Party statutes and party institutionalization

Abstract

Studies of party- and party-system stability have often explored the connection between the party-level property of Party Institutionalization (PI) and parties’ electoral performance and organizational longevity, yet scholars still have not agreed on a standard measure for this concept. This article argues that the length of party statutes could provide part of such a measure, specifically for the extent to which parties have become routinized (a key dimension of PI) through the formalization of their rules and practices. We validate the plausibility of this measure using data on 303 parties from 49 countries, demonstrating that party statute length varies systematically and in ways predicted by our knowledge of how party organizations reflect their institutional environments and the complexity of internal coalitions. We also show that statute length varies in expected ways with attributes often associated with higher or lower levels of party institutionalization. We conclude that statute length offers a conceptually congruent and objective indicator of formalization, one that could be used either alone or combined with measures of parties’ informal practices to advance our understanding of the relationship between PI and democratic development.

Keywords

party institutionalization party statutes party organization

Introduction

Most political parties have formally-constituted extra-legislative organizations, with written rules covering many of their routine operations.¹ These documents have names such as “by-laws” or “statutes” or “constitutions.”² Party statutes answer questions that are fundamental to the operation and viability of parties as collective political actors, such as Who is in charge of the organization? How are leaders selected? How are decisions made? Given the importance of such questions, many parties put great care into debating and revising their statutes. Yet despite their apparent importance in party life, few studies have examined parties’ constituting documents as artifacts in their own right.

This neglect is surprising, because party statutes vary greatly in their specificity and in the topics they address. Some variation may reflect differences in national legal environments, yet party statutes can show big differences in their coverage even within a single country. This suggests that parties write statutes not merely in response to legal rules, but because they find it more or less expedient—and perhaps more or less possible—to formally define their priorities and regulate internal operations. If so, the extent to which parties elaborate their statutes may indicate important cross-party differences in organizational development.

This conclusion is implicit in much of the growing literature on party institutionalization (PI), which is concerned with the democratic benefits of political parties that are embedded in society (see, for example, Dix, 1992; Mainwaring and Scully, 1995; Martínez, 2020). Although there is debate on how best to conceptualize and operationalize PI, it can broadly be thought of as a property describing the extent to which parties are stable political forces. According to Levitsky’s original “unpacking” argument (1998) and Bolleyer and Ruth’s (2018) distillation of the PI literature, this party property has two dimensions: routinization and value infusion. Routinization describes how much party life exhibits regularized behavior structured by formal or informal rules. Value infusion describes how much parties stand for something beyond the outcome of a single election. Routinization in particular would seem to be closely related to the extent to which parties use statutes to elaborate their procedures and highlight their values. However, so far this relationship is unclear, because studies of party institutionalization have paid little attention to parties’ statutes.

This article remedies this omission and proposes that the elaboration of party statutes can serve as a useful indicator of the formalized path to the party routinization dimension of PI. Our analysis proceeds in three stages. First, we discuss previous conceptualizations of PI, and argue that the elaboration of party statutes is likely to be closely related to the processes associated with formalization of party activities. Second, to better understand the mechanisms whereby party statute length might be linked to routinization, and through this to institutionalization, we hypothesize explanations for variations in party statute length. We argue that if statute length captures the formalization that can lead to routinization, it should respond to parties’ contexts and to their internal priorities and challenges. Having demonstrated that this is the case, we then test whether statute length corresponds to two party-level factors commonly associated with PI: party longevity, and persistence after the replacement of the party’s founding leader. For our tests we examine statutes for 303 parties from 49 countries, collected as part of the Political Party Database (PPDB) project,³ combined with political context variables from the PPDB and other sources. Finally, we compare how this measure of one aspect of PI-Routinization (formalization) relates to other proposed measures of the same concept, ones which we argue may better capture the informal paths to routinization. We conclude with a discussion of ways this measure could be employed to improve understanding of how parties’ organizational practices can contribute to (or hinder) their long-term success.

Overall, this research proposes and validates party statute length as a new, objective, and easily replicated measure of one aspect of the party institutionalization dimension of routinization. We suggest that using it alone or alongside other measures of organizational capacity may help illuminate the mechanisms through which parties institutionalize, and may better capture variations in levels of institutionalization. This, in turn, can lead to more precise studies of how party institutionalization influences other political outcomes at the party and country level.

Formal and informal routes to party institutionalization

Previous research has argued that party institutionalization has pivotal consequences for parties’ legitimacy, and for their role and stability as collective social actors. Strongly institutionalized parties can strengthen the quality of democracy by building recognized channels of political recruitment, facilitating electoral choice and peaceful political competition, and providing policy choices (Diamond et al., 1989). In Rosenblatt’s term, they are “vibrant” (Rosenblatt, 2018). Institutionalized parties possess organizational capacity and have predictable interaction patterns among party elites, qualities that can promote the institutionalization of party systems (Mainwaring et al., 2018: p. 26). At the same time, high party institutionalization does not necessarily optimize democratic outcomes, because highly institutionalized parties may inhibit democratic competition and stifle the emergence of party challengers, whether that high institutionalization takes the form of multiple cartelized parties leveraging their access to state resources (Katz and Mair, 1995) or single dominant parties whose recruitment channels are the main alternative for those with political ambitions (Greene and Sánchez-Talanquer, 2018: p. 216). There also may be a curvilinear relationship between institutionalization and responsiveness, with institutionalization aiding party persistence in the face of electoral defeats (Wills-Otero, 2016: p. 759), but very high institutionalization hampering parties’ adaptability to changing circumstances (Levitsky, 1998: p. 81; Rodriguez and Rosenblatt, 2020). For all these reasons, scholars conceive of party institutionalization as a matter of degree, not as a threshold property, with different levels possibly having different implications for parties and party systems.

Discussions of PI have converged towards discussing this property in terms of two dimensions: routinization and value infusion. These dimensions can be traced from Huntington (1968) to Panebianco (1988), Levitsky (1998), Randall and Svåsand (2002), and onwards. While they differ in aspects of their definitions, all speak to the conceptual importance of party organizational activities for defining institutionalization (Mainwaring and Scully, 1995: pp. 4–5). These two dimensions reflect the entrenchment and acceptance of internal rules and processes (routinization), coupled with a collective logic that drives leaders, activists, and members to work towards securing the party’s interests (value infusion) (Bolleyer, 2013: p. 55).

In this paper, we focus on the routinization dimension of institutionalization. Following Levitsky, (1998, pp. 86–7) we agree that routinization may occur informally, outside of rule-based frameworks. For Levitsky, routinization occurs “when rules, procedures, roles, or other patterns of behavior are institutionalized,” whether by formal or informal means, and “come to be repeated and taken for granted, and stable sets of expectations form around them” (Levitsky, 1998: p. 81). Whereas this led Levitsky to distinguish between formal and informal routinization, we conceive of routinization as a single property that is achievable through two potential pathways, one formal, the other not necessarily so.

As illustrated in Figure 1, we suggest that routinization can occur via the creation of practices that become regularized over time, whether or not they are codified (regularization), or through adoption and modification of formal rules, such as party statutes (formalization). The concept of regularization reflects arguments in previous studies of PI that the extensiveness of parties’ formally codified organizations is not necessarily synonymous with their levels of institutionalization, because parties’ informal practices and extra-party linkages are at least as important (Bolleyer and Ruth-Lovell, 2019; Levitsky, 1998; Randall and Svåsand, 2002). In contrast, formalization reflects other arguments that have linked routinization to more formal rules, be it the density of rules governing a party’s relations with its followers (Bolleyer and Ruth, 2018: p. 289), or the reification described by Harmel et al. (2018). In short, these two pathways to routinization are clearly signaled in previous work.

Figure 1.

Party institutionalization unpacked.

These pathways are not mutually exclusive, and most or all parties will rely on both mechanisms. Even the most formalized parties will doubtless have some uncodified routine practices. Moreover, these paths may be mutually-reinforcing, with formalization promoting regularized patterns of behavior even in realms unregulated by party statutes, and regularization increasing payoffs for getting practices enshrined in party rules. Yet while there may be overlap between these two paths, distinguishing between them makes clear that there is not a single path through which PI-Routinization is achieved. This distinction enables researchers to study under which circumstances each pathway comes to the fore, and facilitates the creation of clear and transparent measures. As Levitsky (1998: p. 88) suggests when unpacking the concept of party institutionalization, “Not only would such specification permit scholars to focus more concretely on the mechanisms by which organizations or behavior patterns are reproduced, but it would also improve our analyses of the consequences of such phenomena.”

While there is general scholarly consensus on the importance of the two main dimensions of PI (value infusion and routinization), and some convergence on the sub-dimensions of routinization, there is much less agreement on how to measure these properties in ways that facilitate systematic cross-party and longitudinal comparisons. One notable exception is the work of Bolleyer and Ruth (2018), which uses expert judgment data from the Democracy and Linkage Project (DALP; Kitschelt, 2014) to assess routinization (based on responses to two questions about the extensiveness of local party organization) and value infusion (based on a question about campaign rhetoric). Because the DALP dataset is so extensive, these variables can be used to construct comparable scores for many parties. The V-Party dataset (Lührmann et al., 2020) includes similar questions about organizational extensiveness and activity, and thus offers opportunities to construct similar PI measures. Because these datasets are based on expert judgment coding, their measures potentially capture “what is really happening.” In this sense, they could be particularly valuable for shedding light on the “regularization” sub-dimension, including when parties pursue informal pathways to routinization. However, neither the DALP nor V-Party datasets include expert assessments of the extent to which rules and procedures are formalized, making them less well-suited to capture the “formalization” path to routinization. That is why we propose a complementary measure here. We do, however, return to these expert judgment measures in a later section, to compare them with our proposed measure.

Because prior proposed measures of PI-Routinization have not captured the formalization of party life, we propose a new measure of formalization based on the length of party statutes. We argue that the extent to which parties have elaborated their statutes—measured as statute length—is an objective and easily reproduced indicator that captures levels of formalization within a party. We expect that statute length reflects the extent to which parties turn to formal rules to cope with internal and external pressures, and to thus achieve the routinization which is said to be a key aspect of PI. One advantage of using a measure based on party statutes is that parties generally make such documents public. Even though some parties do not have party statutes, for our purposes that is not a hindrance, because the absence of statutes implicitly denotes very low formalization.

Party statutes are a prime exemplar of “official story” documents (Katz and Mair, 1992), being constituting documents that present an account of who parties are and how they operate. Many party statutes begin with a preamble or other statement of party goals and continue by defining who can join the organization (and how), how the party leaders are chosen, what the leaders may and must do, and to whom party leaders are accountable. Some statutes define the role and rights of sub-groups within the party, such as women’s organizations, and define relations between the national and regional parties. When fully developed, party statutes provide a comprehensive roadmap for party operations.

That being said, we do not assume that the official story is the full story, or that parties always comply with the rules as written. When proposing to use statute length as a measure of formalization, we merely assume that the extent to which parties (and those who are active within them) choose to invest in developing the statutes reflects the extent to which party rules are viewed as valuable resources which can be mobilized to win (or ward off) intra-party disputes. And that, we contend, is one way through which political parties achieve the stable, predictable practices that define routinization.

The value of statute length as a measure of the formalized path to routinization cannot be tested directly because there is no “gold standard” measure of this property. Therefore, we pursue a different strategy to validate our measure, asking whether it performs as we would expect of a good measure of the kind of formalization that would produce PI-Routinization. Drawing on past research on party organizational development and party statutes (particularly Smith and Gauja, 2010), we formulate hypotheses about factors which likely shape how much parties develop their statutes. In making these hypotheses, our concern is with statute length, rather than with parties’ specific organizational choices. We do this because we cannot assume that PI as it relates to the reification of rules is associated with any particular party model or organizational configuration (Bolleyer, 2013: p. 57). That is a separate question, which can only be investigated if the measures of PI-Routinization are not co-terminus with specific organizational characteristics. Therefore, statute length represents a far better measure of formalization than statute content. Because party statutes are generally the prime locus for enshrining party rules, we hypothesize that party statutes will grow as formalization increases.

We group our hypotheses about sources of variation in statute length according to various pressures placed on parties that might foster institutionalization. The first set sees statutes as responses to external environmental conditions; the second set sees statutes as tools to resolve intra-party conflicts. In validating statute length as a measure of formalization, these hypotheses also function to explain the various drivers and processes that underpin a party’s institutionalization across different contexts. Thus, this approach not only provides a strong theoretical argument for why parties may achieve routinization through formalization; it also illuminates some of the underlying factors that contribute to this process.

External drivers of formalization: statutes as responses to environmental conditions

One well-established expectation about parties and other voluntary organizations is that their formal rules are shaped by their environments. Organizational isomorphism means that organizations within the same country—in this case, political parties—generally resemble each other due to their societies’ normative expectations and institutional practices (Bolleyer, 2018; Harmel and Janda, 1982; Meyer and Rowan, 1977).

Regulations affecting parties’ internal affairs may help to drive national convergence in party statutes. Legal oversight of parties as distinctive types of associations has increased dramatically since 1945 (Van Biezen, 2012; Molenaar, 2014), including with the adoption and elaboration of party laws and party finance laws (Van Biezen and Piccio, 2013) although countries still differ greatly in how much they regulate party life. We posit that more extensive legal regulation of parties’ internal affairs will lead to longer party statutes, because these rules expand the topics which statutes must address.

H1a

Party statutes will be longer in countries with more extensive regulation of party organizations.

Other environmental factors may also affect the length and specificity of party statutes, particularly those that dictate the complexity of parties’ political tasks. For instance, statutes may be longer in presidential than in parliamentary systems, because party rules need to consider intra-party roles and selection procedures for multiple types of candidates (Samuels and Shugart, 2010). The same may be true of systems with bicameral legislatures, if bicameralism leads parties to try to improve cross-chamber coordination by centralizing candidate selection (Van Dusky-Allen and Heller, 2014). Operating against this prediction is the possibility that in federal systems (often associated with bicameralism), parties will give sub-national party units more procedural autonomy (Carty, 2004; Deschouwer, 2006; Cross and Gauja, 2014). If so, national party statutes may be shorter under federalism, with procedural details delegated to the party’s geographic sub-units. These considerations lead to the following external environment hypotheses:

H1b

Party statutes will be longer in presidential than in parliamentary regimes.

H1c

Party statutes will be longer in bicameral systems.

H1d

(National) party statutes will be shorter in federal systems.

Internal drivers of formalization: statutes as tools to resolve intra-party conflict

Statutes define relations between different organizational facets of a party and stipulate who is responsible for what. These documents are thus likely to grow in response to situations requiring clarity about such matters, including situations of intra-party conflict. For instance, there may be less pressure for statutory elaboration in parties with a relatively narrow policy agenda or support base, such as niche or small parties which appeal to a well-defined constituency (Meyer and Wagner, 2013; Bischof, 2017). Conversely, parties which unite diverse coalitions may face greater pressures to regulate relations between party sub-groups or factions (Alexander, 1998); therefore, we would expect them to have longer statutes. Electoral success may also sharpen intra-party conflicts and thereby expose gaps in party mechanisms to resolve internal disputes. As a result, parties that enter government may expand their statutes to further address relationships between the voluntary party and the party’s elected representatives (Katz and Mair, 1993). In parties that give their leaders strong roles, we would expect statutes to contain less detail, leaving more discretion for the leader. Statutory latitude for the leader may also result from views about the value of hierarchy and strong leadership (e.g., right-wing populist parties, see Heinisch and Mazzoleni, 2016). These considerations lead to four hypotheses:

H2a

Niche parties will have shorter statutes than catch-all parties.

H2b

The greater the number of recognized sub-groups within a party, the longer its statutes.

H2c

Parties which have served in government (or whose candidate has been elected national president) will have longer statutes than those that have never done so.

H2d

Right of center parties will have shorter statutes.

Data and measures

Having set forth our hypotheses, we now test them by examining current and recent statutes from 303 parties in 49 electoral democracies from six continents. These statutes were collected in conjunction with the PPDB project and are available on that project’s webpage.⁴ The sample includes parliamentary and presidential regimes, federal and unitary systems, and established and newer democracies. The countries have varied population and geographic sizes, and use diverse electoral systems. Most parties in the sample had at least one representative in the national legislature when the statute was collected; very new parties and electorally small parties are under-represented.

Our key dependent variable for the first part of our analysis is statute length, defined as word count as measured by MS Word.⁵ Because our statutes are written in many languages, a simple word count comparison is problematic, given that some languages are inherently wordier than others. We thus standardize these counts by applying a language length algorithm that adjusts for relative length of comparable documents when translated into English.⁶

Figure 2 lists the included countries and shows that our sample displays considerable within-country variation in statute length. This supports our initial expectation that statute length cannot be explained solely by national environments, however important these may be.

Figure 2.

Statute length variation by country for full dataset (Source: PPDB Statutes page: https://www.politicalpartydb.org/statutes/).

Three of our country-level independent variables (presidentialism, bicameralism and federalism) are binary variables drawn from the V-DEM database. Our fourth country-level independent variable, the extensiveness of state regulation of political parties, is a scale derived from Janda’s Database of Party Laws. This variable reflects the number of areas of party activity covered by laws, public regulations, or court rulings. Potentially these could cover 11 categories of party activity: definition, legal status, membership, organization, selecting candidates, activities, public subsidies, party finance, prohibited members, history, and other. Our party-level independent variables come from a variety of sources. Parties with relatively compact appeals, which we refer to as niche parties, are distinguished here as parties which received a 7% or lower vote share at the most recent election. The number of recognized organizational sub-groups within the party comes from the PPDB Round 1 and Round 2 data sets, as do party family designations. “Chief executive” is a binary variable referring to whether a leader of this party has held the top executive office since 1990 (be that president or prime minister, depending on the system). This information comes from the Inter-American Development Bank Database of Political Institutions. The coding scheme and sources for each of these variables can be found in the Supplementary Appendix.⁷

The responsiveness of party statutes

To give context to the key variable of interest, the mean adjusted length of our sample of party statutes was 9892 words. The Five Star Movement in Italy had the shortest statute, at 715 words (adjusted), while the longest statute was the Labour Party in the Netherlands at 63,636 words (see Supplementary Table A3 and Figure A1-A2 for descriptives). Because our variables of interest are at the party-level and country-level, we use multilevel models, which allow us to compare the variance explained nested at the party-level relative to the country-level.

Many of our expectations are met concerning the responsiveness of party statute length to external conditions and internal challenges. All the country-level factors have relationships with statute length which run in the expected direction, and two of these meet conventional thresholds of statistical significance. At the party level, the model shows support for the hypothesis that government experience may prompt parties to elaborate their statutes; here we find relatively large effects. A party having served in government with a chief executive is associated with an effect that is both statistically significant at the 0.01 level and substantively large, corresponding with a 3958-word increase in adjusted statute length. In addition, the coefficients for the other posited relationships are in the expected direction, though only the hypothesis about internal groups crosses even a modest threshold of statistical significance.

Given that the relatively small sample size reduces the likelihood of finding statistically significant relationships, we think the totality of the evidence (direction of all the relationships and strength of some of them) makes a convincing case that statute length is responsive to party circumstances. Thus, our multilevel multivariate tests provide support for the idea that parties’ investment in developing their statutes reflects both their institutional contexts and the extent to which parties use rules to regulate intra-party conflicts. This conclusion is reinforced by the breakdown showing how the national- and party-level variables contribute to the model fit, with party-level variables doing much of the work: 80% of the variance is explained by party-level variables, as shown by the ICC statistic in Table 1. This finding matches the underlying theoretical foundations of how formalization could contribute to PI-Routinization. While broader institutional factors such as legal regulation play a role in facilitating formalization, statutory length appears to bear the larger imprint of parties’ internal incentives to codify internal power-sharing arrangements and move towards more elaborate and permanent structures.

Table 1.

Context, intra-party conflict regulation, and routinization with country-level fixed effects.

	Statute length adjusted
Predictors	Estimates
Country-level effects
Regulation of party organizations	1435.23** (654.21)
Presidentialism	2733.37 (2197.77)
Bicameralism	6505.09*** (1881.23)
Federalism	−2852.39 (2175.96
Party-level effects
Niche party	−1534.45 (1780.46)
Org sub-groups	398.31** (184.73)
Chief executive	3958.73*** (1002.06)
Right of center	−3113.52 (4651.99)
Intercept	−2125.52 (3173.43)
σ²	45381449.50
τ₀₀ _COUNTRY	16490668.38
ICC	0.20
N _COUNTRY	41
Observations	250
Marginal R²/Conditional R²	0.248/0.330

*p<0.10, **p<0.05, ***p<0.01

Party statute length and party institutionalization

Having established that there is a good case for using party statute length as a measure of the formalization of party operations, because of the forces that seem associated with that length, we now more directly test whether the formalization we claim to measure has the expected connection with party institutionalization. As discussed above, the difficulty in conducting such a test is that researchers have not converged on an accepted measure of PI. Nevertheless, two measures correspond to features that are often associated with institutionalization (or lack thereof). On the one hand, parties which are still led by their founding leader are presumed to be relatively less institutionalized (Huntington, 1965: p. 396). This is because they can rely on the leader’s charisma or authority to ward off many potential disputes, and because such leaders may be uninterested in creating structures that will outlive, or might even shorten, their personal careers; they also have not dealt with the challenges of leadership succession. Thus, we would expect statutes to be shorter in parties that are still led by their founding leader. On the other hand, PI is said to promote party longevity; thus, party age is sometimes used as a proxy or at least partial measure for PI (Huntington, 1965: p. 395; Levitsky, 1998: p. 88; Bolleyer and Ruth, 2018: p. 288). Thus, we would expect older parties to have longer statutes. If such a relationship exists, it is probably bidirectional. Older parties have more time to expand their statutes, while changing circumstances may produce new reasons for statutory revisions. Thus, age itself may be a driver of statute length. But on the other hand, the formalized processes embedded in the longer statutes may help parties to weather (or prevent) internal conflicts, as may the rule-following habits that are nurtured in parties that invest in developing their statutes. Our purpose is not to disentangle the direction of such effects (which our current data do not allow), but rather to further assess whether variations in statute length have the expected associations with one measure of PI.

For these analyses, we introduce two new variables. Party age (as of 2020) comes from parties’ websites; these figures are truncated at age 50 and logged, because we assume a diminishing relationship between formalization and PI. Using information from party webpages, we also created a bivariate measure of whether a party is still led by its founding leader. Not surprisingly, these two variables are correlated at modestly high levels, so we assess these relationships separately.

Table 2 shows the association between our measure of formalization and our two indicators of party institutionalization. The first column in Table 2 shows that a one unit increase of statute length (measured in this table as statute length/1000 for a more substantive interpretation) is associated with an increase in logged party age of 0.025; this relationship is significant at the 0.01 level. The second column shows that a one-unit increase in statute length is associated with a decrease in likelihood that the party is still led by a founding leader; this relationship is significant at the 0.05 level. Both relationships are relatively strong and are consistent with what we would expect if statute length captures the impact of formalization on PI-Routinization.

Table 2.

Party statute length on indicators of PI with country fixed effects.

Predictors	Logged party age	Founding leader
Predictors	Estimates	Estimates
Statute length adjusted	0.025*** (0.007)	−0.143** (0.052)
Intercept	3.44*** (0.415)	−17.976 (5344.717)
Observations	303	303
R²/R² adjusted	0.310\0.177	0.273

*p<0.10, **p<0.05, ***p<0.01

Measuring formalization and regularization: comparing measures.

Finally, we compare our statute length measure of the formalization dimension of PI-Routinization with previously used measures of PI-Routinization, looking specifically at the DALP measures used by Bolleyer and Ruth (2018), and at a similar measure drawn from the V-Party dataset. The Bolleyer and Ruth measure is based on expert judgment responses to two questions about the extensiveness of local party organization and its operation outside of electoral periods, and about party reliance on local intermediaries (such as religious leaders or neighborhood leaders).⁸ Like the DALP dataset, the V-Party dataset includes a question about the extent to which parties maintain local party offices outside of national election periods.⁹ We contend that both measures are well-suited to picking up the “regularization” pathway to routinization, which reflects party behavior, but do not give direct evidence of formalization, since the documented behavior could be governed by informal understandings rather than formal rules. If that is the case, the correspondence between these measures and the statute-length measure will not necessarily be high. In contrast, since regularization is one pathway towards routinization, we would expect that these regularization measures are associated with our two indicators of PI, as was the formalization measure. As such, we would expect that a model including formalization plus regularization measures would show the strongest associations with these PI indicators. To see whether this is the case, we consider both measures of regularization—the Ruth/Bolleyer DALP measure and the V-Party measure—because they are similar conceptually in what they measure, but they differ in the extent to which the available data overlap with our measures of statute length.

Before testing these assumptions, we compare the two expert judgment measures, to assess how much they agree among themselves. A drawback of expert judgment measures is that they rely on aggregation of categorical estimates, and thus may truncate differences (Luna, 2014); furthermore, experts have asymmetric information when reconstructing past practices. Thus, expert judgment measures may display some disagreement with each other. To assess the match between the relevant V-Party and DALP measures, we compared responses to their respective questions about the extent to which parties maintain local offices outside of national election periods. We started with the 303 parties included in our analysis of party statutes, but have a much smaller subset of cases (147) for which both the V-Party and DALP measures were available. We use V-Party data, primarily for 2008–2014, which is roughly the same period as the DALP measures.¹⁰ After standardizing the scores, we find a correlation of -0.615 (negative because the coding runs in opposite directions). This general agreement is reassuring, but the extent of differences is also a reminder that despite the apparent precision of such scores, they may nevertheless reflect expert disagreement, and/or subtle differences in question wording.

We then compare the V-Party measure to the DALP 2-variable measure used by Bolleyer and Ruth and the Statute Length measure, asking how well they correspond. Table 3 shows that the two expert judgment measures (which we have labeled measures of regularization), still show relatively high correspondence with each other. However, neither of them is highly correlated with statute length, which we have labeled a measure of formalization. This reinforces our argument that these measures are looking at different routes to routinization.

Table 3.

Correlation matrix of party institutionalization routinization: measuring formalization and regularization.

	Formalization (Statute length)	Regularization 1 (DALP local offices)	Regularization 2 (V-Party local offices)
Formalization (Statute length)	1	--	--
Regularization 1 (Bolleyer and Ruth measure)	−0.14	1	--
Regularization 2 (V-Party local offices)	0.07	−0.54	1

Finally, we ask whether there is support for viewing the separate measures as indicators of complementary pathways to PI-routinization. We do this by re-running our models from Table 2, alternatively adding each measure of local party capacity (regularization) to our party statute measure (formalization). For Table 4, we run an OLS regression (with our logged indicator of party age as the dependent variable), and for Table 5 we run a logistic regression (with our dichotomous dependent variable of being led by the founding leader). As Table 4 and Table 5 show, the relationships are in the expected direction (bearing in mind that the DALP measure runs in a negative direction), although even applying a generous 0.10 threshold, the significance of some of the relationships with statute length disappears, probably due to the sizeable drop in cases compared to Table 2. More importantly for our question here, moving from using a single measure of PI-routinization to using measures of regularization and formalization modestly increases the model fit in three of the four specifications. While these relationships are not large, they do point in an encouraging direction for future research.

Table 4.

Assessing different measures’ effects on pathways to PI via party age logged.

	Party age logged
Predictors	Estimates	Estimates	Estimates	Estimates
Regularization 1 (Bolleyer and Ruth measure)	−1.27*** (0.20)	—	−1.27*** (0.20)	—
Regularization 2 (V-Party measure)	—	1.82*** (0.24)	—	1.76*** (0.24)
Formalization (Statute Length/1000)	—	—	0.00 (0.01)	0.01* (0.01)
Intercept	4.67*** (0.25)	3.35*** (0.32)	4.66*** (0.27)	3.23*** (0.32)
Observations	147	231	147	231
R²/R² adjusted	0.510/0.343	0.460/0.318	0.510/0.337	0.469/0.326

*p<0.10, **p<0.05, ***p<0.01

Table 5.

Assessing different measures’ effects on pathways to PI via founding leader.

	Founding leader
Predictors	Estimates	Estimates	Estimates	Estimates
Regularization 1 (Bolleyer and Ruth measure)	3.42 (2.23)	—	3.09 (2.86)	—
Regularization 2 (V-Party measure)	—	−3.63** (1.26)	—	−3.13* (1.26)
Formalization (Statute Length/1000)	—	—	−0.12 (0.08)	−0.15* (0.07)
Intercept	−23.23 (8769.46)	−19.85 (8780.02)	−21.64 (8700.85)	−18.29 (8784.23)
Observations	147	231	147	231
R²/R² adjusted	0.385	0.398	0.415	0.435

*p<0.10, **p<0.05, ***p<0.01

Concluding thoughts

In recent years, scholars have turned more attention to the concept of PI to help explain why and when individual parties succeed or fail, and to explore the broader implications of this for the quality of democracy in both established and emerging settings. However, progress in this field has been hampered by the lack of agreed upon measures that fit the conceptual dimensions of PI. Bringing a new perspective to the debate, we suggest that the length of party statutes is a compelling and easily replicable measure of formalization, which can be used separately or in conjunction with indicators that capture regularization. Used together, the two types of measures should capture the extent to which political parties have institutionalized through the routinization of their organizations and practices.

After comparing statute length with factors that seem likely to promote or retard the formalization of party approaches, we find strong evidence that statutory length responds as expected to many of the external and internal factors that were identified as likely drivers of formalization. Interestingly, however, the statutory measure suggests that formalization is more likely to occur in response to resolving internal conflicts and managing intra-party groups than through external influences. Because of this, there are important within-country differences in parties’ levels of routinization, and these differences cannot be explained solely by country-level institutional or political factors. This finding has interesting implications concerning the limits to efficacy of measures that countries can adopt, such as legal regulation, to promote party institutionalization. Because of their responsiveness to internal pressures, and because statutes are the primary mechanism for articulating power relationships between intra-party actors, the extensiveness of party statutes (as measured by length) therefore seems a logical indicator of the degree to which political parties have pursued rule-bound and predictable behaviors.

We are not arguing that formal rules tell the whole story of how parties operate. Nor do we suggest that statutes of similar length reflect similar organizational forms. Rather, we contend that statute length offers an important indication of differences in how much political elites invest in their parties’ extra-legislative organizations as one pathway to achieving party institutionalization. While we have only examined party statutes at one point in time, it would be possible to use historical versions of party statutes to track and analyze institutionalization (via formalization) over time. The objectivity and relative accessibility of the measure lends itself well to both temporal and geographic comparisons. Our approach also provides valuable insights into some mechanisms of party institutionalization by pointing to some factors that push parties towards formalization. Future research using these measures may reveal more about the circumstances under which internal challenges do—or do not—spur party rule-making. Statute length as a measure of formalization might also be fruitfully used to examine the consequences of institutionalization through formalization, and to investigate its relationship with systemic level attributes of party institutionalization such as electoral stability. In short, given how much some parties invest in developing their statutes, there are good reasons for scholars to pay more attention to these official stories as indicators of permanent and predictable party structures.

Supplemental Material

sj-pdf-1-ppq-10.1177_13540688211071066 – Supplemental Material for Party statutes and party institutionalization

Supplemental Material, sj-pdf-1-ppq-10.1177_13540688211071066 for Party statutes and party institutionalization by Susan E. Scarrow, Jamie M. Wright and Anika Gauja 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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is supported by Deutsche Forschungsgemeinschaft (PO 370/11-1), Economic and Social Research Council (ES/L016613/1), National Science Foundation US (1419401).

ORCID iDs

Susan E. Scarrow

Jamie M. Wright

Anika Gauja

Supplemental material

Supplemental material for this article is available online.

Notes

Author biographies

Susan E. Scarrow is Professor of Political Science at the University of Houston. She is a founding co-director of the Political Party Database project. She has published extensively on party membership, and on the causes and consequences of party organizational development.

Jamie M. Wright is a PhD candidate at the University of Houston in the Department of Political Science. Her research interests center on political behavior, political parties, and political communication, intersecting in the way that our technology and media-infused environments shape political attitudes and behavior particularly among younger Americans and other under-represented groups.

Anika Gauja is Professor of Politics at the University of Sydney. She researches in the areas of party membership, partisan participation, and party regulation. She is the author of Party Reform: The Causes, Challenges, and Consequences of Organizational Change (OUP, 2017).

References

Alexander

(1998) Environmental constraints and organizational strategies: complexity, conflict, and coping in the non-profit sector. In Powell

Clemens

ES.

, (eds) Private Action and the Public Good. London, UK: Yale University Press, 272–290.

van Biezen

(2012) Constitutionalizing party democracy: the constitutive codification of political parties in post-war Europe. British Journal of Political Science 42(1), 187–212.

van Biezen

Piccio

(2013) Shaping intra-party democracy: on the legal regulation of internal party organizations. In Cross

Katz

(eds) The Challenges of Intra-Party Democracy. Oxford, UK: Oxford University Press.

Bischof

(2017) Towards a renewal of the niche party concept: parties, market shares and condensed offers. Party Politics 23(3), 220–235.

Bolleyer

(2013) New Parties in Old Party Systems: Persistence and Decline in Seventeen Democracies. Oxford, UK, Oxford University Press.

Bolleyer

(2018) The State and Civil Society: Regulating Interest Groups, Parties and Public Benefit Organizations in Contemporary Democracies. Oxford, UK, Oxford University Press.

Bolleyer

Ruth

(2018) Elite Investments in Party Institutionalization in new democracies: a two dimensional approach. Journal of Politics 80(1), 288–302.

Bolleyer

Ruth-Lovell

(2019) Party institutionalization as multilevel concept: base-versus elite-level routinization. Zeitschrift für Vergleichende Politikwissenschaft 13(2), 175–198.

Carty

(2004) Parties as franchise systems: the stratarchical organizational imperative. Party Politics 10(1), 5–24.

10.

Cross

Gauja

(2014) Designing candidate selection methods: explaining diversity in Australian political parties. Australian Journal of Political Science 49(1), 22–39.

11.

Deschouwer

(2006) Political parties as multi-level organizations. In Katz

Crotty

(eds) Handbook of Party Politics. Sage, 291–300.

12.

Diamond

Linz

Lipset

(1989) Democracy in Developing Countries. Boulder, CO: Latin America. Lynne Riener.

13.

Dix

(1992) Democratization and the institutionalization of Latin American political parties. Comparative Political Studies 24(4): 488–511.

14.

van Dusky-Allen

Heller

(2014) Bicameralism and the logic of party organization. Comparative Political Studies 47(5), 715–742.

15.

Greene

Sánchez-Talanquer

(2018) Authoritarian legacies and party system stability in Mexico. In: Mainwaring

(ed). Party Systems in Latin America: Institutionalization, Decay, and Collapse. Cambridge, UK: Cambridge University Press, 201–226.

16.

Harmel

Janda

(1982) Parties and Their Environments: Limits to Reform? New York, NY: Longman Publishing Group.

17.

Harmel

Svåsand

Mjelde

(2018) Institutionalisation (and de Institutionalisation) of right-wing protest parties: the progress parties in Denmark and Norway. Maryland, US: Rowman & Littlefield.

18.

Heinisch

Mazzoleni

(eds) (2016) Understanding populist party organisation: The radical right in Western Europe. Palgrave Macmillan.

19.

Huntington

(1965) Political development and political decay. World Politics 17, 386–430.

20.

Huntington

(1968) Political Order In Changing Societies. London, UK: Yale University Press.

21.

Katz

Mair

(1992) Introduction: The cross-national study of party organizations. In: Katz

Mair

(eds). Party Organizations: A Data Handbook. California, US: Sage.

22.

Katz

Mair

(1993) The evolution of party organizations in Europe: the three faces of party organization. American Review of Politics 14, 593–617.

23.

Katz

Mair

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

24.

Kitschelt

(2014) Democracy and Accountability Project Dataset. https://sites.duke.edu/democracylinkage/files/2014/12/DALP_Codebook_2014-04_01.pdf

25.

Levitsky

(1998) Institutionalization and peronism. Party Politics 4, 77–92.

26.

Luna

(2014) Party system institutionalization: do we need a new concept? Studies in Comparative International Development 49(4), 403–425.

27.

Lührmann

Düpont

Higashijima

, et al. (2020) Varieties of Party Identity and Organization (V-Party) Dataset V1. Varieties of Democracy (V-Dem) Project. DOI: 10.23696/vpartydsv1

28.

Mainwaring

Bizzarro

Petrova

(2018) “Party System Institutionalization, Decay, and Collapse,” pp. 17–33 in Mainwaring

. (ed). Party Systems in Latin America: Institutionalization, Decay, and Collapse. Cambridge: Cambridge University Press.

29.

Mainwaring

Scully

(1995) Building Democratic Institutions. Party Systems in Latin America. California, UK: Stanford University Press.

30.

Martínez

(2020) Presidential instability in Latin America: why institutionalized parties matter. Government and Opposition 56: 683–704. doi: 10.1017/gov.2020.18.

31.

Meyer

Rowan

(1977) Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology 83(2), 340–363.

32.

Meyer

TM.

Wagner

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

33.

Molenaar

(2014) Legitimising political party representation: party law development in Latin America. International Political Science Review 35(3): 324–338.

34.

Panebianco

(1988) Political Parties: Organization and Power. Cambridge, UK: Cambridge University Press.

35.

Poguntke

Scarrow

Webb

, et al. (2016) Party rules, party resources and the politics of parliamentary democracies: how parties organize in the 21^st century. Party Politics 22(6): 661–678.

36.

Randall

Svasand

(2002) Party institutionalization in new democracies. Party Politics 8(1), 5–29.

37.

Rodriguez

Rosenblatt

(2020) Stability and Incorporation: toward a new concept of party system institutionalization. Party Politics 26(2): 249–260.

38.

Rosenblatt

(2018). Party Vibrancy and Democracy in Latin America. Oxford: Oxford University Press.

39.

Samuels

Shugart

(2010) Presidents, Parties, and Prime Ministers: How the Separation of Powers Affects Party Organization and Behavior. Cambridge, UK: Cambridge University Press.

40.

Smith

Gauja

(2010) Understanding party constitutions as responses to specific challenges. Party Politics 16(6): 755–775.

41.

Wills-Otero

(2016) “The electoral performance of Latin American traditional parties, 1978-2006: Does the internal structure matter?” Party Politics. 22(6) 758–772.

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

0.02 MB