Economic internationalization and the decline of the left

Abstract

This article examines the continuing importance of the left–right dimension for voting behavior in Western Europe. We test the extent to which economic internationalization may affect the capacity of this dimension to structure party preferences. We explore two dimensions of internationalization, long-term openness and short-term changes, assessing, respectively, the impact of international trade and foreign investments on voters’ preference formation. To study the influence of changing context, we use four waves of the European Election Study (1999, 2004, 2009, and 2014). We show that openness to international economic exchanges tends to weaken the left–right cleavage. At the same time, long-term economic openness appears to soften the impact of short-term shocks for the relevance of left–right politics.

Keywords

European Election Studies economic internationalization globalization economic voting left–right dimension

Does economic internationalization diminish the importance of the left–right dimension? As most apparently simple questions, this one does not have a simple answer. For a long time, it seemed that international issues only marginally affect individual political attitudes, except for armed conflict. Even European integration, arguably the most ambitious effort of regional integration to date, has long been considered to hardly affect political competition in the domestic arena. As time went by, however, people have learned more about the ways in which international trade, rules, or conflict may affect their lives.

This article tests and compares two perspectives on the possible effects of internationalization on individual political attitudes. More specifically, we try to advance knowledge concerning the effects of those processes on the explanatory power of the left–right dimension for voting behavior.

As internationalization increases economic interdependence, it decreases national governments’ autonomy to implement independent economic policies. Making a difference, once in power, thus becomes increasingly difficult. Put differently, governments have a hard time implementing policies historically associated with left-wing or right-wing governments. Party leadership, consequently, is likely to be more hesitant to emphasize those issues and policies. Ultimately, this should certainly weaken the structuring effect of the left–right dimension in political competition.

If internationalization is perceived to be weakening states, we argue, then voters should concentrate less on issues associated with the left–right divide to evaluate potential candidates. Other issues and aspects, for example, candidate characteristics or cultural issues, should become more important compared to the classical left–right dimension. In other terms, we expect internationalization to moderate the structuring role of left–right positions in determining preferences for parties. A country’s degree of economic interdependence should thus affect voters’ behavior in a nontrivial fashion.¹

Unlike previous studies in this area, however, we assume that globalization does not have a univocal effect. In some countries, “openness” is deeply enshrined and the compensation of potential losers has been institutionalized. Other countries are historically less exposed, and we may expect that the consequences of internationalization may be more disruptive. Put differently, in countries with long-standing trade openness, short-term variations in internationalization should have no visible effect on the importance of the left–right divide. However, in those countries that have been more reticent to internationalization, this impact should be more important.

To test our main assumptions, we pool four waves of the European Election Study (EES)—1999, 2004, 2009, and 2014—covering the 14 Western European countries that are available across all four waves,² forming a robust basis to assess the structuring role of the left–right dimension in a comparative perspective. Although the time span of this study is rather long, compared to existing work, this article does not focus on historical changes. Rather, we concentrate on the varying importance of the distance between left–right self-placement and party positions for the voting propensity for different domestic parties. We contrast two indicators of economic integration to test the relative effect of short-term and long-term economic internationalization on the importance of the left–right dimension.

The rest of the article proceeds as follows. The first section will briefly survey previous work and present our own approach to internationalization and its effects on the left–right dimension. The second section presents our main assumptions. The third section discusses our data and models, and the final section concludes and develops possible directions for future research.

Internationalization and the left–right dimension

The profound nature of internationalization and its consequences for politics have long been debated. Several rival definitions and approaches coexist. There are—at least—two traditions. The first, focusing on individual attitudes, looks at internationalization as an external shock that individuals have to cope with, as it will affect their economic opportunities. A second tradition is more interested in system-wide effects and strategies. The unit of analysis are countries that may adopt long-term strategies toward internationalization. Under these circumstances, internationalization may thus not be viewed as an external shock but may result from a deliberate strategy. We argue that both approaches are needed to explain what internationalization does to politics. We assume that short-term changes will affect individuals differently, depending on the long-term strategies of the countries they live in. Before developing this argument, we will briefly discuss the meaning of the left–right dimension.

The left–right dimension in Western Europe

Issue competition among parties varies over time in response to problems, the emergence of new issues, and the evolution of the electorate (Green-Pedersen, 2007). The left–right dimension has regularly been questioned in this context, especially since the fall of popular democracies in Eastern Europe and the recent trends regarding the decline of ideology and party membership. Yet, it appears to remain fundamental, irrespective of the emergence of other campaign themes.

The existence of a strong relation between left–right positioning of voters and vote choice is not questioned (Aldrich and Dorobantu, 2010). Especially in Western Europe, voters do position themselves on the left–right scale, perceive party positions in a coherent way (Aldrich and Fernandez, 2013; Aldrich and Popa, 2008), and act accordingly when picking out candidates to support. There is, however, considerable variance in the use and the strength of the left–right divide across time and across countries. Context plays a role in shaping the strength of this correlation, with a particular focus in the literature on the role of party system polarization (Freire, 2008; Lachat, 2008; van der Eijk et al., 2005). In the same logics, we look in this article at the role of the economic context and more especially at the role of economic internationalization.

There is certainly considerable debate as to which is exactly the dominant political dimension and how many dimensions there are. Kitschelt (1994) demonstrates that political competition in most countries is structured by a mix of the socialist/capitalist and the authoritarian/libertarian dimensions. While both dimensions exist, they neatly reduce into a single dimension that covers the majority of countries. Another recent study of dimensionality in Europe (Bakker et al., 2012) confirms that a single dimension dominates political competition in virtually all Western European countries but also that Eastern Europe features very different patterns. This is one of the reasons that we will concentrate on Western Europe only in the empirical part (cf. infra). In a similar line of thinking, Marks et al. (2006) show that there is a unique correlation between economic left and liberal politics in Western Europe.

A slightly different argument is made by Mair and Thomassen, who argue that left–right remains the common structuring dimension of politics in the European Union (EU) (Mair and Thomassen, 2010). A study on values and left–right placement, moreover, shows “that political ‘right’ and ‘left’ have coherent meanings” across Western Europe—though not compared to Eastern Europe (Piurko et al., 2011: 558). Following Jim Adams and different colleagues (2004, 2006, 2009), we also believe that the left–right dimension is mainly and, in fact, almost exclusively structured by the economic left–right dimension. A quick analysis of the correlation of party positions in the Chapel Hill Expert Survey (CHES) confirms this. Data from the CHES trend file, covering European elections from 1999 to 2014, show that the economic left–right and general left–right dimensions are correlated at 0.9, compared to 0.77 for general left–right and “gal-tan.” Things are very different for Eastern Europe again. This is another reason to concentrate on Western Europe only in the empirical part (cf. infra).

More importantly, as illustrated by CHES data, patterns in this area are relatively stable over time within countries, even though they may vary strongly between countries. The left–right dimension remains a privileged entry to the study of political attitudes, because change is slow. Short-term changes in the relative importance of the left–right dimension must thus mainly be the result of contextual factors. Voters are strongly constrained by supply-side politics. Parties and governments will provide cues and frames that will help voters make sense of new challenges, problems, and issues. Internationalization arguably represents one of the major such challenges today.

Internationalization and elections

Work on the effects of internationalization on individual political behavior has usually transposed some variant of Ricardian comparative advantage to individual-level political preferences (Frieden, 1991; Gourevitch, 1978; Rogowski, 1989). Internationalization has thus mainly been conceptualized as decreasing trade barriers and increasing exchanges and the development of international legal frameworks. This is said to influence political attitudes and voting behavior to the extent that it opens up new economic opportunities for some social groups, while making other groups less competitive. All will thus depend on the relative comparative advantage of individuals in an internationalized labor market. Following a similar line of thinking, Gabel (1998) developed a “utilitarian” approach to explain attitudes toward the EU.

These approaches proved insufficient, after the signing of the Maastricht Treaty, when EU issues quite suddenly were at the heart of domestic political debates (Eichenberg and Dalton, 2007; Hooghe and Marks, 2009). While the success of Eurosceptic parties at the 1994 and the 1999 European elections proved short-lived (Taggart, 1998), it represented a break with the tradition of “permissive consensus” (Hooghe and Marks, 2009). For a long time, however, changes seemed minimal, as things appeared to return to normal. A growing number of authors have argued that things may finally be changing (e.g. Meijers, 2015; van Elsas and van der Brug, 2015). Miklin (2014) has argued that the Eurocrisis may have enhanced the salience of globalization in domestic political debates. The prominence of debates concerning the EU in recent national elections in the Netherlands or France tends to confirm this awakening (Schön-Quinlivan, 2017).

A series of studies have demonstrated effects on political competition, such as the convergence of mainstream party programs for many EU member states (Dorussen and Nanou, 2006; Nanou and Dorussen, 2013). Beyond convergence, the other major direction of change is the emergence of alternative cleavages and the rise of new issues, such as sovereignty, migration, and related issues. This may favor, first of all, the emergence or the rise of populist parties, especially on the right (Mudde, 2011; Swank and Betz, 2003). But mainstream parties have not been inactive: They have tended to shift attention to new issues and cleavages, away from the traditional left–right issues (Adams et al., 2009; Hellwig, 2014; Hellwig and Samuels, 2007). Third, some have argued that internationalization has diminished the very utility of voting and thus led to lower turnout (Marshall and Fisher, 2015; Steiner, 2010; Steiner & Martin, 2012). As parties converge, voting makes less of a difference, thus decreasing voting utility and, possibly, increasing the appeal of populist parties (Abou-Chadi, 2016). In a related fashion, internationalization should undermine economic voting, as blame attribution becomes more difficult in a more interconnected world and increases the relative importance of noneconomic issues in party choice (Hellwig, 2014; Kayser, 2007).

At the same time, it has become increasingly clear that the economic judgment of voters is much less “egotropic” than assumed by earlier work (cf. Nannestad and Paldam, 1997). Voters are embedded in national economies and these strongly influence voters’ perception of the economy and internationalization (Hellwig, 2014; Van der Brug et al., 2007). This, we argue, speaks to research in the area of comparative political economy looking at countries as a unit of analysis. This type of approaches adopts a longer term perspective. Internationalization is thus viewed as a gradual process rather than a shock.

From this perspective, party convergence and diminishing voter utility are not necessary consequences. They have to be understood in the wider historical context of internationalization. Countries have adopted diverging long-term strategies with regard to economic internationalization. As a matter of fact, some countries compensate the “losers of globalization” (Kriesi, 2008) much more generously than others (Rodrik, 2011). Consequently, the effect of important shocks related to internationalization should be buffered more or less efficiently by domestic institutions. Long-term “openness” in trade or other may favor the integration of internationalization into existing political cleavages (cf. Katzenstein, 1985). Under these circumstances, short-term variations in internationalization should not affect the importance of the left–right dimension. The story should be quite different for countries with less long-term experience of internationalization. Here, short-term shocks are likely to have much more disruptive effects on political competition and behavior.

In a nutshell, we assume that economic internationalization has a diverging short-term and long-term impact on political attitudes. In the short-term, it should weaken the explanatory power of the left–right dimension in political competition and, thus, for individual electoral preferences. In the long run, institutions—and parties—may adapt to changing contexts, compensate potential losers, and thus insulate political competition and voters from the disruptive effects of economic internationalization.

Assumptions

Our central hunch derived from the above discussion is that internationalization does change the domestic political game. As internationalization increases, it will have an effect on government autonomy and thus on the capacity of ruling elites to make a difference once in power. This should affect voters’ attitudes and thus weaken the importance of the left–right dimension for vote choice.

H1: Economic internationalization hypothesis: As internationalization increases, we expect the link between left–right self-placement and party preference to weaken.

However, we expect certain countries to be better prepared for internationalization than others. Hence we expect short-term and long-term patterns of internationalization effects to vary. Internationalization is likely to have been accommodated into domestic political debates in countries that have a long tradition in openness. Short variations, such as sudden shocks, for example, a sudden drastic deterioration of the balance of payments, should have a weaker effect on the power of the left–right dimension in those countries, compared to less open countries. Our second hypothesis thus qualifies hypothesis 1:

H2: Differential internationalization hypothesis: Countries with long-term patterns in economic internationalization should experience hypothesis 1 in a weaker form than less open countries.

We want to test these assumptions for Western EU member states. The EU has clearly been a major test case for the study of the consequences of internationalization on political behavior. The main reason probably is that there is a wealth of data and surveys that facilitate this type of study. It is a highly internationalized region with an exceptionally high level of political integration. But differences in the degree of internationalization persist. Some of this is linked to the size of countries, as smaller countries are systematically more open to trade than larger ones (Katzenstein, 1985, 2003). Certain countries are more competitive and manage to attract greater amounts of foreign direct investment (FDI). In sum, the reality of internationalization varies strongly from country to country. Taking the country-specific context as well as individual-level factors into account, we want to assess the extent to which variations in the level of internationalization affect the relative importance of the left–right dimension.

Data and methods

To test the mediating impact of economic internationalization on the strength of left–right politics in the EU, we created a data set based on four main sources: mass surveys conducted after each European election since 1999³ to get information on citizen’s positions and preferences; expert surveys of party positions;⁴ macroeconomic data from the World Economic Outlook database compiled by the International Monetary Fund⁵; and data from the World bank.⁶

The resulting data set allows us to assess the impact of internationalization over 15 years. To ensure comparability, we limit the study to the 14 West European countries⁷ that are present in the four waves of the EES that we examine. For each country, about 1000 respondents were asked questions in the EES, including their stance toward the main political parties in their countries alongside key indicators of their sociodemographic characteristics. We pooled these data with information about political parties from the CHES so as to get an external source for party positions rather than relying on individual perceptions of party locations only.⁸ If we retain all cases for which information is available both on voters’ attitude toward given parties in EES and information about these parties in CHES, we obtain a database comprising 56,031 individuals, 130 different political parties, and 56 country-by-year observations.

Detailed sources and descriptive statistics are available in the Online Appendix. The sequel presents the operationalization of the key variables and methods used in this article.

Propensity to vote at the individual level

We use propensities to vote (PTVs) as the dependent variable of this study. All respondents were asked in the EES surveys how likely they would be to ever vote for each of the national parties that was deemed significant at the moment of the study. PTVs have been shown to represent a reliable measure of preferences toward actors of any given party system (see Tillie et al., 1995; Van der Eijk, 2002; Van der Eijk et al., 2006). They are correlated to vote or sympathy toward candidates. They offer the advantage of mapping voters’ preferences for all parties, as most respondents answer this type of question for most parties. This explains the rapid development of their use in electoral studies (e.g. Harteveld et al., 2015; Van der Brug et al., 2007). It has to be noted that virtually all the individuals in our sample are able to formulate PTVs for at least some of the parties present in the system.

Using PTVs as the dependent variable means that individuals are not the unit of observation in our study. We use a stacked version of our data set, each individual evaluation of each party representing one observation. This means that the data set we use is in fact composed of 306,604 individual assessments of the PTV for specific parties.

This kind of stacked data set implies a number of constraints in terms of modeling. This is especially the case if we want to explain PTVs by the characteristics of voters. Indeed, voters’ characteristics would in this case explain the average level of PTVs across parties by types of voters rather than the impact of an individual characteristic on the perception of a specific party. Van den Brug et al. (2007) developed an empirical strategy to overcome this limitation. They propose to consider rather the “affinity” between individual characteristics and PTVs by a two-step procedure. We follow their method: In a first step, we predict PTVs party by party, using individual characteristics of interest. We then store the predicted scores based on these regressions and center the values in the complete database around these predictions to keep “only the variations caused by the independent variable” (Van den Brug et al. (2007)). These predicted and centered values are then stacked to generate a generic independent variable which is party-specific.

We use this procedure to generate our main control variable at the individual level. As the process is rather cumbersome, we chose to keep these controls as simple as possible.⁹ We restricted our attention only to gender, age, church attendance (coded as a dummy for those who attend more than once a year), education (dummy identifying those with a level of education superior to lower secondary), and social class (dummy identifying the working class, unemployed people, and unskilled workers vs the rest). All are coded as dummies (except age which is continuous), and all predicted scores are considered as constant over years of elections. Each of these variables has implied running 130 separate regressions.

Our key interest, however, is the impact of the ideological distance between parties and voters on PTVs. These measures of distance are based on a scale representing the left–right dimension, that is, a 11-point scale, based on EES for voters and CHES for parties. We simply compute the Euclidean distance between voter and party positions on these scales.¹⁰ We then log these values to prevent the results from being driven by extreme values and to allow for a decreasing impact of distance on support.¹¹ These distances are of course party-specific and do not require any further transformation.

Economic internationalization

The central hypothesis of this article assumes a mitigating effect of economic internationalization on the structuring impact of the left–right distance on the PTVs for parties. Economic internationalization presents a complex set of issues in terms of empirical operationalization. We looked at a host of possible indicators of economic internationalization. In line with Marshall and Fisher (2015), we understand internationalization as a rather multifaceted process that does not have a single or univocal impact. There are aggregate measures such as the Swiss Economic Institute (KOF) globalization measures (Dreher et al., 2008). While these are very informative, they also mix up potentially rather diverging elements. Hellwig has usually preferred a measure of “trade openness” (Hellwig, 2007; Hellwig and Samuels, 2007). Adams and colleagues (2009) distinguish short-term and long-term indicators of international economic integration. Similarly, we are interested in the way in which long-term and short-term variations interact. Internationalization is obviously nourished by both. We thus define two different parameters of interest:¹²

Trade, of goods and services, which probably represents the most self-evident measurement of cross-border economic transactions; we use exports over gross domestic product (GDP) as this measure of trade openness.

Capital flows in investments (usually known as FDIs), which have come to the fore in political debates in the past few years. We use the yearly flux of incoming FDIs¹³ (as a proportion of GDP) for this dimension.

These measurements are treated as tapping the same latent construct, that is, economic internationalization. We thus expect the same type of consequences from higher dependence on international trade or higher dependence on FDIs. We know, however, that these dimensions are only weakly related. They are also substantially different as we know that the yearly variance of these variables is very different. As illustrated by Figure 1, exports are on average stable over time. On the contrary, Figure 2 shows that FDI can vary very much from one year to another, depending on the global economic context as well as country specific factors. It comforts us in not choosing to group these indicators (with otherwise a risk of regression to the means). It also means that these differences fit well with the idea of contrasting internationalization both as a long-term perspective and short-term shock.

Figure 1.

Exports over time by country (in % of GDP, log values). GDP: gross domestic product.

Figure 2.

FDIs (inward flux) over time by country (in % of GDP, log values). FDI: foreign direct investment.

The mediating role of internationalization is then achieved by interacting these variables with the distance on the left–right dimension.

These key measurements of internationalization are also controlled for by three series of indicators. At the country level, we consider interest payments for public debt (in % of the GDP) to control for the impact of the “debt crisis,” key in our period of interest. At the party level, PTVs can be directly affected by the average support for the party. We thus control for the vote share of each party at the European election. We also consider party families, as delineated by CHES, to consider the likely variation in the weight of the left–right distance in shaping party utilities. It is, for instance, expected that parties outside the most traditional understanding of the left–right axis (ethnoregional parties, radical right parties, etc.) would be less affected by this dimension. We also include parties’ participation to the current national government to consider the impact of retrospective evaluation of the incumbent. At the level of individuals, as already stated, we control for basic sociodemographic indicators (gender, age, church attendance, level of education, and profession).

Modelling

Because we are most interested in the impact of macrofactors on microbehaviors, we estimated models encompassing our entire database. We took into account the multilevel structure of our data set by differentiating three levels. Level 1 units are the respondent by party observations. Level 2 units are political parties. Level 3 units are countries by year of observation. We have then a cross-classified multilevel design as level 1 observations are embedded simultaneously in levels 2 and 3. This makes it possible to take into account the continuity of parties across time in our design.

Results

We try to assess the mediating impact of globalization on the weight of the left–right cleavage to shape party preferences (PTVs) in the Western Europe. Before a direct test of our hypothesis, Table 1 offers a benchmark with all variables and controls except for our variables of interest. The left–right distance does the main job in explaining the formation of individual party preferences. If most coefficients appear as significant, their power is generally limited. We can also note that parties from the left gather higher PTVs. Unsurprisingly, larger parties in terms of vote attract also more support. Notice that the economic crisis has a significant negative impact especially in 2009, while propensities tend to be on average higher in 2014. In the same vein, higher levels of reimbursement of public debt tend to decrease the overall support for parties.

Table 1.

PTV for parties in accordance with left–right distance.

	Model 1
	b	SE	p
Left–right distance (ln)	−1.78	0.01	***
Gender (pr)	0.07	0.01	***
Age (pr)	0.09	0.00	***
Church attendance (pr)	0.13	0.01	***
Education (pr)	0.06	0.01	***
Social class (pr)	0.09	0.01	***
Year (ref. 1999)
2004	−0.02	0.10
2009	−0.30	0.10	***
2014	0.35	0.10	***
Vote in EP elections	0.08	0.00	***
Part. in govt	0.08	0.07
Interest payments (ln)	−0.24	0.10	**
Party family (Ref) radical right)
Conservative	0.41	0.22
Liberals	0.71	0.20	***
Christian democrats	0.57	0.22	**
Socialists	1.01	0.20	***
Radical left	1.24	0.20	***
Green	1.60	0.21	***
Regionalists	0.01	0.23
No family	0.13	0.27
Confessional	0.19	0.32
Agrarian	0.70	0.40
Constant	−14.3	0.71	***
Var. (party)	0.20	0.04
Var. (country year)	0.26	0.03
Log likelihood	−766,004.33
n	306,604
n_2	130
n_3	56

PTV: propensity to vote; SE: standard error; EP: European Parliament.

***p < 0.005; **p < 0.01;* p < 0.05.

Does the inclusion of our different measures of economic internationalization change the results? Generally, these measures and their interaction with left–right distance do not alter significantly the coefficients of most of the controls included at the individual and party levels. Table 2 displays only the results for the macrovariables, omitting the other variables included due to space constraints (cf. Table 1).

Table 2.

PTV for parties in accordance with left–right distance and interaction with indicators of internationalization.

	Model 2			Model 3			Model 4
	b	SE	p	b	SE	p	b	SE	p
Left–right distance (ln)	−1.91	0.01	***	−2.32	0.09	***	−2.54	0.13	***
FDI (ln)	−0.11	0.03	***				−0.69	0.29	**
Exports (ln)				0.28	0.10		0.42	0.14	***
Left–Right × FDI (flux in)	0.12	0.01	***				0.69	0.09	***
Left–Right × Exports				0.15	0.02	***	0.16	0.03	***
Left–Right × FDI × Exports							−0.14	0.02	***
Controls (individuals)	^			^			^
Controls (party)	^			^			^
Controls (country)	^			^			^
Var. (party)	0.22	0.05		0.17	0.03		0.15	0.04
Var. (country year)	0.23	0.02		0.28	0.02		0.24	0.02
Log likelihood	−762,256			−772,098			−782,225

PTV: propensity to vote; SE: standard error; FDI: foreign direct investment.

***p < 0.005; **p < 0.01; *p < 0.05.

Sources: EES 1999, 2004, 2009 and 2014, CHES trend file, World Bank and IMF.

Overall, interactions between all our measurements of economic internationalization and left–right distance are significant. Reading coefficients associated with interaction remains tricky. This is why substantive effects are presented in Figures 3 to 5, depicting the predicted propensities to support a party in accordance with distance on the left right scale in various economic contexts. For each of the models, we propose two graphs displaying in different fashions the same information. Traditional marginal effects for continuous variables are positioned on the left-hand panels of Figures 3 to 5. They show the predicted propensities to support a party in accordance with distance to this party at given levels of FDI, if we consider Figure 3. We see easily that propensities decline at a steeper pace when FDI is negative. This means that the left–right dimension is more “structuring,” that is, distance on this dimension makes more of a difference in terms of the chances to support a party. On the right-hand panels of Figures 3 and 4, predicted probabilities are directly shown as a factor of distance and economic context with discretization of PTVs. In Figure 3, for instance, the PTV for a given party that is at distance 0 from the respondent varies from the interval [5.8, 10] to the interval [5.1, 5.8] when the log of the FDI varies from −3 to +4. Again in Figure 3, we see then that propensities vary more for lower values of FDI than for higher values. This confirms that the left–right cleavage plays more when FDI is low.

Figure 3.

PTV for parties in accordance with left right distance and FDI level. Note: The left panel on this figure depicts the marginal impact of a change in distance on the left–right scale on the propensity to support a party. The different lines correspond to different economic contexts. On the right panel, we plot the predicted propension to support a party in accordance with the distance between voter and party on the left–right scale and the value of the economic indicator. The substance of both graphs is the same, except that the left panel indicates changes in values of PTVs, whereas the right panel shows absolute values of predicted PTVs. PTV: propensity to vote; FDI: foreign domestic investment.

Figure 4.

PTV for parties in accordance with left–right distance and exports level. Note: The left panel on this figure depicts the marginal impact of a change in distance on the left–right scale on the propensity to support a party. The different lines correspond to different economic contexts. On the right panel, we plot the predicted propension to support a party in accordance with the distance between voter and party on the left–right scale and the value of the economic indicator. The substance of both graphs is the same, except that the left panel indicates changes in values of PTVs, whereas the right panel shows absolute values of predicted PTVs. PTV: propensity to vote.

Figure 5.

PTV for parties in accordance with left–right distance, FDI, and exports levels. Note: The left panel on this figure depicts the marginal impact of a change in distance on the left–right scale on the propensity to support a party. The different lines correspond to different economic contexts. On the right panel, we plot the predicted propension to support a party in accordance with the distance between voter and party on the left–right scale and the value of the economic indicator. The substance of both graphs is the same, except that the left panel indicates changes in values of PTVs, whereas the right panel shows absolute values of predicted PTVs. PTV: propensity to vote; FDI: foreign direct investment.

Following the same logic, Figure 3 and 4 lead to same conclusion: Higher values of internationalization lead to less impact of the left–right cleavage on PTVs. Given the variance of both variables on the period and in the countries considered,¹⁴ it is, however, impossible to say that one of the effects is significantly stronger than the other. Exports are probably a bit more powerful in explaining propensities, but the effect is not substantially more significant than the impact of FDI.

The panels in Figure 5, corresponding to model 4 in Table 2, examine the interaction between left–right distance, exports, and FDI. This represents a test of our second hypothesis stating that more long-term economic internationalization may mitigate the impact of short-term shocks in terms of flux of FDI. Figure 5 proves this is indeed the case. Predictions of PTVs decline more quickly with distance when both exports and FDI are at low level. But when exports are at a high level, we see now that predictions are declining now more steeply when FDI is positive. In other terms, exports not only mitigate but even invert the impact of short-term fluctuations of FDI. The right-hand panels on Figure 5 make it clearer that variance is also more important in the context of lower levels of exports.

Summing up, according to hypothesis 1, we expected internationalization to weaken the effect of left–right distance. This is what we find with any of the two indicators of internationalization taken into account. This clearly confirms our hypothesis: Economic internationalization does mitigate the effect of left–right distance on PTVs. Yet, short-term fluctuations and long-term trends interact significantly. Less exchanges in the short-term implies more polarization in countries with lower levels of exports while it increases polarization on the left–right scale in countries with a stronger tradition of exports.

Conclusion

This article aimed at testing two hypotheses about the impact of economic internationalization on the structuring role of the left–right dimension on the formation of party preferences, measured by PTVs. One the one hand, internationalization was expected to decrease the salience of the left–right cleavage for vote choice. On the other hand, we assumed that long-term economic internationalization may mitigate the effects of short-term variations in the degree of internationalization.

Regarding the first hypothesis, results confirm our expectations for the two indicators that we have used here, that is, trade and FDI. Both indicators have a robust moderating effect on the consequences of left–right distance on the formation of party preference. The more internationalized an economy, the less individual voters will rely on their judgment of the distance between given party and themselves to make their vote decision. This confirms some earlier work that has shown that globalization indeed affects voting behavior (Hellwig, 2014; Marshall and Fisher, 2015). The current article extends those previous works by covering a much larger sample across 20 years and 14 countries. Results are robust and conceptually significant: Voters do adapt to a changing world by adjusting the criteria they use to select parties. The hitherto dominant left–right dimension is losing some of its explanatory power. As individual voters have to compete for jobs in more internationalized settings, their personal skills and assets will restructure their political attitudes.

Regarding the second hypothesis, we indeed found that the impact of short-term fluctuations in terms of FDIs is moderated by long-term openness to trade. Put differently, countries with a long tradition of trade openness buffer short-term shocks more smoothly than other countries. We see two possible and probably complementary interpretations of this.

First, while this article does not explore the micromechanisms underlying this hypothesis, we can assume that at least part of this effect results from the capacity of national economies to adapt to volatility in FDI. Short-term variations or shocks should then have less-disruptive consequences for both the national economy and political competition as redistributive institutions buffer the effects for individual citizens. Individuals are more insulated from the risks related to internationalized labor markets. Therefore, they ought to be less likely to radically change their political heuristics.

The second interpretation is related to the politics of FDI itself. FDI is of course linked to trade openness, but it is also much more volatile. As to inward FDI, the measure we are actually using in this article, it has become a major stake for contemporary economic policies. Governments try to attract FDI through preferential tax policies and other accommodating measures. The differential impact of FDI across levels of trade openness may be related to different politicizations of the issue. While FDI flux in open economies is absorbed in preexisting political divides, it becomes a divisive issue in its own right in less open economies.

Overall, our results confirm the political impact of the contemporary transformations of capitalism on political behavior. This impact is significant, both substantively and statistically. We also show that economic internationalization is indeed a multidimensional phenomenon.

The analyses of the dynamics of economic internationalization and political behavior presented in this article suggest several future research directions. The channels and mechanisms of the impact of internationalization are still to be explored. A lot has still to be known about the impact of internationalization on the content of the left–right dimension as well as actual government discretion in policy decisions, how parties react to this context, how internationalization is politicized in domestic political debates, or how voters themselves are aware of and perceive internationalization. We believe that this article has laid the foundations for future research in these directions.

Supplemental material

Supplemental Material, Appendix - Economic internationalization and the decline of the left–right dimension

Supplemental Material, Appendix for Economic internationalization and the decline of the left–right dimension by Jan Rovny, Stephen Whitefield, Emiliano Grossman, and Nicolas Sauger in Party Politics

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material for this article is available online.

Notes

References

Abou-Chadi

(2016) Niche party success and mainstream party policy shifts–How green and radical right parties differ in their impact. British Journal of Political Science 46(2): 417–436.

Adams

Ezrow

(2009) Who do European parties represent? How Western European parties represent the policy preferences of opinion leaders. Journal of Politics 71(1): 206–223. DOI: 10.1017/S0022381608090130.

Adams

Clark

Ezrow

. (2004) Understanding change and stability in party ideologies: Do parties respond to public opinion or to past election results? British Journal of Political Science 34(4): 589–610.

Adams

Clark

Ezrow

. (2006) Are niche parties fundamentally different from mainstream parties? The causes and the electoral consequences of Western European parties’ policy shifts, 1976–1998. American Journal of Political Science 50(3): 513–529.

Adams

Haupt

Stoll

(2009) What moves parties? The role of public opinion and global economic conditions in Western Europe. Comparative Political Studies 42(5): 611–639.

Aldrich

Popa

(2008) Clarity of party positions on ideology: a cross-national analysis. In: CSES conference: Political institutions—rationality—electoral behavior, Warsaw, Poland.

Aldrich

Fernandez

(2013) Uncertainty or Ambiguity? Sources of Variation in Ideological Placements of Political Parties. Chicago: Annual Meeting of the American Political Science Association.

Aldrich

Dorobantu

(2010) The Use of the Left-Right Scale in Individual’s Voting Decisions. Washington, DC: Annual Meeting of the American Political Science Association, p. 28.

Bakker

Jolly

Polk

(2012) Complexity in the European party space: exploring dimensionality with experts. European Union Politics 13(2): 219–245.

10.

Bakker

de Vries

Edwards

(2015) Measuring party positions in Europe: The Chapel Hill expert survey trend file, 1999-2010. Party Politics 21(1): 143–152.

11.

Dorussen

Nanou

(2006) European integration, intergovernmental bargaining, and convergence of party programmes. European Union Politics 7(2): 235–256.

12.

Dreher

Gaston

Martens

(2008) Measuring Globalisation: Gauging Its Consequences. Springer Science & Business Media.

13.

Eichenberg

Dalton

(2007) Post-Maastricht blues: the transformation of citizen support for European integration, 1973-2004. Acta Politica 42(2–3): 128–152.

14.

Freire

(2008) Party polarization and citizens’ left-right orientations. Party Politics 14: 189–209.

15.

Frieden

(1991) Invested interests: the politics of national economic policies in a world of global finance. International Organization 45(4): 425–451.

16.

Gabel

(1998) Economic integration and mass politics: market liberalization and public attitudes in the European Union. American Journal of Political Science 42(3): 936–953.

17.

Gourevitch

(1978) The second image reversed: the international sources of domestic politics. International Organization 32: 881–911.

18.

Green-Pedersen

(2007) The growing importance of issue competition: the changing nature of party competition in Western Europe. Political Studies 55(3): 607–628.

19.

Harteveld

Van Der Brug

Dahlberg

. (2015) The gender gap in populist radical-right voting: examining the demand side in Western and Eastern Europe. Patterns of Prejudice 49(1–2): 103–134.

20.

Hellwig

(2007) Globalization and perceptions of policy maker competence: evidence from France. Political Research Quarterly 60(1): 146–158.

21.

Hellwig

(2014) Globalization and Mass Politics: Retaining the Room to Maneuver. Cambridge University Press.

22.

Hellwig

Samuels

(2007) Voting in open economies the electoral consequences of globalization. Comparative Political Studies 40(3): 283–306.

23.

Hooghe

Marks

(2009) A postfunctionalist theory of European integration: from permissive consensus to constraining dissensus. British Journal of Political Science 39(1): 1–23.

24.

Katzenstein

(1985) Small States in World Markets: Industrial Policy in Europe. Ithaca: Cornell University Press.

25.

Katzenstein

(2003) “Small states” and small states revisited. New Political Economy 8(1): 9–30.

26.

Kayser

(2007) How domestic is domestic politics? Globalization and elections. Annual Review of Political Science 10: 341–362.

27.

Kitschelt

(1994) The Transformation of European Social Democracy. Cambridge: Cambridge University Press.

28.

Kriesi

(2008) West European Politics in the Age of Globalization. Cambridge: Cambridge University Press.

29.

Lachat

(2008) The impact of party polarization on ideological voting. Electoral Studies 27: 687–698.

30.

Mair

Thomassen

(2010) Political representation and government in the European Union. Journal of European Public Policy 17(1): 20–35.

31.

Marks

Hooghe

Nelson

. (2006) Party competition and European integration in the east and west. Different structure, same causality. Comparative Political Studies 39(2): 155–175.

32.

Marshall

Fisher

(2015) Compensation or constraint? How different dimensions of economic globalization affect government spending and electoral turnout. British Journal of Political Science 45(2): 353–389.

33.

Meijers

(2015) Contagious Euroscepticism: the impact of Eurosceptic support on mainstream party positions on European integration. Party Politics 23: 413–423. DOI: 10.1177/1354068815601787.

34.

Merrill

III Grofman

(1999) A Unified Theory of Voting: Directional and Proximity Spatial Models. New York: Cambridge University Press, p. 213.

35.

Miklin

(2014) From “sleeping giant” to left–right politicization? National party competition on the EU and the Euro crisis. JCMS: Journal of Common Market Studies 52(6): 1199–1206.

36.

Mudde

(2011) Radical right parties in Europe: What, who, why? Participation 34(3): 12–15.

37.

Nannestad

Paldam

(1997) From the pocketbook of the welfare man: a pooled cross-section study of economic voting in Denmark, 1986-92. British Journal of Political Science 27(1): 119–136.

38.

Nanou

Dorussen

(2013) European integration and electoral democracy: how the European union constrains party competition in the Member States. European Journal of Political Research 52(1): 71–93.

39.

Noël

Thérien

(2008) Left and Right in Global Politics. Cambridge: Cambridge University Press, p. 24.

40.

Piurko

Schwartz

Davidov

(2011) Basic personal values and the meaning of left-right political orientations in 20 Countries. Political Psychology 32(4): 537–561.

41.

Polk

Rovny

Bakker

. (2017) Explaining the salience of anti-elitism and reducing political corruption for political parties in Europe with the 2014 Chapel Hill Expert Survey data. Research & Politics (January-March): 1–9.

42.

Rodrik

(2011) The Globalization Paradox: Democracy and the Future of the World Economy. WW Norton & Company.

43.

Rogowski

(1989) Commerce and Coalition: How Trade Affects Domestic Political Alignments. Princeton: Princeton University Press.

44.

Schmitt

Bartolini

van der Brug

(2009) European Election Study 2004, 2nd ed. GESIS Data Archive, Cologne. ZA4566 Data file Version 2.0.0 DOI: 10.4232/1.10086.

45.

Schmitt

Popa

Devinger

(2015) European Parliament Election Study 2014, Voter Study, SVoter Study, Supplementary Study. GESIS Data Archive, Cologne. ZA5161 Data file Version 1.0.0, DOI: 10.4232/1.5161.

46.

Schön-Quinlivan

(2017) “The elephant in the room” no more: Europe as a structuring line of political cleavage in the 2017 presidential election. French Politics 15(3): 290–302. DOI: 10.1057/s41253-017-0041-x.

47.

Steiner

(2010) Economic globalization and voter turnout in established democracies. Electoral Studies 29(3): 444–459.

48.

Steiner

Martin

(2012) Economic integration, party polarisation and electoral turnout. West European Politics 35(2): 238–265.

49.

Swank

Betz

(2003) Globalization, the welfare state and right-wing populism in Western Europe. Socio-Economic Review 1(2): 215–245.

50.

Taggart

(1998) A touchstone of dissent: Euroscepticism in contemporary Western European party systems. European Journal of Political Research 33(3): 363–388.

51.

Tillie

(1995) Party Utility and Voting Behavior. PhD Thesis, University of Amsterdam. Available at: http://dare.uva.nl/personal/pure/en/publications/party-utility-and-voting-behavior(9a4c841d-c184-439b-b36c-00adf39537cb).html

52.

Van der Brug

Van der Eijk

Franklin

(2007) The Economy and the Vote: Economic Conditions and Elections in Fifteen Countries. Cambridge: Cambridge University Press.

53.

Van der Eijk

Franklin

Schoenbach

. International Research Group (1999). European Election Studies. Hamburg, Germany: IPSOS, (primary investigator) European Election Study. DANS. DOI: http://doi.org/10.17026/dans-z9j-vy6m.

54.

Van der Eijk

(2002) Design issues in electoral research: taking care of (core) business. Electoral Studies 21(2): 189–206.

55.

Van der Eijk

Schmitt

Binder

(2005) Left-right orientations and party choice. In: Thomassen

(ed), The European Voter. Oxford: Oxford University Press, pp. 134–158.

56.

Van der Eijk

Van der Brug

Kroh

. (2006) Rethinking the dependent variable in voting behavior: on the measurement and analysis of electoral utilities. Electoral Studies 25(3): 424–447.

57.

Van Egmond

Van der Brug

Hobolt

(2013) European Parliament Election Study 2009, Voter Study. GESIS Data Archive, Cologne. ZA5055 Data file Version 1.1.0. DOI: 10.4232/1.11760.

58.

van Elsas

van der Brug

(2015) The changing relationship between left–right ideology and Euroscepticism, 1973–2010. European Union Politics 16(2): 194–215. DOI: 10.1177/1465116514562918.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.33 MB