Beyond policy positions

Abstract

Do parties adapt their programmatic strategies in times of heightened economic globalization? Are these changes captured by right-left positional changes or do parties go beyond policy shifts and enact more comprehensive programmatic overhauls? Furthermore, are such changes linked to traditional party family classifications and, if so, do different party types re-program their manifestos differently? Finally, what role does radical right competition play in the changing programmatic strategies of mainstream centre-right and centre-left parties? This paper addresses these questions by developing a theoretical framework that accounts for economic globalization, cleavage change, and programmatic supply. Using Giebler et al.’s (2015) measure of programmatic clarity, the analysis reveals clear differences in party responses to economic globalization. Additionally, the results show that parties go beyond right-left positional changes and adapt their programmatic supply on a more general level. For social democratic parties, however, such adaptation hinges on whether a radical right competitor is present.

Keywords

comparative politics party manifestos political parties

Introduction

Parties are core actors in representative democracies who forge a crucial representative link between citizens and governments. Yet there is widespread speculation that parties are failing to fulfil this core democratic function as well as they once did (Dalton and Wattenberg, 2000). Much of this failure is attributed to the sluggish way that traditional parties have responded to ongoing transformations of Western European societies brought about by continued globalization, particularly economic globalization, and the increasing salience of new cleavages (Kriesi et al., 2008, 2012). While there is a consensus that economic globalization has affected parties’ strategies somehow, there is less agreement concerning the exact nature of this effect. Scholars examining links between economic globalization and shifting party positions focus predominantly on the link between voters’ preferences and party positional shifts (Adams et al., 2004; Ezrow et al., 2011) or on how parties adapt a specific policy stance (e.g. trade policy, see Milner and Judkins, 2004). Existing research relies heavily on right-left ideological change to identify and explain shifting party strategies. There is less emphasis on examining other possible ways that parties can adapt their programmatic profiles to account for economic globalization independent of positional changes and whether these strategies are dependent on party type (although recent work by Ward et al. (2015) and Adams et al. (2009) provides two examples of work that attempts to remedy these shortcomings).

This article makes three contributions: empirical, methodological, and conceptual. Empirically, the results of the analysis counter Adams et al.’s (2009) findings concerning the interaction of left party status and economic globalization. While that analysis showed little support for the so-called “leftist exceptionalism” of left party responses to globalization pressures (Adams et al., 2009; 630), the current analysis does find evidence of left exceptionalism. Explicitly, the findings of the current analysis show that left parties do respond to globalization pressures differently than other types of parties, but that this response is only significant when moderated by the presence of a radical right competitor.

Methodologically, by applying Giebler et al.’s (2015) measure of programmatic clarity, the analysis is able to test how parties from different ideological backgrounds engage in different types of manifesto reprogramming (Weßels, 2001) in response to heightened levels of economic globalization. Conceptually, two types of reprogramming are introduced: obscuring and clarifying programmatic supply. Obscuring occurs when parties’ simultaneously support both sides of the same issue (i.e. statements both for and against welfare state spending in the same manifesto) or simply avoid discussing a policy altogether. Clarifying occurs when parties stress only one side of a policy issue in their manifesto. Finally, concerning the role of competitor parties, there is excellent theoretical work discussing the role of radical right competition in mainstream party strategies and, explicitly, the strategies of social-democratic parties (Bale et al., 2010). Therefore, in a final step, the analysis tests whether the presence of a radical right competitor affects how social democratic parties adapt their programmatic strategies.¹

The findings reveal significant differences in how party type conditions a party’s response to heightened economic globalization independent of, although possibly in addition to, right-left positional changes. Specifically, the analysis shows that left parties obscure their programmatic stances when economic globalization increases and clarify when economic globalization decreases. However, non-left parties rarely adapt their programmatic supply in the face of changing levels of economic globalization and, for this group, economic globalization has little effect on the clarity of their manifestos. Finally, the results reveal clear differences between the programmatic responses of social democratic parties facing radical right competitors and that of social democratic parties running in elections where no radical right party is present. When social democratic parties face radical right competition they significantly obscure their programmatic supply, but when there is no radical right party competitor social democratic parties opt to clarify their positions.

Toward a comprehensive theoretical framework: Economic globalization, cleavages, and parties

Research on economic globalization and political parties typically adopts one of two perspectives: a political economic perspective or a political parties perspective, that, until very recently at least, followed the general pattern of two ships passing in the night.² Most party scholars approach the problem from the direction of cleavages, explore how globalization affects the traditional Rokkanian cleavage space, and, in a next step, examine how the decreased salience of traditional cleavages affects party transformation and the strategies parties adopt to deal with these changes. A smaller subset of scholars approaches the question from a political economy perspective. While these scholars also examine how party strategies change, they begin with the assumption that globalization, and specifically the shifting economic paradigm from the Keynesian phase that characterized the period directly following WWII toward neoliberalism beginning in the 1960s and 1970s, pressured parties on both the left and right to adapt their traditional policy stances. Of key importance is that while the two perspectives may differ in their starting points (i.e. cleavages or the economy), the predictions they make concerning how parties adapt their strategies are similar. What is needed, then, is a comprehensive theoretical framework that links globalization and changing party programmatic strategies with what we know of the root causes of party decline and transformation—namely, the changing nature of cleavage structures.

Globalization: Shifting economic paradigms and changing party strategies

Held et al. (1999) conceptualize globalization as an ongoing process that has, in the post-war era until today, increased significantly in terms of both intensity and velocity. According to Held et al., the contemporary period of economic globalization began during the 1960s and 1970s and marked the end of the Bretton Woods era that had been in place since the signing of that agreement in 1944 (Held et al., 1999: 201). In terms of party politics, the Bretton Woods period was characterized by high national autonomy and, in general, the popularity of Keynesian economic policies (Kitschelt et al., 1999: 200–201) within (often Social Democratic) national governments and governing coalitions (Kitschelt et al., 1999; Przeworski and Sprague, 1986). Conversely, the perceived failure of Keynesian economic policies led to two outcomes: neoliberalism and renewed attempts to further European integration (Kitschelt et al., 1999: 6).

It is during this same period that the story of party decline and party transformation begins. Shifts in the dominant economic paradigm were mirrored by the end of the “Golden Age” of political parties (Reiter, 1989) and here, too, left parties were seen as the biggest losers (Bartolini, 2007; Brooks, 1983). This makes sense as the Golden Age of parties was, similar to the Bretton Woods era, dominated by social democratic-led governments supporting and implementing Keynesian economic policies (Kitschelt et al., 1999). Nevertheless, while the party transformation and decline literature surely acknowledges the role of shifting economic paradigms, the real emphasis is on the decreasing salience of traditional cleavages (Duverger, 1963; Katz and Mair, 1995; Kirchheimer, 1965; Kitschelt, 1995; Mair, 1997).

Beyond citizens: Parties as winners and losers of (economic) globalization

The class cleavage outlined by Lipset and Rokkan (1967) was primarily an economic cleavage that pitted the interests of workers against owners and “proved more uniformly divisive” than the other three cleavages identified by Lipset and Rokkan’s work (Bartolini and Mair, 2007: 60). However, the changing nature of societies saw the rapid disappearance and decline of traditional left (i.e. worker) constituencies (Kitschelt et al., 1999; Przeworski and Sprague, 1986) and an increased salience of post-materialist issues (Ingelhart, 1977). While globalization has generally led to a restructuring of the traditional European political space (Kriesi et al., 2008), not all European societies and/or voting constituencies were equally affected by these changes and not all parties responded in the same way. Instead, just as we can conceptualize the “winners and losers” of globalization on an individual level (cf. Kriesi et al., 2008, 2012), we can also conceptualize parties of certain ideological backgrounds as “winners” and “losers” of this process (Burgoon, 2012). Explicitly, the rise of economic globalization is directly linked to a shift in parties’ positions on social policies, especially welfare state policies (Boix, 1998), and here the findings clearly suggest that social democratic parties and conservative parties take very different approaches when compensating for globalization pressures (Boix, 1998). If social democratic parties were the clear losers, as much of the literature suggests (Bartolini, 2007; Merkel et al., 2008; Przeworski and Sprague, 1986), due in large part to the fact that they “disproportionally represent those facing globalization’s risks” (Burgoon, 2012: 607), then party families whose economic profiles were more aligned with neoliberal ideologies, such as Liberal parties and Christian Democratic or Conservative parties, should be the winners (Burgoon, 2012).³

Another beneficiary of the post-Bretton Woods era and the declining salience of traditional cleavages were the non-mainstream parties, and studies show that new parties—usually in the form of radical right or extreme right populist parties—and niche parties (Bornschier, 2010; Meguid, 2005; Wagner, 2012) benefitted from these societal changes over their mainstream counterparts. While traditional parties maintained their ties to the class cleavage, an ever-growing group of citizens no longer shared these ties (Dalton, 2000). There are clear indications from the theoretical literature that the presence of radical right contenders (and to some extent radical left contenders (Bale, 2003)) affects the strategies of mainstream parties (Bale, 2003, 2010), and most recent findings from the empirical literature confirm that niche party competition affects mainstream party strategies (van der Wardt, 2015; Meyer and Wagner, 2013). Therefore, ecology and nationalist party families fall solidly into the “winners of economic globalization” camp. Likewise, it is important to note that not all countries were similarly affected by the shift toward neoliberalism and increasing economic globalization.⁴

Taking together findings from the economic globalization literature and the party transformation and decline literature, we can formulate a clear set of predictions concerning how different party families should be differently affected by rapidly increasing economic globalization. Specifically, we have clear reasons to suspect that traditional left-of-center parties, namely Social Democratic and Socialist parties, should be the most negatively affected. Liberal parties, on the other hand, should be the clear winners of these changes while Christian Democratic and Conservative parties should also be clear beneficiaries, albeit to a lesser extent than Liberal parties. Finally, while it is possible to conceptualize winners and losers of economic globalization within the context of traditional, mainstream party families, the party transformation and decline literature makes clear that another classification is also possible: between traditional, mainstream parties (i.e. the losers) and non-traditional, new, and/or niche parties (i.e. the winners).

Party strategies: Responses to economic globalization and cleavage change

Recent work connecting shifts in party strategies explicitly to economic globalization focuses on change in overall ideological positions as evidenced by shifts in right-left positions (Adams et al., 2004, 2006; Ezrow et al., 2011; Schumacher et al., 2013), hereafter RILE. However, ideological change is not the only type of programmatic change that parties can make. Recently, more scholars have begun to move away from explanations that focus solely on positional changes (De Vries and Hobolt, 2012; Rovny, 2012; Ward et al., 2015) and instead include explanations stressing how parties emphasize some issues over others (Kluever and Spoon, 2015; Ward et al., 2015), blur their positions on certain issues (Rovny, 2012), or change the number of dimensions on which they compete (de Vries and Hobolt, 2012). Such work shows the importance of moving away from only focusing on right-left positional changes in order to conceptualize programmatic change. The theoretical framework proposed here relies on Weßels’ (2001) concept of Umprogrammierung, or ‘reprogramming’ of party manifestos and policy positions.

The concept of reprogramming is closely related to Rovny’s (2012) work on issue blurring versus issue emphasis. He argues, on the basis of recent findings about the effect of multidimensionality on party strategies (De Vries and Marks, 2012), that under certain conditions—explicitly ones of increasing dimensionality of competition—positional blurring may be a beneficial strategy. This strategy, according to Rovny, will be most beneficial to parties for whom an important issue dimension crosscuts the preferences of their constituency or where “the party is especially hard pressed to amalgamate the disparate views on the [crosscutting] dimension” (Rovny, 2012: 274). By all indications, globalization has increased the dimensionality of political competition (de Vries and Hobolt, 2012) as well as changed the foundational underpinnings of these dimensions (Hooghe et al., 2002; Kriesi et al., 2008, 2012) and has proven to be a crosscutting issue for traditional left-of-center party constituencies but not for Liberal and right-of-center party constituencies. Unlike Rovny’s conceptualization, though, the concept of reprogramming as Weßels conceives it is much broader than simple positional shifts or than blurring positions on a single issue dimension. Instead, reprogramming constitutes a near-complete overhaul of a party’s program and can be thought of as a more general structural change to a party’s programmatic supply.

Obscuring

The obscuring strategy occurs when parties deemphasize an existing issue/policy position or even ignore some issues all together. Opting for this strategy means that parties decrease the clarity of their programmatic supply. This ‘watering down’ of the manifesto will not necessarily take into account new issues (although this is certainly possible, too) so much as offer less clear stances on traditional programmatic policies. While it is clearly possible that parties will moderate their overall ideological positions (i.e. right-left positional changes), it is also possible that parties will choose to water down their programmatic offerings in ways not fully captured by RILE and/or in addition to RILE positional shifts. It is important to note that, theoretically, the expectation is that parties will engage in this strategy intentionally with the expectation that such a strategy will be helpful. In this way, obscuring, like Rovny’s position blurring differs from the unintentional ambiguity of candidate positions much lamented in American politics research (Alvarez, 1998) and from the ambiguity of policy positions that results from infighting among party elites (Rovny, 2012). Parties can obscure programmatic supply via two methods: first, they can address both pro and con sides of a policy (e.g. welfare state policies); secondly, they can simply de-emphasize issues where there is no clear constituency preference by not mentioning these issues in their manifestos. The measure discussed below—the programmatic clarity index—captures both possibilities.

Given the above discussion about how globalization creates party family winners and losers, the first expectation is that parties belonging to left-of-center Social Democratic and Socialist party families will engage in an obscuring strategy as the level of economic globalization increases (H1). Additionally, Bale et al. (2010) predict that social democratic parties especially will be threatened by radical right competitors. Specifically, they expect that radical right parties “own” issues that “command broad support across Western European electorates” such as immigration; and that these electorates—namely the lower educated working class—were those that traditionally formed the basis of social democratic support (Bale et al., 2010: 411), thereby crosscutting the traditional social democratic constituencies. Therefore, it is expected that social democratic parties who are facing a radical right competitor may make different strategic decisions in adapting their programmatic supply than those parties who do not face a radical right competitor (H2).

Clarifying

Those parties not tied to constituencies that were not crosscut by globalization related issues may find it more fruitful to make clear their manifesto positions rather than follow an obscuring strategy as they still benefit from clarity regarding their constituencies’ interests. This strategy should be especially attractive to the winners of globalization—namely Liberal parties and, to a lesser extent, right-of-center Christian Democratic and Conservative parties whose traditional policy stances are complemented by a neoliberal economic paradigm —who should choose to clarify their existing manifesto positions as the level of economic globalization increases (H3).

Data, modelling specifics, and operationalization

The data is structured so that each party represents a panel (total = 142). Observations are parties in a given election (N = 847) ranging between 1970 and 2009. The nature of the data structure (parties nested in countries and elections) clearly lends itself to a multi-level modelling approach and, indeed, this was the first model tested for this analysis. However, very low ICC levels (0.01 for the party level and 0.001 for the country level) revealed no justification for applying such a model, and a subsequent test of a time series regression with fixed versus random effects confirmed that there was little need for a fixed effects modelling approach based on the non-significance of the Hausman test. Still, given the panel structure of the data, there is a theoretical reason to suspect that the standard errors will have panel-specific first-order autocorrelation and controlling for this is necessary. Therefore, the models reported below are all time series regression models with panel corrected standard errors (Beck and Katz, 1995). The time series component of the model is calculated by the first election for a country in the dataset and then a running number from that point forward. For example, if a country has five elections included in the data set the time variable will range from 1 to 5 and does not depend on year or election date, which would leave gaps in the time series.⁵ However, despite the calculation of the time variable, there remains an unbalanced panel structure (not all parties run in all elections) which necessitates a forced pairwise case inclusion.⁶

Given that the real phenomenon of interest is whether and how parties change their strategies, all variables have been calculated into change variables. This makes interpretation of the results less straightforward, but conceptually such a choice is justified. As we are also interested in long-term rather than short-term change, the change variables are calculated as the change between three election periods: t, t-1, and t-2. Three election periods were selected as this represents, for the majority of the sample, approximately a 10-year period of change. The calculation takes the sum of the difference between t and t-1 and the difference between t-1 and t-2, thereby smoothing out large short-term variations. The descriptive statistics for the variables used in the models below are shown in Table 1. Figure 1 shows the temporal variation in the economic globalization variable for each country. Figure 2 shows the bivariate scatterplot of the programmatic clarity index (the DV) and the economic globalization variable (the main IV) and provides a linear fit line for left parties and non-left parties.

Table 1.

Descriptive statistics and sample overview.

Continuous variables ^a
		Mean		Std. Dev		Min	Max
Δ Programmatic Clarity (DV)		0.025		0.187		–0.696	0.774
Δ Vote Share (%)		–0.28		5.479		–28.02	34.50
Δ Right-Left Position (min = left; max = right)		0.977		18.65		–73.78	76.58
Δ Economic Globalization		5.695		4.749		–3.99	20.53
Categorical Variables
Left		0 = Not Left (baseline) 1 = Left					541 307
Mainstream		0 = Not Mainstream (baseline) 1 = Left 2 = Liberal 3 = Right					211 307 122 208
Social Democrats and Radical Right Parties (RRPs)		0 = Not–Social Democrat (baseline) 1 = Social Democrat/No RRP Competitor 2 = Social Democrat/RRP Competitor					652 125 71
Country		Sweden (baseline) Denmark Finland Belgium Netherlands Luxembourg France Italy			Spain Greece Portugal Germany Austria Switzerland Great Britain
Sampling Information
Obs(Party/Election)	N = 847	Time	Running count from election 1 for each country N = 215 T = 19 Gaps: 10 Years: 1970–2009
Panel (Party)	N = 142 (min = 1, avg = 6, max = 13)	Time

^a All change variables calculated as Δ = (t – t₁) + (t₁ – t₂).

Dependent variable: Programmatic clarity

The programmatic clarity index (PCI), developed by Giebler et al. (2015), provides a way to examine the level of clarity of parties’ programmatic supply based on how many times they mention either the pro or con position of a specific policy issue in their electoral manifesto. The measure uses data from the Manifesto Project (Volkens et al., 2012). The Manifesto Project coding scheme includes 56 policy categories of which 26 (13 matched pairs) are antipodal positional categories—meaning one positive and one negative policy category (e.g. “Welfare State: Positive” and “Welfare State” Negative” (Werner et al., 2011)).⁷ Giebler et al. (2015) begin with the assumption that parties with clear programmatic supply will consistently stress one side or the other of an issue rather than stressing both. Parties who wish to obscure their positions will include statements addressing both sides of the issue or will refrain from addressing the issue at all. Recall that the concept of reprogramming encompasses a broad change in programmatic supply that is not related solely to changes in economic policy positions despite the theoretical importance of economic globalization. Therefore, the PCI is particularly well suited for this analysis as it includes not only economic policies but provides a more all-encompassing measure of programmatic change on social and cultural issues as well thus allowing for a direct test of the theoretical prediction that economic globalization affects the reprogramming of the entire manifesto and not only a subset of it.⁸

The technical calculation of the PCI is given in the Appendix as well as in Giebler et al. (2015). On a more general level though, the PCI is an election-specific measure and the calculation starts first with the calculation of a party’s relative size to other parties (via vote shares) in a given election so as not to bias small parties.⁹ Then, 26 categories are rescaled so that the sum of all positive categories and all negative categories is equal to 1. Next, this sum is weighted by the relative measure of a party’s size in the election so as not to bias small parties (who are expected to have less ‘catch-all’ manifestos) and if the party does not mention the policy area at all in the manifesto the weight is set to zero. The programmatic clarity index (PCI), then, is the grand sum of the absolute difference of positive and negative mentions of a party in a given election divided by the sum of positive and negative mentions multiplied by the previously calculated weight. The final index runs from 0 (completely obscure program) to 1 (completely clear program).

Independent variables

Party family classifications

The analysis consists of three separate models. The variables for each model remain the same, except the party family variable. Three separate party type variables are calculated, and all are based on the party family groupings found in the Manifesto Data. First, a dummy variable for left-of-center party was created using the party family classifications of the Manifesto Project dataset and is used in Model 1. Left-of-center parties are social democratic and socialist/communist parties as defined by the Manifesto Project party family classification (which is based on International Party Family membership). All parties not designated as Social Democratic, Socialist, or Communist by the Manifesto Project are classified as non-left parties. This binary classification yields 307 left-of-center parties and 542 non-left parties in the sample. Model 2 then includes a second, broadened party family variable which, also based on the Manifesto party family classification, codes whether a party is: a) Social Democratic/Socialist/Communist, b) Liberal, c) Conservative/Christian Democratic, or b) non-Mainstream. Finally, and based on the explicit predictions that social democratic parties will behave differently contingent on radical right party competition, Model 3 recodes the party family variable into a three-part categorical variable for Social Democrats and Radica Right Parties (RRPs) which provides information on whether a party is a) a social democratic party with a RRP competitor or b) a social democratic party without a RRP competitor. The variable is calculated exactly in such a two-step process and the zero category represents non-social democratic parties and is used as the baseline category.¹⁰

Economic globalization

The key independent variable is a measure of change in economic globalization. As a system level variable, economic globalization is a constant across all parties within a national party system but the effects clearly vary across countries (cf. Merkel et al., 2008: 16–17).¹¹ The variable is measured with the KOF index of globalization (Dreher, 2006; Dreher et al., 2008) using the economic globalization measure provided as part of the dataset.¹² Economic globalization is defined as “long distance flows of goods, capital, and services as well as information and perceptions that accompany market exchanges” (KOF, 2013: 1). The index is a compilation of data from several sources and is created by combining actual trade flows and restrictions on trade. For more information, see Dreher (2006) or Dreher et al. (2008). The reasons for only including economic globalization are twofold. First, as discussed previously, previous work has led to clear predictions as to how economic globalization should affect party positions. Second, despite the focus solely on Western European countries, there are still differences in how individual national governments responded to the demands of the ‘contemporary era’ and, thus, there is still variance in the level of economic globalization—something that is not true for political globalization where variance in the KOF data is minimal.¹³ The change in economic globalization variable is truncated¹⁴ to range from –4 to 22, so there is not only ample variation but, importantly, there are also a few cases (albeit a relatively small number) where the change in economic globalization is actually negative (see Figure 1).

Interaction terms

Finally, the models include several interaction terms to account for the conditional nature of the theory. First, I predict that the effect of economic globalization on programmatic clarity is moderated by whether a party is a left party or not. Therefore, an interaction between left party and change in economic globalization is included. The second model includes a similar interaction between mainstream party type and change in economic globalization while the third model includes an interaction between social democratic party with or without radical right competition and change in economic globalization.

Control variables

As discussed previously, current research points to a clear link between increasing levels of globalization and right-left positional changes (Adams et al., 2009; Ezrow and Hellwig, 2014; Haupt, 2010; Ward et al., 2011). However, this analysis is concerned with changes to programmatic clarity separate from such right-left shifts and therefore change in RILE position is a control variable only. Like the programmatic clarity index, the RILE index includes 26 categories—13 Right and 13 Left. Of these 26 categories, only half overlap with categories also included in the programmatic clarity index. However, while there is 50% overlap in categories, only three of the 13 pairs included in the programmatic clarity calculation are also found (as pairs) in the RILE index. Furthermore, it is also expected, given the findings of Adams, Haupt, Ward and others, that there is an interaction between RILE and increasing economic globalization levels. Therefore, a second interaction between economic globalization levels and programmatic clarity, change in economic globalization and change in right-left position, is included in the model as a control variable. Finally, in lieu of controlling for several possibly country specific indicators such as SMD vs PR and the level of affluence of a country, the models all include country dummies. These dummies control for such country specific variation.

Results

Table 2 presents results for all three models. On a general level, the analysis provides only partial evidence for H1 and H3, but H2—concerning the effect of radical right parties on social democratic strategies—is confirmed.

Table 2.

Prais-Winsten TSCS regression with panel corrected standard errors.

	Model 1 ^{a, b}		Model 2		Model 3
	β ^c	s.e. ^d	β	s.e.	β	s.e.
Intercept	0.056†	0.029	0.070**	0.030	0.060**	0.028
Δ Right-Left Position	–0.002**	0.001	–0.002**	0.001	–0.002***	0.001
Δ Economic Globalization	–0.000	0.002	–0.001	0.002	–0.002	0.002
Left
Left	0.018	0.020
Mainstream
Left			0.002	0.025
Liberal			–0.040	0.033
Right			–0.011	0.025
Social Democrats & RRPs
Social Democrat/No RRP Competition					0.045†	0.024
Social Democrat/RRP Competition					–0.072*	0.036
Left * Δ Economic Globalization
Left	–0.005†	0.003
Mainstream * Δ Economic Globalization
Left			–0.004	0.003
Liberal			0.010	0.005
Right			–0.003	0.003
Social Democrats & RRPs * Δ Economic Globalization
Social Democrat/No RRP Competition					0.001†	0.003
Social Democrat/RRP Competition					–0.012**	0.004
Δ Right-Left Position * Δ Economic Globalization	–0.00**	0.00	–0.000**	0.000	–0.000*	0.000
ρ	0.070	0.034	0.035	0.035	0.08	0.033
df	20		24		22
R²	0.1306		0.1344		0.1371
χ²	105.86***		110.18***		117.10***
N	847		847		847

^a For All Models: DV = Δ Programmatic Clarity; ^b For All Models: Country dummies not shown for clarity; ^c † p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001; ^d All models include control for panel-specific AR1 correlation (142 estimated autocorrelations).

Concerning H1, that left parties will adopt an obscuring strategy in the face of rising economic globalization, the results of Model 1 are straightforward. The interaction between left party and globalization shows that at increasing levels of changing economic globalization left parties do significantly obscure their programs. The overall programmatic clarity of left parties (as shown in Figure 3) begins near the maximum level but, as the change in economic globalization variable increases from the minimum to maximum level, programmatic clarity of left parties drops quite drastically. For non-left parties, however, there is little change in programmatic clarity as changing economic globalization becomes greater. While the coefficient for the left*change in economic globalization variable is marginally significant as reported for Model 1 (p = 0.06), it is important to note that the full story shown by the marginal effects in Figure 3 changes the interpretation somewhat. Here we see that the significance of this variable is actually at the mean level of economic globalization change and below. This means that left parties who produced programs in times where the changes in economic globalization were just beginning were also the parties who significantly obscured their programmatic supply. The mean level significance, while interesting, also shows no significant difference between left and non-left parties at this point. The real driver here seems to be when economic globalization increases from below average levels to average levels—at least for the binary left/not-left comparison.

When the left/not-left variable is instead recoded into a mainstream left, liberal, right versus non-mainstream categorical variable in Model 2, we see further differences between the party types—even if the main interaction terms fail to reach statistical significance.¹⁵ However, the general direction of the coefficients conforms to the theoretical expectation that Liberal parties clarify their programmatic supply when facing increasing levels of economic globalization, contrasting to the tendency of left parties to obscure. Charitably, one could argue that such a result offers at least partial support for H3: the one party family, the Liberals, for whom globalization supports the traditional programmatic profile clarifies their programmatic supply in contrast to those parties for whom globalization directly counters traditional programmatic offerings.

Turning to the final model, the results are very much as expected: social democratic parties facing a radical right competitor under conditions of rapidly increasing economic globalization significantly obscure their programmatic profiles while social democratic parties where no radical right competitor is present clarify, although not significantly, their programmatic supply. Figure 4 shows the marginal effects of the change in economic globalization on a party’s programmatic clarity conditioned by whether the party was a social democratic party facing a radical right competitor or a social democratic party without a radical right competitor. Under conditions of below average changes in economic globalization and at the highest levels of change in economic globalization, social democratic parties facing radical right competitors also differ significantly from other social democratic parties that do not have such competition.

Discussion

The analysis presented here tests the extent to which parties adapt their programmatic supply—either by obscuring or clarifying it—in the face of changing levels of economic globalization, as an additional or alternate strategy to well-researched positional changes. Using an innovative measure of programmatic clarity, the results show that left parties—namely social democratic or socialist/communist parties—are affected differently by increasing levels of economic globalization, and that these parties respond by obscuring their programmatic positions. This is exactly what we would expect for a party whose traditional core policy profile is undermined by the shift to a neoliberal economic paradigm and, additionally, by a party whose core constituency is no longer sufficient to guarantee electoral success. Generally, the assumption of both the economic globalization literature and the party transformation literature—that left parties receive a lion’s share of the pressures arising from economic globalization—is largely confirmed by the findings. Going a step further, the narrow focus on social democratic parties only, as used in the third model, and the findings from that model, confirm the expectations of Bale et al. (2010) and others that these parties in particular are facing a dual threat: first, changing economic globalization threatens the underlying core principles of their traditional policy stances and, secondly, radical right competitors also cause social democratic parties to obscure their programmatic supply.

While the results largely support the basic assumptions outlined in the theory section, they also show that the relationship between traditional ideological underpinnings and reprogramming due to globalization pressures is not straightforward and that there is a clear need to extend the analysis even more. Further analyses could fruitfully focus on including a variety of additional variables in the model. In particular, the inclusion of the Index of Party Cohesion (IPC) of Jahn and Oberst (2012: 225), which measures the ‘ideological spread of a party’, would allow for the exploration of whether obscuring strategies are also, in part, linked to what are essentially disagreements within the party concerning the party’s right-left ideological profile. The premise of such a comparison would be that those parties facing greater internal strife should obscure their programmatic supply more so than when there is higher internal agreement between party elites.

More broadly, the results are important in terms of electoral competition as it seems that parties are adapting their programmatic supply in ways that are, at the very least, not completely captured by right-left positional changes. While the models presented here treat right-left positional changes as control variables only, the continued significance of these variables in the models suggest that these changes are important to the broader picture, and future analysis should work on combining both positional changes and changes in programmatic clarity in the same model. Finally, these findings carry implications outside the field of party scholarship, as the natural continuation of the research should examine voter responses to such adaptations and the extent to which the concepts of reprogramming and programmatic clarity influence overall patterns of electoral competition in the rapidly changing landscape of Western European societies.

Footnotes

Appendix: Calculation of programmatic clarity

The measure of programmatic clarity is developed by Giebler et al. (2015).

The index begins by calculating the election-specific relative measure of party size (RMPS):

where i is party, k is election, n is the total number of parties competing in the election, and j is a running number (1–n). This ranges from 0 to 1.

Once the RMPS is created, the index is built in three steps. First, the 26 policy categories (forming the 13 pairs) are rescaled so that:

Where n is the total policy categories, j is each individual policy category, i represents each party, and k represents each election. This is then weighted by the relative measure of party size in order to control for saliency of issues in a given election:

For $W_{jk}$ , if a party does not mention a policy area in the manifesto, then the weight of the policy area is zero. Once the weights are calculated, this is combined to create the final programmatic clarity measure :

Where n is the total number of policy areas, j represents the individual policy area, i is for party, and k stands for election. The final measure runs from 0 (completely obscure program) to 1 (perfectly clear program which only includes one antipode per policy area).

Author’s note

A previous version of this paper was presented at the 2014 General Conference of the European Consortium of Political Research, Glasgow, UK, 3–6 September 2014.

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.

Notes

References

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.

Alvarez

(1998) Information and Elections. Ann Arbor, MI: University of Michigan Press.

Bale

(2003) Cinderella and her ugly sisters: The mainstream and extreme right in Europe’s bipolarizing party systems. West European Politics 26(3): 67–90.

Bale

Green-Pederson

Krouwel

. (2010) If you can’t beat them, join them? Explaining social democratic responses to the challenge from the populist radical right in Western Europe. Political Studies 58(3): 410–426.

Bartolini

(2007) The Political Mobilization of the European Left, 1860–1980: The Class Cleavage. Cambridge: Cambridge University Press.

Bartolini

Mair

(2007) Identity, Competition, and Electoral Availability: The Stabilisation of European Electorates 1885–1985. Colchester: ECPR Press.

Beck

Katz

(1995) What to do (and not do) with time-series cross-sectional data. American Political Science Review 89(3): 634–647.

10.

Boix

(1998) Political Parties, Growth and Equality: Conservative and Social Democratic Strategies in the World Economy. Cambridge: Cambridge University Press.

11.

Bornschier

(2010) Cleavage Politics and the Populist Right: The New Cultural Conflict in Western Europe. Philadelphia, PA: Temple University Press.

12.

Brooks

(1983) Left-wing mobilization and socioeconomic equality: A cross-national analysis of the developed democracies. Comparative Political Studies 16(3): 393–416.

13.

Burgoon

(2012) Partisan embedding of liberalism: How trade, investment, and immigration affect party support for the welfare state. Comparative Political Studies 45(5): 606–635.

14.

Dalton

(2000) The decline of party identifications. In: Dalton

Wattenberg

(eds) Parties Without Partisans: Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press, pp. 19–43.

15.

Dalton

Wattenberg

(2000) Parties without Partisans: Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press.

16.

de Vries

Hobolt

(2012) When dimensions collide: The electoral success of issue entrepreneurs. European Union Politics 13(2): 246–268.

17.

de Vries

Marks

(2012) The struggle over dimensionality: A note on theory and empirics. European Union Politics 13(2): 185–193.

18.

Deegan-Krause

(2007) New dimensions of political cleavage. In: Dalton

Klingemann

H-D

(eds) The Oxford Handbook of Political Behavior. Oxford: Oxford University Press, pp. 538–556.

19.

Dreher

(2006) KOF Index of Globalization. Zürich: Konjunkturforschungsstelle ETH Zürich.

20.

Dreher

Gaston

Martin

(2008) Measuring Globalization: Gauging its Consequences. New York: Springer Press.

21.

Duverger

(1963) Political Parties: Their Organization and Activity in the Modern State. New York: Wiley.

22.

Ezrow

Hellwig

(2014) Responding to voters or responding to markets? Political parties and public opinion in an era of globalization. International Studies Quarterly 58(4): 816–827.

23.

Ezrow

De Vries

Steenbergen

. (2011) Mean voter representation and partisan constituencies representation: Do parties respond to the mean voter position or to their supporters? Party Politics 17(3): 275–301.

24.

Giebler

Lacewell

Regel

. (2015) Niedergang oder Wandel? Parteitypen und die Krise der repräsentativen Demokratie. In: Merkel

(ed.) Demokratie und Krise: Zum schwierigen Verhältnis von Theorie und Empirie. Wiesbaden: Springer VS, pp. 181–219.

25.

Haupt

(2010) Party responses to economic globalization: What is left for the left and right for the right? Party Politics 16(1): 5–27.

26.

Held

McGrew

Goldblatt

. (1999) Global Transformations: Politics, Economics, Culture. Stanford, CA: Stanford University Press.

27.

Hooghe

Marks

Wilson

(2002) Does left/right structure party positions on European integration? Comparative Political Studies 35(8): 965–989.

28.

Ingelhart

(1977) The Silent Revolution: Changing Values and Political Styles Among Western Publics. Princeton, NJ: Princeton University Press.

29.

Jahn

C Oberst (2012) Ideological party cohesion in macro-comparative politics: The Nordic social democratic parties from a comparative perspective. Scandinavian Political Studies 35(3): 222–245.

30.

Katz

Mair

(1995) Changing models of party organization and party democracy: The emergence of the cartel party. Party Politics 1(1): 5–28.

31.

Kirchheimer

(1965) Der Wandel des westeuropäischen Parteiensystems. Politische Vierteljahresschrift 6(1): 20–41.

32.

Kitschelt

(1995) The Radical Right in Western Europe: A Comparative Analysis. Ann Arbor, MI: University of Michigan Press.

33.

Kitschelt

Lange

Marks

. (1999) Continuity and Change in Contemporary Capitalism. Cambridge: Cambridge University Press.

34.

Kluever

JJ Spoon (2015) Who responds? Voters, parties, and issue attention. British Journal of Political Science. DOI: 10.1017/S0007123414000313.

35.

KOF (2013) 2013 KOF Index of Globalization. Available at: http://globalization.kof.ethz.ch

36.

Kriesi

Grande

Lachat

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

37.

Kriesi

Grande

Dolezal

. (2012) Political Conflict in Western Europe. Cambridge: Cambridge University Press.

38.

Lipset

Rokkan

(1967) Cleavage Structures, Party Systems, and Voter Alignments: An Introduction. New York: The Free Press.

39.

Mair

(1997) Party System Change: Approaches and Interpretations. Oxford: Clarendon Press.

40.

Mair

(2005) Cleavages. In: Katz

Crotty

(eds) The Handbook of Party Politics. London: Sage Publications, pp. 370–375.

41.

Meguid

(2005) Competition between unequals: The role of mainstream party strategy in niche party success. American Political Science Review 99(3): 347–359.

42.

Merkel

Petring

Henkes

. (2008) Social Democracy in Power: The Capacity to Reform. New York: Routledge Press.

43.

Meyer

M Wagner (2013) Mainstream or niche? Vote-seeking incentives and the programmatic strategies of political parties. Comparative Political Studies 46(10): 1246–1272.

44.

Milner

Judkins

(2004) Partisanship, trade policy, and globalization: Is there a left-right divide on trade policy? International Studies Quarterly 48(1): 95–120.

45.

Poguntke

(2014) Towards a new party system: The vanishing hold of the catch-all parties in Germany. Party Politics 20(6): 950–963.

46.

Przeworski

Sprague

(1986) Paper Stones: A History of Electoral Socialism. Chicago, IL: Chicago University Press.

47.

Reiter

(1989) Party decline in the West: A skeptic’s view. Journal of Theoretical Politics 1(3): 325–348.

48.

Rovny

(2012) Who emphasizes and who blurs? Party strategies in multidimensional competition. European Union Politics 13(2): 269–292.

49.

Schumacher

de Vries

Vis

(2013) Why do political parties change their position? Institutional conditions and environmental incentives. Journal of Politics 75(2): 464–477.

50.

van der Wardt

(2015) Desperate needs, desperate deeds: Why mainstream parties respond to the issues of niche parties. West European Politics 38(1): 93–122.

51.

Volkens

Lacewell

Lehmann

. (2012) The Manifesto Data Collection, Manifesto Project (MRG/CMP/MARPOR). Wissenschaftszentrum Berlin für Sozialforschung. Available at: www.manifesto-project.wzb.eu

52.

Wagner

(2012) When do parties emphasise extreme positions? How strategic incentives for policy differentiation influence issue importance. European Journal of Political Research 51(1): 64–88.

53.

Ward

Ezrow

Dorussen

(2011) Globalization, party positions, and the median voter. World Politics 63(3): 509–547.

54.

Ward

Kim

Graham

. (2015) How economic integration affects party issue emphasis. Comparative Political Studies. DOI: 10.1177/001041405576745.

55.

Werner

Lacewell

Volkens

(2011) Manifesto Coding Handbook. 4th edition. Available at: www.manifesto-project.wzb.eu

56.

Weßels

(2001) Die “Dritten Wege”: Eine Modernisierung sozial-demokratischer Politikkonzepte? In: Schroeder

(ed.) Neue Balance zwischen Markt und Staat? Schwaibach am Taunus: Wochenschau Verlag, pp. 43–64.