Tom Ginsburg, “Constitutions as Products”
Articles Blog

Tom Ginsburg, “Constitutions as Products”

September 18, 2019


PRESENTER: This lecture is held
every year at the Law School to honor Nobel Laureate and
Clifton R. Musser Professor Emeritus of Law, Ronald Coase. It’s supported by grant from
the John M. Olin Foundation and by the newly named– this is the first lecture
since we named the Coase-Sandor Institute for Law and Economics. Now, Lorne economics
was a discipline that was born here
at the law school. And Ronald Coase
was, as many of you already know from your
coursework here, as well as perhaps your education
prior to coming here– one of the major pioneers. Perhaps the best known
article that Ronald wrote, “The Problem of
Social Cost”, was published in the Law
School’s own journal of law and economics and is the
most cited law review article in history. At age 102, Ronald is
still actively pursuing his research agenda,
something that we should all be so lucky to do. He’s living in Chicago. Just recently published a
coauthored book entitled, How China Became Capitalist. And indeed, Ronald
will be here in March, when we have an
event to celebrate the naming of the
Coase-Sandor Institute and will hopefully participate
in some of the events that day here at the law school. Now, when Ronald came to
the Law School in 1962, law and economics
already had a foothold. But he expanded its influence
so that the law school became the dominant school in
this area shortly thereafter and has remained
that way ever since. The discipline spread to
every law school and economics department in the country
and has now become– perhaps thanks to
our own program– internationally known. And it’s a way of rethinking
US legal systems and the law and how they’ve transpired,
how they’ve taken place over the past 50 years. It is indeed striking to
sit in a faculty meeting and hear legal scholars
of all stripes, of all disciplines talk about
the language of incentives, preferences, markets,
and utility maximization, as well as other norms. Now, perhaps the major
insight of law and economics besides the Coase
theorem that, of course, Ronald will actually
deny that he ever came up with the Coase theorem. So you should always understand
that when you talk with him. But it’s something that, for
those of you who are students, you should have
memorized already. It is the application of
social scientific tools to research in
other legal fields. For example, quantitative
and econometrics have become the norm in
many fields of study. Not just economics but
sociology, political science, and other disciplines. For this reason
and many others, I am especially delighted to
welcome Tom Ginsburg, the Leo Spitz Professor of International
Law and the Ludwig and Hilde Wolf Research Scholar and
Professor of Political Science today to give the Coase lecture. Thomas is distinct among
many of our Coase lecturers, because he is someone who–
in addition to using law and economics as a method– does a significant amount of
comparative and international law using legal
scholarship, using social scientific methods. His work is interdisciplinary. He typically works
on research involving comparative constitutions
and is a wonderful example of the dialogue so unique
to the University of Chicago between law and economics,
political science, and other disciplines. A story that I sometimes think
about is before I came here, I was dean of
another law school. And we looked at
all of the scholars that we could get
who knew something about international law. And Tom was at the
very top of my list. At that time, he
was visiting here. He broke my heart and came to
Chicago instead of the school that I was at. So that led me to have
to come to Chicago so we could be reunited here. In any event, I am so pleased
that Tom is on this faculty, that his scholarship is
of such a high level, and really delivering
enormous insights into international comparative
constitutional law. Now, following Professor
Ginsburg’s lecture today, I’d like to invite all
of you to join us out in the atrium, where we will
have refreshments and engage in continued discussion. But for now, I’d like to
introduce Tom Ginsburg to deliver the Coase lecture. TOM GINSBURG: Thank you, Mike. And thanks to all of
you for coming today. You know, I know the
lecture is supposed to be directed at
first year students. And many of you
first year students will have already
been introduced to constitutional law
through the mechanism– through the method
that we usually use to introduce it
to students, which is to have you read cases. This is the way that we
teach constitutional law in law schools. It’s the way we think
about constitutions. I sometimes analogize this
focus on cases and case law as somewhat like a baseball fan. A baseball fan who
shows up at a baseball game around the seventh
inning watches the inning with intense concentration,
focuses especially on the umpire– the balls and the strikes. And then, when the inning is
over, the fan leaves the game. That is when you look at a case. When you read a case closely,
you are looking at the umpire. But you leave. You don’t see what happens
after the case is over. We often don’t spend
much time looking at the actual
implementation of decisions. We also don’t spend
much time talking about the political and social
interests that are at dispute in any particular case. Nor the even more
fundamental question, which is constitutional order. Where does it come from? Now, when one is looking
at a comparative context, I think it’s
especially important to focus on the game
outside the seventh inning. And to do that, of course,
requires tools outside law. You have to use the
tools of social science. And the reason it’s so important
in a comparitive context is because one cannot always
take constitutional order for granted. It is not obvious in many
countries of the world that court decisions
will be obeyed. It is not obvious that
judges will be deciding them in a neutral way. And so, this requires
broadening our lens. Now, of course, there’s many
social sciences one can use. And economists have been focused
on constitutions for some time. And they’ve come up and
made many contributions to the study of constitutions. Economists tend to focus on
the functions of constitutions. What do I mean by the functions? Well, from an economic
perspective, of course, government is a device for
generating public goods, those goods which cannot or
are unlikely to be produced in sufficient quantity
on the private market. Things like national defense or
good environmental regulation. And, of course, if government
is a device for doing that, we need rules to
structure government. A particular concern
from economic theory is the idea of agency control. That is, the people who are
running the government for us are, in some sense, our agents. We hire them to do
some work for us. But as with any
agent you hire, you have to make sure
that that agent is going to be performing
in accordance with your instructions
and not simply acting on behalf of their own benefit. And so, from an
economic perspective, a lot of the rules
of constitutional law and constitutions can
be explained as devices to help govern the selection
and the controlling of agents. Think about regular elections or
criminal procedure rules, which limit what the police can do. Many of things–
these things are quite understandable from
an economic perspective, from that point of view. Another idea that
economists have contributed is the idea of pre-commitment. That is, it may sometimes
make sense for us to limit our collective choices. This is not really about
the agents we select, but as taking certain things
out of the decision function, certain things that
we cannot decide. Because we are afraid
that if we are allowed to decide these
things, we might do so with our short-term
interests in mind, rather than our long-term interests. It might make sense
for all of us, for example, to agree to
confiscate Dean Schill’s apartment and split up the
proceeds among ourselves. But of course, that wouldn’t
be good for Dean Schill. Nor would it be good
for the society, because people in
such circumstances would be less likely
to invest in property, they would be less likely to
invest in their own futures. And so, many times
constitutional rules take things off the table. Now, these are all
standard things out there in the literature. And they’re all
focused, notably, on the functions
of constitutions. I want to look at constitutions
from another angle. And that is not to look at their
function so much as their form. And that’s why I’ve titled
this Constitution as Products. If you think about it,
products have functions. But products also have form. Take this cup, for example. It has a function,
that’s to hold water. But it also has other
attributes which it might be interesting
to interrogate from the point of view
of understanding the cup and trying to get a
complete picture of it. It was produced
in a certain way. And that might tell
us something about it. We might want to
know, for example, if it was mass
produced or if it was produced– as this one
was– by hand by someone. That might make a difference
in its value and such. And similarly, we can
constitutions around the world from this point of view. We can look at the
production process and see if it has any impact
on the actual product that is produced that way. The cup also has
certain features, right? It’s got a color. It’s got a shape. It’s got a handle. Some of these
features, of course, go to the core of
what the cup is. But some of them are
quite peripheral. We could get rid of
them, they’re optional. We could change the color,
it would still be a cup. Probably even take
off the handle. You could change
the shape a bit, but if you change it too
much, it would no longer be recognizable as a cup. So using this idea of core
and peripheral features, I want to interrogate
what it is that really is the essence of constitutions. Or what is conventional,
if you will, to put in the product
called a constitution. In addition, constitutions,
like the cup, have substitutes. That is, if I want
to drink water, I can drink it out of a cup. I can also drink
it out of a glass. Pretty much the same function. There are many things out there,
many kinds of institutions which will play similar
functions to those that we’ve identified for constitutions. That is, if we want
to control agents or we want to tie
our hands, there’s other technologies out there
which are available to us. We could use norms, for example. We could have unwritten
understandings that would function the same
way, to limit the people we select to govern us. We could rely on case law. We could rely on statutes. We could rely on a political
program and a political party to control agents and such. So there’s substitutes
for what constitutions do. But there’s also
some features which seem to be quite essential. And so, the fact that
there are substitutes doesn’t mean that there aren’t
a core set of attributes of a constitution. One way that people–
constitutional scholars have talked about this is to contrast
the small C constitution– those various things
which are outside the formal written document– with the large C constitution,
which is the single document that most countries have. And I want to focus
on that, actually. I want to focus on the latter. So I recognize
that the product– which I’m going to
call the constitution, the written, formal
constitution– is obviously not the only thing
which can play the function. But I think it will be
interesting to interrogate its features nevertheless. Now, one final
introductory point. What defines a product
is a certain kind of mix of function and form. So suppose my cup here had
a little hole in the bottom. It didn’t serve its function. It was unable to hold water. If I held it up, most
of you would still identify it as a cup, even if it
was leaking water right on me. Now, what that
suggests is that it’s some combination of
function and form, which would be the essential
defining attributes. If, however, we were
to change the shape of the cup or we might– significantly– it might
just have the same function, but no longer be a cup. Let’s say it might be a bowl. So the point is that
it doesn’t have to work to still be a constitution. In fact, what’s one of the
things I want to interrogate? Under what circumstances
do constitutions make a difference
for actual behavior? Under what circumstances
are their provisions actually followed? The cup, pretty durable. It’s made of ceramics. It’ll last forever, as
long as I don’t drop it. It was made of
paper, it would not. So that’s one other
feature we might look at when we interrogate
constitutions as projects. Some of them last a long
time, some of them don’t. And I want to ask, what are
the things which allow them to focus, to be more effective? Now, for all of this, I’m
drawing on a large data project that I’ve been running
for several years. A couple of you in the room have
actually worked with me on it. And it involves
looking at the contents of written constitutions
across time and space. The effort’s ambitious, it’s to
look at every constitution ever written for an independent
nation state since 1789, for all independent countries. For most countries, it’s pretty
easy to identify what this is. There’s a single
written document called the Constitution. There are a small
number of countries which don’t have them. Israel, for example, has
a series of basic laws. In Saudi Arabia, the position
is that the Constitution is the holy Quran, so you can’t–
man can’t make constitutional law. For countries like that, we
have a secondary decision rule. That is, if the law is called
a basic law, a bill of rights, or if it sets up a
branch of government, we call it constitutional. For virtually every analysis
we’ve ever undertaken, you can throw those countries
out and it’s still– results seem to be quite robust. I should add that
the United Kingdom is the one country which I
don’t include in the analysis. Every time I ask a British
constitutional lawyer, what is the British Constitution? I get a different answer. So I figure if they can’t
figure it out themselves, I’m unlikely to be able
to improve on that. All right, the
constitution as a product has been an incredibly
successful one. If you analogize
countries to firms, firms can choose to use any
particular kind of governance technology that they want. But constitutions– formal,
written constitution- are a really popular one. This figure gives– for
each year since 1789– the number of new constitutions
from 0 to 25 written in any particular year. The top line here,
the dark line, is the number of countries
in existence in the world. And as you can see, it’s
roughly quadrupled since 1789. Breakup of empires and
such, many new countries. The bottom line
here is the number of countries which have
a written constitution. And so, you can see at
the outset of the period, it was quite rare. United States, France,
Poland, a couple of countries. As time went on, though,
the two lines converge. So that now every new country
that comes into existence is likely to– as one of
its very first acts– write a new constitution. Indeed, many countries are
preparing constitutions in advance of
becoming independent. Scotland just
announced this week that this is their program. They’re trying to become
independent of the United Kingdom, and so
they’re preparing a constitutional
draft just in advance. So the point is that the product
has been incredibly successful. And that’s an interesting
thing to try to understand. I want to look at the
production process a little bit, so I want to start with that. How are these things produced? In any given year, there’ll
be five or 10 constitutions produced. There are various kinds of
ways that one goes about this. Sometimes they’ll
just be written by a very distinct or very small
group of government elites. The constitution of Burma
passed a few years ago was written by just a
handful of people in secret. No one knew what was going on. In other cases, they’re
adopted in very open processes. That is, they are produced in
constitutional assemblies which might have been
elected by the public. They might have public input. And they might even feature,
at the end of the process, a kind of public
referendum, an up or down vote on whether or not
the Constitution should be brought into force. So there’s a great
variety of mechanisms. One way just to contrast is to– to illustrate the
extremes might be to look at the Constitution
of Egypt, which many of you will have noted was
passed in December in a very violent process. In which the majority
decided that it didn’t want to
include other voices and rammed through
a constitution of dubious
constitutional legality itself right before a
Constitutional Court could declare the
process unconstitutional. That’s a kind of pure
majoritarian model. And that’s one approach. At the other extreme, you
have very open processes. Iceland, for example,
last year has adopted a constitution in a very
novel, very 21st century way. They were quite upset
with their government, as the whole country was a
giant, sub-prime mortgage problem itself. The economy collapsed after
the great financial crisis of 2007 and 2008. And the people were very
angry, so they set up a process to draft a new constitution. They first randomly
selected 1,000 citizens to generate ideas for what this
Constitution should contain. They then held a public
election in which anyone could run, as long as
you were not a politician. And then those people then
drafted a constitution which was put before the
public again in a referendum. And it may be passed
later this year. So the other extreme,
a very open process. By the way, they had a
Facebook page and somewhat of a wiki process, where
people could contribute to drafts in a very open way. So this is the
21st century model. Does it make a difference? Well, it’s hard to know. The traditional
view, of course, is that constitutions are going
to be written by statesmen. James Madison,
Alexander Hamilton, you have these great figures
acting in the public interest. The economists who’ve worked
on constitutional design tend to be very
skeptical of that. Economics growth is
about incentives. And we’re always very
skeptical of the idea that anyone might be
publicly motivated. And so, there’s
somewhat of a tension. In particular, if
constitutions are designed to limit
agency costs, if they’re designed to produce good
governance outcomes, there is a prior question as to
how we get a constitution that will do that. Notwithstanding the fact
that it’s also produced in a political process. That is, why don’t agency
problems also exist at the constitutional level? There is an assertion in the
constitutional literature that constitutional politics
are truly different, that they are higher order. But I wonder why we
would think that. Well, one reason we might think
that is that constitutions are, of course, great symbolic
documents and they’re adopted in very public ways. The visibility
might itself serve to constrain agency
costs, because everyone can see what’s going on. Everyone will be motivated–
because of the high stakes– to watch. It also might serve to
facilitate group organization. So another staple of the
public choice literature is that groups, large groups– such as the public or
taxpayers or consumers– it’s very difficult to organize. Because their individual stake
in any particular public choice decision is very small. If the Constitution is
at stake, though, you might be able to overcome some
of these collective action problems, which might lead
to constitutional politics being of a different quality. On the other hand, you
might rebut that by saying, the very fact that
the stakes are so high might increase the incentive
for interest groups, for private interest groups to
also mobilize if you can get a rule into the constitution. Well, then, that’s going
to be very valuable and it’s going to be entrenched. And so, you’re likely to see
some great effort to organize and great expense to be exerted,
in terms of trying to get rules into the constitution. So there’s a bit of a
tension in the literature. But I do think that one
implication of all this is that the more open processes,
the more Iceland-style processes are likely to produce,
if you will, better documents, more open documents. Ones that are likely to involve
the public down the road. To test this, I
and some colleagues looked at trying to relate
process with outcome. Process and product. So we categorized
constitutional design exercises in a variety of ways. We asked if it had a
constituent assembly, if there was a public referendum
thereafter, whether it was executive-centered. And we looked at the
outcomes, looking at the contents of
the constitutions that were produced in these ways,
and tried to examine whether we could find any relationship. Interestingly, we didn’t
find many relationships. And this is somewhat surprising. I mean, it’s become a norm
to think that participation is always good. But we did find two
things in particular. First of all, those
constitutions which were dominated by the executive
in the production process tended to be– to produce executive-dominated
governments. That’s not particularly
surprising. Similarly, those
constitutions in which there was a public referendum
at the end of the process tended to involve more
instruments of direct democracy down the road. So public involvement in
the production process leads to public
involvement in the product. And so, that’s a
kind of relationship which seemed to hold. In addition, when the
public was involved, there did seem to be a
greater number of rights in the Constitution. So that suggests that
there may be something good about
participating, that might be able to overcome some of
the production– the problems that one might think of using
the simple economic theory that the constitution makers
are likely to be motivated by self-interest. Another question I’ve been
interrogating is innovation. Innovation. If you think about a product,
we understand somewhat about where
innovation comes from. If you can design a better
cup, a cup that looks better is more attractive. Wouldn’t be too hard. A cup that is more
durable, perhaps one that holds water in
a more efficient way. You’re likely to do so
in a market setting, because you’re likely to
be rewarded for doing so. Constitutions and
public choice generally doesn’t have this advantage. The fact is that there
is no ongoing revenue stream for the designer
of a constitution, nor do they necessarily get
much credit for their work. There is no intellectual
property right that we know of in public law. In some areas of
legal production, I think this is resolved by
the concept of reputation. So if I’m a judge
deciding cases, I might want to
have a reputation for being an innovative judge. And I might come up
with good legal rules. And that might be good
for me down the road. Similarly, we sometimes have
legislators who take the lead. They get to claim credit
before their constituents, they put their name on the bill. Constitution makers,
by and large, are not in this category. They’re mostly people
who do this one time. They’ll never do it again. They’re mostly faceless people. They’re sometimes technocrats. For every James Madison
that’s out there– whose name everyone knows– there is Jared Ingersoll, who
I’m sure no one in the room can identify, but was
one of the signers– one of the designers of
the American Constitution. It’s not clear that
people are going to get the kind of
reputational benefit. And that, I think, is a
general problem for public law, particularly difficult
in this context. Furthermore, the way
constitution writing works is different from other
examples or other things we might look to
in comparative law. Famously in
corporate law, right? We know that in
the United States, corporate law is a
feature of state law. And there’s a lot of
theory out about– out there about how
competition among the states will induce law makers
to produce better law. there will be a race to
the top in some theories. Others think there will
be a race to the bottom. But the point is
that competition over mobile population
corporations or citizens can induce better
law, can induce innovation, good innovation. We don’t have that in the case
of national constitutions. The other thing is
that we’re operating in an environment
of tremendously incomplete information. It’s very hard. What you’re doing when
you’re writing a constitution is picking a set of
institutions to apply to a particular context. We know very little,
really, about what it is that makes some things
work in some contexts and not others. If you adopt an
institution that works, you don’t really know if
that’s a feature of the context or a feature of the institution. And social science,
I have to say, has not done a
particularly good job here. The entire discipline
of comparative politics has been fighting for 30
years about the merits of presidentialism
and parliamentarism. And the current
state of knowledge is that, well, we
don’t really know. Doesn’t matter much. Or we can’t tell that it– we’re not sure that it
does make a difference. So imagine you’re a
constitutional designer and you’re operating under
this great circumstance of incomplete information. It’s unlikely that you’re going
to invest much in coming up with novel institutions. It’s unlikely that you’re
going to necessarily produce a very good set of choices. It is true, though, that
we’re operating also in a global environment. And so, most constitution
making processes today involve
international actors, in which there are people who
are international organizations who are promoting their
particular point of view, the United Nations and such. You might view this
as a kind of subsidy for certain forms of knowledge,
subsidizing certain positions, making them easier to adopt. But all in all, my
suggestion is that there’s likely to be a kind
of underproduction of new innovations. That we stick with the
institutional categories, the institutional menu
that we’ve inherited. And there’s probably
too little innovation, certainly relative
to a private market. Where, again, producers are
likely to have an incentive to try out new things. One implication of this
is that constitutions are likely to be similar in a
particular context across time. And that’s one thing we observe. So in these figures,
what I’ve done is look at the features,
100 different variables, for a particular constitution
and compare it with– for any particular
constitution– with its predecessor. You could do this
for any two dyads. I could take any two random
constitutions out of the pool and compare them. And it would be likely to– they would likely
have some correlation, but not a very high one. One way to think about
this is a correlation. And the baseline level is
about 0.6, about this level. When you look within
a country– so this will represent Brazil’s
constitution of roughly 1900 relative to the earlier
one of the 1860s. Venezuela’s had a
lot of constitutions. But the point is that
these figures, these dots are all pretty high, right? That is, once a
set of institutions is adopted in a
particular country, it tends to be somewhat sticky. It tends to last. And that, itself, is interesting
and worthy of interrogation. It suggests, again, that maybe– because maybe these are not
sort of optimal situations. With it, it’s hard to tell. But the fact is that
we don’t really– that constitution
makers are likely– seem to be sticking with
the choices they know, notwithstanding great
changes that are going on in these societies. Sometimes society will
move between a dictatorship and a democracy. Most of the institutions will
stay really quite stable. The other thing is
that if you look at the entire
corpus of products, there is a lot of stickiness. There’s sticky
defaults, as I put it. The American
constitutional designers, when they came up
with their design, they decided that
the president should be at least 35 years
of age and legislators should be at least 25. Probably a pretty
undemocratic thing, actually, if you think about it. There’s not that much
rationale for limiting it. But in any case,
that’s what they did. It turns out that to
this day, that is still the most common number
to put in there as an age limit, notwithstanding
the fact there has been tremendous
improvement in life expectancy since the 18th century. Right? There is a dramatic change
in underlying circumstances, yet we stick with the
rules that we came up with. And other countries seemed to
borrow it, seem to copy it. It’s a pretty minor thing
and so not much incentive, I suppose, to
invest in fixing it. One thing I was
looking at recently is phrases in preambles. And it turns out we the people
is a tremendously popular phrase, perhaps not surprising. But that over time,
it’s risen and fallen. Now it seems to be quite a norm. Notwithstanding that,
there are occasionally new institutions
that come into being, new technologies
one might think of and one might track
their spread over time. So this looks at
the institution of independent regulatory
institutions, things like the Federal Communications
Commission and such. And in many countries, those
are constitutionalized. One of the things
I’m puzzled over– and struggling with as more
of a question than an answer, then, for now– is why things which are
invented in a very early period take quite a while
before they take off. But it’s not an
uncommon pattern. We see it in
innovation that occurs. But innovation, of course, is
separate from the diffusion or separate from the
spread of the institution in terms of popularity. And we often see a big gap. And this is true here. So the point is that there is
some innovation that occurs. It’s not clear how
much and it’s not clear what the forces are
that cause countries to borrow institutions
from abroad. We’re not really sure if
when independent regulatory institutions spread,
for example, if that’s a function of learning. Right? If that’s us collectively
getting better as a new handle design might
spread in the world of cups, or whether that’s
simply just a fashion, as a new color might be
if that were to spread. And that’s a research agenda,
really, going forward, is trying to figure out what is
the factors driving diffusion. One of the great trends
in form in constitutions is the spread of rights. So here I have
time on this axis, the number of rights in any
given constitution found in particular countries. And what you see is
that line over there captures the trend
that over time, there have been more and more rights
in national constitutions. There’s greater variance,
I suppose, but certainly more rights. And so, that’s
certainly a trend. This one, I think there is
somewhat of an explanation. And I think part of it has to do
with the transnational context in which constitutions
are increasingly produced. If one looks at the menus of
rights drafted by national– adopted by national drafters,
and one compares them with international instruments– Universal Declaration
of Human Rights, the International Covenant of
Civil and Political Rights– one sees a lot of convergence. You know, you might think about
this in terms of a boilerplate set of rights that are
adopted by constitution makers from abroad. But it might also, of
course, be a product of some sort of
subsidy or lobbying that occurs from outside. But it’s certainly
a big trend and one that’s worthy of explanation. Again, there is a
large difference between rights that are– between the time a right
is innervated, if you will, a time it enters the corpus
and its subsequent spread. So this is just to look
at a few different rights. Right to freedom of expression
is always very popular, starts very popular
and stays that way. Right to health care
emerges in the early 1900s and really takes
off after the 1950s. It’s kind of considered a
second generation right. The right to education is also
considered a second generation right, but notice that
it’s popular much early on. It really takes off earlier. And finally, there’s
the one in the news these days, the
right to bear arms. In the early 1900s,
a lot of countries were just copying the American
template, if you will. And it was fairly popular. Over time, one might
say they’ve learned. Others might say that
they’re just following some international fashion. But it’s certainly the one
right in our Bill of Rights, which has decreased in
popularity over time. All right, so all of
this goes to the contents of constitutions. And I want to just
step back a little bit and think about the
determinants of this. As I said, there’s
a lot of stickiness, in terms of the first
institutions you adopt. But there are also these kind
of time pressures, right? And that might
affect, if you will, the shape of the
constitutional product. In this figure,
what I do is I look at the popularity of
particular features over time. And one way to– I realize it’s hard to read,
for those of you in the back– but this column here
looks at the percentage of constitutions in force which
have any particular provision. So does the Constitution
refer to– require oaths? Very popular now. It has trended over time. And you could see
that oaths have always been kind of a core
feature, if you will, of the constitutional product. Same with provisions
on amendment. Other things, though,
which are now very popular have really only
become popular later. So does the Constitution
refer to political parties? In the 19th century, very
few constitutions did so. Now, many constitutions do. And it’s become a kind of norm. And again, we
don’t know, really, if that’s a response to
trends or just borrowing. Or if that’s actually learning. My guess is the latter. Political parties weren’t
well-known in many countries in the 19th century. As they’ve emerged
and they’ve taken on a constitutional
function, the products have responded, if you will. One could compare
this with things which are much less core. They’re the equivalent
of the color or the handle design
of the cup, right? They’re optional. They’re not essential features. Does the constitution contain
a motto, for example, I ask. Well, some of them do. About 20% in force now. And it’s been pretty
constant over time. Does the Constitution refer
to a specific individual? Well, it’s interesting. A lot of individual–
many constitutions do. The Chinese
constitution mentions Mao and the Vietnamese
constitution mentions Ho Chi Minh. A few constitutions
mention Jesus or Muhammad. But it’s not a very
popular thing, 9%. It’s not what we would
call a core feature. Sometimes you have things
which were not core features but now seem to be on the rise. A Human Rights Commission
was invented, really, in 1986 and has started to take off. Victims’ rights is
another one which seems to be a trend
in that direction. On the other hand,
some things are disappearing from the corpus. Bankruptcy law,
for example, used to be a big thing
in constitutions. And that’s because, of course,
they used to lock up debtors. And so, it was a major
human rights issue. Once we stopped doing
that, much less pressure, I suppose, to
constitutionalize it. Now, the point is that there
are certain features which are of core, certain features
which are peripheral, and they’re not necessarily
corresponding to what you might think of as
the essential functions. The form is quite different
than the function. And the form seems
to be produced in a process which is somewhat
unrelated to function, very much like a product. There are trends
in fashion which will dictate the shape and
the style of the cups for sale in a particular time. Similarly with constitutions,
there are trends. When you look at
all of this, when you look at all analyzes of
the contents of constitutions, there are three things
which really stand out in terms of the determinants
of any particular document or of any particular product. The first is the era in which
the document was written. Time seems to be
really important. And again, that seems
quite intuitive, from a fashion perspective or
with thinking about products. Region is also really important. There are regional trends
in constitutional design. And so, countries
are likely to be more similar to those that
are their neighbors than those that are not. And finally, the first
constitution in the country’s history, that seems to have
a lot of sticking power. And when you add all those up,
really, the essential question then is, when does the country
enter the international system? The period and the region
in which it does that is going to dictate a lot
of the contents thereafter. All right, now I want to
turn to another topic, which is duration or endurance. And of course, the cup, as
I said, will last forever. If it was paper, it would not. We tend to think of
durability as being important in some products but not others. And products obviously
vary, in terms of how long they will endure. Wine seems to get better for
a certain amount of time. And then it gets worse. Some products, again,
will last forever. So one question we have
to ask as a threshold is whether or not duration
is good or bad. Right? Whether we want constitutional
products, if you will, to last a long time. This is something that’s been
debated intensely, really, since the era of the
founding fathers. Our own Thomas
Jefferson, for example, thought that constitution
should not last a long time. His view as a radical Democrat
is that the dead should never govern the living. And so, that all constitutions
should die at the age of 19, they should be torn up. And then, we the people
would discuss them again. If we chose the same
institutions, that was OK. But we would be forced
to make our own choice. And that was his position. He got the 19, by the way,
from a table of European life expectancies at the time. He did the calculus for when
today’s majority would be replaced by tomorrow’s. And so, that was his view. You might intuitively say
that that does make sense for some context, right? We know that products
are purchased but the world changes. And these exogenous
sources of change might mean things that
you buy no longer fit. How many of you people over
40 would still fit into a suit that you bought when
you were 20 years old? Probably not very many. Something’s changed and you
might need to buy a new one. And so, the constitutions, I
think, have the same feature. One example I often
use is the Constitution of Lebanon, which,
in the 1940s, decided on the basis of the
current demography that offices would be divided
up amongst Sunni, Shiite, and Maronite Christian. The president was going to
be a Maronite, the parliament speaker a Sunni, et cetera. And this made a certain
amount of sense at the time. It was a bargain that was
produced in a factual setting. But on the other
hand, of course, it should have been
foreseeable to those folks that the demographics of the
time might not be stable. And indeed, they were not. The Shiites, it turned
out, had many more children than the others. And there was
pressure, if you will, on the bargain in
order to change. The pressure led to
a bloody civil war. So that’s an example of a
constitution which would have, in some sense–
things would have been better had there had
been a forced renegotiation after a certain amount of time. So that’s the Jeffersonian
view, that they ought not last a long time. On the other hand,
there are those who argue for stability
as an inherent good, a kind of Burkian view. Or one might think about
it as from an institutional perspective, as the
value of stability on other institutions. So we know in modern
democracy that many of the things that
help democracies to work and to
function are not going to be found in the four corners
of the constitutional text. Things like political
parties or a free press or an independent judiciary,
some of those things might be influenced by the text
or by the institutions adopting that constitution. But a lot of them involve
collateral institutional development. And for those
things, change that occurs in the
fundamental institutions too frequently is likely to
lead to great disruption. Furthermore, adopting a
constitution is a costly thing. The negotiation isn’t easy. And that’s sunk cost
every time you do it. So from that point of view, we
might think that constitutions ought to last a long time. Well, what would
an economist say? I suppose one thing we
might look at is what works. Right? Empirically, can we identify
any things that are associated with– or perhaps caused by– long lasting constitutions? And this figure gives
some information on this. So this, for every
constitution in the Nth year– from zero to 50– this is the average level
of wealth for all countries, the average level of
democracy for all countries reaches zero right here. The average level of
political stability and the frequency of crisis
is defined in the literature. And what it suggests is that
it’s good for many things to have a long-lasting
constitution. Now, we’re confronted
immediately here with a problem which
empiricists will recognize, which is this is
probably a situation of reciprocal causation. It’s not necessarily clear that
just because your constitution lasts a long time, you
will get longer democracy. It’s likely that
having more democracy is going to contribute to
constitutional endurance. So this is just a
mere correlation. But it’s a suggestive
thing which certainly invites further inquiry. When one looks at duration,
one sees great, great variety. And it seems to be– there
are very many different kinds of national patterns. The Unites States looks– each color here in this
figure, I should say, is a different constitution
in the country’s history. Some countries are like
Australia or the United States, single document. That’s pretty rare. Really common pattern
is Mexico, where you have lots and lots
of experimentation with institutions before you
get a stable constitutional bargain. Panama looks a lot like
Jefferson’s fantasy, right? They seem to change the
product, if you will, pretty frequently,
about every generation. And some countries, like
the Dominican Republic, just can never get it together. Indeed, I want to
show you a figure. Each year up here is a different
constitution in the Dominican Republic’s history. This is the level of
democracy as measured by political scientists. And what you see is that
their changing the product is not having much effect. It would be nice
if I could show you a figure like this
with a country that had a lot of
constitutional change and a high level of democracy. Doesn’t seem to be out there. There does seem to be something
about this endurance feature for democratic performance. I think what I’ll do
is go to this slide. One way to think about this
is, how durable is the product? How likely is it to live? And one thing you
might– we might look at is what economists
call hazard rates. So a hazard rate is
basically a level of risk. If I was to look at the
hazard rate for a human being, this shows you the probability
of death over a particular age. So for a human being– whoops. We have a high hazard
rate at infancy. There’s a lot of
infant mortality. If you can survive to age two
or three, I think it is is, you’re likely to live
for quite a while, till you get to whatever it is–
age 70 or so– in which case, it would spike sharply up. The older you get,
eventually, we all go. That’s not true
of constitutions. The risk seems to
be highest early on. But, then, if you can
make it to, say, age 100, it’s likely you’ll
live to be forever. Now the confidence
intervals indicate that there’s not very many
countries in this category. But it does suggest that
these things are not quite like human beings, that
they are enduring once you can survive some risk at infancy. In a book I published
a couple of years ago, we examined what are
the causes of endurance and the likely
correlates, if you will. And we used a whole bunch
of statistical models called hazard models. I’ll just summarize
them for you. Rather intuitively, we put these
things into three categories. If you think about
the product, a cup can be well-designed
or not well-designed. There are certain things
which will happen to this which will break any cup. A nuclear bomb is
going to destroy it. Shocks are what
will cause the death of constitutional institutions. External shocks like
war, economic crisis. And we indeed found that
defeat in war or coups are likely to lead to the
death of constitutions. Interestingly,
though, we also found that there seemed to be a trend
in replacing your constitution. So if all of your neighbors
were replacing theirs– even controlling for
things like wealth, controlling for
all other factors– it’s still more
likely that you’re going to adopt a new
constitution yourself. So again, the analogy to
fashion is pretty easy. If everyone on the
block is doing it, I’m going to start wearing
a pink suit as well. There are features
of the product itself which would matter. Or– excuse me– not of the
product, but of the country. Richer countries tend to be more
favorable soil, if you will, for enduring constitutions. Those with lots of
ethnic groups tend not to do so well, at least
if they’re dictatorships. They do better in democracies. The key question for
us, actually, though, in thinking about products– and
in thinking about the takeaway from all of this for law
and for normative thinking– is there anything you can do
in the product design that will make it more enduring? Is there a way to build a
better cup, if you will. Better materials and such. And we identified
three factors which are within the control of
constitutional designers, which seemed to facilitate endurance. The first is flexibility. It’s very intuitive. If you’re in a world
which is always changing, you need a document,
a constitution which also can change. And we found, generally
speaking, the constitutions with lower amendment
rates tended to– with simpler
amendment rates right would be easier to change
when things change. And thus, would
be more enduring. We also found in that
regard that countries which had constitutional
review seem to have enduring constitutions as well. So the court is a kind of
mechanism, if you will, to facilitate
constitutional endurance, to facilitate flexibility. And that’s quite
consistent, I think, with conventional understanding. A second feature was inclusion. It’s much better to be like
Iceland than to be like Egypt. I predict the Icelandic
constitution will last longer than the Egyptian one. The Egyptian one might
not even last this year. But if the Icelandic
one is adopted, I could plug it into
the model, the model predicts it’s going to
last 60-something years. It’s adopted in a very open way. And this seems to be important
in the sense of giving interest groups, giving people a kind
of investment, if you will, in maintaining the
constitutional bargain. So when to subject to some sort
of challenge, they want to– they want to protect it. The final feature is rather
counterintuitive for Americans. And that’s what we
call specificity. It really reduces to
the level of detail. Oddly, the more detailed a
constitutional document is, the more likely it is
to survive the storms to which it will be subjected. And again, that’s quite
counterintuitive for Americans. Our theory is that
this has something to do with sunk investment. That is, if it’s costly
to write a constitution– and it is– then the more
detail you have in there, in some sense the
less– it’s going to be more costly to replace. It also means that
particular groups are likely to have
significant benefits that flow from that constitution. In a way, this goes back
to where I was starting. And it suggests that
interest groups may actually have some role to
play in facilitating constitutional endurance. When you have a challenge
to a constitutional order, it’s good if you have a certain
amount of interest group activity of interest
groups that will want to preserve it
and fight for it, rather than renegotiate
another one. And when one looks
at particular cases, this seems to have some– it seems to play out. So, for example, the
Constitution of Mexico has been very enduring. It’s survived the transition
from dictatorship to democracy. It’s a very detailed
constitution. It’s got a long labor
rights provision, which is being
changed all the time, in giving the labor unions– which are a big part of the
coalition that runs Mexico– more and more benefits
in the Constitution. And this, I think,
is one of the reasons that there was kind
of agreement not to rewrite when Mexico became a
democracy and labor units were worried about losing things. India, another example of
a long-lived constitution which is flexible. Certainly, it
changes every year. But it’s also got
a lot of detail about rights that are going to
be given to particular groups. And so, this seems to be
somewhat counterintuitive I think, for Americans. But these three factors
of constitutional design seem to help the product,
seem to help it endure over time in the face of complicated
and changing circumstances in the world around you. Now, I just want to say a
word about the United States. Because our constitution, of
course, is the most enduring. It’s not very flexible. It was adopted in a
very uninclusive process by contemporary standards. And it’s not very detailed. Indeed, the world is full of
constitutional consultants going around telling countries
like Egypt or Iceland that they ought to adopt
one that looks very much like the United States. I think this is quite wrong. I think in the book, we
analogize the United States to a woman named Jeanne Calment. Jeanne Calment was
a woman who died at the age of 122 years
old in France in 1997. Her diet consisted
of regular shots of port wine, olive oil, and
two pounds of chocolate a week. She was a lifelong
smoker, though she gave it up for a
year at age 117, then went back after she
decided that it didn’t make her feel any better. So whatever it was she
was doing, it was– may have worked for her for
reasons that we can’t explain. Not necessarily good
advice if you’re Egypt or Yemen or Iceland. And so, that’s our take
on the United States. Now, I just want
to say one last– make one last set of points. And that has to do with
constitutional efficacy. I started by talking
about, would it still be a constitution
if it didn’t work? And if we think
about constitutions having form as well as function,
I think the answer is yes. But this, then, leads
to the question of, is what is it that makes
constitutions work? What is it that
makes them effective? And this is an enormously
difficult research problem, which I have not in
any way solved yet. But I want to talk about why
it’s a difficult research problem in order to
encourage those of you who are empirically
inclined– and some of you first year law
students– to think about how we might get some
traction on the problem. So I’m going to show you
a figure, first of all. So this is an indicator. On this column, we have the
level of actual civil rights protection– civil
liberties protection– in any given country as
defined by Freedom House. And here, we have
a de jure index produced from our
constitutions data set. And each one of these three
letters here is a country. And so, you can
see that countries have very differing relations
between what’s in the– on the paper and how things
actually work in practice. Many countries have a
lot of promise on paper but don’t deliver
much in practice. This is Uzbekistan,
Turkmenistan, Sudan. So paper doesn’t seem
to be doing much good. On the other hand, we
have countries up here– Australia, New Zealand, United
States is up here as well– where we don’t actually
promise very much relative to other countries,
but we deliver a lot. We’re over-performers,
if you will. An interesting
question for me is, what leads countries
to be on this line? On this progression
line, if you will. Up here, we have countries
that promise a lot and actually deliver a lot. Hungary, Portugal,
Slovenia, Slovakia– these are countries
which are, in fact– generally speaking
have democratized in the last 10 to 15 years. They’ve written a
new constitution to sort of codify their
understandings of what democracy requires. They’ve adopted rights from
the international instruments. And they seem to be
following those things. Now, you might say
that they would have this high level
of rights protection without having
adopted those texts. And that’s the puzzle, right? Whenever we are trying to
compare text and practice, we face the– and this is true generally
in legal studies, but particularly with
constitutions– we face the problem of how to
handle missing variables. It may be that what happened,
the fundamental change which happened in Portugal in 1975,
was that it became a democracy. Society changed. That led to a change
in the text and also to a change in practice. That’s what we would call a
missing variable problem, which is a missing variable which
we’re not able to look at, which we can’t observe. Which is affecting
both the outcomes that we’re interested in. It may also be that there’s
kind of a reverse causation problem, where a country
changes its practice and then changes its law. Or sets in motion some things
which will change practice and that, then, changes the law. So that’s another empirical
problem that we confront. And it’s very
difficult and there’s a lot of statistical methods
to try to sort this out. But I do think it’s
worth interrogating what it is that leads
countries to be around here. My own intuition
is that, in fact, for certain kinds of provisions,
we can identify some causality. But before I– before I
turn away from this slide, I just want to
point out down here. So here, we have countries
which don’t promise much and don’t deliver much. Whatever you want
to say about them, at least they’re honest, right? You have Saudi Arabia down
here not promising much, not delivering much. And that itself is a kind
of interesting question. Why are some dictatorships
like this and some like this? Another puzzle. Well, without trying to
determine or without really being able to fully answer what
it is that leads constitutions to actually be
implemented, I do want to make a final point,
which is that we can compare the level of practice and the
level of performance over time. And one of the things
that we observe when we do that is that
as a constitution ages, the gap between the
level of practice and the level of
performance actually grows. It becomes much more like
these countries up here. These are countries that tend
to have older constitutions. If you just focus on the
gap between the observed and the promised
constitutional rights, well, you might say, well,
these have a really large gap. But the interesting thing is
that the gap grows over time. The older the constitution,
the more you sort of move to this quadrant. So the gap grows, but it
grows in a positive direction. We also can look at
this not just for rights but for institutional provision. So there are ways of codifying
levels of parliamentary power. In particular context,
levels of judicial power. And we observed the same thing– that over time, the gap
grows between the level of parliamentary
or judicial power in a constitution and
the level of actual power in any particular
constitutional structure. And it turns out that they also
grow in a positive direction. It suggests again that there is
something about endurance which might have some tie to efficacy. Well, just to conclude,
constitution clearly are very successful
product, in terms of doing what
they’re trying to do. And in thinking about them
as having form and not just function, one can interrogate
the particular production processes which lead to
particular constitutional designs. And one sees the process
does seem to matter, at least in some circumstances. In particular, inclusion seems
to be a very important trend in constitutional design. And I think a positive
one, in terms of values of democracy and rights. Another point is
that when one wants to understand the
contents of constitutions, you can’t simply look
at the local context. Although we the
people come together and we think of constitution
as being expressing our fundamental values, there
are trends across countries. There are trends over time. There are trends in regions. And these things are
going to likely influence the shape and the
design of products, at the constitutional product. Just as with fashion
or with cup designs, we’re likely to see things that
are very difficult to trace, but that do come from outside
the designer themselves. Finally, in terms of
constitutional endurance, I want to leave
you with the idea that they seem to endure
most when they are least constitutional. As I said, if a
constitution maker produces a document
that’s flexible, that’s relatively
detailed, that’s inclusive, it’s likely to endure. In some sense, that
suggests that constitutions are likely to last
more when they’re much more like statutes, when
they involve interest groups. When they provide benefits
to particular groups who will have a stake
in their endurance. You need to get the producer,
in some sense, involved and have an interest
in the product in order to make it endure. Finally, the final point is
that sometimes the old products are the best. And with that, I will
close and look forward to your questions and comments. Richard? AUDIENCE: I didn’t
notice you saying much about federalism or other
issues of sub-national units. And I guess the
second question is I’m just wondering if there’s
some implicit move you’re making lasting a long time
is good, whereas it might be better to have– It might be constitutional to
achieve more of what they’re supposed to achieve when
they don’t last a long time and they’re replaced. TOM GINSBURG: Right. So in terms of
that second point, there’s obviously a lot
of different purposes. Sometimes constitutions
are designed to resolve a particular
political problem in a very short-term kind of way. We have an ethnic conflict. We want to stop that war. And the Constitution
will have the character of a peace treaty. And I think that for
that kind of situation, lasting five or 10
years is actually a sign of great success. It’s longer than the modal
survival rate and probably not a bad idea. So I think that it’s hard
to evaluate universally, but that being said, I
am looking at averages across all countries. And suggesting that–
on average– it is generally good
to last a long time. Though, of course, that might
be context-specific evaluation. In terms of federalism it’s
a quite interesting thing about constitutional design. The number of countries
that are formal federalisms is actually relatively small. The number of countries with
highly decentralized government is going through the roof. It’s a major trend in
constitutional design, is to sort of implement what
we call in the literature subsidiarity, to try to locate
decisions about public goods at the lowest level possible. And so, this is
something where I think looking at a category
like federalism or non-federal actually is not that helpful. It’s a little bit like
looking at a category of parliamentary or
presidential systems. When you look at the actual
institutional designs, there’s a lot more convergence
across those categories than one might think, given the
dominance that those categories have had in the literature. So one of the parts of
this research program is to try to come up,
perhaps, with new categories. Or to identify the way in
which these hybrids are emerging over time. I’ll take a student. AUDIENCE: Do you think that
perhaps the causation may work in a way that instead of
the Constitution determining outcomes, the
constitution reflects the values of the society? For example, conclusion, you
look at Iceland versus Egypt. Or have Iceland with a
more inclusive constitution because it has more
democratic cultural trades, while Egypt is a more religious
and fragmented tribal society. And that perhaps trying to do in
Egypt what they did in Iceland perhaps wouldn’t
translate very well? TOM GINSBURG: Good. Great point. I mean, so in general,
there’s not– we don’t really know much about best
practices universally. And I certainly would not
assert that it be a good idea to try the Icelandic
approach in Egypt. I think that would be a
recipe for complete disaster. It works in Iceland, because
everyone’s very educated. They’re Nordic, so they
all tolerate each other, although they have this long
history of blood feuds as well. In Egypt, I think the best
approach would actually be an elite level
bargain, where you get the leaders of the various
communities and you get to– and they really
hammer out something which is likely to
work for everybody. That’s not at all what happened. It was one group just
dominating the process. But whenever you
have domination, you’re not likely
to get stability. And so, I do think that one’s
normative recommendations do have to be taken. So I accept the point. Martha? AUDIENCE: Yeah, I wanted
to ask about specificity, because it seems to me
it’s in a lot of tension with transparency. I was speaking
enthusiastically in one class about the Indian constitution–
as I sometimes do. And a student– who was from
India, but now in the US– he pulled out of his
jacket pocket the document. He said, this is
the US Constitution. And I can carry
it around with me. I can look it up
whenever I want to, I can know whatever is in it. If I had to do that with
the Indian Constitution, it would be a huge book. And the result is
that no one really knows what’s in it
unless they’re extremely hard-working and educated. Now, of course,
Americans don’t know what’s in the US
constitution, but they could. And so I guess– I mean, I know you were talking
about transparency ex ante, but ex post, if
it’s very specific, it does seem to
militate against– TOM GINSBURG: Well,
that’s certainly- that’s certainly true. There’s some optimal level
of specificity from the point of view in a society. And it might be different
from the optimal level for a particular set
of interest groups. So if I’m a privately
motivated person or interest group trying to get
some benefits out of the constitution, I want to
make it as obtuse as possible, so that no one can
see what I’m doing. And indeed, the Indian
Constitution, pretty hard to read, as you know. It’s lots of cross
references and such, as well as being something
like 178,000 words. So I think there
is a tension there. Normatively, I’m not sure I’m
convinced about the benefits of specificity. As a positive matter, it
helps induce interest groups to support
constitutional endurance. It seems to provide
a certain level– I think has an independent
good and independent value. On the other hand,
you want something from a normative, democratic
theory perspective that can be understood by citizens. So maybe a tension between the
positive and normative analysis here. Yes, in the back? AUDIENCE: Are you tracking,
as one of your variables, how constitutions
are interpreted? So, for example, the conflict
between originalists and people who view the purposes. TOM GINSBURG: Yeah. So that’s obviously a
very interesting question and a major topic of
comparative analysis. Although, surprisingly– perhaps
for Americans– originalism hasn’t had much purchase
outside the United States. In part, that’s because
in many countries, the constitutional
design drafters are still alive and can say what they– they thought about
things at the time. But in any case, one of the
things about this project– and it’s one of the
implications about looking at the constitution
as a product– is merely looking at the
four corners of the document without trying to look at all
the subsequent interpretations. Or at least not to do
it as systematically. And that’s partly a
practical decision. It’s very difficult,
obviously, to track how every particular rule
of constitutional law has been interpreted across even
in one country, much less many, many others. But it’s a supplement. And there are other
scholars who are working on that thing, which will
be quite complimentary, I think, to this work. Yeah? AUDIENCE: I thought what you
said about innovation was really interesting. But when I think
about this, I would compare it much more to
rather than a product that you’re selling to something
you own, like a home ownership. So people all the time make
investments in their home, even though they can’t
recoup all the investment from the market. And I think, then,
if that’s true, the big concern is
more of the cost side and the collective
action problems. People are incentivized,
but there’s collective action problems. And that would
mean we’re probably more likely to see small, little
changes where people are making changes that they’re really
passionate about, rather than completely remaking it. So I guess my
question is, do you think that’s a valid metaphor? And also, are you doing anything
to measure to what extent you can have changes that
actually might change a constitution
through amendments or constitutional construction. We’re really talking about
separate constitutions. TOM GINSBURG: Good, OK. So, I mean, conceptually,
it’s something I’ve been wrestling with. What is an innovation? Is an innovation
actually coming up with something new that
no one’s ever thought of? Or is it packaging things
that are already out there in new ways? And of course, this gets to
some of literature on design and such and questions about
where we grant intellectual property rights
and where we don’t. What kind of
innovations we really want to encourage versus
those that we think are going to function
fine in the market. One way to think about it
is that every constitution, what changes anything
is a new product. It’s a new packaging
of existing elements. One thing we can look at and my
similarity measures suggested is you could see how much any
particular amendment changes a constitution
versus a replacement. Turns out that replacements
tend to be pretty large relative to amendments. Amendments tend to be very
minor things on average. We can actually
measure, conceptually, the degree to which any
particular amendment is introducing new elements. So I do think that that’s right. The metaphor works. And there’s a certain amount of
conceptual sort of fuzziness, if you will, around
what’s an innovation? When I look out there for things
which are true innovations– which have never been adopted
in any other constitution, which aren’t being borrowed– those seem to be quite rare. And they seem to flow from
really serious ruptures in the previous society,
significant social movements. And they can almost
always be explained in self-interest terms. Someone had to be very
motivated to be an entrepreneur to bring in a new institution. But my general sense is there
aren’t very many of them. We ought to do much
more experimentation. One I was thinking of is– no society has really
ever experimented with the Greek solution,
randomly selected citizens to run the thing. I don’t know that we could
do any worse in this country if we were to do that. But in any case, I think– I believe there’s a
sub-optimal level. AUDIENCE: I have a question
about [INAUDIBLE] particularly in your description
of the US Constitution and why you wouldn’t
recommend it. I don’t think that if you
just look at the document, you get a proper measure
of what the US Constitution [INAUDIBLE]. We have enormous
quantity of case law. It can make the
documents incredibly– constitutional law is
incredibly specific. It’s quite flexible,
judges can do quite a bit to [INAUDIBLE] the laws. So if that’s the
case– and that’s just for the United States. That just happens to
be the one I happen to know a little bit about. I wonder what happens
in other countries. I wonder if it’s a wise thing
to use the text of the document to make predictions about what
would work in other countries, even putting aside the
[INAUDIBLE] of culture. TOM GINSBURG: OK,
that might be right. It’s clearly the case
that part of the reason to focus on just the written
text is a practical one. And also, I think it’s valuable
in terms of comparability. Once you start
getting into case law, one might look at the
other sort of substitutes for the constitution which
one could talk about. What are the
unwritten norms that are operating in Zimbabwe
that are constraining the government or not? And so, there’s, I
think, value in looking at a discreet set of texts. The other thing is that
constitution makers aren’t producing case law. They’re not producing that whole
body of constitutional law, they’re producing texts. These are discrete moments of
purposive institutional design. And they occur in
many countries. And I think we can say
something just focused on them that will have some
normative implications. Mike? AUDIENCE: So early
on, you said something about– that there’s
little competition for constitutional product. And presumably, that
may have something to with immobility of
people across boundaries. But capital is perfectly mobile,
or it can be perfectly mobile. So do you see any competition
among jurisdictions– with respect to constitutions–
on issues like property rights, to try to create the most
desirable location for capital? TOM GINSBURG: Yeah,
one would expect that. But I guess there’s two things. First of all, there’s a
really good set of substitutes in terms of attracting capital. It’s a set of institutions
called bilateral investment treaties, where you– and there’s also, of course,
some multilateral treaties, which would do the same
thing– where you provide for free flow of capital. Just at the treaty level. If you think about yourself
as a foreign investor, you don’t really care what
the local constitution says if you think you have a
credible treaty right that can be enforced internationally. So there may be less pressure,
in a way, to do that. One of the surprising
things, though, that we found is that constitutions that
promise greater property rights protection actually
protect property less well. There’s a negative correlation. And my only theory of
that is that there’s been a great deal of
pressure on countries to improve their form of
property rights protection. So countries that
have poor protection are selecting into
the population by increasing their
property protection. When, in fact, as we
know, to protect property requires a whole bundle of
institutions– collateral institutions– which might
not be available early on. [INAUDIBLE]? AUDIENCE: So
originally you wanted to study election systems,
which were better about– you would look at
all the elections and see the type of
results that they generate. But I imagine that you
wouldn’t count places where they run sham
elections, where the Secret Service actually decides
what’s in the ballot and who gets the vote. Even though, formally,
they have a system, you would rule it out. And I have a little
problem with the inclusion in your study of constitutions
in places where it seems like a sham, like South– North Korea. TOM GINSBURG: Right. Well, the interesting
question is– I think it’s an empirical
question, to what degree a constitution is
actually a sham. One simple approach might be to
say, only look at democracies. It turns out that
constitutions actually do matter in many dictatorships. The rules for executive
selection, for example. Mubarak, Sadat, they
spent a lot of time tweaking their constitution in
order to get certain benefits. By the way, one of the reasons
Egypt’s a kind of disaster today is they had to do a deal
with the Muslim Brotherhood at one point in order
to get that through. Even in a dictatorship, they
were constrained a little bit. And the constitution did
matter in terms of some things. Maybe not rights protection, but
in terms of executive selection and things like that. Now, in a country
like North Korea, I’m sure it is a total sham. It literally has no meaning
or no actual application. Though that itself is a kind of
interesting question as well. It’s like, why,
then, is it produced? Even dictatorships are spending
significant political energy on it. And sometimes, we don’t know,
really, what their motives are. The Burmese government,
which I mentioned before, took 17 years to draft
the constitution which emerged in 2008. I and everyone else
thought that this was just a sham constitution. It turns out to have
been the culmination of an invisible process,
whereby a decision was made to liberalize. So I actually think
those are the ones which may be most
important to look at, because we’ll be able
to try to figure out what clues we can see about the
society in what are generally very obtuse or
obscure situations. Yeah, with the hat? AUDIENCE: My fellow student’s
question piqued my interest on the difference
between the cup maker and the actual
design of the cup. I was wondering if you
have any statistics saying the relative– or
judging the relative importance. Whether, you know,
Iceland is bound to make a more
successful constitution, just because it’s
the people of Iceland than the people of Egypt. Even if they do everything right
and they spend a year on it, versus the people in Iceland
who spend a week doing it, sort of ad hoc. TOM GINSBURG: Yeah. Well, there’s intuition,
of course, which says– again, evaluating what’s
good and what’s bad is very difficult
outside of the context. We don’t know if the fact that
Iceland’s last constitution lasted 60 years was
a result of it being a good set of institutions,
of people being Icelanders, or some interaction of the two. And so, it’s a bit hard to
isolate exactly what would be making the difference there. One way to look at it is
to take designs and look across cross-sectionally
how they are functioning in different countries. And that’s one thing that
I suppose this research program is trying to
address, is to look at a set of like countries. And then, where you might have
some different institutions which vary. And then, that will allow
us to make some inferences, I suppose. about the efficacy or the
quality of the institution. But I don’t want to deny
that there’s something that we might call a
constitutional culture that people do have. Their set of preferences
and their social conditions, which are likely to
make some institutions work better or worse. That seems to
certainly be the case. Jeff? AUDIENCE: If you were
called in as a consultant to a generic country
that was about to draft its first constitution,
which of the three exististing constitutions
would you draw upon most as your models and why? TOM GINSBURG:
Which of the three? AUDIENCE: Which of the three
that exist would you draw upon? TOM GINSBURG: I’m sorry– AUDIENCE: If you
were called upon to draw– help draft
a constitution. TOM GINSBURG: Yeah, South Sudan. Newest country in
the world, right? AUDIENCE: Which is? TOM GINSBURG: South Sudan. AUDIENCE: OK. And the question is, which three
constitutions that you know exist– TOM GINSBURG: I see. Oh, which would be a model. OK, so that’s a–
that’s a good one. Well, first of
all, I’m generally a fan of a kind of
generic parliamentarism. I think it’s really
good for societies which have lots of
different ethnic groups like South Sudan does. So I think that’s good. I would say Germany. It’s a very influential model,
good constitutional accord and all that. But I’m not sure that
that’s likely to suit a country like South
Sudan, which doesn’t have a grand legal tradition. It doesn’t have a lot– it can’t really tolerate a
very complex constitution. So I might say
something much simpler. So, actually, the Icelandic
one is one I kind of like. It’s pretty short, it’s easy to
understand– to Martha’s point. It features a lot of
popular involvement. Not only a lot of rights,
but a lot of opportunities for the people to influence
major decisions and things like initiative and referenda. And I think that’s a
pretty good model, even for a very poor country
like South Sudan. Maybe especially for a poor
country like South Sudan. Because one of the
challenges the country faces is to get everyone to think of
themselves as South Sudanese. And if you’ve got a
mechanism where they’re forced or encouraged to be
involved in ongoing government, I tend to think that would
be a pretty good model. AUDIENCE: And I
take it, though– for a nation with a long legal
culture that’s sort of stable– that Germany’s the one? TOM GINSBURG: I like Germany. I think it works pretty well. Maybe it’s because
they’re Germans, but I’m a fan of the
German Constitution. You asked for a third, so now
I have to come up with a third. A lot of people are big
fans of the Canadian one. I don’t think it’s that great. AUDIENCE: The cultural history
and my whole camera roll for card to constitution has
manifested itself pop culture looking at me representations. And so forth to get a sense
of how a document permeates the popular culture in
general and he kind of tried to make sure that he had it all TOM GINSBURG: One of
the major challenges of this in general,
when you’re trying to look at
constitutional efficacy, is the lack of good
dependent variables, if I could call them that. Good indicators of
constitutional performance. We rely on these
political science ones. They’ve got lots of
measurement error or they’re sometimes crude
and sometimes subjective. So in the era of
big data, I think there’s great
opportunities to try to get at this in greater depth. So imagine if we could,
for example, survey the number of mentions of the
constitution in newspapers in particular countries
and times and places. That’d be a pretty
good indicator of how important it was for
popular consciousness, right? There are various
surveys also, values surveys– world values
surveys and such– which might get at this a little bit. And there’s actually
a new effort– promoted by the American
Bar Association, of all organizations–
to try to measure the values of the rule
of law in a whole bunch of different countries. So there’s more
and more dependent variables, which allow
us to get at that. I don’t have a good one
now, but I’m looking. AUDIENCE: One more. AUDIENCE: It may be too late in
the day to ask this question, but I wonder, what
is a constitution? Because when you compare
across countries, some have much slimmer
constitutions than others. But the longer ones cover
rights that are often covered by statute elsewhere. I know the Cuban
Constitution has lots of provisions that
are covered by statute here in the United States. And the process for
enacting legislation in the United States might mean
that once a statute is enacted, it’s actually very
difficult to get rid of it, so that it tends to
have the durability of a constitutional provision. So as you compare
across countries where it’s not obvious what
counts as a constitution– because there might be things
like statutes that function much like a constitution– I was wondering, how do you
handle that in the project? TOM GINSBURG: Well, that’s
such a great question, and it goes to this
issue of substitutes. There’s an old joke about
someone who goes to the library to ask for a copy of
the French Constitution and is told by the
librarian that they don’t stock periodicals. If you go talk to French people
about their constitution, it’s not like
talking to Americans where we treat it as this
quasi-religious document handed down by God on the seventh day. There’s a much more– there’s a sense that
the constitution is just about the politicians, really. And the fundamental things
that govern French society are things like the French
Declaration of the Rights of Man, the administrative
law regime– which has governed, really, since
the ancient regime, actually. And so, there’s a lot
of stable institutions and a lot of sources of rules,
which don’t have anything to do with the constitution. That’s a generic one. That’s the problem
of substitutes. I suppose my only
answer is to say that until we have a good handle
on the formal constitution, we can’t do a systematic
comparison of what– even if we can
measure them– all the actually operative rules
were with the formal text. What percentage, if
you will, conceptually, of the truly small C
constitutional rules are contained in
the formal text? That’s a question
I’d like to answer. But until we got the
first measurement– or the second measurement,
now we can’t do the first. AUDIENCE: Thank you.

Leave a Reply

Your email address will not be published. Required fields are marked *