Summary Political Economy Articles

2019

1
Week 1: Introduction to comparative political economy
- Arts, W. and Gelissen, J. (2002) Three Worlds of Welfare Capitalism or More? A State-of the
Art Report. Journal of European Social Policy 12(2): 137-158.
- Starke, P. (2006) The Politics of Welfare State Retrenchment: A Literature Review. Social
Policy & Administration 40(1): 104-120.
- Green-Pedersen, C. (2004) The Dependent Variable Problem within the Study of Welfare
State Retrenchment: Defining the Problem and Looking for Solutions. Journal of
- Comparative Policy Analysis 6(1): 3-14. Plümper, T., Troeger, V.E. and Manow, P. (2005) Panel
data analysis in comparative politics: Linking method to theory. European Journal of Political
Research 44(2): 327-354.

1. Arts & Gelissen (2002) Three Worlds of Welfare Capitalism or More?

- Aim/question: Esping-Andersen defined three types of welfare regime (conservative, liberal
and social-democratic) based on degree of commodification and kind of social stratification
and solidarities. The article aims to find out if there are more than three ideal types.
- Main argument/hypothesis: three factors are missing, the Mediterranean states, the
antipodes and a gender-dimension /late female mobilization framework.
- Data/methods: Comparative research into ideal-typical welfare states. Overview of the
debate during the last decade (1990s) and review the state of the art typifying welfare states
at the turn of the millennium.
- Outcome(s): From the literature, five broad types of welfare regimes emerge: the three of
Esping-Andersen plus a category of Mediterranean states and a category for the Antipodes
(Australia + NZ). Real welfare states are hardly ever pure types and are usually hybrid cases,
meaning that most countries cannot be put into (just) one of the categories. Inconclusive
outcomes.

Criticism: Typologies as such have no explanatory power and, therefore, the work of Esping-
Anderson does not contribute to proper theorizing about what is happening with and within welfare
states. Theoretical/empirical value? Only if they are means to represent reality.

Definition welfare state regime: “The institutional arrangements, rules and understandings that
guide and shape concurrent social policy decisions, expenditure developments, problem definitions,
and even the respond-and-demand structure of citizens and welfare consumers”.

According to Esping-Andersen, historical and political factors behind regime differences:

- The nature of class mobilization (especially of the working class)
- Class-political action structures
- Historical legacy of regime institutionalization

Three models/ideal-types:
- Conservative-corporatists
legacy of Catholic social policy,2 on the one side, and corporatism and etatism on the other
side
 moderate level of decommodification, married woman are discouraged to work (regime is
influenced by the Church; preserve traditional family structures)
 principle of subsidiarity: the state will only interfere when the family’s capacity to service
its members is exhausted.

2
  Direct influence of the state is restricted to the provision of income maintenance benefits
related to occupational status: narrow solidarity.

- Liberal
 individualism and the primacy of the market, little redistribution of incomes, limited realm
of social benefits, low level of decommodification, division in the population, women are
encouraged to participate in the labour force

- Social-democratic
 high level of decommodification, stratification
 generous universal and highly distributive benefits not dependent on any individual
contributions, crowds out the market, universal solidarity in favour of the welfare state,
woman are encouraged to work, focus of individual independence, full employment
 to maintain a high-level welfare system

Mediterranean
- Esping-Andersen did not systematically include the Mediterranean countries
- Subcategory: welfare states in their infancy according to Esping-Andersen. However, other
commentators see the South European countries as a separate cluster (southern model of
social policy).
- minimum social protection, but some benefit levels are very generous

Antipodes
- According to Esping-Andersen, the Antipodes countries are representatives of the liberal
welfare state regime (marginal commitment to public welfare and strong reliance on means
testing: a payment available to people who can demonstrate that their income and capital
(their 'means') are below specified limits.). However, according to Castles (1998), these
countries have a more particular and a more inclusive approach to social protection than the
standard liberal form.
- ↓ It seems that the Antipodean countries represent a separate social policy model
Gender, familialism and late female mobilization
- Several authors have tried to reconceptualise the dimensions of welfare state variation by
incorporating gender  Esping-Andersen misses whole areas of social policy.
- Missing: a systematic discussion of the family’s place in the provision of welfare and care +
no serious treatment of the degree to which women are excluded from or included in the
labour market.
- Late Female Mobilization welfare state regime  addition to Esping-Andersen’s
classification (resembles the Mediterranean welfare states)

3
2. Starke (2006) The Politics of Welfare State Retrenchment: A Literature Review.
Welfare state retrenchment = cutbacks in the welfare state. Interest sparked again for this concept
since Thatcher and Reagan cut back in programs as they saw the welfare state as the source of
economic and social problems.
Welfare state = no shared definition. Dependent variable problem: social expenditures as GDP vs real
cutbacks or expansion
Resilience thesis = welfare state is there to stay. Becomes clear from Pierson’s analysis. 2 central
considerations: popular support (path dependency) + voter preference/politicians’ office seeking
Golden Age = of social policy, expansion of the welfare state during post-war prosperity

- Aim/question: Why and under what circumstances do cutbacks take place in highly
developed welfare states despite formidable political obstacles that support welfare state
resilience? This article reviews the literature on the politics of retrenchment, namely on the
impact of socio-economic problem pressure, political parties, political institutions, welfare
state structures and ideas.
- Main argument/hypothesis: Retrenchment is unpopular, making it an electoral risks for
elected politicians who are not just policy-seeking, but also office-seeking. And for voters
who react more strongly to losses (benefits cutbacks) than to gains (lower taxes). And losses
concentrated on few but organized beneficiaries, while gains are widely dispersed.
Several theories try to explain why cutbacks still take place, but they are hindered by some
persistent issues: the dependent variable problem (what is it/how do you measure it), actors’
orientations and motives and the means of retrenchment (blame avoidance).
- Data/methods: Overview of the literature on welfare state development (neo-functionalism,
conflict theories, institutionalism and the role of ideas). According to Pierson, theories about
the ‘old politics’ of expansion, including socio-economic functionalism and class-based
power resources theory, fail to account for developments after the end of the ‘Golden Age’.
- Outcome(s): There has been progress in the study of retrenchment. There is a great
awareness of the possibilities and limitations of measuring retrenchment and there is a
consensus on the powerful causal mechanism of ‘blame avoidance’ as a means for
retrenchment. But, the study of retrenchment will remain pluralistic, both in theoretical and
in empirical terms. A great awareness of the possibilities and limitations of measuring
retrenchment.

Why and under what circumstances do cutbacks take place in highly developed welfare states,
despite these formidable political obstacles?
- Socio-economic problems > austerity > cutbacks
- Institutions & Path dependency: The extent of retrenchment possible depends on specific
institutional configuration of a political system and the path dependence of existing welfare
state structures.
- Pierson: politics of blame avoidance

Neo-functionalism
- Argument: economic globalization leaves only little room for a comprehensive welfare state;
deep economic integration among countries will require a change in the role of the state in
pursuing social protection.
- Retrenchment appears as the only possible solution unless the process of economic
integration is reversed.

Conflict theories
- Social policy is not – or at least, not only – the result of socio-economic shifts but rather of
political struggles about distributive decision.

4
 - Pierson: under the conditions of the ‘new politics of the welfare state’ the explanatory
relevance of parties and left power resources had faded  a lot of authors disagreed

Institutionalism
- Political institutions vs welfare state institutions (social provisions)
- In a system with a high degree of vertical and horizontal fragmentation of power, such as the
United States, large-scale retrenchment should be less likely, due to the higher number of
veto players – i.e. Individual or collective actors whose agreement is necessary for a change
of the status quo (e.g. a strong upper house).
- However: concentrated powers make blame avoidance difficult + no evidence
- Institutional welfare state arrangements already in place today are often depicted as highly
path-dependent

The role of ideas

- Policy ideas: specific policy alternatives (for example, personal savings account) as well as the
organized principles and causal beliefs in which these proposals are embedded (e.g. neo-
liberalism).
- Agenda-setting + framing + policy leaning

The Means of Retrenchment

A number of means have been added to Pierson’s initial three strategies of blame avoidance/success
of retrenchment initiatives (i.e. obfuscation, division and compensation):
- The importance of framing
- The justification of cutbacks
- Targeting cuts to more privileged constituencies
- Combining reductions with a ‘progressive’ restructuring of elements of the welfare state
already considered inefficient
- Tripartite social pacts

3. Green-Pedersen (2004) The Dependent Variable Problem within the Study of

Welfare State Retrenchment: Defining the Problem and Looking for Solutions
- Aim/question: This article addresses an aspect of the retrenchment debate sparked off by
Pierson’s work, namely what can be labelled the ‘dependent variable problem’.
Dependent variable problem: defining the object of the entire retrenchment debate. what is
welfare state (retrenchment) (definition) and how can it be measured.
- Main argument/hypothesis: problem of theoretical conceptualization rather than a problem
of data. Different theoretical perspectives on retrenchment should lead to different
conceptualizations of retrenchment. What should the theoretical definition of retrenchment
be? How can retrenchment actually be measured? There are two broad ways of looking at
retrenchment: as cutbacks or as institutional change.
- Data/methods: Overview of the literature.
- Outcome(s): Whether retrenchment is defined as cutbacks or institutional change matters,
but how much depends on which institutional traits are in focus. A significant cutback in
entitlements is possible without changing institutional traits and vice versa. This does not
mean that there is no relation. Cutbacks and institutional change are also not unrelated.
The main solution to the dependent variable problem is to be very clear-cut about one’s
theoretical perspective and research question

5
Esping-Anderson: degree of decommodification
Most of the research into the growth of the welfare state was based on aggregate expenditure data.
Esping-Andersen (1990) challenged this by basing his analysis of the three worlds of welfare
capitalism on, among other things, a measurement of the degree of decommodification resulting
from different social security schemes. The expansion of the welfare state should be analyzed as a
question of higher levels of decommodification, not as a question of higher levels of social security
expenditure. Esping-Andersen underscored that the question about defining the dependent variable
cannot be detached from a theoretical perspective and a specific research question

Welfare state
No definition of the welfare state is a priori better or worse than others. However, when one moves
away from the mainstream policy definition of the welfare state either by including certain outcomes
or other policy areas, one changes focus considerably.

1. Retrenchment as Cutbacks
Different ways to operationalize cutbacks:
1. Expenditures (outcome measure)
- Advantage: no problem in using social security expenditure as the operational definition of
retrenchment
- Disadvantages: practical problems with expenditure data
- Expenditure data = outcome measures and other factors intervene between political
decisions and actual outcomes such as expenditures
- Expenditures on, for instance, unemployment benefits can rise due to more
unemployed without any changes to legislation
- The ‘‘time-lag’’ problem = many retrenchments are designed to have gradual rather
than immediate effects. Consequently, many enacted retrenchments are not yet
visible in expenditures

2. Average replacement rates (outcome measure)

- Advantage: not affected by, for instance, changes in the level of unemployment
- Disadvantage: still time-lag problem.

3. Output measures
- index measuring changes made in social security in relation to the level of benefits, eligibility
criteria and so forth >>> using microdata
- Advantages: minimization of the time-lag problem + microdata more whole explanations
- Disadvantages: validity problems when calculating expected budgetary effects + time
consuming

2. Retrenchment as Institutional Change

Retrenchment as cutbacks is too narrow because it does not measure changes to the content or
institutional structure of welfare schemes. (Pierson)
Advantage: more general than retrenchment as cutbacks. Maybe cutbacks a type of institutional
change.
Disadvantage: not so straightforward operational definition of retrenchment conceptualized as
institutional change. Qualitative assessments dominant but fuzzy judgement criticism.

6
4. Plümber, Troeger & Manow (2005) Panel data analysis in comparative politics:
Linking method to theory
- Aim/question:
- Main argument/hypothesis: Address the problems of panel data analysis with a lagged
dependent variable and period and unit dummies. There are four potential sources of
problems:
o Absorption of cross-sectional variance by unit dummies
o Absorption of time-series variance by the lagged dependent variable and period
dummies
o Mis-specification of the lag structure
o Neglect of parameter slope heterogeneity
- Data/methods: different methodological approach
- Outcomes: we show that partisan politics and socioeconomic factors such as aging and
unemployment as expected by theorists have a strong impact on the timeseries and cross-
sectional variance in government spending

Panel data: observations of multiple phenomena obtained over multiple periods for the same units.

Conclusion
1. unit fixed effects turn out to be problematic if variables are time invariant or if the theory at
test predicts level effects
2. the inclusion of a lagged dependent variable and/or period dummies tends not only to
absorb large parts of the trend in the dependent variable, but likely biases estimates
3. simply assuming a uniform lag structure may cause biased estimates and wrong inference.
4. if the time dimension in panel analyses exceeds a rather limited number of time periods, it
becomes extremely important to think about and test for structural changes in slopes and
error variance

Our findings support a number of hypotheses:

1. partisan effects matter. However, party preferences’ influence on government spending is
not stable over time
2. Second, our results say that unemployment and the aging of the society tend to put an
upward pressure on government budgets, while growth reduces the government share of
the economy.
3. Third, international economic openness does not seem to have a similarly important
influence on government spending

7
WK 2: Political parties, corporatism and socio-economic reforms

- Pierson, P. (1996) The New Politics of the Welfare State. World Politics 48(2): 143-179.
- Allan, J.P. and Scruggs, L. (2004) Political Partisanship and Welfare State Reform in Advanced
Industrial Societies. American Journal of Political Science 48(3): 496-512.
- Rueda, D. (2006) Social Democracy and Active Labour Market Policies: Insiders, Outsiders and
the Politics of Employment Protection. British Journal of Political Science 36(3): 385-406.
- Emmenegger, P. (2009) Barriers to entry: insider/outsider politics and the political
determinants of jobs security regulations. Journal of European Social Policy 19(2): 131-146.

1. Pierson (1996) The New Politics of the Welfare State

- Aim/question: This essay seeks to lay the foundations for an analysis of welfare state
retrenchment with an emphasis on critical constraints on reform resulting from the role of
supportive interest groups and, ultimately, the voters.
- Main argument/hypothesis: old (pre-1980) vs new (post-1980) politics of the welfare state
Classical argument: left > labour working class > welfare expansion while right > higher class
> don’t want taxes > retrenchment. However, according to Pierson this was only relevant in
the context of the era before 1980s: welfare state growth.
Welfare expansion involved implementing popular policies in a time of relatively
underdeveloped interest groups. After 1980s this is not relevant anymore, it is not about
classes so much. Welfare retrenchment is unpopular and happens in a time of highly
developed interest groups. Because of this difference, variables used to understand one
cannot help to understand the other. Welfare state retrenchment is not the mirror of welfare
state expansion. This shift has created ‘new politics’ marked by blame avoidance. Pierson:
how can welfare retrenchment take place? Blame avoidance: compensation, obfuscation or
division.
- Data/methods: A review of the evolution of welfare states in four affluent democracies
(Britain, Germany, Sweden, US) since the 1970s.
- Variables:
o Retrenchment: a combinations of quantitative data on expenditures and qualitative
analysis of welfare state reforms.
- Outcome(s):
o Welfare state is resilient based on path dependencies
 they create their own programs. Each group of recipients will organize
themselves in interest groups. They will become actors in politics and try to
prevent government of cuts.
 Long-term commitments that have been established and cannot be changed
easily
 left right does not matter much anymore
o To explain the apparent lack of wide-ranging retrenchment – in other words, of
welfare state ‘resilience’ – he develops his theory of the ‘new politics of the welfare
state’. Paul Pierson’s claim that today’s welfare state politics are fundamentally
different from earlier versions of the same

o There is little evidence for broad propositions about the centrality of strong states or
left power resources to retrenchment outcomes.
o The unpopularity of left power resources makes major cutbacks unlikely except
under conditions of budgetary crisis.
o Governments generally seek to negotiate consensus packages rather than impose
reforms unilaterally.
o Cutbacks tend to replenish support for the welfare state.

8
Two important contributions by Pierson:
- Path dependency
- New politics of the welfare state (

2. Allan & Scrugss (2004). Political partisanship and welfare state reform in advanced
industrial societies
- Aim: investigate welfare retrenchment using program specific development and look at the role
of partisanship. Using empirical data.
- Main argument: Investigate 2 claims
Claim 1: Welfare states have remained quite resilient in the face of demands for
retrenchment
Claim 2: Partisan politics have ceased to play a decisive role in their evolution
 Found more evidence of welfare retrenchment during the last 2 decades than to recent
cross-national studies
 Traditional partisanship continues to have a considerable effect on welfare state
entitlements in the era of retrenchment
- Data/methods
o Allan & Scruggs argue that expenditure data is not appropriate to capture welfare state
change:
 Changes in the structure of the dependent population can overwhelm (and hence mask)
real cuts at the individual level
 Differences in economic growth rates
 Differences in tax treatment of transfers distort degree to which social spending
translates into disposable income for program recipients
o Replacement rates for programs: unemployment & sickness insurance
 Evidence for retrenchment
 Some convergence in replacement rates across countries
 contrasts with the recent emphasis on path dependence in comparative welfare state
programs
 consistent with income inequality trends
 However, the existing data has several major shortcomings, particularly with regard to
evaluating partisan effects

o Does partisanship still matters? Yes,

 measured by right/left-party strength in government, usually the share of cabinet seats.
 Partisanship: before 1980, left governments are associated with larger, statistically
significant increases in replacement rates. However, since the recessions of the early
1980s, lev ft governments are not significantly associated with increases in replacement
rates. For right governments, we see the mirror image of this effect
 Partisanship does still matter critically for retrenchment: new constraints (globalisation,
unemployment) determining spending priorities and these are correlated with
partisanship and the size of the welfare state. government by parties of the neo-liberal
right since the 1980s (or ever since the mid-1970s) has tended to result in greater
retrenchment

- Results
o Retrenchment is not simply the mirror image of welfare state expansion
o More evidence for retrenchment than Pierson
o Left/right partisanship still matters

9
- Conclusion: shifting the focus from expenditure measures to data that better reflect the
consequences of welfare state policies. Before this article: everybody used spending, they were
one the first to use program specific development. Program specific spending. Taxation: after
taxes benefits, net benefits. Problem: difference in countries. Some countries unemployment
benefits are taxed. If other countries don’t have that, it comes incomparable. Look at after tax
benefits.

Critique
- Problem with net benefits, differences in countries can make it incomparable. In some countries
employment benefits are taxed and in others not. Therefore look at after tax benefits.
- Dependent variable problem: some things still not included
- To some extent it indeed does reflect policy change if the policy wants lower benefits. But if the
policy has other goals: duration or something else it can’t measure this. Eligibility criteria. Still not
included in this indicator.

How to operationalize the welfare state

- Many studies use aggregate spending data as a proxy for the size of the welfare state
-

Why express in terms of percentages of GDP:

- 1) Sizes of economies differ
- 2) inflation and 3) exchange rates are taken out of the equation

Disadvantages:
- Does not tell you anything about the policy changes
- Spending is driven by the number of benefits to recipients

3. Rueda (2006). Social Democracy and Active Labour Market Policies: Insiders,
Outsiders and the Politics of Employment Protection
Active labour-market policies = supply-side policies that can be used to promote employment,
growth and equality in an environment characterized by increasing levels of internationalization

 Aim: explaining the party strategy changes that result rom new voter demands and political-
economic conditions.
 Main argument: Classic argument: left spend more on ALM. We should also take into
account the organization of the labour market = insiders/outsiders (many definitions of this,
but basic: insiders people who have a job and long-term contract, outsiders are all other
people. Important role in recent labour market debates).
However, Rueda: left wing parties (social-democratic governments) are not always in favour
of active labour market policies.
o Not always in interests of their electorate. Within labour, outsiders tend to be less
politically active and electorally relevant than insiders. When faced with the choice
between insiders and outsiders, social democrats will side their core constituency
with the outsiders.
o ALMPS mainly benefit outsiders. Insiders are apposed by ALMPS. Labour is
disproportionately affected by unemployment. If you increase spending on ALM
more unemployed people will enter the labour market, supply of labour will
increase. Than wages will go down. Also the raising of ALMPs will increase the tax
burdens of insiders. Because of these two arguments the insiders don’t want to
spend more on ALM. Insider opposition: social democratic government will not be
associated

10
 o Why would insiders want then want ALMPS, and why would left-wing parties do this?
When insider/outsider interest become more similar. When insiders become more
vulnerable: unemployment rate (instability) goes up, or protection is less strict, it is
also in the interest of the insiders to have more ALM. Then social democratic
governments increase employment promotion policies to satisfy insiders.
 Data/methods: If Employment protection goes up, the support for ALM goes down. When
this happens, the preference of left-wing parties will change. Cabinet Partisanship X
Employment, this gives an interaction effect. If something changes in employment
protection, what will happen than to the preference of the parties?
an additional factor that may moderate (or exacerbate) the influence of insider-outsider
differences on partisan strategies. It is the relationship between unions and social democratic
 Results: labour is divided into those with secure employment (insiders) and those without
(outsiders) and that the electoral goals of social democratic parties are sometimes best
served by pursuing labour-market policies that benefit insiders while ignoring the interests of
outsiders.
o Whether a government is social democratic or conservative makes no difference to
the levels of ALMPs promoted
o Unemployment level does not affect AMPs
o Unemployment growth positively associated with ALMps
o Negative association between employment protection and ALMPs
 The interaction between unemployment growth and social democratic
government positively associated: rise in vulnerability of insiders, makes
governments more likely to promote ALMPS
 Critique: someone who pointed out some weak points was Emmenegger. Three points of
critique. General point: assumptions are too demanding. Or people don’t vote only for the
labour market items. Insiders/outsider differences are overestimated.
Empirical part: you could argue that the formulated questions are not specific enough.

4. Emmenegger (2009) Barriers to entry: insider/outsider politics and the political

determinants of jobs security regulations
Job security regulations = restrictions on hiring and firing

 Aim/question: examine the insider/outsider theory

 Main argument/hypothesis: insider/outsider theory argues that job security regulations
benefit the labour market insiders. Insiders will fight all reforms and are supported by social
democratic parties which only represents the insiders. Insiders disproportionally vote for
Social Democratic parties, which, in collaboration with trade unions, push for more
regulations
Implications insider/outsider theory:
o expects insiders to emphasize job security regulations to a greater extent than any
other group in the labour market
o insiders vote disproportionately for social democratic parties compared to outsiders
o Social Democratic parties will represent the interests of their core
constituency(insiders) and promote policies that increase the level of job security

According to Emmenegger this is wrong: no use in explaining differences in preferences for

job security regulations. There are three reasons for this:
o suffers from too demanding rationality assumptions
o Misconceptualizes electoral politics and the role of policy packages offered by parties
in election campaigns. People don’t vote only for the labour market items.

11
 o overestimates the differences between insiders and outsiders. Outsiders can be just
as supportive for restrictive job regulations
 long-term expectations, household relationships and the power balance
between capital and labour, etc.
 Power balance argument: Deregulation increases the power resources of
‘upscales’ and self-employed people (whom he refers to as capitalists), as
opposed to labour market insiders and outsiders. Why ‘upscales’ and self-
employed people should prefer less job security regulations than labour
market insiders. However, it is less clear why this should also be the case for
labour market outsiders.. These reforms, however, are neither beneficial for
the insiders nor for the outsiders.
 Data/methods: don’t neglect households.
 Results: the empirical evidence for the insider/outsider theory of employment and
unemployment is mixed. Compared to labour market insiders, temporarily, part-time
employed, ‘upscale’ and, especially, self-employed respondents are rather critical of job
security regulations. However, at odds with the theoretical expectations, unemployed
respondents are very supportive of job security regulations. part-time and temporarily
employed persons may be less supportive of job security regulations. But this has less to do
with labour market status than with intra-household relationships and work preferences
Voting preferences: labour market insiders and outsiders will disproportionately vote for Left
parties as opposed to ‘upscale’ employees and self-employed people. Labour market insiders
do not more often support Left parties than labour market outsiders. In contrast,
unemployed people seem to be the most reliable supporters of Left parties.
 Outcome(s): Insiders can be expected to be equally supportive of job security regulations
and social democratic parties as outsiders. Therefore, deregulatory reforms should not be
designed with the idea in mind that job security regulations are supported by labour market
insiders while they are rejected by labour market outsiders.

12
WK3 Globalization I
- Rodrik, D. (1998) Why Do Open Economies Have Bigger Governments? Journal of Political
Economy 106(5): 997-1032.
- Burgoon, B. (2001) Globalization and Welfare Compensation: Disentangling the Ties that
Bind. International Organization 55(3): 509-551.
- Swank, D. and S. Steinmo (2002) The New Political Economy of Taxation in Advanced
Capitalist Democracies. American Journal of Political Science 46(3): 642-655.
- Genschel, P., A. Kemmerling and E. Seils (2011) Accelerating Downhill: How the EU Shapes
Corporate Tax Competition in the Single Market. Journal of Common Market Studies 49(3):
585-606.

1. Rodrik (1998) Why Do More Open Economies Have Bigger Governments?

 Aim/question: positive relationship between trade exposure (openness) and the scope of
government, because government expenditures are used to provide social insurance against
external risk. A negative correlation is expected, but this does not seem to be true. There is a
positive and robust partial correlation between openness, as measured by the share of trade
in GPD, and the scope of government, as measured by the share of government expenditure
in GDP.
 Main argument/hypothesis: in line with the compensation hypothesis (increased demands
because of globalisation). Positive correlation between an economy’s exposure to
international trade and the size of its government. One explanation is that government
spending plays a risk-reducing role in economies exposed to a significant amount of external
risk. The relationship between openness and government size is strongest when terms-of-
trade risk is highest. The paper provides a range of evidence consistent with this hypothesis.
Why does this relationship exists?
o Economies subject to greater amounts of external risk necessarily experience more
risk in total income as well > household.
o The government sector is ‘‘safe’’ in the sense that an expansion in it would reduce
aggregate income risk for households.
 Dependent variables: Main dependent variable is government consumption. Government
size is measured by the share of government expenditures in GDP. Most relevant indicator
would be welfare spending, Rodrik is aware of this and uses it in a way.
 Independent variable: openness and diversification. For openness he looks at
imports/exports and their share as percentage of GDP. But enormous argument that this is
not sufficient: terms of trade. Openness matters to the scope of government because of the
role played by external risk. Government risk-mitigating role: Governments consume a larger
share of domestic output in economies subject to greater amounts of external risk. Once
external risk is controlled for, openness does not seem to exert an independent effect on
government consumption.Because of globalisation you are exposed to a broad market:
volatile (insecurity), more volatility in consumption. So economic risk should be combined
with openness.
The other indicator he uses is concentration of production circles. What matters is not the
global stability, but the stability of the stream of earnings from domestic production. With
concentration you are more vulnerable if something happens on the market. More
diversified production/economy gives you a smaller risk. > interaction term between
openness (terms of trade) and diversification.
 Data/methods: include not only OECD but also developing countries. In developing countries
there is not a (great) welfare state. In these countries there can be compensation through

13
 creating more public jobs. But Rodrik admits that for OECD countries welfare spending would
have been best
 Outcome(s): positive and significant effects, except for social security and welfare spending.
Spending on social security and welfare is significantly more sensitive to exposure to external
risk than government consumption, which is consistent with our theory
a small (permanent) increase in government consumption (as a share of GDP) would result in
more stable incomes in the overwhelming majority of countries.
 Conclusions/implications: Public spending is a risk-reducing instrument on which there is
greater reliance in more open economies.
o More difficult to disentangle the relationship between government size and
openness, on the one hand, and economic growth, on the other
o Same bias would exist in a regression of growth on trade
o there may be a degree of complementarity between markets and governments
 Governments have expanded fastest in the most open economies, because
governments appear to have sought to mitigate the exposure to risk by
increasing the share of domestic output they consume.
o International trade has expanded significantly during the post-war period, and so has
the scope of government activity in most countries of the world  no coincidence.
o Scaling governments down without paying attention to the economic insecurities
generated by globalization may actually harm the prospects of maintaining free
trade.

2. Burgoon (2001) Globalization and Welfare Compensation: Disentangling the Ties

that Bind
 Aim/question: Many lines of research follow the compensation hypothesis. All perspectives,
however, overlook important details about the nature of openness, welfare efforts, and the
politics connecting them. Greater economic openness should constrain some elements of the
welfare state, spur others, and leave still others unaffected.
 Main argument/hypothesis: twofold argument.
Development: it matters if import competition comes from developed countries or
developing countries. Developing countries are more vulnerable/low wages, employees in
developed countries face tougher competition. This competition will spark more
concentrated demands for compensation and similar/less concentrated opposition to such
compensation.
o low wage openness ought to inspire more welfare expansion or less contraction than
general openness.
Groups at risk. The second argument states that groups put at risk by greater openness will
demand more public spending generally but will tend to focus most on active and passive
labor market policies, less on health, and less still on retirement and family benefits geared
to the elderly and youth dependents.
o Groups put at risk by greater openness demand more public spending generally,
specifically on active and passive labor market policies, less on health care, less
retirement and family benefits for elderly and young dependents. From the side of
the employees: argument trade-unions and employees organisation. Not just that
they ask for more spending, not on all programs: higher spending unemployment
insurance and job training. But because of globalisation they don’t want more health
care benefits.
o Investors/producers and exposed will respond to greater openness by opposing most
expansions of public economy. Higher spending means higher taxes, position is
opposed spending welfare. But lower opposition against programs that are leading

14
 to higher productivity, programs that favour labor training and relocation policies. Do
not favour family and retirement programs, and strongly oppose passive labor-
market policies.
 If we combine both sides, hypothesis would be higher spending on job-training and
relocation assistance.
 Dependent variable: He uses government consumption, spending on the welfare state, but
problem of dependent variable. He uses broad range of welfare programs. Also theoretical
argument: just looking at the correlation of some measure of globalisation and some
measure of welfare spending is too simplistic. Result is decision making in politics. He briefly
discusses the argument of Pierson (WK2).
 Independent variable: it isn’t just about import/export, but distinction between countries of
origin. It matters whether imports come from developed countries/developing countries.
Because developing countries are more vulnerable. Employees in domestic economy are
exposed to tougher competition if you compete with low wage countries
 Data/methods: Author distinguishes between general trade openness and low-wage trade,
and among government programs that more or less directly and immediately aid vulnerable
groups and that are more or less costly to exposed producers and investors.
 Four different patterns of compensation politics. Distinguish between general trade
openness and low-wage trade, and among government programs that more or less directly
and immediately aid vulnerable groups and that are more or less costly to exposed producers
and investors.
1. For welfare programs subject to strong compensation demands and investor
support or acceptance openness should inspire harmonious, one-sided politics that
expand welfare effort.
2. For welfare elements diffusely connected to compensation demands and fostering
high investor concern openness should inspire different one-sided politics that
retrench welfare.
3. For programs subject to strong compensation demands and high investor hostility
openness should spark more combative politics whose outcomes depends on
exogenous political forces like the power of left parties.
4. For welfare elements remote from both compensation demands and investor fears
greater openness should spurt few political struggles and little change.

15
  Outcome(s): different elements of openness and welfare may be in tension for some kinds of
openness and some segments of welfare, in harmony for others, and wholly independent for
still others.
o Suggests that openness encourages bigger changes in some "worlds" of welfare
capitalism than in others.
o the evidence shows that openness has a slight effect on welfare outcomes and that it
is far from the most important determinant of welfare efforts in OECD countries. But
the links between openness and welfare are meaningful and broadly consistent with
a soft version of the arguments
o the relationship between openness and welfare is more predictably divers than
existing scholarship has led us to believe

3. Swank & Steinmo (2002) The New Political Economy of Taxation in Advanced
Capitalist Democracies
 Aim/question: tries to test the efficiency hypothesis. Argue that internationalization,
domestic economic change, and budgetary pressures each prompt significant changes in tax
policy; yet, together, they create a system of constraints on altering the level and distribution
of tax burden
 Dependent variable: taxation, tax rates. Story of the part of efficiency hypothesis. Tax
competition, a race to the bottom because of globalisation. They analyse a number of tax
rates, corporate income, consumption and from labour.
 Main argument/hypothesis: The general idea/argument: because of the tax competition and
because capital is relatively mobile: tax rates will go down. Things that are less mobile like
consumption/labour their tax rates would go up, in order to compensate losses in tax
revenues on corporate income. Tax competition on tax rates, go down, but tax bases are
broadened to compensate for this. Tax base has been broadened because this is much less
transparent: complicated (tax laws, all kinds of deductions). So not clear what happens with
tax base, but in many countries the tax base has been broadened.
 Data/methods: a remarkable stability in the levels and distributions of tax burdens. They
argue that the tax impacts of internationalization are important, but in more complex ways
than globalization theory suggests. Three factors – internationalization, domestic economic
change, and budgetary forces – simultaneously constrain changes in tax burdens and,
together, help explain the complexity of tax policy outcomes.
The majority papers on tax papers is on corporate income, and they add consumption and
labour. 1981 to 1995 data from fourteen developed democracies to analyze the
determinants of taxation.
 Outcome(s): we argue that the tax impacts of internationalization are important, but in more
complex ways than globalization theory suggests. Three factors: internationalization,
domestic economic change, and budgetary forces simultaneously constrain changes in tax
burdens and, together, help explain the complexity of tax policy outcomes.
o Negative effects of globalisation on taxes. Capital mobility and trade are associated
with cuts in statutory corporate tax rates.
o Positive effect of tax rate on consumption. but not with reductions in effective
average tax rates on capital income.
o No positive effect of taxes on labour income, because reduce labour taxes in order to
stimulate the labour market (and not unemployment). Capital mobility is negatively
associated with the tax components of labor costs. Domestically structural
unemployment leads to reductions in labor and capital taxes, while public sector
debt and societal needs raise taxes.

16
 Capital mobility has not led—and is not likely to lead—to a “race the bottom” or the
evisceration of the revenue-raising capacity of the state: governments can (and do) pursue
moderately extensive social protection and public goods provision when they and their
electorates so choose. Overall, policy makers in contemporary democratic polities have
faced intensifying pressure to reform tax policy to promote economic efficiency. They have,
however, little room left for manoeuvre.

4. Genschel et al. (2011) Accelerating Downhill: How the EU Shapes Corporate Tax
Competition in the Single Market
 Aim/question: quantitative evidence to suggest that tax competition is stronger in the EU
than in the rest of the world, and explores qualitatively why tax coordination and tax
jurisprudence have failed to prevent a race to the bottom in tax rates
 Main argument/hypothesis: tax competition in the EU is shaped by 4 partly opposed
institutional mechanisms with a net effect to accelerate tax competition.
o Market integration > increase competitive pressures
o Enlargement > increase competitive pressures
o Tax co-ordination > decreases competitive pressures
o ECJ’s tax jurisprudence > may increase/decrease competitive pressures
There are two intersecting competitive effects:
o within group competition between EU15 states: integration effect
o Between groups competition, like with Asia.
 Data/methods: quantitative analysis
Outcome(s):
o Statutory taxes have fallen faster in the EU than in the rest of the world since 1990s.
Evidence that the incidence of preferential tax regimes is high in the EU. Finding for
effective tax rates on labor income and consumption are consistent with
expectations. Internationalization has not resulted in a shift of the tax burden to
labor and consumption, but evidence suggests that international capital mobility has
sensitized policy makers to potential economic costs of high labor taxes. Domestic
economic and budgetary forces play significant roles in tax policy change, especially
in the case of labor taxation.
o General tax rate competition is more pronounced in the single market (EU) than in
other parts of the world. While the integration effect of one market (single market
programme), one currency (monetary unification) and one law as well as the
enlargement effect of the recent accession of eastern countries have both given a
boost to corporate tax competition since the 1990s, the co-ordination effect of
corporate tax harmonization and the judicialization effect of the tax jurisprudence of
the ECJ have done little to counter it.
Concerns about efficiency and revenues have seemingly eclipsed the goal of redistribution
through remarkedly steeply progressive rates. But with stability in levels and distribution of
tax burdens.
 Conclusion: internationalization mattered in much more complex ways. This article shows
that corporate tax competition in the EU is both different and stronger than the rest of the
world.
 Predictions: future depends on development of the 4 institutional effects.
o The integration effect is unlikely to relax very much.
o the enlargement effect also seems unlikely to relax. Eastern enlargement was the
EU’s only major success in recent years. It is unpopular in some old Member States,
but the foreign policy rationale for further enlargement is strong. Poor countries up

17
 for members: once admitted to the EU, the incentive for them to engage in
aggressive tax competition would be very high.
o The co-ordination effect is difficult to predict.
o The judicialization effect also difficult to predict. A lot depends on whether the
recent (moderate) softening of the ECJ’s corporate tax jurisprudence marks just a
temporary aberration from its earlier more activist case law, or a permanent change
towards more judicial self-restraint
 Critique
o Inadequate comparison of the EU with the rest of the world. EU is very homogenous
and a specific area: specific case
o Yes based on the descriptives you could have seen that there is a trend going down
in the tax rates. You wouldn’t need regression to show that, but one big if. It could
have been the case that the downward trend in the EU was changed by something
else. Advantage of regression analysis is that you can control for these other effects.
But they don’t do that actually.
o Swank and Steinmo have a more advanced model taking in more effects. Genschel
only two times two years, small number of observations and without any control
variables.
o Exclude Luxembourg and Malta. Yes small countries, but they could have included
them. But they have an argument about small countries/large countries and then
they exclude them.

18
WK4 Globalization 2
 Hays, J.C. , S.D. Ehrlich and C. Peinhardt (2005) Government Spending and Public Support for
Trade in the OECD: An Empirical Test of the Embedded Liberalism Thesis. International
Organization 59(2): 473-494.
 Iversen, T. and Cusack, T.R. (2000) The Causes of Welfare State Expansion: Deindustrialization or
Globalization? World Politics 52(3): 313-349.
 Yong Kwon, H. and J. Pontusson (2010) Globalization, labour power and partisan politics
revisited. Socio-Economic Review 8(2): 251-281.
 Walter, S. (2010) Globalization and the Welfare State: Testing the Microfoundations of the
Compensations Hypothesis. International Studies Quarterly 54(2): 403-426.

1. Hays, Ehrlich & Peinhardt - Government Spending and Public Support for Trade in
the OECD: An Empirical Test of the Embedded Liberalism Thesis
Liberal thesis = universal expectation held by citizens in the developed democracies that their
governments will limit the costs and distribute the benefits of open markets through some kind of
government intervention and spending, and that public support for liberalism depends on the
willingness and ability of governments to do this successfully. This implies that governments
committed to free trade provide insurance and other transfers to compensate those who lose
economically from expanded trade.
Underlying idea of compensation is solidarity. In the embedded liberal thesis it is a compensation in
order to maintain political support for international trade. If you would not compensate them, they
would vote for parties that want to close the borders. Fewer economic gains is not in the interest of
the globalisation winners, so the winners want to redistribute to keep them on board. That is the
argument to the paper.

 Aim/question: variation on the compensation hypothesis: embedded liberal thesis: social

security as a tool to increase/not lose political support for economic openness. They find
empirical support for the liberal thesis. Focus on supply side: what governments can do to
maintain support.
 Main argument/hypothesis: Liberal hypothesis: employees that are heart by globalisation.
You should look at sectors that are heart, that is the manufacturing industry. If you have a
large manufacturing industry and high levels of import, than you should expect more
compensation. Then you should expect higher spending. Negative coefficient, because it
industrialisation = 100 - the manufacturing sector. Positive effect if you look at it the other
way around. The industry.
 Independent variable: key is economic openness. They look at free-trade. The abundant
factors are highly skilled labor and capital.
 Data/methods: using micro data (asking respondents) to test the effect of imports versus
effects of exports (was quite unique). If people are hurt by trade, they are hurt by imports. If
export increase it means more gains. Why would you take the sum of imports and exports of
GDP? What they do is they use the interaction term of imports and deindustrialization. Not
just the imports, but the interaction. interaction effect: imports x industrialisation
Deindustrialization of Iverson and Gusack. Used as proxy to test tradables vs non-tradeables.
 Outcome(s):.
o the effect of imports on a country’s level of government spending depends on how
exposed the domestic labor market is to free trade. An increase in a country’s
imports is associated with an increase in government spending. All controls are
statistically significant

19
 o effects of increasing imports on government spending. The results are similar for all
three types of government spending. The short-term effect on increasing a country’s
imports is larger for industrial than postindustrial economies.
 Conclusion: Government- policies that remedy the negative effects of trade increase support
for economic openness. Very supportive of the embedded liberalism thesis. Thus, our micro
results suggest that the movement of workers from tradable to non=tradable sectors will
increase support for trade. Globalization and deindustrialization should have interdependent
effects on government spending
o the strongest opposition to trade comes from individuals employed in sectors of the
economy that have the highest levels of import
o politicians can, in fact, build support for trade, even among these sectors
o Workers who compete against imports tend to oppose free trade, but their
opposition can be reduced with policies designed to protect them like
unemployment insurance and ALMPs.
o Politically feasible policy reforms can offset declines caused by increased exposure to
international competition

o Hypothesis: In post-industrial economies, the effect of imports on spending should be

smaller in magnitude
 the effect of imports on a country’s level of government spending depends
on how exposed the domestic labor market is to trade
 the short-term effect of increasing a country’s imports is larger for industrial
than post-industrial economies (post-industrial has mostly service sector)
 relationship between trade and government spending is robust when
imports and exports are distinguished and the interactive effects of trade
and deindustrialization are recognized
o If it is such a good idea politically to compensate t hose who lose from trade, why
does it not happen more often?
 It happens more often than many realize
 Governments are less responsive to imports today than in the past
 National institutions play an important role in determining if, when, and how
the bargain of embedded liberalism is implemented in different countries.

2. The Causes of Welfare State Expansion: Deindustrialization or Globalization?

Iversen & Cusack
 Aim/question: an alternative explanation for the compensation hypothesis is
deindustrialization. convince the reader that growth in both transfers and government
consumption can largely be explained as a function of the severity of internally driven
employment losses in the traditional sectors, not by forces in the global economy. Two views
are challenged:
o Increasing openness of national economies leads to growing economic insecurity
which then fuels demands for larger welfare spending in the form of insurance.
o Challenges promoted by globalization when met by strong left-labor power within
the domestic political system combine to produce a compensation strategy that
entail a large and vibrant welfare state.
 Main argument/hypothesis: Our argument in short is that structural changes in the
domestic economy resulting from technological progress lead to economic insecurity, which
increases the demand for social security. Increasingly productivity, changing consumption
patterns, and saturated demand for products from the traditional sectors of the economy are
the main force of change. It is these structural sources of risk that fuel demands for state
compensation and risk sharing. People feel ore insecure, higher demand for social

20
 protection, not insecure because of globalisation but because of domestic developments on
the labour market mainly as result of technological process.
Key assumption: it takes some time to change the sector-specific skills. If you are replaced,
you can’t find a job in another sector immediately. Specific skills for those jobs can’t be find
easily in new jobs, because of that insecure, higher demand of social protection.

• Data/methods: Empirical contribution: non-discretionary spending. They operationalized

deindustrialization. Changes in welfare spending not only driven by welfare spending, but
also unemployment rate and number people pensions. But they say if you want to look at
the effect of deindustrialization, look at that part of spending that can change. Take out the
non-discretionary spending: part that is already fixed unemployment/elderly spending. They
took unemployment rate times the replacement rate (that part of the income replaced by
the benefits), then you have a rough indication of the spending that is mathematically spent
on the unemployed. They compare the effect of trade openness and globalisation on capital,
positive effect of deindustrialisation.
• Outcome(s):
o The shift in the labor market (deindustrialization) has been the driving force of
welfare state expansion since the early 1960s. The importance of changes in the
occupational structure depends on the transferability of skills and social benefits.
Workers can protect themselves against these risks only through mediation of the
state.
o While the authors maintain that deindustrialization is a crucial source of welfare
state expansion, they do not imply that political and institutional factors are
unimportant.
o None of the globalization variables registers a statistically significant impact on
government transfers. Thus, while growing exposure to competition from low-wage
countries raises the uncertainty for those already at high risk, trade may well
improve welfare for all others
o The effect of deindustrialization runs along two different paths:
 Deindustrialization raises the generosity of transfer payments as
governments respond to electoral pressures from insurance against labor-
market risks (path a).
 Insofar as redundant labor in the traditional sectors are not being picked up
by new employment in private services (path b) governments are likely to
face electoral pressures to create employment by expanding public provision
of services (path c).
 Conclusion: Labor markets have shifted dramatically and individuals face significant risks as a
result. Jobs are not always available and if they are, their skills are not always applicable. For
many, loss of employment in the traditional sectors entails complete removal from the active
labor force. Broadly speaking, governments have responded to transformation in three ways:
1. Some have promoted employment in private services, often by deregulating product
and labor markets and allowing greater wage dispersion.
2. Regulation of private services as well as relatively compressed wage structure while
simultaneously expanding employment in public services.
3. Heavy regulation of labor and products markets hampered a major expansion of
private sector service employment and public sector has not been allowed to grow
 limiting labor-force participation.

21
 • Critique: : weak part of whole argument, deindustrialization itself could be the result of
globalisation. Some parts of the production lines can be offshored. What is good, they bring
up this weak point themselves an analyse to what extent this is a problem. In table 4 they
analyse this. But deindustrialization is driven by domestic things, but in table 4 they do find
that deindustrialization is to some extent driven by globalisation. But they say it is a small not
that significant effect. Still result of domestic changes.

Globalization, labour power and partisan politics revisited. Socio-Economic Review

Kwon & Pontusson
 Aim/question: efficiency/compensation effect: policy dilemma has influence on the variance
in left-wing parties. Does government partisanship matters is a relevant question for this
paper.
Whether the policies of Left-leaning and Right-leaning governments have converged or
diverged. We hypothesize that Left parties have a particularly strong incentive to respond to
the demand for more social protection because unskilled and skilled workers are part of their
core constituency and that strong unions render Left parties more prone to respond to
globalization by increasing social protection
• Main argument/hypothesis: Effect of left varies across decades. Effect of left is dependent
on globalisation. Subsequently, this effect is dependent on the strength of labour unions.
Distinction left and right maybe not that relevant anymore. No effect for left wing parties,
but split in three periods, the coefficient for left-wing parties varies over the years. In line
with the paper of Scrudge (also split the period). Why does the effect of left-wing parties vary
over time? Because of the effect of globalisation they claim: efficiency/compensation effect,
policy dilemma. So left-wing party would then opt relatively more for higher spending.
Therefore interaction term between left-wing parties and globalisation. This effect varies
over time. Why is that? Contingent is dependent on the strength of labour unions. This goes
back to the idea of the Wassenaar Agreement: agreement that labour unions have with the
governments. In the past if the unemployment rate is high, governments like to stimulate the
economy and ask the labour unions to lower their wage demands. In exchange labour asks
for higher spending on unemployment benefits. Why would governments do that? 1 result
they got political support for policies 2 if they increases spending higher levels inflation,
keeping wages constant lowers inflation. At the moment no high level of inflation, but in the
past this was a problem for the government. So therefore deal with Labour Unions. Only
then.
Some decades there were strong labour unions, in some decades there were no strong
labour unions. Positive effect of strong labour unions on welfare spending, if you have weak
labour unions left-wing parties if they could not get political support from the unions they
choose welfare retrenchment.
Labour markets nowadays do not have many members, but are relevant in collective
agreements of the sector. In case of strong labour unions: expansion, now at the moment
retrenchment should be happening. And if you have globalisation and left-wing parties.
 Data/methods: Effects on social spending growth in OECD counties in the period 1971-2002.
Interaction term of 3 variables: globalisation, left-wing parties and labour unions.
 Outcome(s): Our results clearly contradict the claim that partisan conflict over social policy
declined in the 1970s and 1980s.
o Partisan effects increased from mid-1970s to late 1980s and then disappeared in the
1990s
o Partisan effects rose with globalisation in the 1970s and early 1980s a period
characterized by rising labour strength in many OECD countries but this is not true
for post-1990 period when labour strength declined

22
 o Globalisation was associated with declining partisan effects in countries that
experienced union decline in 1980s and 1990s, but associated with rising partisan
effects in countries in which unions remained strong.
Pierson would say that the government partisanship has declined because of the growing
importance of welfare-state clienteles and onset permanent austerity. Globalization
might be invoked to explain the decline of partisan effects in the 1990s, but why did
globalization not have this effect already in the 1970s and 1980s? Could argue that the
effects of globalization are conditional on domestic political configurations and that
many OECD countries passed some threshold of globalization in the late 1980s or early
1990s. Insider–outsider argument does not provide any ready explanation of the
previous rise of partisan effects insider–outsider argument does not provide any ready
explanation of the previous rise of partisan effects.
 Conclusion: thesis that globalization generates pressures on Left parties to expand the
welfare state when unions are strong and pressures in the opposite direction when unions
are weak provides a coherent framework for explaining the rise as well as the decline of
partisan effects. indicates that globalization was associated with rising partisan effects in
countries where the strength of organized labour held up in the 1980s and 1990s, but it was
associated with falling partisan effects in countries that experienced extensive union decline
 Critique: assumes a common causal logic across all OECD countries and that the effects of
changes in government partisanship are monotonic.

Globalization and the Welfare State: Testing the Micro-foundations of the

Compensations Hypothesis.
Walter
 Aim/question: compensation hypothesis: government in open economies insure potential
globalization losers against the risk associated with increased international competition and
volatility (demand side). Walter provides a test of the whole causal chain to test whether
globalisation led to welfare state expansion using a micro-level focus.
 Main argument/hypothesis: globalisation leads to welfare expansion.
 Data/methods: Uses Switzerland as a case. micro level focus (survey). Determinants of
globalisation losers/winners: low skilled/high skilled + occupation sector. The effect of
globalization is beneficial for some but not for others.
 Outcome(s): Results: in line with compensation hypothesis. Left vote among low skilled
individuals (losers of globalization). High skilled are less likely to vote for left parties, unless
they become more insecure.
 Conclusion: basis for compensation hypothesis on microlevel. the effect of globalization
exposure is highly dependent on individuals’ skill levels. The results also suggest that a direct
relationship between individuals’ vulnerability to economic internationalization and their
vote choice exists. And notion that the individual economic situation strongly affects people’s
policy preferences.
 Critique:
o Switzerland is the only case. Is this really a comparable country? Unique country in
the world. Not a country with problems of globalization.
o Doesn’t take into account welfare institutions/political parties. Why doesn’t she take
that into account? Only micro, only one country. She cannot do it. One moment in
time, individuals. Why only 1 country? She could have taken or another country, or
more countries than that one. Taking only one year, argument compensation:
increase globalization, increase protection demand. Here argument: people have
higher exposure, higher demand for protection. But the original argument would
have some time dimension in it. With only one moment in time it is not possible.

23
o Right wing parties? When people lose their jobs they can also vote for right wing
parties, but would the policy preference/agenda be the same? Parties could have
other preferences, like close the borders/limit globalisation/limit trade.

24
WK5 Migration
 De Giorgi, G. and M. Pellizzari (2009) Welfare migration in Europe. Labour Economics 16(4): 353-
363.
 Kvist, J. (2004) Does EU Enlargement Start a Race to the Bottom? Strategic Interaction among EU
Member States in Social Policy. Journal of European Social Policy 14(3): 301-318.
 Gaston, N. and G. Rajaguru (2013) International migration and the welfare state revisited.
European Journal of Political Economy 29: 90-101.
 Burgoon, B. (2014) Immigration, Integration, and Support for Redistribution in Europe. World
Politics 66(3): 365-405.

1. De Giorgi & Pellizzari (2009) Welfare migration in Europe

 Aim/question: the extent to which welfare generosity affects the location decisions of
migrants in the 15 countries of the pre-enlargement of the EU. Whole argument whether
migration flows are mainly driven by labour market conditions or the welfare state benefits.
 Main argument/hypothesis: generosity of the welfare state might act as a migration magnet
across countries of the EU. Concerns: threat to the welfare state posed by enlargement of
the EU.
Argument of the welfare magnet hypothesis. Argument stemming from the US, where states
are very similar like culture and language, but the tax rates can differ. The labour mobility in
the US is much higher than in the EU. The EU labour mobility is lower, reasons of language,
cultural differences, etc. The argument would be thinking about welfare state generosity as
an important driver. If this is the case, the bad risk in terms of insurance, high risk of
becoming unemployed and receiving benefits will go the countries with good unemployment
benefits. The low risk, good people with high education etc. will go to low taxes and low
contributions, low public benefits. So one country good risks and one country bad risks.
Welfare state not sustainable in the long run. Benefits have to be cut. Social race to the
bottom.
 Milton Friedman: you cannot have welfare state and open borders.
 Data/methods: panel data sources of EU and OECD. Describing migration decisions. Only
3000 individuals, but quite a number of observations. They do use microdata, but in the table
it looks like aggregated data, no control data for individuals, macro level could have included
more variables. Why such a small model?
 Outcome(s):
The size of welfare magnets (reasons?) is relatively low, compared to the role of labour
market conditions.
Small effect for women. Less sensitive for attractiveness of welfare benefits. Italy: attracted
by the kind of jobs (household jobs etc). Why are women relatively less attracted to welfare
benefits than men? Guess: men migrate first. This data of 1980s, women did not worked that
much. Also imply that women are less sensitive to jobs or benefits, because they just joined
their partners. But we don’t know this for sure.
 Conclusion: welfare benefits do determine migration locations, but effect is small. Labour
market conditions are more important. This is important for the influence of the extent of
variation in welfare institutions across countries will generate distortions in flows of
migrants. These distortions can be large enough to reduce the potential benefits of acquiring
a more mobile labour force (= most important advantage of increased migration in immobile
labour force of pre-enlargement EU).
 Critique:

25
 o Suggests that migrants use the welfare state relatively more than natives, hence
migration is driven potentially by the generosity of the welfare. But this causality is
flawed. A lot of heterogeneity about this causality
o Authors did not include the period after 1994? They tried to frame the paper as
labour migration as result of the enlargement of the EU. But the say mostly
something about seventies and nineties?
o Illegal immigration is not included. But how could you include illegal immigration.
Not well documented because the people are illegal. But in the media the illegal
issues are most prominent.

2. Kvist (2004) Does EU Enlargement Start a Race to the Bottom?

 Aim/question: whether Eastern enlargement has led the EU 15 member states to enter
strategic interactions implying a race to the bottom.
 Main argument/hypothesis: ‘race to the bottom’ = downward bias in restriction and benefit
accessibility and generosity. Privileged groups tend to view enlargement more optimistically
than less privileged groups, and positive attitudes to enlargement vary with education and
socio-economic position. Mobility concerns related to enlargement can be summed up under
three headings; social tourism, social dumping and social raids.
These concerns share a belief that national labour markets and welfare systems become
open to nationals from other countries, and that this may result in people shopping around
to get the best mix of benefits, wages and taxes. Question whether indeed persons and
labour are mobile across borders and, if so, whether they are rational and in a position to
perform such cost–benefit calculations when deciding upon a future host country?
 Data/methods: more qualitive analysis of the years before 2004 when the EU enlargement
happened. Two assumptions are important not to forget: great uncertainties in projections
on migration flows as reflected by varying estimates in different studies & migration may be
a problem for the sending countries and a solution for the receiving countries.
 Outcome(s): Little/no empirical evidence to support the assumption that welfare states with
generous benefits and accessible labour markets will become magnets for welfare migration.
Nevertheless, the study demonstrates that the EU 15 member states do enter strategic
interactions as if such migration would occur. In sum, national adjustments have been made
in social policy with explicit reference to mobility issues, including Eastern EU enlargement.
EU 15 member states have: decided on policy measures when others have done the same;
justified policy measures with reference to policy stances of other member states; and
sought inspiration abroad. Overall, however, these changes are marginal in nature. >>> not
because of actual pressure but because of prevention.
 Conclusion: EU 15 member states with the least restrictions are the ones most active in
adjusting their social policies. Strategic interactions in social policy may thus intensify in the
future as transitional periods come to an end and future enlargements come into place. How
can such a race to the bottom be avoided? In principal this can be solved through
coordination or harmonization.
biggest barrier for a race to the bottom may be institutional resilience and the vested
interest that welfare-state institutions and programmes cater to. Dramatic retrenchment
unlikely.
- Remarks: Tricky in the EU: free movement, you can’t discriminate. Policy changes that
actually do that, but can’t be framed like that. Requirements that you should have worked in
the country for 1 year if you want unemployment benefits. Also: distinction between types of
policies: contributory versus non-contributory social policies. Another remark: the game is
more complicated for the EU. With competition between EU and competition outside EU.

26
3. Gaston & Rajaguru (2013) International migration and the welfare state revisited.
European Journal of Political Economy
 Aim/question: about the relationship between migration and the welfare state. Whether a
large intake of immigrants reduces welfare state effort.
 Main argument/hypothesis: Key hypothesis is that downward convergence leads to a social
race to the bottom. The usual presumption is that welfare states in relatively wealthy
countries involve net transfers from natives to immigrants. This presumption is undermined,
to some extent, by two factors: the actual pattern of immigration; and the valuation of
lifetime effects
exposure or insurance effect = Workers (and voters) exposed to unemployment risk prefer
higher welfare state expenditures(andtaxation!) (= compensation hypothesis)
redistribution/tax effect = workers with a higher probability of working always prefer lower
taxes and fewer publicly-provided services in response to an increase in immigration. (=
efficiency hypothesis).
Overall, whether more open immigration places downward pressure on welfare state
spending very much depends on who it is that immigrates. In fact, if immigrants are fiscal
contributors, then the political support for higher welfare spending could actually rise
 Data/methods: data for 25 OECD economies for the years 1980–2008
 Outcome(s): Little evidence for social race to the bottom. While immigration does have a
relatively modest effect on the welfare state, if anything there is some support for the view
that a greater influx of immigrants has lead policy-makers to increase welfare state spending.
They find a small effect of migration: a positive effect.
They find a negative effect of globalisation, efficiency hypothesis. Positive effect between
immigration and spending on the welfare state.
It is important what type of country they come from. OECD not significant effect. Highly
educated immigrants: negative effect. Chose to migrate to countries with lower taxes. They
argue it is in line with the compensation hypothesis (exposure effect). As share of total
migration, opposite with low-skilled you would find a positive effect. But there is an
alternative explanation, this could also be a selection effect; highly educated people will
move to countries with low benefits/taxes, that would also imply support for the other
argument that it would lead to a social race to the bottom.
Than the causality would be reverse.
 Conclusion
Immigrants are easy targets for populists and xenophobes when economic conditions
deteriorate
exposure effect and a redistribution effect which always work in opposite directions. Our
findings support the view that immigration has had a relatively modest effect on welfare
state spending. Some effect of immigration on welfare spending. Some support for the
exposure or compensation effect. However, as recent studies have also emphasised, this
finding is sensitive to the group of countries

4. Burgoon (2014) Immigration, Integration, and Support for Redistribution in Europe.

 Aim/question: seeks to clarify the relationship between immigration and social policy by
exploring how social and economic integration of immigrants in host societies mediates that
relationship
 Main argument/hypothesis: policy reactions because of voters. Governments may have
incentives to reform welfare states, because voter are concerned about the impact of
immigration. Are they concerned because of the actual impact or because of perception?
Burgoon tries to disentangle the actual pressure and the perception of the problem.

27
 Higher immigration should more negatively affect support for government redistribution to
the extent that immigrants have higher rates of unemployment than natives, are more
dependent than natives on social benefits, or do not share the native population’s
sociocultural values.
 Data/methods: five waves of European Social Survey data on individual attitudes in twenty-
two European polities between 2002 and 2010
 Outcome(s): exposure to higher foreign-born percentages tends to diminish support for
redistribution and social protection but also that this effect is substantively and statistically
significantly more negative when migrants have proportionately higher unemployment rates
and greater dependency on social benefits. In contrast, the gap between foreign-born and
native respondents in sociocultural values has a more modest and less consistent negative
effect on the relationship between immigration and redistribution.
 Conclusion: economic integration, more than sociocultural integration, may be an important
factor in dampening immigration’s negative effects on welfare states in Europe. Important
mechanisms for such effects: economic non-integration, again more than cultural non-
integration, exacerbates how immigration can spark concerns about the fiscal viability of
welfare states, while doing little to alter how immigration affects altruism or individual
economic risks. But greater economic integration of immigrants into the labor markets can
meaningfully diminish such negative effects. Better social integration does not have this
effect.

28
WK6 Income inequality
 Autor, D.H., D. Dorn and G.H. Hanson (2015) Untangling Trade and Technology: Evidence
from Local Labour Markets. The Economic Journal 125(584): 621-646.
 Iversen, T. and A. Wren (1998) Equality, Employment, and Budgetary Restraint: The Trilemma
of the Service Economy. World Politics 50(4): 507-546.
 Mahler, V.A. (2004) Economic Globalization: Domestic Politics, and Income Inequality in the
Developed Countries. Comparative Political Studies 37(9): 1025-1053.

1. Autor, Dorn & Hanson (2015) Untangling Trade and Technology: Evidence from
Local Labour Markets
 Aim/question: tried to distinguish the effects of globalisation and SBTC by looking at the
effects of trade and technology on employment in US local labour markets between 1980
and 2007
 Main argument/hypothesis: globalization and technological progress have developed
simultaneously since 1980s and those effects are very similar. But do they have the same
effects?
 Data/methods: Autor measured in commuting zones, local markets in the US. Geographical
area where people live and work. In the US 740 or so. That are local labour markets. Between
that markets the mobility is low. They will not directly accept a job in another zone. Autor
combines this with variation in type of industry ánd combines it with type of task (routine or
managerial job) and combines it with level of skills people have (high/low education). . All
combined this gives you enough variation in order to disentangle these two effects
(globalisation and technology).
Didn’t look at globalisation in general, but imports from China. Because argument that
globalisation is slow process, but look at a very specific shock and for that he looked at
imports from china because from 2001 China’s market opened and this had a big impact on
the US. . In the EU we already imported this stuff from other countries, but the US before
2001 produced a lot of this themselves in Walmarts etc. They had much larger manufacturing
industry. So the impacts of China has had a much larger impact there. Important exogenous
shock for international trade.
 Outcome(s): Both globalisation and SBTC are relevant processes in different ways.
o Effect of globalisation: employment declined in sectors that are exposed to
international trade. Employment goes down because sectors are transferred to
China. Whole sectors disappeared: overall effect of unemployment. The employment
decline is not limited to production jobs but instead affects all major occupation
groups, including a notable decline in managerial, professional and technical jobs.
o Effect of SBTC: neutral effect on overall employment of a sector, but within a sector
there is a lower demand for low-skilled employees and a higher demand for high-
skilled employees. No overall effect of unemployment, but a shift from low to high
skilled job within a sector.
o So: both channels: factor intensity matters (as shown by non-college workers being
the skill group most impacted by trade) but so does the nature of the task (as shown
by routine occupations being most affected by exposure to import competition)
 Conclusion: it is a very important insight that both globalisation and SBTC matter in a
different way. Autor did this for the US. Could you see similar effects in the EU? Yes you can.
Imports from china give similar effects, but a bit smaller.

29
2. Iversen & Wren (1998). Equality, Employment, and Budgetary Restraint: The
Trilemma of the Service Economy
 Aim/question: since 1970s the advanced democratic societies changes: more
unemployment, inequality and fiscal constraints. Dominant approach is to look for causes in
global economy. This paper takes a different approach: the most important change in the
advanced liberal democracies over the past three decades has been the transition from an
economy dominated by (exposed) manufacturing production to one dominated by
(sheltered) services production.
 Main argument/hypothesis: trilemma of the service economy: fiscal discipline, employment
growth or income equality. There are in general 3 types of strategies. The whole problem
there is that there is a trilemma. You can only achieve two/three at the same time but you
want all of them. Why is it impossible to get all three? Which two you pursuit is matter of
political preference.
Importance service economy. Because globalisation and technological processes the service
economy is becoming more important. If jobs in manufacturing industry disappear, people
should find employment somewhere else. Low skilled people should find job in service
economy, because the sectors in the service economy are not vulnerable sectors to
globalisation or technological processes (e.g. the Horeca).
Productivity problem. Service industry you can’t see that much growth in productivity. Large
part of our economic growth is result of growing productivity. But in some sectors this isn’t
really possible, it takes exactly the same time and number of people for some services, there
is no growth in productivity but their incomes have gone up. They are coupled to more
general levels of the wages. Same applies to for instance nurses, same productivity: relatively
their work has become more expensive. In the past it wasn’t a problem, because small part
of economy. But now service industry is becoming larger compared to manufacturing
industry, it is becoming a problem. Service industry is becoming very expensive. This is what
they saw in 1998, and this is what is going on right now in 2019.
 Trilemma 3 models: three possible strategies that governments can pursue in the face of the
trilemma
o Neoliberal model
 Fiscal discipline
 Employment growth
 No earnings equality: in order to get employment growth you let the wages
go down in the service sector so that sector can grow.
o Christian democratic model
 Fiscal discipline
 Earnings equality
 No employment growth because you can’t stimulate the market with lower
wages and you can’t create more jobs by spending extra money.
o Social democratic model
 Earnings equality
 Employment growth
 No fiscal discipline: spend extra money to stimulate the labour market by
creating more jobs in the public service. Not extra jobs in service industry,
because you want earnings equality.
 Data/methods: economic argument implies that the expansion of service sector
employment is negatively affected by both wage equality and budgetary restraint.
expectation is that high earnings equality will inhibit expansion of employment in the private
service sector
 Outcome(s): main thing is that the services sector productivity is not rising in productivity as
the manufacturing sector did, coupled with rising prices is this a problem.

30
  clear negative relationship between a change in public sector employment ratios and
budgetary restraint, the price of labor is a variable that determines the budget effects of
employment expansion
Economic booms- whether politically engineered or resulting from swings in the business
cycle-do improve employment performance, but the effects are cyclical and do not affect our
main conclusions. Positive effect of an increase in trade can be explained by the fact that
domestic markets are effectively being enlarged by such trade.
The strongest effect of capital-market liberalization is clearly on budgetary restraint. has little
effect on employment, and trade seems to improve performance in manufacturing. Most
importantly, the engine of employment growth over the past three decades-services-appears
to have been largely unaffected by trends toward greater globalization."
Thus, while the effect of wage equality is substantially smaller (though statistically
significant), productivity has a strong and significant negative effect on employment.
 Emphasize that existing institutions and ideological platforms are themselves placed under
an entirely new set of strains by the growing tensions of the trilemma: equality and
employment expensian
o Social democrats: new trade-off between equality and employment expansion,
which did not exist in the golden age of manufacturing (example of Danish case:
distributional conflict between high- and low-productivity workers)
o Christian democrats: traditionally minimizing labor force participation to keep
unemployment rate from rising, dilemma when discouraging participation becomes
politically infeasible (Dutch case)
o Liberals: increase in inequality and poverty (GB)
 Conclusion: supports the existence of a trade-off between intersectoral wage equality and
the expansion of employment in the private services sector. Given the declining capacity of
the manufacturing sector to generate employment in most OECD countries, the existence of
this trade-off presents governments with a trilemma. Severity of the trilemma depends on
the continued inability to increase productivity in the services sector. ncreases in services
productivity would allow prices to fall and wages to rise simultaneously, as in the golden age
of manufacturing.
 Remark: similar paper on deindustrialisation of Cusack.

3. Mahler (2004) Economic Globalization: Domestic Politics, and Income Inequality

in the Developed Countries
 Aim/question: Looks at the relationship between globalisation and income inequality. Offer
empirical assessment of relative impact of international and domestic factors on the
distribution of income, etc.
 Main argument/hypothesis: 2 specific theories:
o Positive relationship between economic globalization and income inequality: winners
& losers bc globalisation. Race to the bottom. More income inequality
o Negative relationship: more competition, lower prices, benefit for lower-income
groups. Embedded liberalism: not an increase of income inequality.

 Data/methods:
Examines three major modes of international integration—trade, direct foreign investment,
and international financial flows—as well as four domestic political variables—the partisan
balance of national cabinets, electoral turnout, union density, and the centralization of wage-
setting institutions.
it matters which dependent variable you take for income inequality. Mahler is very
transparent: both mechanisms are relevant, therefore using all the three variables (market
income, disposable income and fiscal redistribution). But doesn’t include interaction

31
 variables. You want to combine ideally all the variables in one estimation, but Mahler uses
separate analysis.
Using household-level data. Mahler makes the argument that there are differences between
measuring at the individual level or household level. He uses household incomes because of
economies of scale. Households can share costs so that they need less money than 1 person.
Effect of globalisation is a bit blurred if you look at 2 persons in 1 household. One person can
be in a sector that is exposed to globalisation and perhaps the other one is not. Then
individual data is better.
 Outcome(s): Mahler didn’t find a big effect of globalisation but he did for labour market
institutions? Only scattered relationships between global integration and income distribution
or redistribution, but reasonably strong positive relationships between several domestic
political variables and an egalitarian distribution of income and /or extensive state
redistribution. Emphasize the resilience of domestic political factors in the face of global
economic globalization
 Conclusion: little impact of economic globalization and income inequality. Domestic political
factors are still relevant.
the fact that domestic policies continue to be relevant would lead to: reject the claim of anti-
globalization leftist and to reject the claim of pro-globalization rightist. There appears to be
middle ground that combines a broad commitment to global liberalism with a recognition
that economic globalization is compatible with a wide variety of political dynamics that can in
turn lead to a wide range of distributive outcomes.

32
WK 7

1. Global capital markets, housing prices, and partisan fiscal policies

Ansell, Broz & Flaherty
 Aim/question:
 Main argument/hypothesis: including housing markets, one of the ways that financial
markets have an impact on average citizens and daily behaviour. They argue that there is a
causal relationship between housing prices and capital inflows. The international capital flow
has an impact on housing prices. This affects fiscal policy preferences of both voters and
political parties. Argument is that citizens, relying on housing as private insurance, will be less
(more) supportive of social insurance where rising house prices push up (down) the value of
this asset. And we argue that global capital flows have a direct impact on housing and other
asset prices. Policy preferences: Capital flows > substantial variation in house pricing +
housing as permanent income. The effect of partisanship on fiscal policy is conditional on
capital inflows and house price appreciation:
o Large inflows > house prices are rising > demand for lower taxes + less social services
> opportunity for right parties
o Declining house prices > individuals exposed to market risk: demand for greater
social insurance > opportunity for left parties
o Fiscal preferences of mortgagors more sensitive to house price changes than
homeowners that acquire property without leverage

 International capital flow: Deficit countries: attract substantial foreign capital + show large
run-ups in house prices. Surplus countries: exported capital + show flat or slow house price
growth
o Global capital markets channel savings from surplus countries into real estate
markets of deficit countries have significant political consequences for fiscal
preferences and policies.
o Importance of asset markets, such as housing. This matters because housing affects
macro-economic environment + individual demands of government.
Capital flows  housing prices  fiscal preferences  partisan fiscal policies

 Capital flow bonanza = an unusual shift in inflows. House prices typically sky-rocket during
capital inflow bonanzas and when the money leaves, the downturn is sudden and sharp.
o Money flows in: Foreign borrowing  more foreign money floods in  increase
amount of money people have  raised demand for goods  elastic traded goods
(import)  inelastic non-tradables: price is driven up, i.e. higher home prices.
o Money flows out  lower demand for non-tradables  pushing prices down

 Preferences rising of homeowners in case of rising house prices

o Tax averse
 Pocket-book mechanism
 Psychological mechanism
o Less redistribution spending
 Lower demand for spending themselves
 House as ‘nest egg’: self-insurance against labor market
misfurtunes/retirement
 Data/methods: showing that capital flows have important economic consequences for
homeowners

33
 Outcome(s): When large capital inflows enter an economy, the demand for assets that are in
fixed supply increases, and asset prices rise
o Capital inflows have important economic consequences for homeowners - a one
percent of GDP capital inflow is associated with an increase in house prices of around
two-thirds of a percent
o the composition of capital inflows also matters: bonanzas vs FDI. Bonanzas of debt-
related investment put strong upward pressure on house prices
o the effect of capital flows on house prices was amplified in Eastern Europe, which
experienced wild swings in home prices as the capital flow cycle ran its course during
the 2000s.
o Large inflows: demand for assets in fixed supply increases, asset prices rise

Less public support for taxation + redistribution > government cut taxes + spending >
amplified when right-wing governments are in power bc of ideological preferences. Left-wing
governments are less likely to make such cuts.
Direct effects of changes in house prices
o Right-wing parties: house prices lead to reductions in government consumption
o Left-wing or center parties: house prices lead to increased government consumption
o Center parties: effects of house prices increases are essentially nil
 Conclusion: global capital markets that channel savings from surplus countries into the real
estate markets of deficit countries have significant political consequences for fiscal
preferences and policies. In this way global economic forces shape citizens’ fiscal policy
preferences.
Political economy argument: homeowners view the equity they build up in their houses as a
private substitute for publicly provided insurance. Therefore homeowners will demand both
lower taxes + less publicly provided social insurance, where capital inflows are large and
housing prices are rising. The same holds for redistributive spending: citizens with higher
prices homes express less support for redistributive government spending. This effect
appears larger in countries experiencing larger house price booms.
increases in house prices appear negatively related to government consumption; that this
effect is magnified in countries that have been experiencing sizable medium-term capital
inflows; and that the effect is also driven largely by right-wing parties. We argue that this
partisan dynamic is a function of right-wing parties being more likely to represent
homeowners, whose preferences over social spending tilt negatively as house prices rise.
Global flows that may have been most important in affecting fiscal policy preferences and
outcomes are those that drove asset prices. In other words, instead of increasing labor
market insecurities, globalization may have created a (possibly false) sense of income
security through the asset price channel.
> Globalization may indeed undermine the welfare state, but it has done so through an
unexpected channel. Opportunity for the Right to cut public spending.
 Critique: They try to combine a number of dimensions in the literature: globalization, left
right wing parties, etc. if you put it in that context, it would make sense to control some
control variables, like exposure to globalization. Walter: argument employees exposed to
globalisation or not, more insecure or not and whether that had an effect on their welfare
state preferences. The article should at least have controlled for that, it is almost if the whole
impact of globalization was run by the house market and that is not likely.
 Remarks: interesting article. Comparative to what we have seen before. Idea of globalization
and capital flows. Other hand: preferences with respect to taxes and social policy. nuanced
argument in between.

34
2. Colantone & Stanig () Global Competition and Brexit
 Aim/question: Compare the effects of global trade international trade and immigration with
actual voting behaviour. Looks at international trade versus immigration.
Support for the Leave option in the Brexit referendum was systematically higher in regions
hit harder by economic globalization. These areas witnessed poorer performance in terms of
real wage growth and a lower employment rate in the past two decades. Colantane and
Stanig used imports from China as main structural driver of globalisation and divergence in
economic performance across U.K. Regions.
 Main argument/hypothesis: Important: globalisation without compensation. Globalisation
led to displacement with the absence of effective compensation of the losers. The inability of
governments to do so, might have led to a Crisis of Embedded liberalism hypothesis: buying
support for the losers in order to maintain/keep their political support. This breeds isolatism
and neonationalism.
In line with Heckscher-Ohlin/Stolper-Samsuelson theorem: idea of factor endowments.
- Low skilled labour: winners in China, losers in western countries > imports
- High skilled labour: winners in western countries > exports
- Net welfare gains, although distributed highly unequally
 Data/methods: impact China on UK. Similar approach like David Autor where he compares
US zones, argument the same: examine local labour markets. Chinese imports would replace
the domestic manufacturing industries. Because of historic reasons those manufacturing
areas where geographically concentrated. Therefore in some regions the effect would be hit
harder. Colantone and Stanig: same approach as Autor in the regional-level robustness. And
indeed they find an effect: people more affected by imports of china vote for Leave the EU.
 1980s China: economic reform
unexpected, structural catch-up, comparative advantage in
manufacturing/abundance of labour after collectivization of agriculture
 imports from China in the UK rise rapidly
 import pressure from China
 downward pressure on prices (benefit consumers)
 dramatic supply shock for developing countries in manufacturing
 displacement of manufacturing activities
 unemployment + lower earnings disproportionally on low-skilled workers in
import competing industry
 losers of globalization
 historically manufacturing geographically concentrated in regions
 geographically concentrated economic distress
 ‘left behind’ areas of globalization
 Higher support for leaving
Globalization-induced shock: Import shock is an important determinant of divergence across
regions. Three possible mechanisms through which the import shock might lead to higher
support for Brexit:
1. As vote against political elites + businesses
2. As vote against international integration + in favour of national sovereignty
3. As vote against immigration
 Overall causal effect of the import shock, which is a key structural determinant of
discontent, by means of divergence in economic performance across regions
 Outcome(s): immigrants are not so relevant, import shocks are!
o Clear positive relationship between import shock + support for leave
o No clear relationship between immigration + support for leave.

35
  Individual attitudes towards immigration are systematically worsened by the
import shock, while they are not related in a clear way to the actual extent of
immigration in a region.
 Self-reports do mention immigration because:
 Scarcity of employment
 Scapegoating
 Concerns that immigration creates overcrowding + congestion of
public services
o With EU immigrants they find positive effect, more immigrants from this area is
associated with more votes for leave. Why is that logical? In the beginning of the
Brexit story that is what it was about. Cameron: argument, immigrants from central-
eastern-european countries and how that affected the welfare system in the UK.
From a more substantive perspective, this is the only variable that would make
sense. If you think that this is a problem, then leaving the EU could be the solution.
Then you don’t have the free movement anymore. For the other variables, if you
have high levels of immigration general, like from Africa, leaving the EU wouldn’t
matter
 Conclusion: if we look at the impact of globalization on actual voting behaviour, in the Brexit
case international trade was more important than immigration. Immigration does matter,
but more based general on the economic situation rather than the specific effects of
immigration on an individual. General argument that if you are more affected by
globalisation you feel more insecure, and you would vote for leave. This brings all kinds of
arguments together, international trade compared with immigration.
 Critique: Brilliant paper, it is very good that they could actually examine typical behaviour
and link it with globalisation in stead of using survey data. There are however 2 main
limitations
o Article last week: impact globalization or technological process. It would have been
strong to control for technological process in this paper as well. In the beginning
specific areas with manufacturing industry, it would have been good to control for
the effect of technological innovation in these areas.
o Second main limitation: argument that people who voted for leave where negatively
affected by globalization and not compensated enough about the welfare state. In
their empirical analysis there is no variable for welfare state compensation or
something. To what extent are the losers actually compensated or not?

3. Weisstanner & Armingeon () How redistributive policies reduce market inequality:

education premiums in 22 OECD countries
Very important: Education premiums = wage differentials between high-educated and low-educated
workers. Indicate differences in the returns to education. Measured in the paper as the percentage
of difference of median wages of full-time workers with tertiary education and the remaining full-
time workers. Education policies concern education finance in the first place
 Aim/question: Onwijs PvdA achtig paper over verschillen in onderwijs en onderwijskansen
die kunnen lijden tot inkomensverschillen. Belastingen en publieke uitgaven aan educatie
kunnen helpen verschillen te verkleinen en maken de hele samenleving zo veel mooier.
Weisstanner and Armingeon provide evidence that taxation levels and public education
spending particularly affect education premium levels and changes within countries. They
argue that public education spending and tax-transfer policies can contain the spread of
education premiums (wage differentials). They can do this through material incentives
(decomofidication) and attitudinal responses (motivation, etc.).
 Main argument/hypothesis: Taxation and education policies can contain the rise of
education premiums  Less inequality

36
 o Large tax: reduces material incentives + attitudinal motivation of highly qualified
workers to exploit their strong position in the labour market.
 Income taxed away
 Reduced legitimacy of maximum return because of policy-feedback
mechanism??
o Large public spending on education: advantageous for non-tertiary workers > skills
lifted upwards
 Enable higher productivity for workers
 Raising total supply of skills for economy
 Lowering the wages
Skill biased technological changes and globalisation reinforce wage inequalities between
workers with varying educational attainment. Conversely, if the supply of high-educated
workers increases, returns to education would decrease. Education premiums are a crucial
source of wage inequality, etermined mainly by factors of supply and demand in the labour
market.
o Explaining international variation of education premiums:
 Redistributive policies (welfare state and education policies)
 Regulative policies
o Hypothesis
 H1: strong public education finance policies reduce education premiums
 H2: strong redistributive welfare policies such as a large tax state reduce
education premiums
 H3: strong regulative labour market policies reduce education premiums
 Data/methods: Study 22 OECD countries between 1989 and 2014. large variation of
education premiums across countries, in terms of both levels and changes over time
- Education premiums are comparatively large in the Anglo-Saxon countries
- Education premiums remained lowest in the Nordic countries across the whole
period
- Southern and Eastern European countries, education premiums started from high
levels in the mid-1990s but decreased throughout the 2000s
- Education premiums increased significantly in continental European countries such
as Austria, Germany, the Netherlands and Switzerland

seven indicators for education policies, other redistributive policies, and regulative policies
- public expenditures on education
- private expenditures on education
- taxation
- non-elderly social expenditures
- bargaining coverage rate
- employment protection legislation (EPL) for regular contracts
- EPL for temporary/fixed-term contract
 six of seven policy variables are significantly related to education premium levels in
the expected direction.
 substantial support that policies contribute to explaining the international variation
in education premiums
 These descriptive statistics are in support of a strong attitudinal policy feedback.
 Outcome(s):
o Very large education premiums are problematic if they reproduce existing social
inequalities and reduce mobility across generations. This has an impact on quality of
life and social closure. With modest education premiums, up- and downward social
mobility will be greater

37
 o Pay more systematic attention to redistributive policies shaping the “market”
distribution of incomes. They are highly relevant for international and intertemporal
variation of education premiums
o Education premiums not only caused by social and economic structures and
processes, but also by political choices. The choice of public policy (politics and
instutions) on wage inequality is important for whether education premiums are
bigger/smaller

Extra info for with the hypotheses:

Education policies: difference in private and public financing  H1

- Private financing results in higher educational inequalities
- Rationale for maximizing individual utility
o Private financing: high wage as compensation of
investment/student debt
- Policy feedback on attitudes and norms
o Maximum of returns becomes less convincing
o Major difference in educational credentials will be
due to talent and intellectual effort

Other redistributive policies  H2

1. generous welfare policies provide options to maintain a living outside the labour market and
therefore increase the bargaining power of low-educated workers
2. the larger the public sector, the larger segment of workforce with more equal wages,
because wages are more compressed in the public sector than in the private sector
3. A large tax state is another form of decommodification: reduce market inequality of net
wages
4. if non-wage labour costs are relatively high, this discourages employers from creating new
jobs
5. Finally, there may again be a feedback effect on attitudes

Regulative policies  H3
- create legal or informal norms that actors must follow.
o effectively limit market forces > strong trade unions
o employment protection > improves bargaining
position of (low-skilled) workers who are in danger
of being laid off if they do not accept lower wages

38