Mid-Term Notes
Mid-Term Notes
Political Economy, broadly defined, is the study of the interaction between politics and economics (the
state and the market). What is the difference between governments and markets? Both can be seen as
two types of decision-making organizations and most conflicts are about the allocation of goods that are
valuable for society. So what is the difference between how politics and how markets decide? In markets,
every individual is a rational decision-makers who decides based on the supply and demand of a certain
good or service, whereas in politics, few people are delegated to make decisions for everyone else with
respect to the distribution and redistribution of goods (individual versus collectives). When do and when
should governments intervene in markets? To deal with negative externalities and provide public goods
that the market cannot ensure. This might include welfare states or unemployment insurance, yet it is all
about the distribution of wealth in society. Government involvement is very much a normative enterprise.
The two disciplines (politics and economics) used to be pretty close. Both were interchangeably used by
Adam Smith, John Stuart Mill, combining normative and analytical approaches to understanding society
as a whole. However, in the late 19th century, the two disciplines split. Economics became a formal study
of the mechanics of markets, whereas political sciences came slightly later from previously being a very
qualitative domain and established itself in the ranks of social sciences among psychology, sociology and
economy. In the recent decades, there was a renewal of interest in Political Economy from a substantive
rather than methodological perspective (i.e. to understand how both disciplines influence each other).
With regard to the definitions, economics can be defined as the study of human behaviour as a relation
between ends and scare means that have alternative uses (Drazen, 2000). Likewise, politics is the study
of power and authority, the exercise of power and authority over others and the struggle over authority
in political institutions (Drazen, 2000). Economics is interested in power as well, but only in the context of
monopolies and oligopolies. Power being the ability to achieve outcomes that reflect objectives (Weber).
Political economy starts with the assumption that there is a conflict of interest between economic actors.
Furthermore, conflict of interests are a necessary condition for the existence of political constraints and
the mechanism by which these conflicts are resolved shape the effect of politics on economics.
In a market, conflicts of interest are very easy to reconcile: people have different tastes over goods and
the markets just sort them out. If more and more people hate Apple, some other suppliers will emerge
and start selling other computers. So the conflict of interest is simply a conflict of taste and the market
will sort itself out. This type of conflict of interests is called ex-ante heterogeneity, where different policy
preferences stem from different tastes over goods and relative factors of endowments. In politics, it is a
different story. There are different decision mechanisms that apply to parliamentary legislative processes
and the choice of the way by which to implement a given decision is also open up for contestation. Thus,
ex-post heterogeneity is the idea that once a government has installed a policy, people may have different
interests as to its distribution or the costs of its provision (different types of conflict of interests).
The main difference between the Drazen (2000) and Frieden et al. (2011) text, is that the latter is broader
and we are more embedded in this perspective, as opposed to getting fixated on applying certain tools.
Most economists take Drazen’s view, whereas political scientists would take Frieden’s view of political
economy. The more political science logic of political economy is more in line with Frieden’s view. The
new political economy looks at how politics affect economic outcomes, but it falls into two subtypes: (1)
positive political economy, or how political constraints explain the choice of policies, and (2) normative
political economy that seeks to cope with political constraints within existing institutional frameworks
and design political institutions and policies to better achieve (Pareto-efficient) economic objectives.
Most political economy research today employs some form of rational actor model to study government
decisions in the context of political and economic institutions. The model presumes individuals that try to
maximize their utility and the assumption is that they have preferences over something and they have a
reasonable amount of information (not full information). Once they have enough information they weigh
their options thinking about what would happen if they did A, B or C and in the end establish some sort
of preference. This is still largely the case, for example, the German government’s decision to shut down
Nord Stream II, the politicians are weighting their options, thinking about the most optimal choice among
all other choices. They look at the scenarios and describe the pros and cons of every option and then
select the most utility maximizing option. As such, political decision-making processes are mostly rational
in this sense. It does not mean that they are economically rational (i.e. Putin invading Ukraine), however,
even though it might not be economically rational, this does not mean that it is irrational in the complete
sense of the word. It can be rational in a political sense, it can be rational in the short or long term etc.
The standard criticisms of the rationality argument is that it denies people’s humanity (precludes altruism)
and renders culture, norms and socialization unimportant. Another point is that it assumes intentions
are unique to individuals. However not all decisions are monetary rational i.e. you buying a painting. Is
buying a piece of junk rational? In an economic sense? No. In a sense of personal pleasure and wellbeing?
Yes. Again, this type of rationality takes us beyond economics. Brexit is another example: economically
not rational, but from a sense of it feeling better, it might as well be a rational decision (but here we can
call everything rational, yet where do we draw the line between rationality and emotion?).
With regard to the levels of analysis, most economists focus on the domestic level. Within Comparative
and International Political Economy analysts might disagree with regard to whether the causes of political
and economic trends are found at the domestic or international level, and whether it is the politicians
and political institutions, or private social actors (interest groups, firms, NGOs) that are more important.
The two perspectives open up the following four distinctions: (1) international political perspective: that
emphasises the constraints imposed on states by global geostrategic and diplomatic environments; (2)
international economic perspective, which emphasises the constraints by global socioeconomic factors
(i.e. international developments in technology, telecommunications and production that constrains and
enhances state actions); (3) domestic institutional perspective that emphasises the role of the institutions
of the state at the domestic level (policy makers are autonomous from society and implement policies to
satisfy national goals that are tied to their survival in power/office); and (4) domestic societal perspective,
which emphasises the economic and sociopolitical pressures as major determinants of national policies
(i.e. liberalization and protectionism are preferred by different socioeconomic groups within society).
Lastly, we turn towards international political economy paradigms: Realism, Liberalism and Marxism. In
the 1980s and 1990s, people were debating which perspective is more relevant, however, nowadays it is
not that relevant. The point is to use them as a heuristic tool to frame the explanation of a given event
rather than to prove which perspective is superior. Then there is a debate about whether constructivism
is another perspective or not. According to the professor, it is about falsifiable arguments and empirical
evidence and much less so about which perspective is right. The constructivists have contributed in that
not everything is about a rational utility calculus, it is about ideas and sometimes these ideas are crazy,
but they have real world consequences in the end (i.e. Hitler and communism in the Soviet Union).
1. Realism assumes the superiority of politics over economics: states use economy to pursue power.
The international system is anarchic and changes in the distribution of international power affect
and determine the form and type of international economy.
2. Liberalism looks at economics and politics as separate spheres, there is a superiority of markets,
limited governments, harmony or coordination of interests among states (spread of liberalization)
3. Marxism assumes superiority of economics over politics, where the means of your production,
determine the means of your social reproduction, capital exploits labour and the relationship
between the capitalists and labourers are antagonistic (all in Frieden’s paper).
The three papers assigned reflect each of the perspectives applied. The main argument of the realist text
is that in a multipolar world, security concerns do not interfere with trade given that we have relative
gains (a realist framework). To cut trade would be inefficient for security. The polarity of systems is an
issue of measurement. You trade until there is fire under your roof and then you cut all ties, which is why
Lieberman’s analysis makes sense for multipolar world orders (unlike in bipolar orders, trade happens up
until the very end). But also how do you measure polarity? The absolute gains argument is a bit of a straw-
man. Countries are economically motivated and will trade under all circumstances, looking at absolute
gains (it is almost wrong from the start). But to think that countries will give up huge benefits from trade
even under moderately hostile conditions, that is an equally poor case for trading. The truth is somewhere
in the middle of countries prioritizing relational and absolute gains form trade. In any case, politics are
not separate from economics and if there is a huge political disturbance between countries than this
also harms trade (if this is the realist position, liberals would also agree). Absolute gains and multipolarity.
The Dependencia Theory has its roots in Marxism, but it is not clearly a Marxist theory. The main point is
that economic development is a zero-sum game. The North develops at the expense of the South (yet, it
is very hard to test). At this point, it also ran into a dead end (just like the polarity argument). The North
is to blame for the lack of economic progress in the South. It is a structural and systemic argument. The
main problem is bad data, even if the authors do a good job in using the data that they have. It is the
Marxist argument pushed beyond the class as the unit of analysis. Now the unit of analysis is the state.
Most political economy research is motivated by the liberal perspective. With regard to the Simmons and
Elkin’s (2004) text, the main argument is economic liberalization is driven by two things: the properties
of countries and the choices of other countries (countries similar to the country in question). The question
is how much does it matter what particular others do. It does not matter what the rest of the world does,
but it matters what people close to you do. The whole idea is that you are more likely to do what other
countries close to you do. Where it gets tricky is how do we measure the extent to which your decisions
are influenced by what countries close to you do? The point is to explain what country A does explained
by what its close allies do plus the characteristics of country A. To what extent can we explain country
behaviour based on country characteristics compared to what other countries do (the point is to look at
whether policy choices are domestically or internationally determined).
Political Economy Lecture II
Government Spending
Government size has increased immensely since World War II. The trend reversed in the 1980s, but not it
is back up again because of Corona. Spending per person is highest in Norway, where it is over 30,000 USD
per person. Of course, there is cross-country variation in spending which we will also discuss. The question
now is why has government growth occurred? There are several theories of government size and growth
predominantly following from a microeconomic perspective in economics and political science literature:
1. Citizen-over-state theories of government: citizens ask for programs and the government simply
responds to the will of the people (government grows because of the citizen).
a. Governments as provider of public goods and a reducer of externalities. It is the median
voter who determines the demand for publicly provided goods as a function of income,
the relative price of public goods and taste. The Median Voter Theorem effectively states
that in a two-party system, everyone’s preference are single peaked on a single dimension
and given that few people vote on the far-left and on the far-right, their preferences will
peak somewhere in the middle, so you have a normal distribution. This is why parties try
to appease the median voter, because if they appeal to the extremes, they will lose a lot
in the middle. The median voter spawns as a reaction to the problem of the majority. The
general will is absent, so it is the majority that decides. But how do we incorporate the
losers? The median voter solves this problem because the ideological gap between the
winners and the median losers are close. However, the theory relies on two assumptions.
First is the dimensionality assumption, the fact that preferences are represented along a
single dimension. Secondly, is that preferences might not be defined in spatial terms (i.e.
distance from the ideal point). It might be that there are many issues, there is no single
issue that takes place, i.e. what if people prefer everything or nothing. The assumption is
that there is going to be convergence and politicians will provide something in the middle.
Another definition is that of a Public Good, which is a good that is not exclusionary, so it
cannot exclude anyone and there is no rivalry over it, so if one person consumes the good
it does not reduce the possibility of another person from doing the same. The government
provides goods as an actor in the free market. Alternatively, commons are goods that are
not exclusionary but are prone to rivalry because you destroy them at the end. So for a
public good, everyone wants to consume it, but no one wants to pay for it. An Externality
is a consequence of an action of an individual or groups of individuals that creates benefits
or drawbacks to third-party that has nothing to do with it (i.e. pollution). The provision of
public goods and the reduction of externalities is a reason for government to intervene.
i. Income: every time your income or salary increases, you will spend a bit more. If
the income goes up by 10%, you will increase spending from anywhere between
1% and 10%. As such, the demand for public goods will increase as the income of
the median voter increases. This, however, does not work in practice. There is not
a lot of evidence. This theory does not receive much support.
ii. The price of public goods compared to private goods: public goods are inelastic,
so they do not respond very well to increased prices in the private sector. If the
price of government goods and services increases and the quantity demanded of
the goods does not decrease by the proportionate amount, government spending
increases. The explanation is that the price of production in the private sector has
productivity gains (especially in the manufacturing sector), which decreases the
relative price of private goods compared to public ones, given that productivity
gains are not observed in the public sector. This is especially the case given that
the public sector is predominantly service oriented (i.e. national defense, police
and education). The public sector did not experience the same efficiency gains as
manufacturing and, as such, the relative price of public goods has been increasing
(so governments have to pay more for public goods relatively and thus the size of
the government always expands). This is called the Baumol Effect (relative).
iii. Taste: this explanation is based on the idea of the compensation hypothesis: as
the word has become more globalized and countries became more exposed to
external shocks or outsourcing of many jobs abroad. The fear of losing one’s job
made citizens demand for more securities and employment insurance from the
government. And these problems are costly. If the government were to provide
for such insecurities, the government’s size would surely increase. So the more
open the economy the larger the government to provide insurance to the citizen.
b. Government as a redistributor of income and wealth. The government still responds to
demands from citizens, but this time in light of redistribution. The most famous model is
the Meltzer and Richard (1981 and 1983) model. Alas, the model has assumptions, such
as leisure time being inversely related to the fraction of total time worked, consumption
being inversely related to the tax rate and income being positively related to productivity.
The bottom line is that a higher level of productivity equates to a higher level of income,
and the higher level of income increases consumption and wellbeing (the older income
earner despises taxation while the young dependent loves to receive free services). Thus,
individuals demand a combination of tax rates and lump sum payments that maximize
their well-being. Individuals with lower levels of productivity (lower income) will demand
higher tax rate and a higher lump-sum payment from the government (insurance). Also,
individuals who do not work will simply demand higher lump-sum payments from the
government and higher tax rates compared to those who are working. So governments
grow in part because, over time, new entrants into the voting population are low income
workers and they will cast votes for candidates who levy higher taxes and increase the
amount of redistribution (mostly the left as opposed to the business right). At any given
time young people join the voters and demand transfers, which the government pays up.
If the median age is below the mean, the government will increase as it moves money
between the groups. Increased taxation benefits the new entrants as opposed to the old
taxpayers. However, this theory assumes that every year there is more of the younger
generation coming in with these demands than older workers becoming conservative. We
also have theories explaining redistribution based on social affinity. If the middle class is
closer to the poorer class in terms of incomes, the greater the amount of redistribution
coming from the median voter theorem. Alternatively, if the median voter income is high
and it feels closer to the upper class, the size of the government will not budge (e.g. Great
Depression). Lastly, Peltzman (1980) offers a third model, where candidates promise re-
distribution to groups of citizens in order to gain their support. If income distribution
across classes is similar then candidates must offer a greater amount of redistribution to
gain supporters (however, the empirical evidence is mixed and often contradictory).
c. Interest groups. In democratic systems, interest groups have become crucial to decision-
making processes and their continuous engagement begets for higher government size.
It might the case that what interest groups demand is not beneficial for society as a whole,
so there is a deadweight loss. There are three characteristics that have to be met for a
group to be characterized as an interest group: (1) it needs to be small, (2) it needs to be
powerful enough to penalize freerides, and (3) their public activities are the by-products
of their private efforts. The Gauss Theorem says that if the transaction cost is low, there
is always the possibility that parties reach an optimal agreement without a government,
i.e. environmentalists ask the government to regulate pollution. According to Gauss, the
government is not needed here. If the residents around this plan were able to negotiate
with the industrialists to come up to a common agreement for citizens to shut up about
the pollution and agree on a compensation, that will make both parties happy. Essentially,
the government is not necessary for regulation (a very libertarian notion of government).
This is also because the politicians have vested interest in the industrialists, i.e. sponsoring
their campaigns and also bringing income from taxation, so it is not needed to solve CAP.
2. State-over-citizen theories of government: the administration and bureaucracy is incentivized to
grow for the personal gains of the bureaucrats.
a. Bureaucracy theory. The goods and services provided by the government are created by
a government agency. As such, the supply of government output is also a function of the
demand of government bureaucrats who want to maximize their salaries. For example,
Niskanen argues that bureaucrats maximize the size of their agencies’ budgets according
to their own preferences and are able to do so because of the unique monopoly position
of the bureaucrat. Important here is the principle-agent problem where the government
cannot really observe the cost of the production of its services and bureaucrats within
these services have a monopoly on their production and provision. For example, in the
procurement of weapons the army just gives any figure and the government simply buys
it because there are no other private competitors in the sale of this arm. Of course, there
are some assumptions. First, some of the services cannot exceed the maximum. In Zurich
we have 400,000 people who produce a certain amount of garbage in a day, so we cannot
have millions of waste collectors working in Zurich financed by the government (there is
a limit). Second, only the bureau knows the true cost schedule because bureau funding is
provided by agents external to the bureau. Third, is the assumption that the bureaucrats
can make take-it-ot-leave-it budget proposals to the funding agency. Thus, bureaucrats
expand the size of government to enrich themselves in terms of income or prestige. But
there is another downside to this theory is that the workers who benefit from this are
appointed officials who have a fixed term, so they are not the ones to actually benefit
from expanding the bureaucracy given that they would leave before the effect kicks in.
b. Fiscal Illusion. The government can effectively deceive the citizen about the size of the
government given that citizens perceive the size of the government to be a function of
the taxes they pay, however, there are other ways of collecting taxes less obvious than
others (for example property taxes collected through monthly mortgage payments). This,
however, lacks strong empirical support. There is also tax elasticity: technological change.
3. Monopoly government and the Leviathan: it is the government per se (i.e. parliament) while it is
incentivized to increase spending on its constituency, it will only obtain approval if it increases the
spending on other constituencies represented by other legislators (scratch their backs).
a. The monopoly government refers to the power of the legislature, i.e. the congress to
decide on the budget and the president asks the legislature to approve of the presidential
budget. However, how much the president actually receives depends on the principle of
scratch my back and I will scratch yours. In other words, for approving the presidential
budget, legislatures ask for even more money for their constituents to obtain their vote.
“I will vote for the budget if you give 30 more million to the state of Iowa” for example.
So in the end, the budget is inflated three times as much as the original one, because they
can haggle for the budget. So by bundling narrowly defined issues that benefit only the
individual members of the government (along with the more popular public goods and
services) they increase the size of the government. This is highly inefficient because it is
the civilian who pays the price for the increased government size (government behaves
as a monopolist and engages in vote-trading among legislatures to approve the budget).
b. The Leviathan is different from a monopoly government. Here the government acts as a
whole with the sole objective to maximize revenues. Alas, there are constitutional limits
to the power of the executive, which constrains the government’s authority to tax.
4. Democratic institutions: the different reasons why government spending increased over time, it
is because of the different institutions (i.e. electoral rules and types of government).
a. The Median Voter Theorem does not discern differences between types of democracies
(all of the above applies to all democracies), so here we look at how institutions influence
government spending. We focus on two things: type of government and electoral rules.
With regard to types of government, we have parliamentary and presidential variants of
democracy. In a presidential system, citizens directly elect the executive, whereas in the
parliamentary system, an elected parliament appoints the executive. With regard to the
electoral system, we have proportional and majoritarian systems. In proportional system
parties in the parliament receive seats proportional to the number of votes the party won
in the elections. In the majoritarian system, the politician who gets the most votes wins.
Here district magnitude also matters. Which systems (form of government) will have the
highest government spending and why?
i. Parliamentary: In parliamentary systems we are more likely to have a coalition
government, which means that we rarely have a single party getting more than
50% of the vote. In such systems we have a vote of no confidence, where if one
party does not agree with the larger party, it can leave and break up the coalition.
This gives a lot of power to the minority parties because they can make demands
with regard to how much cash the government is going to spend. These minority
parties have incentive to provide money to their constituencies. So if we have all
these coalition members wanting something for their own consistencies in order
to maintain the coalition, spending is going to increase. The majority party needs
to give in and increase the spending in order to appease the minority parties.
ii. Presidential: In presidential system, the president is elected directly. He needs to
spend a lot to get elected. He needs to provide a lot of public goods. However, it
is the government that asks, but it is the parliament that appropriates. In this case
accountability becomes key. In parliamentary systems, accountability is low, you
cannot accuse specific parties in bringing the government size to the roof. In the
case of the presidential system, the president is accountable to the voters, they
can see what he did. As a result, this will act as a break from very high government
spending (accountable to the population and the population does not want high
spending; the underpinning assumption is that voters are fiscally conservative). It
is not the president, but rather you can put the blame on the main congressman.
They do not want to give a big budget to the president because the citizens will
keep them accountable for this, unlike in a coalition government (this happens
because the president is separate from the legislative unlike in parliamentary)/
b. Which electoral rules result in a higher government spending and why?
i. Proportional Representation: Here you have many parties. In order to be elected
they need to promise a lot of things. The more parties that exist, the greater the
spending they will all try to bargain for and the size of the government is going
to increase. In the end, because all the parties are responsible, accountability will
be lower in the PR system than in the plurality system (low accountability).
ii. Plurality: We have a small constituency and we elect the candidate with the most
votes. We have many candidates, but only one can win. This induces candidates
focus on a specific group that is going to give them this extra vote to win. This is
called the swing voter. In this case, we are not increasing the spending a lot, but
very little. But we will also see very targeted and regional government spending.
This is no the case with PR, where we have broad programs directed to the whole
country, again because it is proportional (so everyone wins), unlike in plurality
systems where the winner can forget about the constituents of the losers. Again,
accountability to the core voters becomes key here (more accountable here).
c. What does the empirical evidence show? The empirical results seem to be in alignment
with the theory. There might be endogeneity, or an omitted confounder in the analysis.
Presidential systems spend about 5% less on the government than parliamentary systems
and this relationship is statistically significant. Likewise, plurality rules also spend 5% less
than PR rules, so combining both presidential systems with plurality voting, we have a
reduction of 10% in the size of the government. When it comes down to the type of the
government, many presidential systems are new coming from Latin America (not that old)
so it might be the case that young democracies as such spend less than older democracies.
i. Accountability: voter’s ability to identify politicians responsible for policy (high in
plurality and presidential systems, low in PR and parliamentary systems).
ii. Representativeness: political decisions reflect the preferences of large spectrum
of voters (high in PR and parliamentary systems, low in plurality and presidential).
d. Finally, we look at federalism and representative democracies. However, the paper by
Funk and Gathmann (2011) do not look at the federal government. Instead they look at
the canton and whether referenda at the canton level will have an effect on the local
level. They use two ways to measure this. The first is mandatory budget referenda and
the second is the amount of signature needed to initiate a popular initiative. The results
show that budget referenda reduce the size of the government by about 8% and for the
signatures, the lower the number of signatures needed to initiate the reform, the lower
the size of the government. However, cantonal and local level are not substitutable. In
direct democracies, voters act as another constraint on the size of the government (here
we assume that the median voter is fiscally conservative, or at least more so than the
government). In this case, we are disciplining the government. Usually, you call referenda
for government spending because you do not like how much money is spent. Bureaucrats
hide the real size of government, so the citizen in referenda acts as an extra check.
Political Economy Lecture III
There has been huge economic growth over the last 200 years, especially after the industrial revolution.
Nowadays, we are at least three times richer than our predecessors. So what explains this huge economic
growth and heterogeneity in growth? In the early years of economy researchers tried putting everything
into a single statistical model and run it. The most important determinant of growth was religion (in fact
Confucianism). The most classical theory of growth is that it is exogenously determined by labour and
capital and if capital and labour goes up it will increase the growth of the country, i.e. Y = F(K,N,H,A,X). For
one, savings rate might cause long-term output to rise, because higher savings allows for more investment
and a larger capital shock. Another theory contends that it is the rate of population growth that causes
long-run output to fall, because with higher population growth, more output must be used to equip
workers with capital, leaving less output available to increase capital per worker (crowds out investment).
Yet the most important determinant is productivity (i.e. Solow’s model), which causes long-term output
to rise because higher productivity directly increases output; by raising incomes it also raises savings and
capital. So the thing that ensures growth over time is technology, but technology is exogenously given.
Would economies converge then? There are three possibilities: unconditional convergence, conditional
convergence and no convergence. Unconditional convergence implies that over time poor countries will
catch up to the rich ones (Uganda will have the same standard living of Switzerland). No convergence
means that there will always be heterogeneity in the standards of living. Conditional converge means that
convergence will only appear that share some fundamental characteristics: so there will be convergence
but among specific countries in groups rather than as a fact. In reality, empirical evidence shows that there
is conditional convergence. One problem to this classical growth theory (the Solow’s model) is that the
rate of productivity is exogenously given. Now we have the New Growth Theory, which says that there
is a possibility to play with technology and the government can do something about it. Governments can
spend a lot and emphasise technological innovation, but in the private sector (so productivity increases
can also be introduced endogenously). Technological innovation by private firms results in productivity
growth. The point being that we increase human capital (knowledge and skills) through education, which
will be more productive and contribute to increased productivity and innovation. Empirical studies reveal
that education and literacy levels are strongly related to the rate at which the economy of a country is
able to growth (Glaser et al. text). In Solow’s model productivity and technological change is exogenous,
whereas in the New Growth Theory, productivity caused by human capital and innovation is endogenous.
These two theories are the most known theories on economic growth: productivity and human capital.
However, apart from this, everything else can affect economic growth (like religion), but the main point
is to find the main determinants. This is where we transition towards institutions as the more fundamental
cause of economic growth as proposed by Acemoglu and Robinson (2008) compared to capital and labour.
Institutions are defined as the rules of the game in a society or, more formally, are the humanly devised
constraints that shape human behaviour (definition from North). They determine the rules of the game
of interaction among agents. They are human made and they are about politics; political institutions (i.e.
electoral rules and checks and balance systems) are different from economic institutions (i.e. free market).
In order to study whether institutions matter we need to know who has the power, what are the limits of
their power and what are the dynamics that determine this power. There are several theories that try to
explain differences in political institutions and their heterogeneity between countries and there are four:
1. The efficient institutions view: society or economic agents can choose whichever institutions will
maximize the size of the pie. They choose appropriate policies and institutions for their conditions.
2. The ideology view: societies may choose different policies because they or their leaders disagree
about what would be good for society and this uncertainty leads well-meaning political actors to
differ what is good for their own people. Strong forces prevent the implementation of policies
that are known to be bad for society at large (theories of belief differences).
3. The incidental institutions view: societies have different institutions because these institutions
were determined by historical accidents at critical junctures.
4. The social conflict view: societies choose different policies, some of which are disastrous for the
citizen, because these decision are made by politicians or politically powerful social groups that
are interested in maximizing their own payoffs, not aggregate output or social welfare. As such,
institutions emerge as a result of economic agents’ conflicting preferences, which does not make
them necessarily efficient. This view is intimately related with political economy. But how does
conflict aggregate and produce the existing institutions that we have? (At the core of Acemoglu).
This has to do with power, which reconciles conflicting preferences. For example, in South Africa during
apartheid, whites could vote while blacks couldn’t. As such, whites have more political power because it
is their preferences that count. A distinction must also be made between de jure and de facto power. Alas,
de jure political power is allocated by political institutions (such as constitutions or electoral systems). Yet
de facto political power emerges from the ability to engage in collective action, use brute force including
paramilitaries, armies, or other channels such as lobbying or bribery (so institutional versus raw power).
However, why is it that agents do not agree on something in the first place? The main source of conflict is
the concept of political commitment. There are no guarantees to enforce any commitment. So actors are
unable to commit to their actions. A lot of the problems that arise are exactly because of this issue. Both
political and economic losers will resist all institutional change. As a consequence, commitment problems
lead to inefficient institutions because the losers do not want to lose what they had. Institutions that
resolve this commitment problem are the ones that will be implemented (institutions solve the CP).
This happens through a chain of processes. Initially, we start with the existing economic institutions that
affect the economic performance of the country as well as the distribution of its resources. Economic
performance will help some groups have more economic power. This distribution of resources to some
groups will affect their de facto political power (i.e. economic growth redistributed de facto power to the
merchant class of the 17th century in contrast to the established monarchy). Those who have the de facto
power like the merchants will establish their own political institutions, which will shift the de jure political
power in their favour. This will lead to new types of economic institutions forming and the cycle repeats.
Again, a good example of this is the replacement of the monarchy by the merchant class in the 17 th and
18th century. Distribution of resources gives you more de facto power and agents will use their power
to reshape economic institutions to give themselves more power (Acemoglu and Robinson example).
Those who have power will establish institutions that reinforce their power to maintain their benefits. In
essence those in power create their own constraints. To resolve their own commitment problems, they
constrain themselves. Everything is based on how to solve the commitment problem, so political elites
would introduce constraints on their own power for more efficient growth. Institutions are better when:
So in the 17th century when the landed nobility started to feel threatened by the change in the de facto
power towards the merchant class, as they could establish a large army and threaten the landowners. As
such, the nobility created new institutions giving rights to the merchants, which also tied their hands in
the future so that the merchants do not take all their power away. So effectively the nobility integrated
the merchant class into its system of institutions. The nobility did not hoard the power so instead of being
violently overthrown, they decided to go down slowly and quietly. By expanding their power to the new
merchant class, this led to economic growth in Europe (incorporating the Industrial Revolution). Power
sharing is also a characteristic of democracy (European democratic power; this is different from Ukraine).
The Acemoglu and Robinson paper is the most popular theory of economic growth right now. They show
that higher the settler mortality, the higher the expropriation risk (i.e. not settling by exploiting). Likewise,
expropriation risk is highly correlated with economic growth. It also has to do with population density, if
the population density of the locals was very high, the colonizers would not settle. If population density
was low and mortality rate not that high, they thought of staying there and brough their institutions to
ensure that they could stay there for the long term. The disease environment was used to measure the
mortality rate of settlers as an exogenous instrumental variable. When it comes to institutional reform,
unsuccessful reform includes changing the economic policy, without changing the underlying balance of
power. It might be useful to change the de facto power. Furthermore, changing the de jure power does
not work (i.e. the slavery example with the landowners retaining their de facto power). Should both de
facto and de jure power change? Even this might not lead to successful reform because of the iron law of
oligarchy, which says that even if both de facto and de jure power is reshaped, if the new elites that come
to power are faced with the same incentives for exploitation based on the political environment and will
engage in the same bad practices as their predecessors. Perhaps changing incentives or the commitment
problems might serve as a solution. In other words to summarize the above:
1. Directly reforming specific economic institutions will not work without changing the balance of
power in society or the political equilibrium (the see-saw effect: replace one inefficient instrument
(patronage) for another (repression)).
2. Reforming political institutions (i.e. altering the distribution of de jure power in society) will not
work either if de facto power persists and overrides the effects of reforms to political institutions.
3. Simultaneously changing both de jure and de facto power might may not achieve real reform
either, because the political equilibrium may be path dependent (a. the iron law of oligarchy:
those who acquire power in the new political equilibrium (new elites) do not have the correct
incentives; b) fighting fire with fire (citizens support a leader as bad as the incumbent one).
Another possible solution is liberalization that could lead to economic growth. This is the Tobellini article
where they try to figure out whether democracy after liberalization or liberalization after democracy leads
to growth. However, they find that liberalization after democracy decreases economic growth because it
will lead to a stasis of among the interest groups. Liberalizing makes sure that no special interest is going
to suffocate the political processes for their benefits. In other words, young democracies in very closed
economic environments are more likely to be bogged down in redistributive conflicts and populist policies
while young democracies in open economies are forced to pay more attention to economic efficiency.
The opening of the economy goes hand in hand with the protection of property rights and the rule of law,
which may be a pre-requisite for a well-functioning democracy. As such, authoritarian regimes ought to
give priority to economic rather than political liberalization. An example is the famous natural experiment
of North and South Korea. In both countries, everything was the same in the 1950s, but after the war, one
became poor, the other became rich because of developing different economic and political institutions.
Even though a dictator, it opened the economy. When people are wealthier from liberalization they ask
for property rights and this leads to demands for political democratization. The other way around, if just
democracy is implemented without any economic freedoms, interest groups will just be bogged down
by redistribution conflicts instead of promoting growth in the country. For China, it was the same thing,
however, what we have seen is a retrenchment of the freedom of the people. If they still keep opening
the economy, will this eventually lead to political democratization? Singapore is another case to look at.
- So democratization after liberalization works, while liberalization after democratization does not.
Likewise, parliamentary and proportional representation systems are more likely to democratize and will
result in greater liberalization, but the authors failed to find any effects for economic growth. Finally, from
the Glaser article, the authors criticized the measurements used citing that most indicators of institutional
quality are constructed to be conceptually unsuitable for testing the proposition that political institutions
cause growth. The risk of expropriation and government effectiveness are outcome measures reflecting
the government’s past behaviour, while the constraints on executives measure reflects the outcome of
the most recent elections (so it is volatile and does not capture something deep). They also argue that
most instrumental variable techniques are flawed as settler mortality and population density predicts
institutional quality and level of economic development today (Acemoglu et al); these instruments are
highly correlated with human capital, human capital is better predictor of growth, and predictors of
settlement patters are not valid instruments for institutions. So, human capital is a more basic source of
growth than are institutions; poor countries get out of poverty through good policies; economic growth
improves political institutions. So autocrats should focus on economic liberalization. Improving human
capital, improving education leads to the establishment of institutions that then foster economic growth.
However, there is a conditional effect. You need institutions to better human capital and both have an
independent effect on economic growth. We need to have institutions that embitter education in order
to have greater growth. So it is an interactive term. What we need is to have education that increases
productivity to increase economic growth. The Soviet Union had good scientists, but they were employed
in the military industrial complex or in the government and not in the private market (where technological
change leads to more productivity). As a consequence, their employment (and the Soviet Union in general)
was inefficient. Back to Acemoglu and Robinson just to summarize:
1. Some legitimacy will be able to implement and sustain policies that may bear high short-term
costs.
2. Various of the institutional characteristics of a democracy (an independent legal system) are also
required for a successful policy implementation (liberalization).
3. Democratization may limit rent-seeking due to its system of checks and balances.
However:
1. Democracies may lead to policies that hamper economic growth (rich-to-poor redistribution,
large public sector, higher taxes) due to majority voting.
2. Influence of interest groups will reduce flexibility of the economy.
However:
1. Dictators may also be forced to follow opportunistic policies if their survival in office is threatened.
2. Authoritarian regimes are not homogeneous = “technocratic” regimes (S. Korea and Taiwan) vs
“kleptocratic” and/or inept authoritarian regimes.
3. Authoritarian rule can mean arbitrary rule and undue inference, which may hinder economic
growth.
4. A strong state and an authoritarian state are not necessarily the same thing.
Political Economy Lecture IV
Sustainable Development
There has been a lot of discussion about how we should measure sustainable development. For example,
measuring GDP is one thing, but there are other things. What about infant mortality? Personal freedoms
or happiness? Are the happiest people on the planet living in the richest countries? There is more than
money involved in development. There have been other factors added to GDP to measure development.
The Human Development Index (HDI) is one attempt that adds education, literacy and infant mortality
rates to GDP and comes up with an aggregate measure of development. GDP does not tell us much about
sustainable development, so we need new things. There have to be alternatives to the GDP (i.e. the SDGs),
and if undertaken with sufficiently broad participation, the hunt for the successor to GDP can be found by
2015. However, why have we not found a substitute yet? Because it is easier to understand for everyone
(especially the public and government bureaucrats) that $100 is better than $10, and as soon as you get
away from money, arguments start about how you would weight and aggregate other measures. There
are also international rules for reporting growth and this is how the global political architecture operates.
Analytically, there is merit to the GDP as well. There are a lot of merits from disaggregating and looking at
indicators separately. Despite this, policy-makers and bureaucrats want complex scales and rankings.
People are preoccupied with rankings everywhere, even universities. As such, there is no agreement of
whether we should move beyond the GDP, however, it might be useful not to lump everything together
just out of a political desire to lump everything together to figure out how sustainable each country is.
The affluence paper is a direct attack on the traditional Kuznets curve. When we think of the Kuznets curve
we traditionally think of territories. All pollutions and emissions are taken on the territory of a country
and we get numbers and look at how well countries are doing. This is the world of the traditional Kuznets
curve. The discussion involved, however, to include planetary boundaries, or the carrying capacity of the
whole system because if we destroy the Amazon, this will not just harm Brazil, this will have fundamental
repercussion for global climate and biodiversity. This evolved research into the consumption perspective.
We but a computer, but a lot of pollution went into creating it. This is called embodied pollution or grey
emissions. We do not see this, but the emissions went to producing this device is substantial (whether it
is produced in China or India or elsewhere). In environmental accounting, the production of this computer
do not show up in the Swiss environmental records, rather they show up in the Chinese environmental
records. This is what the consumption perspective is about. If we move away from considering territorial
boundaries, but look at the global system and the global environmental impacts of local consumption,
then the whole environmental assessment becomes very different. The global environmental impact of
consumption in Switzerland is located 80% abroad. So if we are doing conventional calculations than we
are only capturing 20 – 30% of the actual environmental impact of Switzerland globally. As a consequence
if we are making assessments about how clean and green the Swiss economy is, we are totally mislead if
we look at the pollution produced locally in Switzerland. In fact a new paper by Bernauer seems to suggest
that democracy and income levels are creating more pollution than less. This is the complete opposite
of the paper Bernauer published with Koubi in 2009, where it is the other way around, where democracy
and presidential systems seem to produce less pollution (measured in SO2). However, when we take the
impacts of consumption, we see that democracies at high income levels seem to push out the impacts of
their consumption abroad (the West in a nutshell). Trade is the conveyor belt. What happens is that the
drivers of this outsourcing of pollution are the more democratic countries and the importers are the rest.
Thus, the switch from a territorial to a consumption-based perspective contributes to the discussion on
how we measure sustainable development. Simply taking GDP and adding things to it might not be that
useful (continuing to expand on the Kuznets curve), however, shifting from a territorial to consumption-
based perspective on pollution is both analytically useful and politically relevant. This is the elephant in
the room. The richer democratic countries are becoming greener and cleaner as the Kuznets curve seems
to suggest, but the elephant in the room is that these environmental impacts have not simply gone away,
i.e. there is no endogenous reduction in these impacts, most of them have been transferred elsewhere
and the total environmental impact on the system has continued to go up. We cannot just accuse China
of polluting everything given that it is actually us who consume these goods. This raises normative and
political debates: who should be responsible? Should we allow China to pollute the global environment?
What is happening right now is that if we have this more holistic perspective, there is a lot of evidence
that we are overconsuming the planetary boundaries (we are consuming more than the planet is capable
of sustaining and thus we have climate change). This is driven by international trade, but it is the conveyor
belt, it is not to blame. Our own unsustainable consumption is what is to blame unfortunately.
- A shift away from the Kuznets curve towards a consumption-based perspective. We demand the
poorer countries to sponsor our consumption by polluting. So it is us that pollute even more, not
directly but indirectly by outsourcing our pollution. As such, global pollution levels keep going up.
Everything is organized around territorial boundaries, but it totally ignores the fact that us consuming this
produce is what leads to climate change. As such, there have been significant developments in the EU to
address this. Sustainability cannot just be about adding infant mortality to GDP and measuring it, as we
will be stuck in the old world of territorial measures. We need to move beyond territorial boundaries.
Thus, sustainable development should mean that our own environmental impact should be looked at from
the perspective of consumption. A society should not be sustainable just in its territory, it does not stop
at the border, it should be sustainable globally. We are heading towards this direction. Moving beyond
territorial boundaries and the Kuznets curve with the West overconsuming and outsourcing pollution.
Going back to measuring sustainable development, the UN has created the 17 Sustainable Development
Goals (SDGs), which goes beyond the idea of aggregating various measures into a single complex index.
The expectation is not that these goals develop parallelly in the same direction, but it allows us to try to
understanding the progress humans are making. The Kuznets curve studies are all territorial (amazing).
The idea of the Kuznets curve effectively states that pollution during development initially goes up then
it goes down, the question whether that is true has been hotly debated. What drives the curve? In an ideal
world, we have the Maslow’s pyramid of needs. First we prioritize economic development and use what-
ever technology is available to get growth by any means possible (the London smog is an example with
very little regulation on environment). Economic growth goes up and as income increase you see things
disappear (wildlife and forests), you have highly polluted air and water that you see happening in your
lifetime, which you physically see. You see this in your lifetime and it affects you and this leads to more
demand for environmental regulation (from the mass publics). This leads to a levelling off of pollution in
the developed world and finally a declining curve. Impacts that you can outsource are out of sign and out
of mind. This is why current democracies export and outsources a lot of pollution embodied in imported
goods. This is why we are more likely to notice pollution affecting us than outsourced pollution abroad.
In the environmental economics literature, the EKC has become somewhat popular because it is simple
and stupid. It is empirically easy, a theory that turns out to be empirically alright for the West. There is a
clear hypothesis and increasingly good data. It allows a lot of researchers with little money to get rich.
This is a bit of the explanation why the Kuznets curve is popular (some self-criticism by Bernauer).
It is more interesting when we look at variation across countries. There is a huge amount of variation and
income helps to explain some of it, but we have rich countries that are highly polluting and others that
are rich but not polluting. What explains this variation and what are the drawbacks? Will the functional
form of development for the Chinese in these decades will be the same as that of the UK in the last 200
years? Will the trajectories be the same? The curve now should technically be flatter given the increase
of new greener technologies. However, here consumption also goes up and surpasses the impact of new
technology, so technology does not off-set pollution. Technology produces efficiency gains, but do these
efficiency gains get overcompensates by consumption (i.e. cars become more overpowered, but people
still demand overpowered cars). So if the Chinese curve will become flatter then the UK curve, that is not
so certain. Even though the Chinese can import new technology consumption may surpass the effect of
this new technology. Another dimension is the societal regulatory one. People might start making more
demands for cleaner air or water because they see other cleaner countries and how other people live.
The government itself might see how well other states are doing with cleaner air and try to get their as
well, pushing their policy makers for cleaner regulation. These are boundary conditions that are different
from what the UK had a couple of hundred years ago (reference, comparison and new technology). Also,
if we think of off-loading pollution, at the time the UK was industrializing, it could not offload its pollution
to other countries. Nowadays there is a lot of outsourcing even from China to a lot of the poorer countries
in Asia, South-East Asia and Africa. So there are various boundaries that are different today. So does China
has the prospect of reaching the peak of the curve faster than Europe (for Europe it was the 1960s)?
Of course, even in developed countries pollution still remains, for example noise pollution. It is a big public
health issues. Biodiversity is not doing well in Switzerland and some esoteric problems like pharmaceutical
residues and micro-plastic. How should the EKC look at for different times of history and do the Chinses
have to suffer the same pollution as the UK given the different boundary conditions? Do developing states
today have an opportunity to develop without experiencing the worst of pollution (smog in the UK). Not
just because these countries are polluting themselves, but they are also creating repercussions for the
rest of the world (i.e. do we want the Brazil government to chop down the Amazon forest, or should the
world push Brazil into protecting its forest better). Also if we look at Ocean pollution, there is damage that
poorer countries can do to the rest of the world if they move through the worst of the EKC production.
For developed countries (like Switzerland), cleaning water, setting up the filtration systems and getting
rid of lead pipes costs a lot of money. The estimates are typically that it is a couple of franks per day for
the Swiss, which culminates into billions of franks that go into maintaining the water supply system. In the
end it is a question money and technology, and richer countries can do this easier than developing ones.
This is why you see a decrease in pollution in the richer countries, precisely because they have the money.
Moving to the Bernauer and Koubi (2009) paper, the idea being that politics should matter when it comes
to pollution levels. The main findings concern forms of government (parliamentary versus presidential)
and the effect of interest groups (labour unions and green parties) on pollution (regulation). They were
pushed into looking at whether presidential or parliamentary systems were better, however, the empirics
remain weak. The presidential and parliamentary system argument is obsolete now. The argument is that
presidential systems have a larger winning coalition and so they produce more public goods. Why focus
on forms of government not electoral rules? Would PR produce more public goods? (Yes – increased size).
Authoritarian regimes has a smaller winning coalition than democracies and therefore are more effective
at reducing pollution levels (governments as providers of public goods). Revise the labour union argument.
If we have a democracy and a non-democracy and assume that the environmental pressure is the same,
or constant, why would we expect more demand for protection from pollution than in autocracies? Both
systems are doing a terrible job, but democracies might be doing it better. In a democracy, there will be
more civil liberties (NGOs can protest etc.), so if there is more environmental pressure, there is more
demand from the civil society to do something about it. This bottom-up demand for more regulation
should be stronger in a democracy relative to other systems and when it comes to the supply-side, in a
democratic system there is more accountability towards the citizen and if the policy-makers do not do the
things they were voted on, they will be out of the office. This is not the case in authoritarian regimes, who
only need to satisfy a core group they are dependent on to not rebel. Transparency is another thing. Also
scientists are freer to report and be critical. So there is more pressure both bottom-up and a government
that is more responsive top-down. Based on this demand, we will see more responsiveness with respect
to pollution. So more possible pressure bottom-up combined with a more responsive government.
With regard to the labour union, we wanted to have an argument: if interest groups are pro-environment
and are stronger in society that groups that are not for the environment, we are more likely to see more
environmental action (it is about interest group strength). How can we measure the strength of business
associations and environmental lobbying groups? Data is pretty bad partly because NGOs are not actually
that transparent and also it is not totally clear what green interest parties actually are. In some countries,
we do not have green parties, but all mainstream parties are pretty green. Thus, measuring the share of
seats of green parties in parliaments is not a very good measure. The democracy argument is very strong,
while all the other parts are relatively weak (presidential vs. parliamentary and interest groups). This is
because the data is just not there yet. More data and better indicators should be better. READINGS.
Political Economy Lecture V
Trade is a crucial part of economic systems. There has always been trade between nations (even since the
Roman times), and the question now becomes how best to govern this trade. It is no longer a question of
whether trade is good, but more of a question concerning who benefits from trade (for whom is it good),
what are the aggregate effects and what are the distributional effects. Liberals would argue that all trade
is good, but there are distributional conflicts given that some industries and businesses benefit more than
others. It also about distributional conflicts between countries. As such, trade issues are a key part of PE.
A lot of research goes into the question of who benefits from trade liberalization and there is some sort
of consensus that trade liberalization is driven by large firms that are internationally competitive who
gain most from open trade. So it is a lot about who benefits, who loses and who lobbies. Let us hone in
on the first strand of literature. In the past 40 years, trade has grown and tariffs rates have gone down.
The average tariff for most countries is rather low. There are some non-tariffs barriers to trade (e.g. quotas
or hygiene standards and regulations including technical and safety standards for given products). But in
any case trade is going up and tariffs are going down. This is happening because trade liberalization might
raise the standard of living. Another argument is that it fosters peace (through the Liberal economic peace
model). In the context of Ukraine, the Liberal economic peace is somewhat confirmed in that Russia now
is paying a huge price for the war in Ukraine. It is similar to how you have a law, it does not mean that
everyone will abide by it, but the system might respond and despite the human suffering in Ukraine, the
international system is responding rather well to the war in trying to punish those who violate the Liberal
economic political idea. Moving towards institutions, the role of the World Trade Organization (WTO) is
huge, but also preferential trade agreements (PTAs) are also abundant. Furthermore, the WTO dispute
settlement mechanism is still relevant and effective in brokering trade disputes, which is the most robust.
One liberal principle of trade is that it makes societies better off on average (by no means implies that
everyone benefits equally). It reduces prices of consumption goods due to given countries’ specialization
in production. Sectors in which you have a comparative advantage, economic activity will expand and in
sectors where you are less good in international comparison (too high production costs), there we shall
see the number of imports to increase and the number of exports to decrease. As such, countries gain by
trading goods in which they have a comparative advantage for goods in which they have a comparative
disadvantage. Many countries experience changes in composition of their industries. Of course, countries
that have many natural resources export their resources, which is sometimes called the ‘Dutch disease’,
a form of resources curse where it actually prevents countries from mobilizing and becoming competitive.
This might stem countries from actually growing. Usually, things created for the military is then developed
into consumer goods (e.g. the radio or the internet), but in Russian this did not happen despite a large
military-industrial complex. As such, the resources course may combine with authoritarian politics and
prevent technological innovation (a toxic combination of the resource course and autocratic tendencies).
To summarize, the main issue is the Heckscher-Ohlin model, where comparative advantages arise from
difference in factor endowments. Some countries are more labour abundant, others are capital abundant,
which shapes the difference in production costs and different countries essentially have advantages or
disadvantages in one or another abundant factor, which they can then trade on the global market. At the
simplest level, parts of the labour force who are employed in sectors which are out of sync with this are
ones who ask for protection (out of sync with the global comparative advantage dynamics). MICRO.
How to know who trades how much? The total trade volume is greater between bigger states and smaller
in smaller states. Economic size is a predictor of trade. The EU and US, all else held equal, will trade more
than Fiji and Kenya. Another predictor is proximity. This explains why Germany and Switzerland trade but
not Switzerland and Fiji (both in terms of mass and distance). Dyads with larger economies and closer
proximity are more likely to trade. Trade offers many economic benefits, but it is also conducive to peace.
The Liberal peace thesis suggests that trade and interdependence are good predictor of peace between
states. The main argument is that it is about opportunity costs. War between Ukraine and Russia did occur
so what does it mean for the Liberal peace theory: it does not prove it wrong because the sanctions were
strong (the economic punishment for breaking the peace law was strong). There is a fairly strong liberal
economic peace there and if there is some kind of actor in the system, the system mobilizes fairly strongly
and the punishment is rather severe. The Kantian peace is based on deterrence, however, deterrence did
not work with respect to Russia, so the Kantian peace effectively failed. To reduce Kantian peace to the
post-fact sanctioning for breaking the liberal peace rule is not satisfactory. We move onto the empirics.
The argument from the paper is rather broad and is reminiscent of Karl Deutsch. In essence, peace is the
natural effect of trade. Two nations who traffic with each other become reciprocally dependent; for if one
has an interest in buying, the other has an interest in selling; and thus their union is founded on their
mutual necessities. To empirically measure it, you add the imports or exports of direct trade between two
actors, but the authors also look at indirect trade dependence and also the general embeddedness of the
countries in a trade network. The measures they construct how much overall countries are embedded in
a trade network. They take all of this into account (embeddedness within a trade community – Deutsch).
So they are not measuring direct trade flows, but rather constructing broad measures of the degree to
which countries are embedded in the global network of trade relations (how central or peripheral they
are). This goes way beyond the simple dyadic relationships previously used. What do the authors find?
Direct trade has a statistically significant effect on peace, indirect effect (via third party) was significant
only until the 1970s and network embeddedness it was not significant before the 1970s, but it is significant
nowadays. This is important because when previous literature measured direct ties in a dyadic form, the
results were always a little shaky. The also test the democratic peace theory and find support for it as a
dyadic relationship. They also find that the DTP effect is not bigger than the liberal peace effect (similar
sign effects, which is somewhat surprising). We have to look at two more papers on WTO and PTA.
Trade networks have become denser and denser. One key finding is that tariffs have gone down, but the
debate has shifted towards the under-the-table tariffs. Challenges to multilateral trade follows the decline
of the American hegemony that ensured free trade. Trade is not always good because of the prisoner’s
dilemma and it is a relevant metaphor for what we see. WTOs are negotiated tit-for-tat: trade is a very
complicated reciprocal process and partners expect partners to reciprocate. Because of the prisoner’s
dilemma actors would better protect themselves, even though liberalization is the best option. The WTO
can help with the prisoner’s dilemma given that it can enforce commitments between two parties through
the dispute settlement mechanism. It also benchmarks standards and disseminates information (trust).
The dispute settlement system is rather unique and it functions like a court. Initially it was ineffective as
the accused party could slow down the initiative and stall things. This changed and now if country A brings
to justice country B, a process is opened up and the accused party cannot stall the process. It is also quite
impartial and technocratic. Panel members look at cases and try to figure out who is right. The standard
accusation is that there are some import restrictions that are harming the party economically and one is
violating their WTO commitment. After which the WTO tries to figure out whether these restrictive laws
are compatible with WTO regulation or not. If the country is found guilty, there is a compliance processes
where the punishment is not considered by the court? It disciplines the punishment. The panel tries to
figure out what is the economic damage suffered by the principal while preventing the winning country
from lashing out at the loser. It leaves in the hands of the country how the accused is punished. Punishing
should not be severe as smaller countries that ask too much might instead suffer a lot with their trade
with a bigger country. As such, are there such effect in the international system where the smaller states.
There always the debate between right versus might. Although power should be under the law, does the
WTO dispute settlement live up to that or is there more discrimination against smaller states, i.e. do they
use the system less often because they are worried about it, or even if they win cases, they do not use the
opportunity of punishment as they are worried that the more powerful country that they sued will hit
back at them. The proclivity of trade disputes is proportional to the size of the trade between two states
(and here we go back to the gravity model of size and proximity that determines trade). The punishment
is set by the state that filed the complaint if it won, but it then asks the court as to the correct punishment
it is allowed to carry out if it were to do so. There is a ceiling in the punishment escalation process. This
helps disputes from going out of control. The panel limits and constraints the exertion of punishment. But
the paper by Bernauer really focuses on is the explanation of the use of the DSM (does the system favour
richer countries or no). Essentially what explains the use of the DSM by countries of different sizes?
If we have two small countries, how likely is the conflict? If we have Fiji and the US, Fiji is very unlikely to
attack the United States, and if the US the plaintiff, then it would be more likely to attack Fiji. Likewise,
Fiji attacking another small power is intermediate likely. With regard to power preponderance (measuring
the absolute difference between countries) and if the difference is small, nobody is really deterred from
suing the other, or the more powerful might use the size difference to push an agenda. With legal capacity,
the DSM is in the favour of the more powerful countries. Likewise, with the Gravity argument, as has been
said before, here if the countries trade less, there is a lower chance of the dispute, if the countries trade
more there is a higher chance of a dispute. This is a very interesting model taking into a lot of dynamics.
With regard to the results, the gravity effect dominates in terms of disputes (size and proximity), whereas
if the preponderance effect exists it suggests hidden discrimination. Lastly, there is no effect of income or
legal capacity on the initiation of the DSM. With regard to the empirics, the authors needed to have each
dyad in two different forms: one being the plaintiff the other being the defendant and vice-versa to gauge
the effects. Then we add a time series to each of these data and see what we find. The difficulty is with
the statistics is that we have many dyads with an outcome variable full of 0s because neither party began
a DSM process with another. In disputes like wars between countries are extremely rare events, which
requires a different data analysis (rare events logit). So the authors had two strategies to deal with this,
the first is more complex and the second one is more stupid: just look at the cross-sectional format and
then run simple models and the hard and easy methods produced the same thing. Always be suspicious
about the rocket science model especially when the authors do not find anything significant simply.
So the main point is that gravity matters: how big you are and how close you are matters more for trade
and for starting disputes. If you trade with each other and you have a bigger economy, you will have more
disputes. This is why we see more disputes between the EU and US than between Fiji and Kenya. This is
stupid and trivial, but it works. Likewise, there is preponderance suggesting some type of discrimination.
We would need to do qualitative research to observe it. We find no effect of income or legal capacity.
Finally, we look at the PTA as they are proliferating profusely. So why is it that they proliferate? They have
advantages, for example the fact that it is easier to negotiate with one or two countries than with 100 at
the same time. How can we explain which countries conclude PTAs with one another? It is about policy
diffusion and what is the dynamic that drives it? It is both competition, global value chains and signaling.
The empirical implication of the liberal peace theory are visible. Trade liberalization has economic benefits
and likely contributes to peace. A governments subvert the logic of trade liberalization cannot be taken
for granted. The WTO plays a crucial role in this regard: solving commitment and time-inconsistency
problems. However despite (or because of) its significance, the WTO is not free of criticism and is currently
riddled with problems. Current PE research focused on PTAs. Do PTAs increase wealth? It is a chicken and
egg problem. Increasing wealth can induce PTA, whereas PTAs can induce more trade. PTAs tend to be
associated with increased trade flows between countries. By increasing trade flows, do they a positive
effect on economic growth? There is some evidence in that direction.