that a custom tailored policy approach is required to tackle the
problem of court delay based on the degree of litigation and the level of judicial independence
in a country.124
Inequation 5illustrates how public policies directed solely and exclusively toward expanding
access to the judicial systemthat is, reforms whose stated purpose is only to reduce the cost of
litigation CπCΔ
actually increase litigation without affecting any change in the other var-
iables of the litigation condition. In an already overloaded judicial system, increasing the number
of disputes may constitute a legitimate exercise of citizenship. Nonetheless, we must acknowledge
that it also contributes to court congestion and to the increase in time required to settle any dis-
pute (t). This increase in court delay, in turn, reduces the present value of the lawsuit to the rights
holder (U). In other words, the isolated incentive to litigate through the reduction of litigation
costsaccess to justiceall other things remaining equal, induces court congestion, which, in
turn, reduces the real value of rights.
In addition, along with the increased court congestion, a marginal group of potential users of the
judicial system will refrain from using it to assert their rights because it will not be cost-effective.
De facto, these rights holders will not be able to exercise their legitimate rights, which is exactly the
opposite result initially intended by the implementation of the free-access policies to the judicial
system. Again, we encounter the tragedy of the judiciary in a situation where the granting of free
access to a common resource system, encouraged by public subsidies, leads to an overexploitation of
the resource that, in turn, leads to the exclusion of legitimate users from the system.125
Consequently, the overuse of the judicial system may impose an even more pernicious result
other than the increase in the resolution time of disputes. Judicial delay may constitute an adverse
selection mechanism in which some legitimate rights holders are excluded from the judicial
systembecause of a lower present value of rightswhile agents with no rights are attracted pre-
cisely because judicial delay will postpone compliance with obligationsbecause of a lower
present value of legal duties. In a kind of Gresham Law,126 good litigants are excluded from
the litigation market while frivolous litigatants are drawn to it. The public policies of indiscrimi-
nate access to the judicial system, when isolated, exclude good marginal claimants from the system
due to judicial delay and reduce the social value of the judiciary as a cooperation tool. The judi-
ciary will then tragically become a market for lemons.127
124See Samantha Bielen et al., supra note 123, at 20.
125For an example of an unexpected exclusionary effect resulting from increased access to justice, that does not constitutean
adverse selection problem, see Niblett & Yoon, supra note 87.
126In economics, the Greshams law is the observation that bad money drives out good.In other words, if coins containing
metal of different value have the same value as legal tender, the coins composed of the cheaper metal will be used for payment,
while those made of the more expensive metal will be hoarded or exported and thus will tend to disappear from circulation.
127See George A. Akerlof, The Market for Lemons: Quality Uncertainty and the Market Mechanism, 84 Q.J. E
CON
. 488
(1970). But see Hugh S. E. Gravelle, Rationing Trials by Waiting: Welfare Implications,10I
NT
L
R
EV
.L.&E
CON
. 25570 (1990)
(discussing when rationing courts by delay may be efficient).
German Law Journal 671
To mitigate this problem, legal claims should be entitled to earn interest. Note that the inclu-
sion of adequate interest rates can mitigate the exclusionary effects of judicial delay and the prob-
lem of adverse selection of litigants. We may again alter the model to include the payment of
interests by simply including in Inequation 5the desired judicial interest rate (rjthat will not
necessarily be the same as the market rate rm
:
Pr Litigation with interestsfPπPΔU
1rmrj

t>CπCΔAπAΔ(6)
In this hypothesis, judicial delay continues to reduce the probability of litigation, although to a
lesser extent, assuming that the judicial interest rate (rjis lower than the market interest rate rm
,
that is rj<rm. On the one hand, if rjrm, then the litigation value will be adjusted at the market
ratepresuming this is the opportunity cost of the litigantmeaning there will be no time cost
associated with judicial delay, and it is no longer relevant in determining the probability of a lit-
igation. On the other hand, if rj>rm, the expected value of the litigation rises, because the slower
the Judiciary, the greater the present value of the claim to the plaintiff. In this sense, as a general
public policy, judicial interest rates should be as close as possible to market rates in order to pre-
vent the adverse selection problem of excluding rightful claims and attracting those who want to
postpone the fulfillment of obligations due to court delay. It goes without saying that the fastest
courts are, the less important the judicial interest rate is.
Therefore, the best strategy to increase the number of out-of-court settlements and reduce the
use of the judicial system without excluding users due to system coststhat is, guaranteeing free
accessinvolves investing in legal certainty and the resulting convergence of Pπand PΔthe
expectations gap. Such an investment is even more relevant when the political decision is made
to eliminate or at least substantially reduce the private costs of litigation CπCΔ
, and recent
studies indicate that hiring more judges in already litigious jurisdictions will not reduce court
congestion due to the compensatory effect. In any case, the adoption of a legal interest rate com-
patible with the market interest rate may mitigate and even eliminate the adverse selection prob-
lem resulting from judicial delay. In a world in which free access to the judicial system is a goal,
investment in legal certainty and market interest rates for legal claims seem to be the only mea-
sures without undesirable side effects.
D. Conclusions
Judicial delay is a worldwide phenomenon that has been the subject of mitigating public policies
since the beginning of the twentieth century. At each wave of reform, innumerable measures have
been undertaken to try to alleviate or resolve the issue, not always successfully. I believe part of this
failure is due to the lack of comprehension of the economic nature of law and courts. The eco-
nomic theory of law delineated in this Article may help policymakers understand the dynamics of
legal rights, courts, the interrelationship between these elementsthe bundled productand
access-to-justice policies.
By realizing that law as coercion may be comprehended as a club good and the judiciary as a
private good and that both are perfect complementary goods, we are now able to better understand
why unrestricted access-to-justice policies transforms the judicial system into a common pool
resource that is simultaneously non-excludable and rival. The unrestricted access to this rival
resource will then lead to the court congestion phenomenon that plagues jurisdictions worldwide.
In this sense, traditional access-to-justice policies may lead to increased access to the judicial sys-
tem itself while simultaneously excluding people from having access to their rights, reinforcing the
tragedy of the judiciarya totally counterintuitive result.
Because adjudicative services are merit goods, access restriction implies a restriction to access to
ones rights. Therefore, another kind of policy strategy must be adopted to prevent the overex-
ploitation of adjudicative systems. In this context, the importance of investing in legal certainty is
672 Ivo Teixeira Gico
paramount, since this is the most important approach that may tackle the tragedy of the judiciary
conundrum without generating side effects, while also simultaneously reinforcing the social value
of courts and the law.
If overexploitation of the adjudicative system is a tragic result of access-to-justice policies, then
(1) underinvestment in legal certainty, (2) the exclusion of marginal litigants due to court delay
and (3) the attraction of sham litigantsadverse selectionbecome an even more tragic effect of
this institutional arrangement. Although this problem can be mitigated by the adoption of legal
interest rates compatible with market rates, the only measure that effectively reduces the use of the
adjudicative system as a whole while maintaining free access is the investment in legal certainty.
Therefore, legal certainty should be the primary focus of public policies aimed toward increasing
access to public adjudication services while simultaneously attempting to prevent, or at least to
mitigate, the tragedy of the judiciary.
Cite this article: Gico Jr IT (2020). The Tragedy of the Judiciary: An Inquiry into the Economic Nature of Law and Courts.
German Law Journal 21, 644673. https://doi.org/10.1017/glj.2020.34
German Law Journal 673
ResearchGate has not been able to resolve any citations for this publication.
Thesis
Full-text available
O desenvolvimento do complexo agroindustrial da soja no Brasil deu-se, em parte, como decorrência do surgimento de formas alternativas de crédito, tal como a venda antecipada de soja por meio de contratos. O problema que motivou este estudo foi decorrente das quebras contratuais por parte dos produtores rurais em um momento de expressiva alta do preço e suas conseqüentes disputas judiciais. Foram realizadas análises descritivas e econométricas (PROBIT) sobre 161 decisões judiciais do Tribunal de Justiça de Goiás e uma pesquisa quantitativa com 70 produtores rurais. O estudo tem como hipótese central que o ambiente institucional deficiente eleva custos de transação e reduz o desenvolvimento econômico. Os testes realizados entre decisões judiciais de primeira e segunda instância (Apelação) indicaram larga dispersão, bem como entre decisões no mesmo nível: em comarcas do interior alguns resultados foram 100% favorecendo a manutenção dos contratos e outros exatamente o contrário. Da mesma forma, a análise das decisões de Apelação revelou a diversidade de entendimento das turmas julgadoras entre as Câmaras Cíveis. No entanto, a probabilidade de o contrato ser mantido aumentou ano após ano e uma das variáveis significantes foi a primeira decisão do STJ sobre esse conflito, favorável à manutenção dos contratos. Os impactos das decisões jurídicas no ambiente econômico podem ser percebidos, como a maior exigência de garantias e a redução do número de contratos. Os produtores que não quebraram seus contratos também foram negativamente afetados com as novas estratégias adotadas pelas empresas compradoras de soja. As empresas ficaram receosas de negociar, uma vez que não puderam contar com a segurança jurídica de que o contrato seria cumprido. O uso do conceito de função social dos contratos gera neles um alto grau de instabilidade, portanto, os custos de transação aumentaram para todos os agentes, bem como a importância de sanções econômicas.
Article
Full-text available
To address the problem of court backlogs, policymakers in many countries have been pursuing reforms to reduce case disposition times and the demand for litigation. Yet Priest’s (1989) congestion-equilibrium theory states that reforms aimed at reducing court delays are offset by an increased tendency to litigate. To test the congestion-equilibrium hypothesis, we use biennial panel data from 36 European countries over the period 2006–2012. Specifically, we estimate (i) a repeated cross-section model using conventional (pooled) OLS, (ii) a standard (static) random-effects panel data model, and (iii) a (static) random effects model using the new method of unconditional quantile regression (UQR) to examine the countrylevel relationship between litigation rates and court backlogs (measured by the number of pending cases per judge in each country). In accordance with the congestion equilibrium hypothesis, the UQR estimates suggest a negative impact of court backlogs on litigation rates but only in highly litigious countries, insofar as the latter experience a high degree of judicial independence. This shows the need for custom-tailored policy approaches to tackling court delay based on countries’ existing litigation rates.
Article
Answering one question often begets another. We present a decision-theoretic model that describes how this affects the sequencing of decisions over time. Because answering an easy question may raise a more difficult one, a rational judge may delay resolution even if he has perfect information about the correct decision. Furthermore, because otherwise unrelated questions may raise similar follow-ups, he may optimally clump decisions together. Our theory thus generates an endogenous economy of scale in dispute resolution and contributes to the literature on punctuated equilibrium theory. We illustrate the results of our model with a case study from legal history in the United States.
Article
Priest and Klein's 1984 article, "The Selection of Disputes for Litigation," famously hypothesized a "tendency toward 50 percent plaintiff victories" among litigated cases. Despite the article's enduring influence, its results have never been formally proved, and doubts remain about their meaning, validity, and generality. This article makes two main contributions. First, it distinguishes six hypotheses plausibly attributable to Priest and Klein. Second, it mathematically proves or disproves the hypotheses under a generalized version of Priest and Klein's model. The Fifty-Percent Limit Hypothesis and three other hypotheses attributable to Priest and Klein (1984) are mathematically well-founded and true under the assumptions made by Priest and Klein. In fact, they are true under a wider array of assumptions. More specifically, the Trial Selection Hypothesis, Fifty-Percent Limit Hypothesis, Asymmetric Stakes Hypothesis, and Irrelevance of Dispute Distribution Hypothesis are true for any distribution of disputes that is bounded, strictly positive, and continuous. The Fifty-Percent Bias Hypothesis is true when the parties are veil/accurate in estimating case outcomes, but only sometimes true when they are less accurate. As shown in Klerman and Lee (2014), the No Inferences Hypothesis is false.
Article
Forum shopping is problematic because it may lead to forum selling. For diverse motives, including prestige, local benefits, or re-election, some judges want to hear more cases. When plaintiffs have wide choice of forum, such judges have incentives to make the law more pro-plaintiff because plaintiffs choose the court. While only a few judges may be motivated to attract more cases, their actions can have large effects because their courts will attract a disproportionate share of cases. For example, judges in the Eastern District of Texas have attracted patent plaintiffs to their district by distorting the rules and practices relating to case assignment, joinder, discovery, transfer, and summary judgment. As a result of these efforts, more than a quarter of all patent infringement suits were filed in the Eastern District of Texas in 2014. Consideration of forum selling helps justify constitutional constraints on personal jurisdiction. Without constitutional limits on jurisdiction, some courts are likely to be biased in favor of plaintiffs in order to attract litigation. This Article explores forum selling through five case studies: patent litigation in the Eastern District of Texas and elsewhere; class actions and mass torts in "magnet jurisdictions" such as Madison County, Illinois; bankruptcy and the District of Delaware; ICANN domain name arbitration; and common law judging in early modern England.