Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=cnpe20
New Political Economy
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/cnpe20
Culture & European attitudes on public debt
Alessia Aspide, Kathleen J. Brown, Matthew DiGiuseppe & Alexander Slaski
To cite this article: Alessia Aspide, Kathleen J. Brown, Matthew DiGiuseppe & Alexander
Slaski (2022): Culture & European attitudes on public debt, New Political Economy, DOI:
10.1080/13563467.2022.2143490
To link to this article: https://doi.org/10.1080/13563467.2022.2143490
© 2022 The Author(s). Published by Informa
UK Limited, trading as Taylor & Francis
Group
Published online: 20 Nov 2022.
Submit your article to this journal
View related articles
View Crossmark data
Culture & European attitudes on public debt
Alessia Aspide , Kathleen J. Brown , Matthew DiGiuseppe and Alexander Slaski
Political Science, Leiden University, Leiden, Netherlands
ABSTRACT
Popular media and politicians have often blamed the high public debt of
some EU countries on cultural differences. These claims are most apparent
in the discourse contrasting ostensibly prudent Northern Europeans with
spendthrift Southern Europeans. Despite the prominence of these and
similar narratives and evidence that culture plays a nontrivial role in
other economic outcomes, there is no systematic evidence that culture
influences attitudes towards sovereign debt in the EU. We provide the
first empirical test of this claim using over 233,000 responses to a
Eurobarometer question about the salience of national debt. Our
analysis reveals that national and sub-national differences explain very
little of the variance in debt preferences. Further, the differences that
do emerge do not fit existing cultural narratives. Additional analysis
reveals that established measures of national culture or religious
observance, at the national and regional levels, do not correlate with
debt attitudes as cultural arguments would predict.
KEYWORDS
Public debt; culture; debt
preferences; political
economy
JEL CLASSIFICATION
H63 debt; debt
management; sovereign
debt; Z1 cultural economics;
economic sociology;
economic anthropology
Culture plays a prominent role in European Union (EU) fiscal policy discourse. Notably, cultural argu-
ments frame differences between regions of the EU –particularly in the North and South –as the
cause of different debt burdens and ultimately debt crises. In particular, some pundits and policy-
makers have consistently cast Southern Europeans in an unflattering light, not simply as indebted,
but also as suffering from self-imposed debt crises because of fundamentally inferior values.
For example, on May 30th, 2020 the front cover of Elsevier Weekblad, a popular Dutch weekly
news magazine, portrayed Italians and Spaniards on holiday, drinking wine and coffee or relaxing
by the pool, while Dutch people were working to pay for their holidays (Wiersma, 2020). The title
expressed outrage at Southern Europeans benefiting from the hard work of Northern Europeans:
‘Not a penny more to the South of Europe.’That article was a reaction to an EU fiscal rescue
package that included debt mutualisation and a €500 billion recovery fund to help regions and
sectors hardest hit by the COVID-19 crisis. Four countries –Austria, Denmark, the Netherlands,
and Sweden –already dubbed ‘the frugal four,’were reluctant to support the deal. The group has
become known for its opposition to burden-sharing of public debt with Southern European
countries that have experienced varying degrees of fiscal crises in recent years. In this same vein,
Dutch finance minister Wopke Hoekstra opposed the EU’s so-called ‘coronabond’joint-debt plan
because it could induce moral hazard in Southern European economies (O’Leary, 2020). His alle-
gations played heavily on the cultural narrative of the prudent North and irresponsible South and
sparked a fierce backlash from political leaders in Italy and Portugal, who in turn blamed Dutch
policy on the country’s stubborn Calvinist roots (Khan, 2020).
© 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://
creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the
original work is properly cited, and is not altered, transformed, or built upon in any way.
CONTACT Matthew DiGiuseppe mdigiuseppe@gmail.com Political Science, Leiden University, 2300 RA, Leiden,
Netherlands
NEW POLITICAL ECONOMY
https://doi.org/10.1080/13563467.2022.2143490
This tendency to blame cultural attitudes for macroeconomic outcomes implies that national
culture drives voters’fiscal preferences, which in turn pushes policymakers towards disparate
public debt policy. This is not a new argument: in the aftermath of the European debt crisis, policy-
makers and some media outlets were quick to use cultural stereotypes to impose moral frames on
the causes of the crisis and the bargaining over rescue packages. Portugal, Italy, Ireland, Greece, and
Spain were referred to with the derogatory acronym ‘PIIGS,’which implied a moral judgment about
their public debt levels (Fourcade, 2013). Politicians then reinforced the stereotype: for example,
former Dutch finance minister and then-head of the European Stability Mechanism, Jeroen Dijssel-
bloem, stated this latent message explicitly, saying ‘I attribute exceptional importance to [European]
solidarity. [But] you also have obligations. You cannot spend all the money on drinks and women and
then ask for help’(Khan and McClean, 2017). Other Northern European leaders made similar com-
ments throughout the crisis, with Swedish Minister of Finance Anders Borg remarking that
‘Obviously, Swedes and other taxpayers should not have to pay for Greeks that choose to retire in
their forties. That is unacceptable’(Coleman, 2015). Similarly, Angela Merkel invoked the ‘Swabian
housewife’–a cultural icon of frugality –to contrast German virtues with Southern irresponsibility
(Kollewe, 2012). This cultural signalling was more than just discourse; it played an important role in
EU policy negotiations to resolve the debt crisis (Matthijs and McNamara, 2015), as policymakers
actively claimed that cultural differences exacerbated the severity of the crisis (Bohn and de Jong,
2011).
Furthermore, these insinuations often reflect public opinion. Indeed, surveys and media outlets
continue to show that Northern Europeans have negative perceptions of Southern Europeans,
describing them as ‘lazy,’‘debt-ridden,’and ‘tax-evaders’(Coleman, 2015). Similarly, Del Ponte
(2021) shows that cultural rhetoric powerfully influences preferences for austerity in the EU.
Support for these cultural arguments is important as public opinion directly impacts fiscal policymak-
ing. For example, Diamantopoulou and Pierrakakis (2019) argue that perceptions of the importance
that each country attaches to hard work dramatically influenced the design and implementation of
the EU’s Greek bailout programmes and Rathbun et al. (2019) demonstrate that morality played a
pivotal role in Germany’s hard-line approach.
Cultural arguments persist despite evidence that individuals in Southern European countries are
neither lazy nor fiscally irresponsible. For instance, countries of the ‘frugal four’record the lowest
average hours worked per week in the EU, while Greece records the highest (Eurostat, 2019).
1
The
image of the fiscally careless ‘PIIGS’countries is at odds with their experiences in the European
Debt Crisis during which Italy underwent the largest fiscal consolidation in the EU, followed
closely by Ireland, Portugal, and Spain (Hübscher et al., 2020). Research also shows that the role
of culture has frequently been overstated as a source of conservative fiscal policies in the North.
For example, Howarth and Rommerskirchen (2013) argue that German ‘stability culture’does not
produce fiscal preferences, but is rather invoked ex-post by German central bankers and Christian
Democrat politicians to legitimise conservative fiscal policies. Howarth and Rommerskirchen
(2017) go a step further and demonstrate that the ‘stability culture’divide between Northern and
Southern Europe is a ‘myth’in an analysis of inflation fears. However, such evidence has not
helped to overcome persistent stereotypes.
While some may view cultural narratives around fiscal preferences as a crude and likely inaccurate
portrait, cultural values are not entirely divorced from economic outcomes. A rigorous multidisciplin-
ary research agenda finds that religious and cultural differences play an important role in economic
attitudes and outcomes (Greif, 1994; Guiso et al., 2006; Acemoglu and Robinson, 2021). For example,
Friedman (2021) argues that people’s conceptions of economics are firmly rooted in religious tra-
ditions. In particular, attitudes towards debt have been traced to different religious doctrines
(Graeber, 2011; Dyson, 2014). Evidence also suggests that the effect of culture looms large among
economic policymakers (Van Esch and De Jong, 2019). Further, regional culture has been shown
to have an impact on citizens’trust in key European economic institutions (Angino et al., 2021). In
2A. ASPIDE ET AL.
short, existing research necessitates that scholars take seriously the possibility that culture shapes
economic preferences concerning fiscal policy and debt.
Despite efforts in political disagreements and academic research to link culture to economic insti-
tutions and outcomes, there is little evidence to refute or confirm that culture matters for fiscal policy
concerns broadly or within Europe. We contribute to this discussion by investigating the validity of
the cultural argument. We test whether any correlation exists between cultural explanations and
debt preferences within the EU.
2
We approach this task with the prior that popular stereotypes
are misguided, i.e. that national culture does not induce different fiscal policy preferences. In light
of this, we do not theorise about specific mechanisms through which culture would operate.
Instead, we simply test whether national and regional culture has any relationship with individual
debt preferences even under very generous specifications.
Our research design tests the effect of culture in multiple ways and in models that are charitable
to the cultural argument. We use 20 waves of the Eurobarometer survey (N = 233,953) that ask
respondents about their concern for national debt. Overall, we find that national cultural differences
play a trivial role in explaining variances in individual attitudes towards government debt. Most
importantly, we find that cross-national differences explain less than 4% of the variation in
concern for government debt burdens and that country-level differences in attitudes towards
debt are not consistent with existing North–South narratives. We then proceed to identify cultural
indicators that are plausibly related to differences in attitudes towards fiscal policy. Across a
range of tests, we fail to find evidence that these national and regional cultural measures correlate
with debt policy preferences.
Our findings have clear implications for EU debates regarding debt mutualisation and future fiscal
bailouts. First, our analysis directly counters narratives that suggest cultural differences explain debt
preferences in the EU at the citizen level. If culture plays a role in debt negotiations, it is unlikely that
it begins with the public. Second, our results speak to debates over European integration and fiscal
cooperation (Matthijs and McNamara, 2015; Dalton, 2021) as national cultural norms are often
argued to be an obstacle to agreement on efficient policies, especially during hard times (Guiso
et al., 2016). According to Matthijs (2016) and Fourcade (2013), German responsibility culture and
aversion to moral hazard impeded a timely and effective response to the Eurozone fiscal crisis.
The contrast between the German and other cultural attitudes towards fiscal policy across
Europe, such as the French emphasis on social solidarity, also hindered problem-solving in the Euro-
zone more generally (Brunnermeier et al., 2016). While leaders and policymakers continue to cite cul-
tural divides as barriers, our findings suggest that fiscal cooperation should not be threatened by
heterogeneous cultural attitudes.
We also contribute to scholarship examining the microfoundations of attitudes toward austerity
and government debt reduction. This literature points to the role of political cues, ideology, and ego-
tropic preferences as a source of attitudes toward fiscal policy and debt consolidation (Hübscher
et al., 2020,2021; Bansak et al., 2021; Barnes and Hicks, 2021; Ferragina and Zola, 2022). As
Bremer and Bürgisser (2022) explain, this once quiet area of public policy has become heavily poli-
ticised. As such, the public’s constraint (or lack thereof) on such policies is important for understand-
ing macroeconomic policy decisions.
Cultural explanations for economic outcomes
Culture is notoriously difficult to define as it captures a broad set of values, norms, and behaviours.
One commonly understood definition that we adopt is ‘those customary beliefs and values that
ethnic, religious, and social groups transmit fairly unchanged from generation to generation’
(Guiso et al., 2006). This suggests that culture is a permanent, or at least slow-moving, feature of
a society that is shared amongst members of a group or nation.
This conceptualisation considers culture to be a set of shared experiences. For example,
Putnam (1992) argued that centuries-old differences in Italian regions led to different cultural
NEW POLITICAL ECONOMY 3
attitudes toward cooperative behaviour which influenced preferences for the provision of public
goods and eventually economic growth. More recently, Galor and Özak (2016)arguethata
society’s long-term orientation was an adaptation to particular climates and associated forms of
agriculture.
Various approaches to measuring culture have led to broad empirical research linking culture
with a variety of economic outcomes. In all, such work suggests that it is plausible that culture is rel-
evant for Europe’s economic policy debates. For example, extensive literature demonstrates that
culture impacts economic outcomes in conjunction with and separately from institutions (see
Alesina and Giuliano (2015) for a review). Further, culture holds relevance in behavioural approaches
to economics. Culture is seen as a causal factor in the development of economic preferences and
behavioural traits, like temporal discounting, and thus plays a key role in economic decisions
(Wang et al., 2016; Falk et al., 2018). Other work leverages immigration as an identification strategy
to demonstrate that economic cultures and behaviours, like savings rates, accompany individuals
across borders and last multiple generations (Al-Awad and Elhiraika, 2003; Costa-Font et al., 2020).
There is also some evidence that culture correlates with macroeconomic policy outcomes. For
example, Calzada et al. (2013) demonstrate that citizens’cultural values are weakly linked to
different types of welfare states in Europe, and Pham et al. (2018)find a correlation between thrift
culture and government size at the country level.
Religion has also been used to broadly define cultural boundaries, as it is a shared experience
that shapes cultural values across generations (Lewer and Van den Berg, 2007). Similarly, research
links religious values with economic outcomes, particularly in Europe, where the evolution of
Christianity has contributed to distinct cultural regions. Weber provided one of the most influen-
tial theories attributing the success of Northern Europe and the Anglophone settler colonies to
the hard work, self-discipline, and future-orientedness that are central to Protestant culture and
Calvinism and Puritanism in particular (Weber, 2001). Others, building on Weber’s thesis, have
used religious differences to explain long-term economic growth trajectories (Gorodnichenko
and Roland, 2010)orthedevelopmentofdifferent traditions of economic thought (Friedman,
2021).
Culture, religion, and public debt
Substantial evidence links culture to economic outcomes, attitudes, and behaviours, but is culture
related to the politics of public debt? And if so, which mechanisms would link national and regional
culture and attitudes toward public debt? While the cultural narrative leads to few testable hypoth-
eses, existing research lends several plausible mechanisms.
The first possible connection is via the impact of culture on individual characteristics. Citizens’and
politicians’attitudes towards debt are potentially a product of national and regional cultural factors
that generate variance in time discounting, risk tolerance, and willingness to contribute to public
goods. While academic work has not tested these mechanisms at a cultural level (again, this argu-
ment is the provenance of politicians and popular media), some researchers have argued that
these mechanisms are influenced by cultural values, not individual characteristics.
Notably, culture influences present versus future orientation, which has clear implications for debt
given that it is, at its core, an intertemporal trade-offbetween the present and future (Wang et al.,
2016; Falk et al., 2018). Indeed, time horizons are not unique to individuals but are rather products of
one’s cultural environment (Reno, 2007). Similarly, risk tolerance has traditionally been an individual-
level characteristic implicated in the formation of fiscal policy (Ball and Mankiw, 2007), but some
have argued that this is a product of culture. Wang et al. (2016)find large differences in ‘uncertainty
avoidance’across countries and link this cultural effect to strong hyperbolic discounting that prefers
smaller, immediate, and certain rewards over later, larger, and riskier ones. In terms of debt prefer-
ences, uncertainty-avoidant societies would likely prefer balanced budgets; as events between dis-
bursement and repayment are uncertain, high debt is inherently risky.
4A. ASPIDE ET AL.
Altruism, as concern for the common rather than the individual good, can likewise be linked to
preferences for public spending as a greater willingness to contribute to public goods (Anderson
et al., 2011). However, existing studies do not have unidirectional expectations about how altru-
ism affects preferences toward public debt reduction. Higher altruism could be coupled with a
greater willingness to shoulder the burden of debt reduction for the benefit of future generations
or due to concerns about creditors’rights. Moreover, moral judgments can clash with altruistic
motives, influencing people’s views on public debt repayment (Del Ponte and DeScioli, 2021).
Scholars have argued that debt is potentially influenced by culture through religion (Graeber,
2011; Dyson, 2014). ‘Confession culture’and a reputation for both opulence among clergy and
collective identity have been proposed as reasons that Catholics might be relatively less debt
averse, while Protestantism is associated with restraint and balanced budgets (Nelsen and
Guth, 2015;ArruñadaandKrapf,2019). In other words, since Catholicism professes that sins
will always be forgiven after confession, Catholics should be expected to express less concern
about debt. Conversely, Protestantism is associated with adherence to rules and, concerning
fiscal policy preferences, avoiding moral hazard (Chadi and Krapf, 2017). Historically, Southern
Europe has been predominantly Catholic while Northern Europe has been Protestant, a divide
that loosely maps onto recent experiences of debt crises.Insummary,multiplecausalpathways
plausibly link culture to attitudes towards private and public debt, although often with contra-
dictory expectations about the nature of that relationship. In the next section, we present our
strategy to search for any evidence that cultural arguments can explain contemporary concern
for public debt in the EU.
Research design
Our goal is to test whether national culture influences individuals’public debt attitudes in the EU. For
several reasons, we approach this task with the prior that culture is not very important in the
expression of public debt preferences. While cultural explanations are prominent in popular and pol-
itical discussions, the political economy literature indicates that egotropic and sociotropic factors are
leading drivers of economic policy preferences (Mansfield and Mutz, 2009; Curtis et al., 2014). In
addition, in the EU context, there is considerable evidence that countries labelled profligate and
spendthrift behave more fiscally conservatively than cultural narratives would suggest (Howarth
and Rommerskirchen, 2013; Hübscher et al., 2020).
Given the absence of strong priors, our empirical strategy is designed to give evidence for cultural
explanations the best possible conditions to emerge. If our charitable research design finds evidence
of a cultural effect, we will narrow our focus and test the robustness of the results to confounding
and other biases.
Clearly, we cannot conclusively dismiss a relationship between two variables with econometric
methods. However, if the effect of culture is not present under favourable conditions, it will
provide a stronger argument for the absence of evidence compared to a strategy that attempts
to precisely estimate the magnitude of the cultural effect by removing all possible bias.
Our analysis consists of two approaches. First, and most importantly, we estimate the country-
level differences in individual attitudes towards public debt by assessing if differences correspond
with the implications of existing narratives. Culture is a group variable that is applied to members
of a common nation or region. As such, if culture is relevant, an individual’s geographical location
should be predictive of their fiscal policy attitudes. Consequently, we test if individuals in ‘Southern’
countries are less concerned about national debt than those in ‘Northern’countries. We then use
effect decomposition to probe how much cross-national (and cross-regional) differences matter. Fol-
lowing this central analysis, we directly estimate the correlation between established national cul-
tural variables, from three sources, that are plausibly related to debt attitudes and individual
attitudes toward debt.
NEW POLITICAL ECONOMY 5
Operationalising debt preferences
Cultural arguments about debt and deficits imply that individuals from one culture will hold different
preferences for public debt management compared to individuals from another, distinct, culture. As
such, a test of cultural explanations must rely on comparing individual assessments, whether they
are politicians or voters. Cultural effects might eventually lead to aggregate macroeconomic differ-
ences in debt levels or creditworthiness. However, such differences are insufficient to demonstrate
the effect of culture as they can be driven by other national factors, such as political or economic
institutions, or global economic factors, like the behaviour of other members of a country’s currency
union. Consequently, evidence of a cultural influence has few other places to emerge than in citizens’
preferences.
Cultural variables are, by definition, assigned to groups of individuals whether at a regional or
national level.
3
To compare citizens’opinions over the national debt, we require a cross-national
survey that is nationally representative and asks citizens from multiple countries about their prefer-
ences for public debt management. Towards this end, we rely on a question asked by the Euroba-
rometer repeatedly from 2011 to 2020 and used in previous research on public debt attitudes
(Barnes and Hicks, 2021). The question was administered to respondents in one of two branches
of the survey in each biannual wave. The question is worded as follows: ‘Measures to reduce the
public deficit and debt in (OUR COUNTRY) are not a priority for now.’
Respondents then indicated if they ‘totally agree,’‘tend to agree,’‘tend to disagree,’‘totally dis-
agree,’or ‘don’t know.’We are primarily interested in the distinction between support for and oppo-
sition to debt reduction, rather than differences between degrees of support or opposition. To ease
estimation, we collapse the answers into a binary variable indicating if the respondent agrees (totally
or tends to).
4
In the Appendix we show the geographic distribution of the variable and analysis
employing the full ordinal variable.
The question is consistently answered across multiple countries thus permitting a broad cross-
national comparison. The question is also asked across multiple waves thus reducing concerns
that our results are time-specific. While the question’s breadth and depth are helpful, its formulation
requires an adaptation to our empirical model. The question solicits two potential assessments of
debt. First, by asking whether debt is a priority, it probes respondents’assessments of the objective
reality of whether debt is a problem for the country. However, it also picks up a subjective assess-
ment regarding the urgency of addressing public debt. We are only interested in the subjective
component. Our solution to this problem is to control for a country’s objective debt troubles on
the right-hand side of the equation in our statistical models.
Primary analysis: cross-country differences
We begin our analysis by first estimating cross-country differences in individual attitudes toward
public debt reduction. If cultural preferences matter, we should observe patterns in the preferences
of individuals across countries consistent with cultural narratives (e.g. North–South divisions). As
such, we estimate a multilevel model with country-level random intercepts and a single control vari-
able: the country’s Standard & Poor’s credit rating.
5
The multilevel model allows us to estimate the
intercept of each country while also allowing a decomposition of the variance attributed to those
country differences and other factors (the residual). In our first analysis, we estimate the following
linear probability model (LPM) where Yitc is the response of respondent i, answering in year t, residing
in country c.
m
cindicates the country intercepts and S&P
b
tc is the S&P rating and its coefficient.
Yitc =
b
0+S&P
b
tc +
m
c+
e
itc
The dependent variable captures an assessment of national debt priority. As national debt poses
varied risks across the EU, country-level differences in these responses may reflect objective debt
concerns in respective countries. By including a country’s S&P rating, we estimate differences due
6A. ASPIDE ET AL.
to objective risk separately from the country intercepts. S&P’s credit rating considers various political
and economic factors that determine if a country can repay its debts, and is used by creditors across
the globe to price sovereign default risk. As such, it serves as an objective indicator of how concerned
the average citizen should be about government debt. A drop in the rating is a newsworthy signal
that public debt is cause for concern (Cawley, 2016), and is likely to cause contemporary preferences
for debt to change regardless of culture. Even if individuals are unaware of a country’s credit rating,
the rating objectively captures underlying debt sustainability that citizens are likely to come in
contact with through other sources of information. We argue that while individuals may not directly
observe these macroeconomic indicators, they have a qualitative sense of a country’s debt problems
that is driven by these underlying indicators. Figure 1 presents the sample means for each country
for both the dependent variable and the S&P ratings.
After including the rating, any remaining variation in the country intercepts should then capture
the influence of time-invariant country-level characteristics, such as culture, on individuals’assess-
ment of national debt. We show in the Appendix that the results are substantively similar when
we substitute other indicators that capture objective concerns about debt burdens, including
bond spreads, interest rates. and the debt/GDP ratio.
Figure 2 plots the country intercepts from our statistical model, where higher intercepts indicate
agreement that debt is NOT a priority. The left-hand panel shows intercepts ordered by magnitude,
with the preferences of citizens of the so-called ‘PIIGS’and the ‘Frugal Four’countries isolated at the
bottom. The right-hand panel shows intercepts ordered by the country’s average debt-to-GDP
burden throughout the period. At first glance, there are significant differences among EU countries
in their assessment of national debt; on a scale from 0-1, the distance between the extremes is 0.4.
However, this figure demonstrates that the country intercepts do not align with the most general
cultural debt argument. There is no apparent North–South divide in attitudes towards debt, nor is
Figure 1. Average of Dependent Variable and S&P Rating for each country in the sample.
Note: ‘Frugal four’and ‘PIIGS’countries are shown in bold.
NEW POLITICAL ECONOMY 7
any other geographical or cultural pattern readily discernible. The ‘PIIGS’have greater concern than
most other states. Citizens in the ‘Frugal Four’countries of Denmark and Sweden are actually less
likely to consider public debt reduction a priority, after estimating out objective credit risk.
Cross-country differences in preferences may be driven by concern over contemporaneous debt
levels. Cyprus had very high debt throughout the survey period and is most likely to regard public
debt reduction as a priority, and Estonian citizens are both least likely to consider it a priority and
have the lowest debt burden in the EU. Despite this, when country intercepts are ordered by
average debt burden in the right-hand panel of Figure 2, there is no strong pattern to suggest
that national fiscal context drives cross-country differences in preferences. Furthermore, the coeffi-
cient of the S&P rating variable is minuscule; a standard deviation increase in rating is associated with
a 1% decrease in the probability of agreeing debt is not a priority. As such, we are not very concerned
that individuals’assessments of sovereign risk influence debt attitudes.
While country differences are apparent, it is not yet clear how much they matter relative to other
factors. To answer this question, we use effect decomposition to establish how much of the variance
is driven by between-country, in contrast to within-country, differences. Table 1 presents this analy-
sis. We find that cross-country differences only account for about 3.9% of the variance in attitudes
towards debt preferences. This suggests that there are far greater differences within a country or
across time that explain attitudes towards debt than between countries. Simply, country-level differ-
ences are not very important in explaining individuals’assessments of sovereign debt problems.
Figure 2. Differences in Debt Attitudes Across Countries, Random Intercepts.
Notes: The figure displays the estimated country-random intercepts from a multilevel model in which the only other predictor is the country’s
Standard & Poor’s Credit Rating. The left-hand panel shows the random intercepts ordered by intercept magnitude within each country grouping
(‘PIIGS,’‘Frugal Four,’and others). The right-hand panel shows the random intercepts ordered by the country’s average debt-to-GDP burden
throughout the study period (from Greece at 175% to Estonia at 9%). Dots indicate the estimated intercept and bars indicate 95% confidence inter-
vals. Larger intercepts indicate that respondents in that country are more likely to agree that ‘debt is not a priority’, i.e. they do not consider public
debt reduction a priority. N = 233,953.
Table 1. Explained Variance of Country and Region.
Proportion of Variance
Country Region (NUTS II)
Geographic Unit 3.89% 7.15%
Residual 96.11% 92.84%
8A. ASPIDE ET AL.
One possible critique of our approach is that culture in Europe expresses itself regionally, rather
than nationally. As such, we consider that modern states imperfectly capture culture and that many
cultural differences exist at the sub-national level. To investigate this, we replicate our analysis by
instead estimating country-region level random intercepts based on NUTS II regions. We again
find that geographic boundaries have little explanatory value. Regional differences account for
just 7.2% of the variation in public debt attitudes.
Correlation with cultural measures
While the evidence above casts serious doubt on the idea that cultural differences have a meaningful
effect on public debt preferences, it remains possible that comparing aggregated responses at the
country (or regional) level fails to capture the mechanisms that link cultural values and debt prefer-
ences. To address this concern, we now test the relationship between individual-level debt prefer-
ences and several measures of national culture.
Cultural data
Cultural narratives around fiscal preferences are diverse and there is no standard expectation about
the precise mechanisms by which culture might influence public debt. Consistent with our effort to
give cultural arguments the best chance to present themselves, we use a variety of measures to test
our argument.
Some of the most well-known measures of national culture were established by Geert Hofstede.
His model conceives of national cultural differences along six dimensions and assigns countries a
time-invariant value for each dimension. Two of these dimensions, indulgence and long-term orien-
tation, are relevant to debates on spending and debt. A high indulgence score captures a culture
that values the free expression of impulses, while a low score captures the cultural value of restraint
and of duty and personal control (Hofstede et al., 2005). The long-term orientation score is higher in
cultures that value thrift and saving for the future, and lower in cultures that place a high value on
immediate gratification. Both indices are based on survey responses to questions in the World Values
Survey.
Next, we draw on the Human Values set of cultural variables developed by Shalom Schwartz
and outlined in Bilsky et al. (2011). They were constructed using questions from the European
Social Survey (ESS) conducted between 2010 and 2018. The dataset includes several cultural dimen-
sions. We find that only one - hedonism - has any relevance towards public debt. The hedonism score
is constructed using the weighted average of respondents’answers to two questions about personal
indulgence, scaled between 0 and 1 so that 1 represents a higher hedonism value.
6
As the ESS is
nationally representative, the aggregated responses provide cultural scores for each country.
Lastly, we draw on a data set of economic preferences collected by Falk et al. (2018,2022) from
their original Global Preferences Survey (GPS). We examine the effects of the following cultural
values: patience, risk-taking, and altruism. These variables are scaled from −1 to 1, with higher
values indicating a more positive rating. They are all constructed using weighted combinations of
qualitative survey questions to capture stated preferences and quantitative questions to capture
revealed preferences.
7
We take the country mean for each variable and use these to examine the
effect of national preferences on public debt attitudes.
In addition to broad cultural values, both policymakers and researchers have explained differ-
ences in debt preferences through the different cultural ‘treatments’of Catholicism and Protes-
tantism. As the Eurobarometer survey does not ask respondents about religion, we create
national and regional measures of Catholicism and Protestantism from the European Social
Survey data. This does not allow us to identify the Eurobarometer respondent’s relationship
with Catholicism or Protestantism but rather captures the cultural treatment of their local
context. ESS respondents identify their religion (if any), and so we aggregate survey responses
NEW POLITICAL ECONOMY 9
to the national level to measure the proportion of the population that identifies as Catholic and
Protestant in each country.
These country-level measures might obscure regional differences within countries. Many EU
countries have sub-national cultural regions, such as the majority-Catholic southern regions and
majority-Protestant northern regions of the Netherlands and Germany. As the ESS survey is nation-
ally but not regionally representative, ESS responses cannot be aggregated to the regional level to
obtain accurate measures of regional religion. Instead, we rely on the machine learning algorithm
and post-stratification strategy used in Lipps and Schraff(2021) to predict the Catholic and Protes-
tant populations in each NUTS II region.
8
While this strategy only yields religion and culture measures
for 104 regions across eight countries,
9
this more fine-grained measure allows us to test whether
sub-national culture affects individual preferences when national culture may not. Importantly,
measuring subnational differences allows us to capture potential cultural regions that do not
align with contemporary national borders, and allows us to control for national-level economic vari-
ables. Ideally, we would replicate our analyses of all cultural variables at the sub-national level.
However, the estimation of regional variables is only possible for the variables reliant on ESS
survey data.
Correlation with cultural variables
For each of the cultural variables we estimate three separate LPMs. We begin with a simple pooled
model including a single cultural variable and the S&P credit rating. This specification should offer
the best opportunity to measure any cultural effect, for a simple reason: culture can potentially
impact preferences through many channels. For example, culture can affect education through
economic growth or the value placed on education (Swank, 1996). As such, if we control for variables
like education, we would induce post-treatment bias and may underestimate potential cultural
effects. Next, since culture is assigned to every individual at birth, and the Eurobarometer samples
are representative, there is little risk that selection into the sample biases the effect in a way that
would require covariate adjustment.
Despite concerns about post-estimation bias, our second set of models adds some basic controls
plausibly related to culture and debt to address any concerns regarding the pared-down model.
Namely, we include individual-level controls for years of education, as well as whether the respon-
dent is in an urban or rural location. We include these controls given that concern for debt may vary
by level of education and urban/rural environment. We also include a set of macroeconomic factors
as control variables, notably GDP per capita and economic growth. Much like the S&P credit rating,
GDP per capita and economic growth capture the macroeconomic environment, which may lead
respondents to be more or less concerned about debt. Including these basic controls allows us to
hold constant factors that may influence attitudes towards government debt beyond culture with
the recognition that they may induce post-treatment bias.
The last set of models presents the most conservative estimate. Here, we include country-level
random intercepts to address confounding by non-cultural time-invariant factors. We begin with
the basic models with the expectation that even this specification will find no substantive effect
of culture on debt preferences. If our goal were to provide conclusive evidence of a cultural
effect, we would provide a more rigorous model to identify the cultural effect more precisely. We
present all three specifications to demonstrate the (in)consistency of our results.
Figure 3 presents the results of 18 models, each estimating one of the debt-relevant cultural vari-
ables from Falk, Hofstede or Schwartz. Starting with Schwartz’s hedonism, we find no evidence that
those living in ‘hedonistic’cultures are more likely to believe debt reduction is not a priority. In fact,
in one model, we find evidence for the opposite. Hofstede’s measure of indulgence shows a similar
pattern. The results are the opposite of what one would expect if there were a relationship between
cultural priorities in present satisfaction and citizens’attitudes towards fiscal politics.
10 A. ASPIDE ET AL.
Falk’spatience and Hofstede’slong-term orientation variables are highly relevant given the inter-
temporal dynamics of debt. Sustainable debt burdens require that individuals consider the long-
term consequences of current spending. Yet, the coefficient of long-term orientation is largely
insignificant and substantively small. Similarly, the sign and significance of patience are highly
dependent on model specification. Further, even where the relationship of patience is consistent
with expectations, the effect size is very small. Patience is expected to induce individuals to disap-
prove of debt as a longer time horizon would give more weight to debt repayment over current
spending. In the simple specification, only controlling for credit rating, individuals from countries
with higher patience values are more likely to regard public debt reduction as a priority.
However, a one standard deviation increase in aggregate national patience is associated with a
2.1% decrease in agreeing that debt is not a priority. This effect then flips signs when basic macro-
economic variables and individual attributes are added to the model. Lastly, the inclusion of random
effects wipes out any significance in the results. As such, we find that even the small effect is not
robust. Similarly, where long-term orientation shows a significant effect, the effect size is again
small as a standard deviation increase has a less than 1% decrease in the dependent variable.
Figure 3. Effect of national cultural values on individual preferences of government debt.
Notes: The figure shows the results of 18 separately estimated models estimating the relationship between a single cultural variable and
a respondent’s assessment of the priority of debt (DV(1) = ‘debt is not a priority’), controlling for the country’s credit rating. All variables are stan-
dardised (mean = 0, SD = 1). We report three basic model types: a simple LPM (green), an LPM including controls for GDP per capita, economic
growth, respondents’education, and if they live in a rural or urban area (blue), and a multilevel LPM including country random effects (red).
The dots indicate the estimated coefficient and the lines indicate the 95% confidence intervals. Schwartz models: N = 98,856. Falk models: N =
133,368. Hofstede models: N =229,202. Note that because of the addition of controls, these models have fewer observations.
NEW POLITICAL ECONOMY 11
Next, risk-taking is closely related to time preferences and patience. People in risk-taking cultures
would be expected to prefer more debt, as postponing payment is a bet on future wealth (with
uncertainty-avoiding people preferring present consumption). Despite this, our estimates indicate
that respondents in more risk-taking cultures are, if anything, more concerned about debt levels.
Again, however, the effect size is minuscule (0.02).
Lastly, we consider that altruism should capture willingness to contribute to public goods and
share fiscal burdens across society (Anderson et al., 2011), but we acknowledge that the state of
the literature leaves space for different expectations. Altruism could be associated with higher con-
cerns for future generations’fiscal space, a deeper concern for the societal consequences of default,
or a stronger perception that not repaying debt is a selfish choice that would hurt creditors. Here, we
find that reported results suggest that altruism, across all three specifications, is associated with
concern for the national debt. A standard deviation increase in the measure is associated with a
7% decrease in the DV. We find a significant correlation: people in altruistic cultures are more
likely to consider public debt reduction a priority.
Overall, these results demonstrate that dimensions of national culture have little meaningful
relationship with citizens’prioritisation of national debt. Where measures are associated with prefer-
ences, it is often in the opposite manner predicted by cultural arguments. With the contrarian results,
we are not proposing a new theory for the culture-debt relationship, but rather that these effects
indicate that cultural measures are not explanatory and that other factors explain individual
preferences.
A key feature of relying on the Eurobarometer and its 233,953 responses is that we are unlikely to
commit a type-II error. Our analysis is over-powered and so highly unlikely to report a null result
when in fact there is a significant result. In fact, such large samples provide statistically significant
results even in the presence of very small effects. All of the cultural measures have weak effects,
most in directions contrary to what would be predicted by existing cultural theories, and lose signifi-
cance when country-level random intercepts are included to address unobserved time-invariant
country characteristics. As such, we cannot say with any confidence that even the significant coeffi-
cients are substantively interesting.
Cultural effect of religion
In Figure 4, we present the correlation between debt attitudes and Catholic and Protestant obser-
vance. Again, our measures indicate the proportion of Catholics and Protestants living in the respon-
dent’s country or region and not the actual religion of those respondents. As our test of religious
culture is limited to the divide between Catholicism and Protestantism, we exclude the predomi-
nantly Orthodox Christian Greece, Cyprus, Bulgaria, and Romania which have very small Catholic
and Protestant populations. We use these measures of religion to proxy for the strength of religious
culture in the respondent’s environment. Literature on the salience of religious culture argues that
Catholicism is associated with ‘confession culture’and opulence that favours indebtedness (Nelsen
and Guth, 2015; Arruñada and Krapf, 2019), while Protestantism values thrift and positions debt as a
moral failure (Chadi and Krapf, 2017). Our results offer no support for these arguments and instead
find insignificant or contrary evidence for the effect of religious culture on debt preferences.
We repeat the sequence of LPM models used in Figure 3; here, each measure of religion is
regressed on debt preferences in separate models including first only credit rating, then additional
controls, and finally with country random intercepts. The first set of models at the top of 4 shows that
the number of people identifying as Catholic in a respondent’s country has little bearing on their
attitudes towards debt. The effect gains significance when individual and national controls are
included, but indicates that Catholic culture is associated with higher concern about debt and
remains substantively small: one standard deviation in Catholicism decreases the likelihood to
agree that debt should not be a priority by only 0.02%. Furthermore, the national Protestantism
12 A. ASPIDE ET AL.
measure is correlated with less concern for public debt across all three models, running contrary to
expectations. Again, however, the substantive effect is minuscule.
Next, we replace the national religious culture measures with the regional measures predicted by
the machine-learning method from Lipps and Schraff(2021).
10
Catholic culture at the regional level
has a more significant effect on debt preferences, but again indicates that respondents in regions
with a larger Catholic community believe that reducing the public deficit and debt is a priority.
Regional Protestant culture is associated with less concern about public debt.
These results suggest that concerns about national or regional Catholic guilt culture or Protestant
frugality are not relevant for individual public debt preferences. Again, we stress that we do not
make conclusions about individual religion and attitudes but instead about the prevailing religious
culture in an individual’s country or region.
Discussion
In sum, our findings indicate that national- and even regional-level cultural differences do not sub-
stantively determine individuals’preferences for public debt reduction. Across a wide range of spe-
cifications designed to expose any effect of culture on public opinion, we find no convincing
evidence that culture is correlated with debt policy preferences. One potential concern in this analy-
sis is that we are committing an ecological inference problem by using group membership to predict
individual characteristics. However, there are several reasons why this issue does not apply to our
analysis.
First, we are not using one group’s characteristics to predict another group’s characteristics;
instead, we are estimating the effect of living within cultural boundaries on policy attitudes. More-
over, we argue, consistent with the literature, that culture is not an individual-level characteristic but
is instead shared by members of a group (Brewer and Venaik, 2014). Second, if our analyses commit
an ecological fallacy, it is because the arguments we are addressing are built on the same fallacy.
Testing them involves applying the same assumptions upon which the arguments are built. The
fact that cultural arguments are based on this fallacy is one of the reasons, we argue, that we find
no evidence of this relationship.
Next, we are careful to point out that our analysis does not rule out that ideologies impact econ-
omic policymaking. Our study specifically examines national attitudes. It remains possible that
Figure 4. Effect of Catholic and Protestant values at both the national and regional levels.
Notes: The figure shows the results of 12 separately estimated models, estimating the relationship between a single cultural variable and respon-
dent’s assessment of the priority of debt (DV(1) = ‘debt is not a priority’) controlling for the country’s credit rating. All variables are standardised
(mean = 0, SD = 1). We report three basic model types: an LPM including controls for GDP per capita, economic growth, respondents’education,
and if they live in a rural or urban area (blue), and a multilevel LPM including country random effects (red). The dot indicates the estimated coeffi-
cient and the lines indicate the 95% confidence intervals. Models estimating the effect of national religion: N = 88,521. Models estimating the effect
of regional religion (using BART-predicted religion measure): N = 16,910.
NEW POLITICAL ECONOMY 13
culture operates via policymakers who are exposed to specific schools of thought, influenced by par-
ticular economic events (Oudenampsen and Mellink, 2022), or bring their national culture to Euro-
pean policymaking (Van Esch and De Jong, 2019).
Conclusion
Arguments that economic and political differences across societies can be explained by cultural
divides have a long tradition dating back centuries. Recent crises have seen such narratives deployed
to justify fiscal policy in the EU and lay the blame for unsustainable debt at the feet of certain cul-
tures. Despite the prominence of these arguments in EU fiscal policy discourse, there is little empiri-
cal evidence to establish their validity. We set out in search of any evidence above. None of our
empirical tests, even the most generous, find a meaningful correlation between culture and individ-
ual concern for debt. At the most basic level, debt preferences are not substantially different in
Northern and Southern Europe or even between ‘frugal four’and ‘PIIGS’countries. Debt policy pre-
ferences do not correlate with national or regional cultural variables in any manner consistent with
cultural explanations.
Again, we cannot completely disprove cultural arguments, but the absence of evidence for them,
under favourable conditions, casts serious doubt on the salience of culture for public debt prefer-
ences. Our findings suggest that differences between countries and cultural groups are small and
that differences within cultures or countries (for example, political ideology) provide more leverage
in explaining individuals’preferences for public debt and deficits. This work has clear implications for
EU policy discourse where cultural arguments have been used to resist debt mutualisation and fiscal
rescue packages during both the Eurozone crisis and the COVID-19 pandemic. We suggest that pol-
icymakers are misguided in relying on cultural differences to resist fiscal cooperation and that culture
is not a barrier to successful debt policy.
Notes
1. For example, Greece records 41.7 and the Netherlands records 30.4 h on average worked per week.
2. The United Kingdom is included in the definition of ‘EU countries’in this study as it was part of the EU in the time
period that we analyse.
3. Conceptually, culture could be measured at the individual level, but this is impossible to separate from environ-
mental effects like upbringing or other life experiences.
4. We drop ‘don’t know’answers. The Eurobarometer also asked another question about debt preferences to half
of the respondents in another branch of the survey. We omitted this question because it violates survey design
best practices, which we discuss in the Appendix. We also replicate our main analyses using the alternate survey
question. Our results are unchanged.
5. Credit rating data from Ballard-Rosa et al. (2019).
6. For additional information on the questions used to construct the index, see Bilsky et al. (2011).
7. ’Patience’is constructed by combining the intertemporal choice sequence using the ’staircase’method (quan-
titative item) with a respondent’s self-assessment of their willingness to wait, across increasing monetary trade-
offs between now and later. Similarly, ’risk-taking’combines a lottery choice sequence using the staircase
method (quantitative item) with a self-assessment of the willingness to take risks (qualitative item). ’Altruism’
combines a quantitative question about a donation decision and a qualitative self-assessment of the willingness
to donate to good causes. For additional information on the methodology used to construct the measures and
the exact wording of the questions see the Appendix or consult Falk et al. (2022,2018).
8. This implementation of a Bayesian additive regression tree (BART) is a recent advancement to the challenge of
noisy regional data in nationally representative surveys. Interactive effects between individual and national
characteristics are predicted across partitions in the data and used to adjust national data at the regional
level. We discuss the method further in the Appendix.
9. Denmark, France, Germany, Netherlands, Poland, Portugal, Spain, United Kingdom.
10. Note that these models make use of a more limited sample size and estimate the effect of regional religious
culture in 8 countries. See the Appendix for more details on the estimation.
14 A. ASPIDE ET AL.
Acknowledgement
The authors would like to thank Adina Akbik and the two anonymous reviewers for their comments and suggestions
and Alexandre Diogo and Nina Waals for their research assistance.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
This work was supported by the HORIZON EUROPE European Research Council under grant #852334).
Notes on contributors
Alessia Aspide is a PhD candidate in political science at the Institute of Political Science at Leiden University.
Kathleen J. Brown is a PhD candidate in political science at the Institute of Political Science at Leiden University.
Matthew DiGiuseppe is Associate Professor of International Relations at the Institute of Political Science at Leiden
University.
Alexander Slaski is Visiting Assistant Professor at the Department of Government at Georgetown University.
ORCID
Alessia Aspide http://orcid.org/0000-0001-6013-7445
Kathleen J. Brown http://orcid.org/0000-0003-3187-175X
Matthew DiGiuseppe http://orcid.org/0000-0003-0823-0436
Alexander Slaski http://orcid.org/0000-0001-9335-055X
References
Acemoglu, D., and Robinson, J.A. 2021. Culture, institutions, and social equilibria: A framework. Working Paper 28832,
National Bureau of Economic Research.
Al-Awad, M., and Elhiraika, A., 2003. Cultural effects and savings: evidence from immigrants to the United Arab Emirates.
Journal of development studies, 39 (5), 139–151.
Alesina, A., and Giuliano, P., 2015. Culture and institutions. Journal of economic literature, 53 (4), 898–944.
Anderson, L.R., DiTraglia, F.J., and Gerlach, J.R., 2011. Measuring altruism in a public goods experiment: a comparison of
U.S. and Czech subjects. Experimental economics, 14 (3), 426–437.
Angino, S., Ferrara, F.M., and Secola, S., 2021.The cultural origins of institutional trust: The case of the European central
bank. European Union Politics.
Arruñada, B., and Krapf, M., 2019. Religion and the European Union. In: Advances in the economics of religion. Springer,
295–308.
Ball, L., and Gregory Mankiw, N., 2007. Intergenerational risk sharing in the spirit of Arrow, Debreu, and Rawls, with appli-
cations to social security design. Journal of political economy, 115 (4), 523–547.
Ballard-Rosa, C., Mosley, L., and Wellhausen, R.L., 2019. Contingent advantage? sovereign borrowing, democratic insti-
tutions and global capital cycles. British journal of political science,1–21.
Bansak, K., Bechtel, M.M., and Margalit, Y., 2021. Why austerity? the mass politics of a contested policy. American political
science review, 115 (2), 486–505.
Barnes, L., and Hicks, T., 2021. All Keynesians now? public support for counter-cyclical government borrowing. Political
science research and methods, 9 (1), 180–188.
Bilsky, W., Janik, M., and Schwartz, S., 2011. The structural organization of human values-evidence from three rounds of
the European social survey (ess). Journal of cross-cultural psychology, 42, 759–776.
Bohn, F., and de Jong, E., 2011. The 2010 euro crisis stand-offbetween France and Germany: leadership styles and pol-
itical culture. International economics and economic policy, 8 (1), 7–14.
Bremer, B., and Bürgisser, R., 2021. Do citizens care about government debt? evidence from survey experiments on bud-
getary priorities. European journal of political research.
Brewer, P., and Venaik, S., 2014. The ecological fallacy in national culture research. Organization studies,35 (7), 1063–
1086.
NEW POLITICAL ECONOMY 15
Brunnermeier, M.K., James, H., and Landau, J.-P., 2016. The euro and the battle of ideas. In: The euro and the battle of
ideas. Princeton University Press.
Calzada, I., et al., 2013.Regımenes de bienestar y valores en europa.Revista Española de Investigaciones Sociológicas (REIS),
141 (1), 61–90.
Cawley, A., 2016. From AAA to junk. Journalism studies, 17 (5), 647–666.
Chadi, A., and Krapf, M., 2017. The protestant fiscal ethic: religious confession and euro skepticism in Germany. Economic
inquiry, 55 (4), 1813–1832.
Coleman, J. 2015. Greek bailout talks: Are stereotypes of lazy greeks true? https://www.bbc.com/news/world-europe-
31803814.BBC News.
Costa-Font, J., Giuliano, P., and Ozcan, B. 2020. The cultural origin of saving behavior. Working Paper DP14413, CEPR.
Curtis, K.A., Jupille, J., and Leblang, D., 2014. Iceland on the rocks: the mass political economy of sovereign debt reset-
tlement. International organization, 68 (3), 721–740.
Dalton, R.J., 2021. National/European identities and political alignments. European union politics, 22 (2), 340–350.
Del Ponte, A., 2021. The influence of foreign elite rhetoric: national identity, emotions, and attitudes toward austerity.
European union politics, 22 (1), 155–178.
Del Ponte, A., and DeScioli, P., 2021. Pay your debts: moral dilemmas of international debt. Political behavior,1–24.
Diamantopoulou, A., and Pierrakakis, K., 2019. Work values in politics: the European Union debt crisis as a case study. The
ANNALS of the american academy of political and social science, 682 (1), 222–233.
Dyson, K., 2014.States, debt, and power: ‘saints’and ‘sinners’in European history and integration. Cambridge, UK: Oxford
University Press.
Eurostat. 2019. Hours of work - annual statistics. https://ec.europa.eu/eurostat/ statistics-explained/index.php?title =
Hours_of_work_-_annual_statistics#:∼: text = In%202019%2C%20the%20EU%2D27.
Falk, A., et al. 2016. The preference survey module: a validated instrument for measuring risk, time, and social prefer-
ences. IZA Discussion Paper No. 9674, (4).
Falk, A., et al., 2018. Global evidence on economic preferences. The quarterly journal of economics, 133 (4).
Ferragina, E., and Zola, A., 2022. The end of austerity as common sense? an experimental analysis of public opinion shifts
and class dynamics during the COVID-19 crisis. New political economy, 27 (2), 329–346.
Fourcade, M., 2013. The material and symbolic construction of the BRICS: reflections inspired by the ripe special issue.
Review of international political economy, 20 (2), 256–267.
Friedman, B.M., 2021.Religion and the rise of capitalism. Knopf.
Galor, O., and Özak, Ö, 2016. The agricultural origins of time preference. American economic review, 106 (10), 3064–3103.
Gorodnichenko, Y., and Roland, G. 2010. Culture, institutions and the wealth of nations. Working Paper 16368, National
Bureau of Economic Research.
Graeber, D., 2011.Debt: the first 5000 years. Brooklyn: Melville House.
Greif, A., 1994. Cultural beliefs and the organization of society: a historical and theoretical reflection on collectivist and
individualist societies. Journal of political economy, 102 (5), 912–950.
Guiso, L., Herrera, H., and Morelli, M., 2016. Cultural differences and institutional integration. Journal of international
economics, 99, S97–S113.
Guiso, L., Sapienza, P., and Zingales, L., 2006. Does culture affect economic outcomes? Journal of economic perspectives,
20 (2), 23–48.
Hofstede, G., Hofstede, G.J., and Minkov, M., 2005.Cultures and organizations: software of the mind, Volume 2. New York:
McGraw-hill.
Howarth, D., and Rommerskirchen, C., 2013. A panacea for all times? the German stability culture as strategic political
resource. West european politics, 36 (4), 750–770.
Howarth, D., and Rommerskirchen, C., 2017.Inflation aversion in the European Union: exploring the myth of a north-
south divide. Socio-economic review, 15 (2), 385–404.
Hübscher, E., Sattler, T., and Truchlewski, Z., 2020. Three worlds of austerity: voter congruence over fiscal trade-offsin
Germany, Spain and the UK. Socio-economic review Forthcoming.
Hübscher, E., Sattler, T., and Wagner, M., 2021. Voter responses to fiscal austerity. British journal of political science, 51 (4),
1751–1760.
Khan, M. 2020. Dutch pm faces dissent at home over hardline coronabonds stance. https://www.ft.com/content/
b65da5ba-7873-4d99-8fc9-589d51800f2e.Financial Times.
Khan, M., and McClean, P. 2017. Dijsselbloem under fire after saying euro-zone countries wasted money on alcohol and
women. https://www.ft.com/content/ 2498740e-b911-3dbf-942d-ecce511a351e. Financial Times.
Kollewe, J. 2012. Angela Merkel’s austerity postergirl, the thrifty Swabian housewife. https://www.theguardian.com/
world/2012/sep/17/ angela-Merkel-austerity-Swabian-housewives. The Guardian.
Lewer, J.J., and Van den Berg, H., 2007. Religion and international trade: does the sharing of a religious culture facilitate
the formation of trade networks? American journal of economics and sociology, 66 (4), 765–794.
Lipps, J., and Schraff, D., 2021. Estimating subnational preferences across the European Union. Political science research
and methods, 9 (1), 197–205.
16 A. ASPIDE ET AL.
Mansfield, E.D., and Mutz, D.C., 2009. Support for free trade: self-interest, sociotropic politics, and out-group anxiety.
International organization, 63 (3), 425–457.
Matthijs, M., 2016. Powerful rules governing the euro: the perverse logic of German ideas. Journal of european public
policy, 23 (3), 375–391.
Matthijs, M., and McNamara, K., 2015. The euro crisis’theory effect: Northern saints, southern sinners, and the demise of
the eurobond. Journal of european integration, 37 (2), 229–245.
Nelsen, B.F., and Guth, J.L., 2015.Religion and the struggle for European Union: confessional culture and the limits of inte-
gration. Georgetown University Press.
O’Leary, N. 2020. Fierce backlash to Dutch hard line on Eurobonds. https://www.irishtimes.com/business/economy/
fierce-backlash-to-dutch-hard-line-on-eurobonds-1.4217157. The Irish Times.
Oudenampsen, M., and Mellink, B., 2021. The roots of Dutch frugality: the role of public choice theory in Dutch budget-
ary policy. Journal of european public policy,1–19.
Pham, H.T., Carmignani, F., and Kler, P., 2018. Thrift culture and the size of government. Economic modelling, 70, 571–
578.
Putnam, R., 1993.Making democracy work: civic traditions in modern Italy. Princeton University Press.
Rathbun, B.C., Powers, K.E., and Anders, T., 2019. Moral hazard: German public opinion on the Greek debt crisis. Political
psychology, 40 (3), 523–541.
Reno, B.J., 2007. Rethinking horizon theory: culture vs. nature. Perspectives on political science, 36 (2), 84–90.
Swank, D., 1996. Culture, institutions, and economic growth: theory, recent evidence, and the role of communitarian
polities. American journal of political science, 660–679.
Van Esch, F., and De Jong, E., 2019. National culture trumps EU socialization: the European central bankers’views of the
euro crisis. Journal of european public policy, 26 (2), 169–187.
Wang, M., Rieger, M.O., and Hens, T., 2016. How time preferences differ: evidence from 53 countries. Journal of economic
psychology, 52, 115–135.
Weber, M., 1930.The protestant ethic and the spirit of capitalism. Routledge.
Wiersma, J. 2020. Geen stuiver extra naar zuid Europa. https://www.ewmagazine.nl/ economie/achtergrond/2020/05/
geen-stuiver-extra-naar-zuid-europa-207225w/. Elsevier Weekblad.
NEW POLITICAL ECONOMY 17
