Cultural Values and Economic Resilience: An Empirical Analysis

_______________________________________

Cultural Values and Economic Resilience: An Empirical Analysis ______________________________________________

Keith Tanti Special acknowledgement is due to Prof. Lino Briguglio for the expert guidance and advice he offered me throughout the whole process of this study.

1

Abstract This study sets out to go beyond the analysis of already recognised determinants of a nation's economic resilience by exploring possible cultural dimensions of economic resilience. The latest economic crisis in Europe is used as a context for testing the hypothesis that strong work ethic values enhance national economic resilience. For this purpose, a Cultural Values Index (CVI) incorporating four work ethic dimensions was constructed;with scores for 43 European countries being obtained from the 2008 European Values Study. The average annual ‘recovery growth’ manifested by the 43 countries over the 2008-2013 period was regressed using the Ordinary Least Squares (OLS) method on the CVI and other required control variables.The results suggest that cultural values did indeed have a significant effect on countries’ economic resilience, implying that policymakers seeking to enhance their nation’s economic resilience need to look at both economic and cultural aspects of economic resilience.

JEL Classification: O10, Z10, Z13

2

1. Introduction 1.1 Background to the Study Economic resilience, the nurtured ability of an economy to withstand or cope with adverse shocks (Briguglio et al. 2006), is an essential element for development in any economy. Strong economic resilience is desired by developed as well as by developing countries. The recent global economic crisis highlighted just how important economic resilience is. Some economically powerful nations were brought down to their knees, while other economically vulnerable nations managed to emerge from the crisis largely unscathed. This would suggest that there are certain important factors which determine a county's economic resilience to negative shocks besides their inherent vulnerability. The recent economic crisis provides the opportunity to examine further what these factors may be.

Past studies on economic resilience have focused mostly on purely economic determinants of economic resilience, with little importance being assigned to cultural values. This study seeks to take a different perspective by testing for possible cultural determinants of economic resilience. As in Granato et al. (1996, p.608), in this study culture is defined as"a system of basic common values that help shape the behaviour of people in a given society". While extensive research has been performed on the effects of cultural values on long-run economic development, none has focused on their effects on economic resilience in times of crisis. The effect of cultural values on an economy could well be more influential in times of crisis, and therefore, analysing them in this context may provide more meaningful results. This research attempts to contribute to economic literature by working on this gap in the literature.

Do cultural values have an effect on a country's economic resilience? This is the main question that this research seeks to answer. This question will be analysed in the context of the latest economic crisis in Europe. The analysis is confined solely to Europe since it is the most economically integrated region

3

(therefore less idiosyncrasy in economic shocks), but with still vastly varying distinctive cultures. The research hypothesis is that those European countries whose populations possess strong work ethic values were more economically resilient to the last crisis.A cross-sectional empirical investigation will attempt to assess the validity of this hypothesis.

1.2 Outline A review of the literature is carried out in Chapter 2 to identify those values which, according to the literature, could have an impact on economic resilience. To test the study's hypothesis, a Cultural Values Index (CVI) proxying the strength of these values will be constructed, and scores for 43 European countries will be obtained from survey responses to the 4th wave (2008) of the European Values Study (EVS). To test the significance of these values, this index will be regressed (together with other control variables) as an explanatory variable in an Ordinary Least Squares (OLS) regression, with the dependent variable being a measure of actual economic resilience manifested during the crisis – ‘Mean Annual GDP Growth over 2008-2013’. Chapter 3 will delineate how the Cultural Values Index was constructed, and present the other variables as well as the estimation procedure used to test the study’s hypothesis. The estimation results and some further analysis are then presented in Chapter 4. Chapter 5 will conclude with a discussion of the resulting implications.

4

2. Literature Review 2.1 The Influence of Sociocultural Factors on Economic Development. Cultural explanations for economic phenomena have been discussed since the classical economists. The first systematic investigation of the effect of (religious) values on economic outcomes was however Max Weber's 'The Protestant Ethic and the Spirit of Capitalism' (1905). In this work, Weber contends that the rise of the industrial capitalistic system is largely attributable to Protestant values, particularly those pertaining to the Calvinist and Puritan sects. According to Weber, it is no coincidence that the earliest forms of industrial capitalism originated in predominantly Protestant nations.Many different interpretations have been attributed to Weber's theory but the most 'common interpretation' (Delacroix &Nielsen 2001)is that new attitudes were fostered by the Protestant Reformation which in turn affected behaviours. These new attitudes and behaviours favoured economic development and contributed to the rise of industrial capitalism (Delacroix &Nielsen 2001).

Of particular relevance to this study are the values and attitudes which Weber’s theory identifies as possible conducers to economic development. Weber's central argument is that Protestants developed a distinct work ethic. This work ethic arose mainly from the Calvinist predestination doctrine which asserts that God predestined those elected for salvation and those who are damned. Success in business and worldly affairs was seen as a sign of possible inclusion amongst the 'elected'. Thus, the Calvinist doctrine gave hard work a divine dignity (Schaltegger&Torgler 2010). It asserted that hard work and frugality evidenced possibility of salvation, while idleness and waste of time meant certain condemnation (Modrack 2008). Therefore, at the heart of the Protestant Work Ethic lie favourable attitudes towards hard work and frugality, together with strict avoidance of any idleness or waste of time (Modrack 2008). Following an extensive review of the literature, Modrack (2008, p.6) asserts that it has “become

5

common sense” that work ethic is made up of four constructs: (i) 'belief in hard work'; (ii) 'leisure avoidance'; (iii) 'independence from others'; and (iv) 'asceticism'. These are the beliefs which could have contributed to economic development and which, in this study's perspective, could have an impact on economic resilience as well.

Weber's theory immediately sparked off debate, controversy and extensive research. Besides the empirical battlefield on the validity of the PWE theory, two main points of general consensus emerge from an analysis of the literature:

i)

Values can indeed have significant influence on economic phenomena, and thus deserve

consideration in analyses of economic development. Ntibagirirwa (2009) asserts that cultural values are important factors which need to be taken seriously for the achievement of economic growth and development. In 'TheAchieving Society', McClelland (1969) demonstrated using quantitative methods how certain sociological and psychological factors, mainly the need for achievement (nAch), are generally important for economic development. On similar lines, Modrack (2008) affirms that approaches similar to that undertaken by Weber are not outdated and continue to matter. According to Modrack, such an approach could provide valuable insights on the interactions between culture, institutions and economic outcomes. Granato et al. (1996) have gone as far as to demonstrate that an empirical growth model that incorporates both cultural and economic variables is superior to one which utilises only one set of these variables.

ii)

Any analysis of the effect cultural values on the economy does no longer need to be

necessarily linked with religious affiliation. Some of the literature has retained Weber's original focus on religiously specified sets of values; searching relationships between national religious orientations and economic outcomes (Guiso et al.

6

2006; Schaltegger&Torgler 2010; van Hoorn &Maseland 2013). However, while in pre-industrial times values were closely tied to religion, with the process of industrialisation and modernisation values have become more rational and secular (Granato et al. 1996). Miller et al. (2002, p.453) state that research has “failed to find a consistent relation between religious orientation and work ethic beliefs”. This study, therefore, takes Furnham et al.'s (1995, p.262) position that it would be “more accurate to talk about work values and beliefs rather than the Protestant Work Ethic itself”.This study's analysis therefore, focuses mainly on the effect of culturally specified values and beliefs rather than those of Protestantism or of any other religion.

2.2 Economic Resilience and its Determinants. The majority of the literature in cultural economics has focused on analysing whether values have any impact on the long-run economic growth of a nation or region. There have been no studies as yet that have analysed whether cultural values could have an impact on economic resilience. Briguglio et al. (2006) describe economic resilience as the nurtured ability of an economy to recover from or adjust to adverse shocks to which it may be inherently exposed. Building on this concept of resilience, the authors construct an Economic Resilience Index (ERI) composed of four dimensions: i)

Macroeconomic Stability

ii)

Microeconomic Market Efficiency

iii)

Good Governance

iv)

Social Development.

Baritto (2009) proposed an alternative proxy of resilience – Net Savings per capita. Although Briguglio's (2006) and Baritto’s (2009) indicators cover different aspects, a strong correlation between them was found (r=0.84) given that they are both strongly positively related to income levels. These estimates of economic resilience incorporate factors which are rightly deemed as important determinants of an

7

economy's ability to withstand adverse shocks. However, none of them takes into consideration the possibility that cultural values could in some way affect economic resilience.

2.3 How can Cultural Values affect Economic Resilience? Just as there has been no research on the effects of cultural values on economic resilience, there has neither been any systematic study of the possible channels through which the two can influence each other. Some studies, however, have demonstrated how certain values do affect certain economic behaviours and outcomes. I will now review some of the ways indirectly suggested by literature through which values can affect economic resilience; thus supporting the rationale of my hypothesis.

A strong work ethic, and its four constructs identified by Furnham (1990) and Modrack (2008) - 'belief in hard work'; 'leisure avoidance'; 'independence from others'; and 'asceticism' - are likely to be influential in times of economic crisis. Entrepreneurs and workers who possess these values are likely to exert greater efforts to protect their work and income in times of economic adversity. Workers whose employment is threatened by an economic crisis will seek to secure their workplace by working harder and by being more productive. Those possessing a strong work ethic will seek to work harder even more, even if it means less spare time, since work is a very central aspect of their life.

Another identified channel through which cultural values can affect economic resilience is through the well-being effects of unemployment. Van Hoorn &Maseland (2013) showed that unemployment hurts Protestants more and hurts more in Protestant societies. Hence, it can be expected that in those cultures with a vigorous work ethic, unemployment is going to hurt more. In such cultures, those who become unemployed during a crisis could be expected to: search more actively for a job; accept offered jobs more quickly (lower search unemployment); be more ready to start a self-employment business; be less willing to rely solely on social security benefits; and be less willing to withdraw from the labour force

8

(become discouraged workers). These tendencies could all have positive effects on economic resilience.

Attitudes amongst entrepreneurs could also have an effect on economic resilience. In an experimental study, McClelland (1969) showed how businessmen who underwent achievement motivation development courses subsequently demonstrated heightened entrepreneurial activity. Those businessmen with higher need for achievement (strongly linked with the work ethic) were those who sought improvements of their existing business, expansions into new product lines and industries, and collaborations on new joint enterprises. Such actions during times of economic adversity could very well enhance the economic resilience of a nation.

Another channel through which values could affect resilience is through the savings effect. The asceticism promoted by a vigorous work ethic can lead to higher national savings (Guiso et al. 2006) which, as suggested by Baritto (2009), is an important determinant of an economy’s resilience.

2.4 The Interaction between Cultural Values and Economic Performance An important final consideration to be made is about the interaction between cultural values and economic performance. This study seeks to examine the effect of values on a particular aspect of economic performance – economic resilience to crises. There is evidence, however, that values are in turn influenced by the previous economic performance of a country. Inglehart (1990) asserts that the shift from materialist to postmaterialist value priorities starts to emerge when a nation attains relatively high levels of economic security. This gradual shift in value priorities, with a reduced emphasis on economic achievement, represents an erosion of the work ethic. Supporting this theory, Furnham et al. (1993) reported that people from more developed countries tend to have lower scores on work ethic scales. It can therefore be expected a priori, that the countries’ level of social (and economic) resilience manifested throughout the Great Recession was influenced by their respective stage of development.

9

This two-way interactive relationship between culture and the economy could also imply that the recent economic crisis could lead to significant cultural changes in the near future.

10

3. Variables and Estimation Procedure 3.1 The Estimation Procedure In order to test the study's hypothesis, a Cultural Values Index (CVI) capturing important socio-cultural values which have been recognised by the literature as possibly influential on economic resilience was constructed.

The effect of values on economic resilience was then analysed by regressing in a cross-sectional OLS regression a measure of actual manifested economic resilience during the crisis period (2008-2013) on the CVI and other required control variables. Ideally, panel data estimation techniques would have provided more informative results on the interactive relationship between cultural values and economic resilience. However, given that no other EVS has been conducted yet since the one in 2008, using panel data estimation techniques was not possible, and so OLS had to be used.

The variables described in this chapter are:  The Cultural Values Index (CVI): The Independent Variable  Mean Annual Growth Rate in GDP per capita PPP (08-13) (Recovery Growth): The Response Variable  GDP per capita at PPP 2007 (GDP07): Control Variable  The Economic Resilience Index (ERI): Control Variable  The Economic Vulnerability Index (EVI): Control Variable

11

3.2. The Cultural Values Index (CVI): The Independent Variable To test whether values had an effect on European countries' economic resilience to the last crisis, a Cultural Values Index intended to capture the values (and their associated effects) indentified in the literature was constructed. What follows is a description of the data sources and methods used in the construction of the CVI, which is the independent variable of main interest in this study. 3.2.1

Data Source

The data for the CVI were obtained using survey responses from the fourth wave of the European Values Study (EVS) conducted in 2008. The EVS is a comprehensive cross-national and longitudinal survey research program providing insights into the basic human values, ideas, beliefs, attitudes and opinions of citizens all over Europe (Europeanvaluesstudy.eu 2015). The surveys for the fourth wave of the EVS were conducted between May and November of 2008, which is an ideal timing for this research since it roughly coincides with the start of the economic crisis in Europe. CVI scores for 43 European countries were obtained using the EVS’ survey responses1.

3.2.2

Dimensions of the Index

The Cultural Values Index utilises responses to eight questions in the 2008 EVS to cover four cultural dimensions. Their focus is dominantly on the work ethic concept identified by Weber, and its constructs which were discerned and developed by subsequent researchers. The four dimensions are: i)

Work Centrality

ii) Independence iii) Thrift & Belief in Hard Work 1

The fourth wave of the EVS covers 47 countries but scores to only 43 countries were given due to the unavailability of data for 4 countries - Belarus, Northern Ireland, Kosovo & Northern Cyprus - for one or more of the other variables used in the study.

12

iv) Achievement Motivation What follows is a description of these dimensions, the rationale for their inclusion and a description of how they were quantified using EVS data. These are then summarised in Table 3.1 (pg 22).

i)

Work Centrality

The concept that work lies at the very heart of the life of an individual with a strong work ethic is a central aspect of Weber's PWE theory. 'Leisure avoidance' was also one of the five work ethic constructs identified by Furnham's (1990) factor analysis. Work centrality can influence economic resilience since those who value work highly are more likely to exert greater productivity efforts to maintain their workplace during a crisis, or to spend longer working hours if need be. Also, they are likely to search more actively for alternative employment if they happen to lose their job during the crisis. This dimension is thus intended to capture the importance of work in people's life in a given society. It will be measured from responses to two survey questions in the 2008 EVS: 'How important is work in your life?’ and 'Should work always come first even if it means less spare time?' Since work centrality is a key feature of the work ethic it was assigned a weight of 40% within the final Cultural Values Index.

ii)

Independence

Another work ethic construct identified by Furnham (1990) is independence. People who value personal independence desire to maintain control over their own welfare without depending on others. This dimension is likely to be influential on economic resilience as well since such independent persons would be less willing to depend on social security benefits when they lose employment during a crisis. They are instead likely to actively search for alternative means of sustaining themselves, such as by setting up an independent self-employment business. This dimension will be quantified from two survey questions of the EVS:'Do you consider independence as an especially important quality which children should be encouraged to learn at home?’ and 'Is it humiliating to receive money without having to work for

13

it?’Although the first question does not directly ask the respondentsif they personally value independence, it is deemed reasonable to assume that generally they would not answer positively to such a question unless they do themselves value independence. The second question is intended to reflect the society's reluctance to rely on social security benefits. The 'Independence' dimension holds a weight of 20% within the final composite index.

iii)

Belief in Hard Work & Thrift

According to Weber (1930) two of the key Protestant principles that facilitated economic development in Protestant societies were the devotion to hard work so as to signal the 'calling' and the conduction of a simple self-disciplined ascetic life that encouraged high savings. Indeed, Furnham's factor analysis (1990) found 'belief in hard work' and 'asceticism' to be two of the constructs of the work ethic. The belief that hard work will ultimately bring success (in this case, economic success, not religious) and help overcome any crisis, can produce a hard-working productive society which is more economically resilient. This belief was proxied using the responses to the EVS question: 'Do you consider ‘hard work’ as an especially important quality which children should be encouraged to learn at home?' Quantifying the attitudes towards thrift through the EVS was considered as the best means to proxy ascetic values. Positive valuations of thrift encourage careful wealth management and high saving rates which can in turn enhance a nation's economic resilience (Baritto 2009). Attitudes towards thrift were quantified from people's responses on whether thrift is a quality which children should be encouraged to learn at home. Again, the questions were deemed to give a reasonably accurate reflection of the respondents' own personal beliefs. The weight assigned to this dimension was that of 20%.

iv)

Achievement Motivation

Achievement motivation is another important determinant of economic performance identified by the literature. Research by McClelland (1969) found that the need for achievement (nAch) is an important

14

determinant of economic growth. Similarly, Granato et al. (1996) found cultural attitudes toward achievement to have a positive effect on economic growth. The EVS question on whether children should be encouraged to learn determination/perseverance qualities at home was used to measure these qualities in a given country.

Previous research demonstrates that the shift from materialist to postmaterialist value priorities is linked with a shift away from economic achievement motivation towards the motivation for the achievement of a better quality of life more generally (Inglehart 1990; Abramson &Inglehart 1995). It was therefore decided to incorporate in this dimension also a measure of the extent to which a society gives priority to materialist values concerned with physical and economic security over postmaterialist values concerned with self-expression and self-actualisation. The EVS question 'Which aim of the country is most important?' was used for this purpose. This question was specifically designed and used by Inglehart (1990) to measure the materialist-postmaterialist dimension. Those who answered 'maintaining order in the country' or 'fighting rising prices' were considered to have materialist value priorities, while those who answered 'giving people more say in government decisions' or 'protecting freedom of speech' were considered to have post-materialist value priorities. The achievement motivation index accounts for the remaining 20% of the overall index. 3.2.3

Index Construction Method

The Cultural Values Index is composed of four dimensions, each incorporating two questions from the 2008 EVS. The two questions have equal weighting within each dimension. The questions have different response formats and, therefore, different scoring methods were used: 

For those questions on whether certain qualities should be learned at home (hard work,

independence, thrift & perseverance), the percentage of respondents who mentioned that these qualities should be learned was taken as the score.

15



The two questions in the ‘centrality of work’ dimension and the question on whether it is

humiliating to receive money without working had scaled response categories. For these questions, a weighted mean score was calculated by assigning weights proportional to the scale values.



The percentage of respondents who gave priority to materialist values (order in the nation or

fighting rising prices) was taken as the score for the materialist-postmaterialist question in the achievement motivation dimension.

3.2.4

Rescaling Procedure

Since the scores were calculated in different ways, the scores were normalised so that they are all rescaled to a common scale of zero to one. The following feature scaling method transformation was used: XNij = (Xij – Minj) / (Maxj – Minj) where: XNij is the value of the normalised score for country i for question j; Xij is the actual score for country i in question j; Minj and Maxj are the minimum and maximum scores for question j. One of the possible drawbacks of this technique is that if there is an outlier in the data, then the other 'normal' scores will be clustered into a very small interval. In this case, however, there were no significant outlier scores and the normalised scores were thus fairly spread in all of the four dimensions. The final scores for each dimension were then calculated by taking the simple average of the normalised scores for the two questions which constitute the dimension. The scores for the four dimensions are reported in Table A.2 in the Appendix.

16

Table 3.1: Dimensions of the Cultural Values Index 1. Work Centrality

Description

The importance of work in an individual's life.

EVS Questions

How important it is in your life?: work (Q1A) Do you agree with the following statement?: Work should always come first, even if it means less spare time. (Q18E)

2. Independence

Weight

40.00%

Description

A measure of people's belief that they should carve out their own life/success, and not rely/depend on others.

EVS Questions

Do you consider 'independence' as an especially important quality which children should be encouraged to learn at home? (Q52B) Is it humiliating to receive money without having to work for it? (Q18B)

3. Belief in Hard Work & Thrift

Weight

20.00%

Description

The belief that hard work will bring success; and the belief that careful management of money and other resources is important.

EVS Questions

Do you consider 'hard work' as an especially important quality which children should be encouraged to learn at home? (Q52C) Do you consider 'thrift' as an especially important quality which children should be encouraged to learn at home? (Q52G)

4. Achievement Motivation

Weight

20.00%

Description

The motivation for success or the attainment of excellence. A measure of the extent to which a society gives priority to materialist values over postmaterialist ones is also included in this dimension.

EVS Questions

Do you consider 'determination and perseverance' as especially important qualities which children should be encouraged to learn at home? (Q52H) Which aim of the country is most important? (Q60) 1. Maintaining order in the country (materialist) 2. Giving people more say in government decisions (postmat.) 3. Fighting rising prices (materialist) 4. Protecting Freedom of Speech (postmaterialist)

Weight

20.00%

17

3.2.5

Assembling the Index

As illustrated by Table 3.2, the dimensions of the CVI have been found to be positively related to each other. This is in line with a priori expectations since the dimensions are intended to reflect the main values expected to be possessed by populations with a strong work ethic. Since the correlations are not excessively high, it was decided to retain all the dimensions in the composite index. Table 3.2: Correlation Matrix of CVI Dimensions Work Centrality Independence Work Centrality Independence

Belief in Hard Work & Thrift

Achievement Motivation

1 0.57

1

Belief in Hard Work & Thrift

0.26

0.30

1

Achievement Motivation

0.31

0.28

0.67

1

The final Cultural Values Index was then calculated by taking a weighted average of the dimensions’ scores, based on the weights outlined previously. The data and ranking results for the 43 European countries are presented in Table A.1 in the Appendix. A thematic map of the CVI scores (Figure A.1 in the Appendix) shows how lower work ethic values are generally present in the more developed Western countries. This is in line with Abramson &Inglehart’s (1995) theory that economic prosperity leads to a gradual erosion of the work ethic.

3.3 Mean Annual Growth Rate in GDP per capita PPP (08-13) (Recovery Growth): The Response Variable To analyse the effect of cultural values on economic resilience, a dependent variable measuring economic resilience actually manifested by European countries during the last crisis is required. The

18

mean growth rate in GDP per capita at purchasing power parities (PPP) from 2008 to 2013 was taken as the dependent variable. The data was sourced from the World Bank’s ‘World Development Indicators’ (WDI). GDP growth was considered as the most appropriate available measure of countries' manifested ability to avoid, withstand and recover from the adverse shocks of the economic crisis. Growth was taken in per capita terms and at PPP, so that it is as much as possible a true reflection of real changes in standards of living.

3.4 GDP per capita at PPP - 2007 (GDP07): Control Variable When taking GDP growth as a dependent variable it is very important to control for the convergence effect (catch-up effect). Basing on neoclassical growth models with diminishing returns to capital, the convergence theory states that a country's per capita growth rate is inversely related to its initial level of GDP (Barro 1991). Sala-i-Martin &Barro (1991) reported that the rate of conditional convergence amongst European countries stood at about two percent a year. Thus, due to the convergence effect, developing countries in the data set can be expected to register better growth performances than the more advanced countries, when controlling for other relevant factors. In line with other studies that used per capita growth as a dependent variable (Cecchetti 2012; Czernich 2011), the initial level of GDP per capita at PPP (2007) was taken as a control variable so as to control for the convergence effect. Again, data for this variable was sourced from the World Bank database (WDI).

3.5 The Economic Resilience Index (ERI): Control Variable The literature presents other drivers of economic resilience besides cultural values. It is therefore imperative to control for these other causes of economic resilience so as to be able to isolate the effect of cultural values. Briguglio et al's (2006) framework for the 'Economic Resilience Index' (ERI) was used to control for these drivers. Using this framework, scores for the four components of the index (macroeconomic stability, microeconomic market efficiency, good governance and social development)

19

were calculated for the 43 countries in the data set. The purpose of this control variable is to control for countries’ quality of economic governance which (besides cultural values) should have influenced their economic resilience manifested throughout the crisis. Of relevance to the study is the level of these components just prior to the start of the crisis, and so 2007 data was used to score the countries in the data set. Data was sourced from the World Bank database, Economic Freedom of the World database and the 2007/08 Human Development Report.

As found by Briguglio et al. (2006), countries’ social development component was found to be highly correlated with their level of economic development. Since the initial level of GDP per capita and social development were found to be highly correlated (r=0.8), the social development component was considered redundant and was dropped from the ERI so as to reduce the collinearity between the two control variables. The three remaining components were rescaled and integrated into a single index score. Half of the weighting in the overall index was assigned to macroeconomic stability since it is the component which can have the greatest effect on overall economy-wide resilience. Microeconomic market efficiency and good governance share the rest of the weight. The country scores for the ERI and its components are reported in Table A.3 in the Appendix.

3.6 The Economic Vulnerability Index (EVI): Control Variable Briguglio et al. (2006) emphasize the distinction between economic resilience and economic vulnerability. In their approach, while economic resilience is associated with man-made measures (‘nurtured’ resilience), economic vulnerability is confined to quasi-permanent inherent features which determine a country's ability to withstand or recover from external shocks. Therefore, besides the ERI to control for drivers of ‘nurtured’ resilience, it is also necessary to control for inherent economic vulnerability. For this purpose, Briguglio et al's ‘Economic Vulnerability Index' (EVI) was used in the

20

analysis. The index is based on three inherent economic features: 

Degree of economic openness: A high degree of economic openness makes a country more

vulnerable to uncontrollable negative external economic conditions. The ratio of international trade (sum of exports and imports) to GDP was taken as a measure of economic openness. The data required to measure economic openness was sourced from the World Bank’s WDI.  Export concentration: The dependence on a narrow range of exports aggravates vulnerability associated with economic openness. The UNCTAD merchandise export concentration index was used as a measure of export concentration. 

Dependence on strategic imports: The dependence on strategic imports such as food, energy or

industrial supplies exposes an economy to shocks related to the availability and costs of such imports. Given the rapid rise of oil prices during the last crisis period, this study focused on the dependence on energy imports. 'Net energy imports as a percentage of energy used', sourced from the WDI, was used as an indicator of the dependence on energy imports.

All countries in the study were scored on each component by taking simple averages over the five years preceding the start of the crisis (2003-2007). Five year averages were taken so as to cover for any possible single year anomalies in these flow variables. After the scores of each component were rescaled to a range between zero and one, they were integrated with equal weighting into a single index score. Table A.4 in the Appendix reports the country scores for the EVI and its components.

21

4. Analysis and Results

4.1

Results of the Primary Regression

As outlined in Chapter 3, the research's hypothesis will be tested by regressing average GDP growth over the crisis period on the Cultural Values Index, while controlling for the convergence effect and other factors which affect an economy's resilience and vulnerability to adverse shocks. The analysis is therefore revolved around the following OLS regression model specification: Growth = a0+ a1CVI + a2 ERI + a3 EVI + a4 GDP07 where: 

Growth = Mean growth in GDP per capita in PPP over the period 2008-2013



a0 = Constant term



CVI = Cultural Values Index



ERI = Economic Resilience Index



EVI = Economic Vulnerability Index



GDP07 = Initial level (2007) of GDP per capita at PPP in thousands.

The coefficient of the CVI is positive and statistically significant at the five percent level. This lends support to the research's hypothesis that those countries whose population possessed resilienceenhancing cultural values manifested greater economic resilience to the last crisis, and thus registered superior growth rates. Although the main focus of the study is the effect of cultural values, it is also appropriate to analyse the effects of the control variables as well. As expected a priori, the ERI has a positive coefficient. It is also statistically significant at the five percent level indicating that the quality of economic governance did indeed have a significant effect on the economic resilience manifested during the crisis period. Contrary to expectations, the EVI has a negative coefficient. The result, however, is not statistically significant suggesting that the inherent exposure to negative shocks did not have a significant

22

effect on countries' economic growth performances. Finally, the highly statistically significant negative coefficient of GDP07 indicates that there was a strong convergence effect during the crisis period. An R2 of 0.46 and a statistically significant F-statistic indicate that, on the whole, the model explains a significant proportion of the variation in the recovery growth rates registered across Europe during the crisis.Table 4.1 summarises the results of this primary regression. Table 4.1: Results of the Primary Regression3 Dependent Variable: Average Annual Growth in GDP per capita PPP (2008-2013) Regressor

Coefficient

Standard Error

t-statistic

a0

-0.182

1.666

-0.109

CVI

4.581*

2.212

2.071

ERI

4.298*

1.962

2.191

EVI

2.613

3.335

0.784

GDP07

-0.077**

0.022

-3.464

Summary Statistics R2

0.462

Adjusted R2

0.406

F-statistic

8.170

p (F-statistic)

0.000

N

43

* denotes significance at the 5% level and ** denotes significance at the 1% level using a two-sided test.

23

4.2 Multicollinearity Tests If the regression is affected by excessively high multicollinearity, then the coefficient estimates and test statistics obtained could be unreliable. Since this study's focus is primarily on the effect of cultural values, of particular relevance is the extent to which the CVI is correlated with the other control variables. The correlation matrix in Table 4.2 shows that the CVI is quite strongly negatively correlated (r=-0.55) with the initial level of GDP per capita (GDP07). This negative correlation is consistent with Abramson &Inglehart’s (1995) theories on the inverse relationship between work ethic values and economic development. Although the correlation is quite high, it is not deemed to be excessive; and so it should not have had a serious effect on the results. Table 4.2: Correlation Matrix of Explanatory Variables CVI

ERI

EVI

CVI

1

ERI

-0.36

1

EVI

0.16

0.05

1

GDP07

-0.55

0.67

0.04

GDP07

1

4.3 Robustness Checks and Stability Tests Although the diagnostic tests for multicollinearity suggest that there are no serious problems with multicollinearity, stability tests can be used to verify whether the positive effect of the CVI on economic resilience is stable and persistent. The stability of the CVI coefficient was tested in two ways: 

By making modifications to the specification of the model.



By conducting a formal stability tests: the RESET test.

24

4.3.1

Alternative Specifications

Table 4.4 shows the regression results of two alternative models which were estimated to verify whether the positive effect of the CVI is stable.  In model (2), the EVI was dropped since it was found to be statistically insignificant in the original model. The only resulting difference with the removal of the EVI is that the other variables became slightly more statistically significant. Since the Adjusted R2 of model (2) is greater than that in the original specification, model (2) can be considered superior to the original specification, and it was thus decided to exclude the EVI from any ensuing analysis.  In model (3), the variable controlling for the convergence effect (GDP07) was dropped as well since it is the variable mostly correlated with the CVI (Table 4.2). In this model, the ERI became statistically insignificant while the CVI resulted as highly statistically significant. The Adjusted R2 registered a large decline, indicating that the initial level of GDP per capita is, as expected, a very important control variable that needs to be maintained in the analysis. The results of these two alternative models suggest that the significant positive coefficient of the CVI is stable across different specifications. Table 4.4: Stability Tests – Alternative Model Specifications5 Dependent Variable: Average Growth in GDP per capita PPP (2008-2013) Regressor

(2)

(3)

a0

0.43

-1.374

(0.29)

(-0.891)

4.961*

8.260**

(2.311)

(3.830)

4.298*

0.307

(2.226)

(0.176)

CVI ERI EVI GDP07

-0.075** (-3.412)

25

Summary Statistics R2

0.454

0.291

Adjusted R2

0.411

0.255

p(F-statistic)

0.000

0.001

N

43

43

t-statistics are shown in parenthesis * denotes significance at the 5% level and ** denotes significance at the 1% level using a two-sided test.

4.3.2

Stability Test

i)

Ramsey’s RESET test was conducted to test for any specification errors. RESET is a general test

for specification errors such as: omitted variables, incorrect functional form and correlation between the independent variables and the error term. The RESET test was conducted on model (2), and the null hypothesis was accepted at all significance levels - indicating that there are no particular specification errors in the model. The results of the RESET test are reported in Table A.5 in the Appendix.

The results of the Ramsey RESET test show that the positive effect of the Cultural Values Index on economic resilience is stable and is unlikely to have resulted by chance or through some specification error. The next section analyses further the nature of this relationship while commenting as well on the other observed relationships.

4.4 Concluding Analysis The results of the empirical analysis give significant evidence that cultural values did indeed have an impact on the resilience manifested by European countries during the last crisis. The results thus lend support to the research's hypothesis. Those countries possessing a strong work ethic culture and its associated values generally registered higher economic growth rates during the crisis period (while

26

controlling for the state of development of the country and other economic factors affecting resilience). The implications of these findings are discussed in the following chapter.

27

5. Discussion, Conclusions and Implications

5.1 Linking the Findings to the Literature Max Weber's 'The Protestant Ethic and the Spirit of Capitalism' (1905) was the first real insight into the role of cultural factors in the determination of economic outcomes. Since then, numerous other studies have sought to explore the relationship between cultural values and economic performance, producing diverging conclusions. There is still, however, no consensus on whether cultural factors should be even considered in any economic analysis. The results of this study suggest that they should be. The analysis supports Modrack's (2008) assertion that the consideration of the work ethic construct and other related values is not outdated; but instead a valuable approach which could add to the knowledge on the interactions between cultures, institutions and economic outcomes.

Inglehart (1990) traced the following two economic and cultural developments since the Protestant Reformation:  The Protestant Ethic was conducive to the development of capitalism in certain economies, which as a result experienced rapid economic growth.  With increasing prosperity, however, as a result of the 'scarcity'1 and 'socialisation'2 effects, work ethic values in advanced industrial societies were gradually replaced by postmaterialist value priorities; leading to lower economic growth prospects.

1

2

The 'scarcity' hypothesis asserts that periods of prosperity lead to increased postmaterialist attitudes, while periods of scarcity lead to increased materialist attitudes. The 'socialisation' hypothesis asserts that an individual's basic values reflect the conditions that prevailed during his pre-adult years.

28

The strong negative correlation between countries' GDP per capita and the Cultural Values Index (r=0.55) observed in this study is conformant with Inglehart's 'socialisation' and 'scarcity' hypotheses that prosperity eventually leads to a shift from work ethic value priorities to postmaterialist value priorities. In addition, the results of the analysis suggest that besides lowering economic growth prospects, this cultural shift has also reduced advanced countries' resilience to economic crisis. Therefore, from a cultural perspective, economic crises can accelerate the convergence process across nations since advanced countries possess values which render them less economically resilient. From an economic perspective, however, advanced countries are better able to secure micro and macroeconomic stability (ERI was positively correlated with GDP per capita (r=0.67)) which enhances their economic resilience.

5.2 Policy Implications The observations discussed in the previous section imply that, generally, advanced countries require a different policy approach towards resilience-building from developing countries. Policymakers in advanced countries seeking to enhance economic resilience should focus more on improvements in cultural dimensions rather than in economic factors. The central question however is; ‘How, and to what extent can policymakers change or influence national values?’ Trying to influence national attitudes and values is certainly an arduous task given they are mainly a matter of personal choice. This is however definitely possible given that much of the cultural change over the last centuries can be attributable to changing economic, political and social conditions. Inglehart (1990) attributed the decline in materialist attitudes to increased economic security resulting from prosperity. Therefore, measures to decrease the perceived sense of economic security, although politically undesirable, can enhance the work ethic culture and economic resilience. Scaling down the generosity of some welfare benefits could have a positive effect on economic resilience since it would reduce the sense of economic security, and thus enhance the possession of resilience-enhancing values. Of course, such measures need to be assessed keeping in mind as well any other economic, social or political costs and benefits.

29

An individual's basic values can be mostly influenced during his pre-adult years. Therefore, any efforts to change cultural values need to be mainly concentrated in education. An education that promotes values and attitudes which this study has found to have a positive effect on economic resilience - such as independence, hard work, thrift, and achievement motivation - can be a significant investment which yields greater economic resilience in later years. McClelland (1969) demonstrated that achievement motivation can be developed even in adults. Businessmen who underwent psychological training intended to enhance their need for achievement subsequently showed heightened entrepreneurial activity. Following training; increased business activity, business extensions, expansions into new industries, and more collaboration on new joint enterprises were observed amongst participants. Therefore, while the values of the future workforce could be influenced through education, those of current workers could be influenced through appropriate training programmes which promote the resilience-enhancing values and attitudes identified in this study.

On the other hand, the challenge for policymakers in developing countries is to implement policies aimed at achieving economic prosperity and stability, while avoiding the erosion of the work ethic culture associated with increasing prosperity. In pursuit of rapid economic growth, developing countries may adopt aggressive expansionary policies. This study confirms that this expansionary approach needs to be employed with prudent and disciplined economic governance since macroeconomic stability (fiscal discipline, and control on inflation and unemployment) was found to be one of the major drivers of economic resilience. Economic expansionary policies need to be accompanied by measures seeking to preserve the strong work ethic that these nations possess. Just as in advanced countries; education, training programmes and welfare system design can have a powerful role in influencing cultural dimensions.

30

The distinction made between the policy-needs of advanced and developing countries is only a general observation. Each nation, irrespective of its state of development, needs to assess its economic and cultural aspects of economic resilience and take the measures necessary to improve in those dimensions where it is underperforming while preserving any desirable features that they possess.

5.3 Limitations of the Study and Possibilities for Further Research The analysis provides evidence that cultural values do exert a significant effect on countries’ economic resilience. This evidence still needs to be interpreted with caution given that the analysis had to overcome certain limitations, some of which may not have been fully overcome. Clearly, not all countries were faced with the same economic shock given that they differ in their level of trade openness, sectoral development, financial liberalisation, and in several other factors. The inclusion of the ERI, the EVI and the initial level of capita as control variables was done to, as much as possible, 'equalize' across nations the impact of the crisis. However, these control variables may not have fully controlled for the vast range of cross-country differences that exist. Furthermore, it was not possible to control for any national policies and other developments which may have occurred during the crisis period. Still, however, although some variables could have been omitted, Ramsey's RESET test suggests that there have been no important omitted variables which could have critically altered the results. Future research utilising regional data within the same country rather than cross-country data could reduce such possible confounds in the analysis. However, no extensive regional data on cultural values is currently available.

Another limitation of the study was the limited sample size of 43 countries. The sample could have been extended by including other regions in the world using the World Values Survey besides the European Values Study (although the two surveys are not exactly identical), but it was decided to limit the sample to European nations only so as to avoid any additional confounds. The tests performed suggest that the limited sample size did not have any significant effects on the stability of the estimated effect of cultural

31

values. Knowledge in the area would still benefit from additional analysis on the role cultural values in other regions of the world or from a global analysis spanning different continents.

It can also be argued that the effects of the crisis are still being felt to some extent to the present day, and therefore, considering economic resilience only up to 2013 (latest data available) can be seen as inadequate. A revision of this study using a longer time period could give a truer reflection of economic resilience since greater emphasis would be placed on the ability of an economy to recover from a shock besides its ability to avoid and withstand it.

The intention of this study was to provide a basis for further analysis in an area in which research has not yet focused upon. Although the resulting evidence is by no means conclusive, the results do warrant further research on the role of cultural values in determining economic resilience (and other aspects of economic performance as well).

The literature review identified a number of possible channels through which cultural values can affect economic resilience. This study's analysis found a positive relationship between the values on which these channels rest and manifested economic resilience. This indicates that some or all of the identified channels, or even possibly other channels which have not been identified, have been at work during the Great Recession. A detailed analysis of these channels was beyond the scope of this research since the objective was to search for any final relationship between values and economic resilience. Further research on the channels underlying this observed relationship would certainly enrich the knowledge in the area. Finding through which channels values affected economic resilience would be possible through analyses of changes and differentials in working hours, inactivity rates, discouraged workers, selfemployment, and saving rates.

32

As identified by Inglehart (1990), cultural and economic developments influence each other in a dynamic relationship. This study analysed only the effect of cultural values on economic resilience. When a new wave of the European Values Study is published, it would then be possible to analyse the two-way interaction between the variables. Have those nations most adversely affected by the crisis experienced cultural changes towards a stronger work ethic? Have the developing nations, which due to their cultural resilience managed to converge towards the most advanced nations, experienced a cultural shift towards postmaterialist value priorities? Such questions and others would become analysable through panel data analysis when a new cultural survey on European values is conducted.

5.4 Conclusion A nation's resilience to external shocks is certainly a fundamental element in its long-run economic development. Dedicated theoretical and empirical attention to the effect of cultural factors on a nation's economic resilience had yet to be provided by the literature. This study sought to provide the foundation for such an analysis in the context of the recently experienced economic crisis in Europe. It analysed the effects on economic resilience of both cultural factors, which have been largely neglected in the literature, and of other economic factors which had already been identified by previous studies. The results give strong evidence that both sets of factors are important determinants of economic resilience. This signals that due consideration needs to be given to both economic and cultural factors in any analysis of economic resilience; or of any other aspect of development for that matter. Augmenting any economic analysis with cultural considerations could expand significantly our understanding of observed economic outcomes.

33

References Baritto, F. 2009."Disasters, Vulnerability and Resilience from a macro-economic perspective: Lessons from the Empirical Evidence." Background Paper for the 2009 ISDR Global Assessment Report on Disaster Risk Reduction.

Barro, R.J. 1991.“Economic Growth in a Cross Section of Countries.”The Quarterly Journal of Economics, vol. 106, no. 2, pp. 407-443.

Becker, S.O. &Woessmann, L. 2009.“Was Weber Wrong? A Human Capital Theory of Protestant Economic History”.The Quarterly Journal of Economics, vol. 124, no. 2, pp. 531-596.

Briguglio, L., Cordina, G., Farrugia, N. & Vella, S. 2006.“Conceptualizing and measuring economic resilience.”Building the Economic Resilience of Small States, Malta: Islands and Small States Institute of the University of Malta and London: Commonwealth Secretariat , pp. 265-288.

Cavalcanti, T.V., Parente, S.L. & Zhao, R. 2007.“Religion in macroeconomics: a quantitative analysis of Weber’s thesis.”Economic Theory, vol. 32, no. 1, pp. 105-123.

Cecchetti, S.G. &Kharroubi, E. 2012.“Reassessing the impact of finance on growth.”BIS Working Paper no. 381.

Available at:

[Accessed 3 February 2015]

Czernich, N., Falck, O., Kretschmer, T. &Woessmann, L. 2011.“Broadband infrastructure and economic growth.”The Economic Journal, vol. 121, no. 552, pp. 505-532.

Delacroix, J. & Nielsen, F. 2001.“The beloved myth: Protestantism and the rise of industrial capitalism in nineteenth-century Europe.”Social Forces, vol. 80, no. 2, pp. 509-553.

34

European Values Study 2008: Integrated Dataset (EVS 2008). GESIS Data Archive, Cologne. ZA4800 Data file version 3.0.0

Europeanvaluesstudy.eu 2015.About EVS. [online] Available at: [Accessed 3 April 2015]

Furnham, A., Lewis, A. & Webley, P. 1995.The New Economic Mind: The Social Psychology of Economic Behaviour. Harvester Wheatsheaf, Hertfordshire.

Furnham, A. 1990.“A content, correlational, and factor analytic study of seven questionnaire measures of the Protestant work ethic.”Human Relations, vol. 43, no. 4, pp. 383-399.

Furnham, A., Bond, M., Heaven, P., Hilton, D., Lobel, T., Masters, J., Payne, M., Rajamanikam, R., Stacey, B. & Van Daalen, H. 1993.“A comparison of Protestant work ethic beliefs in thirteen nations.”The Journal of social psychology, vol. 133, no. 2, pp. 185-197.

Granato, J., Inglehart, R. &Leblang, D. 1996.“The effect of cultural values on economic development: theory, hypotheses, and some empirical tests.”American Journal of Political Science, pp. 607-631.

Guiso, L., Spaienza, P. &Zingales, L. 2006.“Does culture affect economic outcomes?”Journal of Economic Perspectives, no. 20(2), pp. 23-48.

Hall, P.A. & Lamont, M. 2014.Social resilience in the neoliberal era.Cambridge University Press, New York.

Inglehart, R. 1990.Culture shift in advanced industrial society.Princeton University Press, Princeton, New Jersey.

35

Marques, A. &Soukiazis, E. 1998.“Per Capita Income Convergence across Countries and across Regions in the European Union: Some New Evidence.”International Meeting of European Economy.

McClelland, D.C. &Winter, D.G. 1969.Motivating economic achievement. Free Press, New York .

McClelland, D.C. 1961.The achieving society.Van Nostrand, Princeton, N.J.

Modrack, S. 2008.“The Protestant Work Ethic revisited: a promising concept or an outdated idea?” No SP I 2008-101, Discussion Papers, Research Unit: Labour Market Policy and Employment, Social Science Research Centre Berlin (WZB). Available at: [Accessed 9 January 2015]

Ntibagirirwa, S. 2009.“Cultural values, economic growth and development.”Journal of Business Ethics, vol. 84, no. 3, pp. 297-311.

Sala-i-Martin, X. &Barro, R.J. 1991.Convergence across states and regions.Economic Growth Center, Yale University.

Sapienza, P., Zingales, L. &Guiso, L. 2006.“Does culture affect economic outcomes?”National Bureau of Economic Research, Working Paper no. w11999. Available at: [Accessed 6 February 2015]

Schaltegger, C.A. &Torgler, B. 2010.“Work ethic, Protestantism, and human capital.”Economics Letters, vol. 107, no. 2, pp. 99-101.

Smith, A. 1759.The theory of moral sentiments ed. A.Millar, A.Kincaid& J.Bell.AM Kelley, London.

van Hoorn, A. &Maseland, R. 2013.“Does a Protestant work ethic exist? Evidence from the well-being effect of unemployment.”Journal of Economic Behavior& Organization, vol. 91, pp. 1-12.

36

Weber, M. 1930.The Protestant Ethic and the Spirit of Capitalism: Translated from German by Talcott Parsons.

37

Appendix Table A.1: Scores and Rankings for the Cultural Values Index6 Country

Cultural Values Index

Rank

Albania

0.586

8

Armenia

0.565

9

Austria

0.465

24

Azerbaijan

0.493

17

Belgium

0.326

40

Bosnia & Herzegovina

0.451

32

Bulgaria

0.647

4

Croatia

0.409

35

Cyprus

0.529

11

Czech Republic

0.497

16

Denmark

0.402

36

Estonia

0.399

37

Finland

0.245

41

France

0.393

39

Georgia

0.600

7

Germany

0.484

20

Greece

0.470

22

Hungary

0.622

6

Iceland

0.418

34

Ireland

0.463

27

Italy

0.426

33

Latvia

0.511

13

Lithuania

0.510

14

Luxembourg

0.512

12 38

Country

Cultural Values Index

Rank

Macedonia

0.818

1

Malta

0.458

30

Moldova

0.555

10

Montenegro

0.486

19

Netherlands

0.213

43

Norway

0.466

23

Poland

0.464

25

Portugal

0.491

18

Romania

0.671

3

Russian Federation

0.463

26

Serbia

0.471

21

Slovak Republic

0.626

5

Slovenia

0.498

15

Spain

0.398

38

Sweden

0.460

29

Switzerland

0.460

28

Turkey

0.780

2

Ukraine

0.455

31

United Kingdom

0.237

42

39

Figure A.1: Thematic Map of CVI Scores

40

Table A.2: Scores for the Dimensions of the CVI7 Country

Work Centrality (40%)

Independence Belief in Hard Achievement Cultural (20%) Work & Thrift Motivation Values Index (20%) (20%)

Albania

0.869

0.151

0.540

0.504

0.586

Armenia

0.875

0.035

0.574

0.463

0.565

Austria

0.569

0.566

0.265

0.355

0.465

Azerbaijan

0.705

0.477

0.303

0.280

0.494

Belgium

0.374

0.148

0.414

0.322

0.326

Bosnia

0.555

0.176

0.384

0.586

0.451

Bulgaria

0.759

0.276

0.678

0.762

0.647

Croatia

0.419

0.256

0.517

0.436

0.409

Cyprus

0.835

0.199

0.297

0.479

0.529

Czech Rep.

0.464

0.556

0.603

0.400

0.497

Denmark

0.479

0.743

0.018

0.289

0.402

Estonia

0.383

0.049

0.679

0.500

0.399

Finland

0.190

0.376

0.106

0.366

0.245

France

0.468

0.089

0.452

0.489

0.393

Georgia

0.782

0.408

0.580

0.451

0.600

Germany

0.550

0.639

0.293

0.389

0.484

Greece

0.668

0.258

0.283

0.474

0.470

Hungary

0.678

0.704

0.656

0.393

0.622

Iceland

0.370

0.775

0.313

0.264

0.418

Ireland

0.387

0.509

0.514

0.516

0.463

Italy

0.583

0.246

0.414

0.302

0.426

Latvia

0.363

0.608

0.724

0.499

0.511

Lithuania

0.408

0.667

0.635

0.430

0.510

Luxembourg 0.581

0.394

0.500

0.505

0.512

Macedonia

0.705

0.868

1.000

0.811

0.818

Malta

0.577

0.170

0.545

0.422

0.458

41

Country

Work Centrality (40%)

Independence Belief in Hard Achievement Cultural (20%) Work & Thrift Motivation Values Index (20%) (20%)

Moldova

0.679

0.344

0.616

0.458

0.555

Montenegro 0.609

0.228

0.512

0.471

0.486

Netherlands 0.278

0.000

0.256

0.255

0.213

Norway

0.533

0.823

0.059

0.380

0.466

Poland

0.465

0.242

0.660

0.490

0.464

Portugal

0.601

0.206

0.597

0.448

0.491

Romania

0.707

0.504

0.814

0.625

0.671

Russian Fed. 0.428

0.158

0.681

0.621

0.463

Serbia

0.612

0.196

0.414

0.518

0.471

Slovak Republic

0.710

0.614

0.716

0.377

0.625

Slovenia

0.564

0.642

0.316

0.404

0.498

Spain

0.580

0.215

0.265

0.347

0.398

Sweden

0.638

0.552

0.294

0.177

0.460

Switzerland

0.539

0.548

0.297

0.378

0.460

Turkey

0.944

1.000

0.575

0.436

0.780

Ukraine

0.509

0.009

0.751

0.495

0.455

United Kingdom

0.093

0.356

0.352

0.289

0.237

42

Table A.3: Economic Resilience Index (ERI) Scores8 Country

Macroeconomic Microeconomic Good Stability (50%) Market Governance Efficiency (25%) (25%)

Economic Resilience Index

Rank

Albania

0.051

1.000

0.200

0.326

34

Armenia

0.094

0.958

0.320

0.366

27

Austria

0.264

0.542

0.920

0.497

12

Azerbaijan

0.441

0.000

0.460

0.336

33

Belgium

0.180

0.354

0.640

0.339

32

Bosnia

0.115

0.479

0.000

0.177

43

Bulgaria

0.401

0.729

0.200

0.433

20

Croatia

0.140

0.292

0.360

0.233

40

Cyprus

0.423

1.000

0.620

0.616

6

Czech Rep.

0.263

0.479

0.420

0.356

29

Denmark

0.520

0.458

0.980

0.619

5

Estonia

0.726

1.000

0.640

0.773

3

Finland

0.353

0.542

1.000

0.562

8

France

0.145

0.500

0.700

0.373

24

Georgia

0.174

0.917

0.320

0.396

22

Germany

0.189

0.604

0.880

0.465

16

Greece

0.000

0.792

0.460

0.313

35

Hungary

0.008

0.354

0.500

0.218

42

Iceland

0.380

0.688

0.960

0.602

7

Ireland

0.254

0.833

0.760

0.525

10

Italy

0.192

0.292

0.380

0.264

37

Latvia

0.443

0.708

0.540

0.533

9

Lithuania

0.262

0.875

0.540

0.485

14

Luxembourg 0.889

0.500

0.840

0.779

2

Macedonia

0.100

0.479

0.200

0.220

41

Malta

0.114

0.708

0.760

0.424

21

43

Country

Macroeconomic Microeconomic Microeconomic Economic Stability (50%) Market Market Resilience Efficiency (25%) Efficiency (25%) Index

Rank

Moldova

0.173

0.771

0.340

0.364

28

Montenegro 0.252

0.583

0.320

0.352

30

Netherlands 0.422

0.417

0.820

0.520

11

Norway

1.000

0.500

0.980

0.870

1

Poland

0.134

0.458

0.300

0.257

38

Portugal

0.115

0.625

0.620

0.369

26

Romania

0.275

0.979

0.360

0.472

15

Russian Fed. 0.579

0.313

0.300

0.443

18

Serbia

0.086

0.646

0.200

0.254

39

Slovak Republic

0.139

0.833

0.440

0.388

23

Slovenia

0.295

0.396

0.420

0.352

31

Spain

0.270

0.688

0.560

0.447

17

Sweden

0.325

0.250

0.860

0.440

19

Switzerland

0.539

0.938

0.920

0.734

4

Turkey

0.147

0.875

0.320

0.372

25

Ukraine

0.258

0.458

0.220

0.299

36

United Kingdom

0.203

0.875

0.700

0.495

13

44

Table A.4: Economic Vulnerability Index (EVI) Scores9 Country

Trade Openness (33%)

Export Dependence Economic Concentration on Energy Vulnerability (33%) Imports (33%) Index

Rank

Albania

0.095

0.339

0.935

0.456

15

Armenia

0.088

0.429

0.959

0.492

11

Austria

0.226

0.025

0.963

0.405

29

Azerbaijan

0.252

1.000

0.701

0.651

3

Belgium

0.431

0.082

0.971

0.495

10

Bosnia

0.257

0.149

0.911

0.439

20

Bulgaria

0.294

0.113

0.936

0.448

17

Croatia

0.180

0.105

0.947

0.411

28

Cyprus

0.208

0.216

0.997

0.474

13

Czech Rep.

0.326

0.066

0.910

0.434

22

Denmark

0.185

0.041

0.819

0.348

38

Estonia

0.429

0.150

0.910

0.497

8

Finland

0.126

0.217

0.945

0.429

24

France

0.019

0.043

0.939

0.333

40

Georgia

0.148

0.226

0.953

0.443

18

Germany

0.121

0.076

0.951

0.383

31

Greece

0.035

0.075

0.959

0.356

36

Hungary

0.387

0.148

0.954

0.496

9

Iceland

0.116

0.535

0.907

0.519

7

Ireland

0.425

0.279

0.986

0.563

4

Italy

0.016

0.000

0.982

0.333

41

Latvia

0.235

0.124

0.952

0.437

21

Lithuania

0.287

0.197

0.941

0.475

12

Luxembourg

1.000

0.132

0.997

0.710

1

Macedonia

0.240

0.208

0.933

0.460

14

Malta

0.489

0.623

1.000

0.704

2

45

Country

Trade Openness (33%)

Export Dependence Economic Concentration on Energy Vulnerability (33%) Imports (33%) Index

Rank

Moldova

0.378

0.273

0.996

0.549

5

Montenegro

0.251

0.084

0.927

0.421

26

Netherlands

0.341

0.081

0.904

0.442

19

Norway

0.095

0.621

0.000

0.239

43

Poland

0.120

0.046

0.900

0.355

37

Portugal

0.075

0.045

0.980

0.367

35

Romania

0.116

0.105

0.913

0.378

32

Russian Fed.

0.029

0.467

0.781

0.426

25

Serbia

0.111

0.084

0.921

0.372

33

Slovak Republic

0.471

0.161

0.958

0.530

6

Slovenia

0.316

0.104

0.943

0.454

16

46

Spain

0.036

0.097

0.973

0.369

34

Sweden

0.166

0.097

0.921

0.395

30

Switzerland

0.166

0.131

0.945

0.414

27

Turkey

0.000

0.063

0.965

0.343

39

Ukraine

0.231

0.135

0.931

0.432

23

United Kingd. 0.029

0.082

0.886

0.332

42

47

Table A.5: Ramsey RESET Test Results10 F-statistic

0.762

Probability

0.474

Log Likelihood Ratio 1.736

Probability

0.420

Variable

Coefficient

Standard Error

t-Statistic

C

1.681

1.996

0.842

CVI

-8.015

10.744

-0.746

ERI

-6.411

8.930

-0.718

GDP07

0.102

0.145

0.701

Fitted^2

0.968

0.810

1.195

Fitted^3

-0.111

0.099

-1.117

R-squared

0.475

Akaike Info Criterion 3.778

Adjusted R-squared

0.404

4.024

Log Likelihood

-75.234

Shwarz Info Criterion F-statistic

Durbin-Watson Stat

1.931

Prob (F-statistic)

0.000

48

6.703