The effect of ruling party change and party ideology on participatory budgeting stability: Survivors and victims of local elections

Abstract

Participatory Budgeting (PB) has demonstrated a remarkable capacity for global diffusion. However, its ability to endure over time and become fully institutionalized remains more limited. This study addresses one of the most frequently cited explanations for PB’s instability: changes in the ruling party. Contrary to prevailing assumptions, we argue that it is not party turnover per se that leads to the discontinuation of PB, but rather the ideological orientation of such changes. To investigate this hypothesis, we draw on an original dataset covering 295 Spanish municipalities, including both cases of PB continuity and discontinuity. Our findings indicate that changes in the ruling party alone do not account for the interruption of PBs. Instead, the key determinant is the ideological direction of government change or continuity: PB is more likely to persist when left-wing parties remain in power or when transitions occur toward left-wing administrations.

Keywords

participatory budgeting instability ideology party change elections

Introduction

Participatory Budgeting (PB) has become one of the most widely adopted mechanisms¹ for incorporating citizen preferences into policymaking. As a democratic innovation, PB constitutes a participatory process through which lay citizens are involved in governmental decision-making, with the aim of broadening and deepening their capacity to influence political outcomes (Elstub and Escobar, 2019; Smith, 2009). According to Dias et al. (2021),² over 14,000 PB processes have been implemented across all continents, encompassing a wide range of socioeconomic and political contexts. Despite this global expansion, PB has faced significant challenges–most notably, frequent abandonment and institutional instability (Alves and Allegretti, 2012; Wampler and Goldfrank, 2022).

While growing scholarly attention has examined the adoption of PB through the lens of policy diffusion (Ganuza and Baiocchi, 2012; Jacquet et al., 2024; Sintomer et al., 2013; Wampler, 2008), less is known about the factors that explain its discontinuation—particularly outside the Brazilian context. Indeed, much of the existing literature on PB instability remains focused on Brazil (Avritzer and Wampler, 2008; Bezerra and Junqueira, 2022; Spada, 2014; Wampler and Goldfrank, 2022). Now, three decades after PB’s introduction to Europe (De Vries et al., 2021), enough cases exist to enable a longitudinal analysis that can provide new insights into the conditions underpinning PB stability. However, advancing research in this direction remains challenging due to the lack of comprehensive data–particularly in the form of a census or a longitudinal panel of PBs.

This study addresses the central research question of which political factors explain the maintenance or abandonment of PB following local elections. While the adoption and diffusion of PB have been extensively studied, the causes behind its discontinuation remain underexplored. Understanding the conditions under which PB disappears is essential for a more comprehensive account of the global trajectory of this democratic innovation—extending the focus beyond its spread to include its potential decline.

We examine a set of external factors—those not directly related to the internal design or characteristics of PBs–focusing instead on local contextual variables such as municipality size, municipal wealth, and indicators of political competition, including types of governing majorities and party ideology. The article seeks to contribute to the international debate on PB stability, while also laying the groundwork for the development of a comprehensive panel dataset on PBs in Spain. Specifically, the analysis sheds light on one of the most frequently cited explanations for PB discontinuation: changes in the ruling party. The findings indicate that party turnover alone does not account for the abandonment of PBs. Rather, it is the ideological direction of the change—or continuity—of government that proves more decisive.

Moreover, this study contributes to ongoing debates on the role of party ideology in shaping local participation policies (Gherghina, 2024; Premat, 2009; Schneider and Busse, 2018). Our findings challenge the prevailing view that PB continuity is ideologically neutral and unrelated to government change. Instead, we demonstrate that PB persistence is closely tied to the distinct values and preferences of political parties, thereby advancing the literature on party–policy linkages.

The analysis draws on an original dataset of 295 Spanish municipalities that implemented PB at some point during the 2015–2019 and 2019–2023 electoral terms. We define continuity cases (n = 151) as municipalities in which PB was implemented at least once during both electoral cycles. In contrast, discontinuity cases (n = 144) refer to those that implemented PB during the 2015–2019 term but did not continue it at any point in the subsequent electoral cycle.

Spain provides a particularly suitable context for examining potential party–policy linkages for two key reasons. First, Spanish local governments enjoy considerable autonomy in the design and implementation of participatory policies, as there is no overarching national legislation mandating or constraining PB–unlike in many other countries. Second, the Spanish tradition of PB has been more ideologically aligned with the original Brazilian model, which has historically been associated with the political left (Sintomer et al., 2013), thereby offering fertile ground for exploring the role of party ideology in PB continuity.

The article is structured as follows. First, we outline the most plausible factors that may explain PB stability, with particular emphasis on variables related to party competition. From this discussion, we derive a set of hypotheses concerning the role of external factors—those unrelated to the design and internal characteristics of PB processes. We then provide an overview of the local PB context in Spain, followed by a detailed explanation of our data collection strategy. The next section presents the empirical analysis and key findings. Finally, we conclude by discussing the broader implications of our results for understanding local party–policy linkages and the role of political dynamics in shaping the trajectory of democratic innovations.

Explaining the stability of participatory budgeting

We adopt a broad and inclusive definition of local PB, understood as a participatory mechanism aimed at incorporating citizens’ voices into the allocation of (a portion of) the municipal budget, based on the outcomes of citizen deliberation (Dias et al., 2021). Rather than following a single, standardized model, PB encompasses a variety of forms that vary across temporal and geographical contexts³. While the causes of PB’s success and diffusion have been more commonly investigated, its failure or disappearance has received comparatively less attention. Yet both dimensions can be seen as two sides of the same coin, each contributing to a better understanding of PB’s stability and institutionalization.

The literature on PB stability distinguishes between internal–such as process design–⁴ and external explanatory factors, including socioeconomic and political contexts (Alves and Allegretti, 2012; Kukučková and Bakoš, 2023)⁵. Recent studies, however, suggest that internal design features are less decisive than expected in explaining key outcomes, including participation rates (e.g., Van der Does, 2022). If changes in PB design do not have an immediate effect on participants, it is unlikely that they significantly influence political representatives’ decisions to maintain or abandon PBs. In other words, although the design may be crucial for aspects such as the quality of deliberation, decisions to abandon PB tend to be driven primarily by ideological and political considerations, as well as contextual factors such as economic crises or political competition (Alarcón et al., 2019; Alves and Allegretti, 2012; Bezerra and Junqueira, 2022; Spada, 2014). This conclusion is supported by one of the few studies employing panel data to analyze the causes of PB discontinuation (Kukučková and Bakoš, 2023).

Accordingly, this paper focuses on indicators related to political competition and municipal characteristics–such as population size and economic wealth–as the most plausible explanations for the continuation or abandonment of PB. The next section reviews the main findings in the literature concerning the relationship between these external factors and PB stability.

Participatory budgeting and political competition

Democratic innovations in general, and PB in particular, lie at the intersection of multiple forces and motivations. For some of their proponents, these mechanisms represent a novel and more effective means of producing efficient or epistemically superior policies. In this context, citizens often value the perception that such processes are removed from partisan disputes and political conflict (Falanga et al., 2021; Gherghina et al., 2023). However, potential social desirability bias– such as concealing strategic calculations–makes it more difficult to assess the true intentions of political actors in promoting these initiatives. Like any other policy or communicative act, democratic innovations are embedded in political life and may carry political and electoral implications (Anduiza et al., 2008). Even if the absence of comprehensive data on these motivations, analyzing their potential influence is essential for a deeper understanding of how participatory processes operate.

PB is one of the potential democratic mechanisms for developing participation policies. Whether these policies should differ depending on which party is in power is a classic research question–traditionally explored in areas such as macroeconomics and tax policy (Hibbs, 1977; Högström and Lidén, 2023), and more recently extended to the field of participation policies (Donaghy, 2020; Gherghina, 2024; Ramis-Moyano, 2025). Political parties are expected not only to show different levels of interest in participatory practices but also to make policy choices that are at least partially shaped by their ideological commitments (Gherghina and Geissel, 2020; Ramis-Moyano, 2024).

Traditionally, PB has been one of the participatory mechanisms where the influence of political party ideology has been most clearly observed (Ramis-Moyano, 2025; Sintomer et al., 2013). There is broad consensus in the literature that the earliest PBs in both Latin America and Europe were primarily promoted by left-wing parties. However, as PB has gained popularity⁶, the parties that promote them have become increasingly diverse in terms of their ideological families (Ganuza and Baiocchi, 2019; Welp, 2022; Falanga et al., 2020). As PBs emerge within increasingly pluralistic political arenas, understanding how various aspects of political competition influence the stability of these participatory mechanisms becomes critically important. In the next section, we focus on three key dimensions of political competition highlighted in the literature on PB stability: ruling party change, party ideology, and the strength of the government.

Ruling party change

Firstly, regarding the effect of ruling party change on PB stability, Alves and Allegretti (2012) observed that in Portugal, between 2001 and 2012, most PBs were suppressed by the same party or coalition that had originally introduced them. Their argument is not that political parties were irrelevant, but rather that other factors—such as austerity measures—played a more decisive role during the period analyzed. A few years later, Spada (2014) found that in Brazil, between 1989 and 2012, political continuity alone did not guarantee the survival of PB; instead, its persistence was primarily influenced by the strength (seat share) of the mayor’s party in the city council. Also in the Brazilian context, Bezerra and Junqueira (2022) showed that between 1996 and 2016, political continuity was not a decisive factor for PB stability. Rather, stability was conditioned by a cumulative effect resulting from the repeated implementation of PB year after year. In a different context, Svidroňová et al. (2024) provide new evidence from Slovakia suggesting that the re-election of the mayor who initially launched the PB is positively associated with its continuity after elections. Although the evidence is mixed and not conclusive, a substantial portion of the literature supports the assumption that changes in government do not significantly affect the likelihood of PB being discontinued. Therefore, in line with this body of research, we expect:

A change in the ruling party does not, by itself, significantly affect the likelihood of Participatory Budgeting being abandoned.

Party ideology

Secondly, regarding the influence of party ideology, the literature on democratic innovations identifies several stages in which ideology may shape the life cycle of PB– adoption, design, implementation, and stabilization. While different political party families adopt and promote democratic innovations (Baiocchi and Ganuza, 2016; Jacquet et al., 2024), they do not necessarily support the same types (Font and Rico-Motos, 2023; Ramis-Moyano, 2025). In contrast, no consistent ideological effects have been found in relation to the implementation of outcomes stemming from democratic innovations in general, or PB in particular. Although evidence on the influence of ideology on PB continuity remains limited–and often derives from qualitative case studies (Fernández-Martínez et al., 2020)– we argue that ideology continues to play a relevant role in explaining the persistence of PB. It shapes preferences for participatory practices among both citizens (Van der Meer et al., 2009) and political elites (Wampler and Goldfrank, 2022), influencing not only the content of these processes but also the decisions surrounding them (Becerril-Viera et al., 2024). In this sense, the preferences of both voters and political elites are likely to help explain the continuity of PB. We therefore hypothesize that the ideological direction of government change or continuity following elections affects the likelihood of PB persistence. Given the stronger ideological commitment to participatory processes among left-wing parties and their supporters, we expect the following:

Participatory Budgeting is more likely to continue in municipalities where left-wing parties remain in power or take office after elections.

Strength of government

Thirdly, the extensive literature on democratic innovations has demonstrated that these policies are not immune to political dynamics, nor to the structures and forms of government through which these dynamics take shape. More specifically, in relation to government strength, solid majorities have emerged as important variables for both the adoption and continuity of PBs. In Belgium, Jacquet et al. (2024) found that launching a PB is particularly attractive to new leading lists due to electoral incentives and the desire to appear closer to citizens’ preferences. Their findings also indicated that governments with repeated absolute majorities are more likely to promote PB. Similarly, Spada (2014, p. 27) found that those governments that feel less threatened in terms of the size of the opposition are more likely to continue with PBs. In this sense, a scenario in which the same further increase the likelihood of PB continuity. Following this reasoning, we expect:

Participatory Budgeting is more likely to continue in municipalities where the same party held an absolute majority both before and after the 2019 local elections.

Most of the aforementioned studies have examined the effects of political competition in conjunction with municipal-level determinants. For instance, Jacquet et al. (2024) reported that in Belgium, PB is more likely to be adopted in larger and wealthier municipalities. Similarly, Svidroňová et al. (2024) find that PB is more likely to persist in municipalities with a healthier economic situation and where the mayor has been re-elected. Similarly, Bezerra and Junqueira (2022) also highlight the relevance of financial factors—such as fiscal constraints—in explaining both the initial adoption and the continuity of PB. Finally, as noted earlier, Alves and Allegretti (2012) observed that the austerity measures in Portugal significantly diminished the influence of political competition on PB stability. In line with these findings, we control our results by including population size and economic conditions to our analysis.

Participatory budgeting in Spain

The introduction of PBs in Spain began in 2001, with only a few experiences preceding the 2003 local elections (Table 1). All these early PB were implemented in municipalities governed by left-wing parties, predominantly (post)communist parties. The periods between 2003-2007 and 2007-2011 marked significant stages in the development of local PBs. By 2007, except for four cases, all PBs had been initiated by left-wing parties (Ganuza and Gómez, 2008). In 2011, the number of PBs launched in Spain increased considerably, reaching 150 cases and becoming more ideologically diverse (Francés et al., 2018).

Table 1.

PB diffusion and the electoral cycles in Spain.

1999-2003	2003-2007	2007-2011	2011-2015	2015-2019
The beginning: 3 cases (Ganuza and Gómez, 2008)	First diffusion: 22 cases (Ganuza and Gómez, 2008)	Second diffusion: 150 cases (Francés et al., 2018)	First setback: exact n unknown (López and Gil-Jaurena, 2021)	First revival: exact n unknown (Pineda and Abellán, 2021)

Source: Own elaboration based on literature review.

The period from 2011 to 2015 marked the first setback for PBs in Spain, possibly due to the success of the conservative party (Partido Popular, PP) in the 2011 municipal elections, along with the effects of the austerity policies derived from the 2008 Great Recession (López and Gil-Jaurena, 2021). Just 1 week before the 2011 municipal elections, the 15M movement (Indignados) emerged on the Spanish political scene (Feenstra and Tormey, 2021; Nez, 2022). This appearance created the seed for new political parties such as Podemos in the next elections. In this way, by the 2015 local elections, numerous citizen-led candidacies came to power in several town halls. This political wave, known as New Municipalism, introduced a political agenda focused on the creation of new participatory mechanisms and the reform of existing ones (Blanco et al., 2021; Fernández-Martínez et al., 2023). The success of these political platforms contributed to a strong revival of PBs, not only in terms of the number of initiatives (Pineda and Abellán, 2021), but also in terms of their expansion across municipalities governed by a range of political parties, often adopting online voting.

Despite internal variation and recent changes, Spanish PB has been characterized—especially in comparative terms—as more closely aligned with the decisional Brazilian model than with the more consultative approaches common in other European countries (Sintomer et al., 2013). The strong ideological imprint of early left-wing PBs has likely shaped current practices, with certain political characteristics still evident in how these processes operate. This makes Spain a particularly favourable context for exploring the influence of political variables—such as party ideology—on the continuity of PB.

It is not only the content and style of Spanish PB that make it a promising case for expecting political factors to matter, but also its institutional context. In contrast to countries where PB expansion has been driven by national legislation—such as the Dominican Republic or Peru (McNulty, 2019)—Spain lacks a national legal framework mandating or promoting PB. This grants local governments significant autonomy to adopt, modify, or abandon PB at their discretion. These national characteristics position Spain as a particularly suitable case for investigating party-policy linkages and the potential ideological imprints on both the adoption and discontinuation of PB.

Methods

Data

To analyze the effect of party change on PBs stability in Spain, we employed a range of data collection strategies. The primary source was an online survey conducted between May 11th and September 15th, 2021. The survey was sent to municipalities, targeting the participation department–or, in the case of smaller municipalities, the general municipal email address–to reach the official responsible for participatory matters. In most cases, the respondents were the participation officers, mayors, or participation councilors. The survey was distributed to all municipalities with populations greater than 1000 inhabitants (N = 684) in two regions: Andalusia and the Madrid region⁷. Municipalities lacking an email address, as well as non-respondents to the online survey, were contacted by phone.

The questionnaire included an item asking whether municipalities were currently (in 2021) carrying out participatory processes, including PB. If the response indicated that PB was being implemented, we conducted a manual internet search to confirm the existence of the PB and to determine whether it was a ‘new case’ or a ‘continuity case’. Similarly, when the response was negative, we performed a manual internet search to assess whether the municipality had previously implemented a PB (‘discontinuity case’) or if it had never done so. Additionally, supplementary data was collected from December 2021 to January 2023.

First, following a census approach, we gathered information on municipalities classified as ‘big cities’ according to the Law of Big Cities in Spain (Ley de Grandes Ciudades en España), as well as municipalities with more than 100,000 inhabitants and the five most populous cities in each province of Spain. Second, we identified additional cases through a literature review of studies on Spanish PBs (Blanco and Ballester, 2011; Gómez del Peso, 2014; Mayor, 2017; Pineda and Abellán, 2021). Finally, we expanded our dataset by collecting data from additional PBs from institutional social media and reports, using a snowball sampling strategy, where previously identified cases led to the next ones. Although an official and comprehensive census of municipalities implementing PBs is lacking, our data collection strategy sought to construct a diverse sample by combining census-based and territorial criteria. This approach allowed us to capture a diverse range of experiences in terms of their (dis)continuity (Seawright and Gerring, 2008) in a final database with 295 PBs, of which 144 are discontinuity cases and 151 are continuity cases (Table 2).

Table 2.

Source distribution for case selection.

	Discontinuity cases% (n)	Continuity cases% (n)	Total cases% (n)
Online survey	50.0 (72)	31.1 (47)	40.3 (119)
Census (big cities)	15.3 (22)	18.5 (28)	16.9 (50)
Literature review	29.2 (42)	32.5 (49)	30.8 (91)
Institutional social media and reports	5.6 (8)	17.9 (27)	11.9 (35)
	100 (144)	100 (151)	100 (295)

Source: Own elaboration based on primary data.

Variables and operationalization

Our dependent variable classifies a PB as a continuity case when it was implemented at least once during both the 2015-2019 and 2019-2023 municipal legislatures (depicted in light blue in Figure 1). In contrast, we define a discontinuity case as one in which PB was implemented at least once during the 2015-2019 period, but not in the subsequent 2019-2023 term (depicted in light pink in Figure 1). As shown in Figure 1, PB continuity and discontinuity cases are geographically balanced.

Figure 1.

Geographic distribution of PBs based on continuity or discontinuity of cases.

To examine the factors affecting PB continuity, we estimate a series of logistic regression models.⁸ Two sets of independent variables are included. First, we incorporate a group of variables related to political competition, which constitute the core of our analysis: ruling party change, ideological change/continuity and government strength. Second, we add two contextual control variables identified in the literature review as relevant: population size and municipal wealth (Bezerra and Junqueira, 2022). We code ruling party change as a binary variable (0 = no change; 1 = change)⁹, based on whether the mayor’s party changed between the two periods. Ideological change/continuity is captured through a categorial variable with three values: 0 = radical left continuity/arrival; 1 = social democratic continuity/arrival; 2 = right continuity/arrival. Government strength is also a binary variable: it takes the value 1 when the same party held a stable single-party majority in both legislatures, and 0 otherwise.

As control variables, we include population and municipal wealth, both included in previous studies on the adaptation and stability of PB (Bezerra and Junqueira, 2022; Jacquet et al., 2024; Spada, 2014). The data was collected from National Statistics Institute (Instituto Nacional de Estadística, INE). Population is divided into five categories: up to 5000 inhabitants = 0; 5,001-10,000 = 0.25; 10,001-20,000 = 0.5; 20,001-50,000 = 0.75; more than 50,000 inhabitants = 1. Municipal wealth is measured in five categories according to percentiles. Tables A1 y A2 in appendix summarize the independent variables operationalization and their distribution across the 295 PB cases in our final sample.

The analysis proceeds in two steps. First, we conduct bivariate analyses between the dependent variables and each independent and control variables using the Chi-square test. Second, we estimate a series of logistic regression models: starting with control variables alone, then adding each explanatory variable individually, and concluding with the full model including all predictors.

Results

This section presents the results of our analyses. Before turning to the regression, we begin with a bivariate analysis to assess the level of association between each independent variable and the dependent variable. Table 3 highlights in bold the percentages that reflect statistically significant differences in distribution.

Table 3.

Chi-Square test (percentages by row)*.

	PB discontinuity	PB continuity	Total N
Population size
Up to 5000	62.8a	37.2b	43
5001 to 10000	49.0a	51.0a	49
10001 to 20000	52.7a	47.3a	55
20001 to 50000	42.4a	57.6a	66
More than 50000	43.9a	56.1a	82
Municipal wealth
Percentile 1	62.7a	37.3b	59
Percentile 2	57.6a	42.4a	59
Percentile 3	45.8a	54.2a	59
Percentile 4	44.1a	55.9a	59
Percentile 5	33.9a	66.1b	59
Party change
No party change	45.5a	54.5a	274
Party change	52.6a	47.4a	295
Ideology continuity/arrival
Radical left	53.1a	46.9a	295
Social democracy	44.2a	55.8a	295
Right	66.1a	33.9b	295
Government strength
Stable majority	61.7a	38.3b	295
No stable majority	44.5a	55.5b	274

Source: Own elaboration based on primary data.

*Each subscript letter denotes a subset of PB stability categories whose column proportion do not differ significantly from each other at the 0,05 level. Bold values are significative differences between categories.

Starting with the variables related to political competition, we find no significant differences between municipalities where a governing party change occurred and those where it did not. However, when considering the ideological direction of the change or continuity in government, clear patterns emerge. In municipalities where right-wing parties either remained in office or came to power, the proportion of discontinued PBs is significantly higher. These findings align with our expectations regarding the influence of party ideology. In contrast, the results related to government strength diverge from our initial hypothesis. Specifically, in municipalities where a single-party absolute majority was maintained across both terms, the percentage of abandoned PBs is significantly higher–contrary to the expectation that government stability would support continuity.

Turning to the control variables, we observe that smaller municipalities–particularly those with fewer than 5000 inhabitants–show a significantly higher proportion of PB discontinuation, with nearly two out of every three such municipalities abandoning the process.⁹ Regarding municipal wealth, the results indicate a linear relationship: PB discontinuation is more common in the poorest municipalities and less frequent in the wealthiest ones. The differences are statistically significant for both extremes of the income distribution.

Table 4 presents the results of five logistic regression models. In all models, population size and municipal wealth are included as control variables. Models 2, 3, and 4 test the effect of different aspects of political competition by introducing, respectively, party change, ideological direction of change/continuity, and government strength. Finally, Model 5 incorporates all explanatory variables simultaneously.

Table 4.

Logistic regression models (PB discontinuity 0; PB continuity 1).

	M1			M2			M3			M4			M5
	Coef.	std.Err.	Odds R	Coef.	std.Err	Odds R	Coef.	std.Err.	Odds R	Coef.	std.Err.	Odds R	Coef.	std.Err.	Odds R
Population size	.211	.375	1.235	.281	.394	1.325	.517	.451	1.677	.236	.400	1.267	.409	.467	1.505
Municipal wealth	1.075***	.377	2.931	1.198***	.396	3.313	1.133**	.452	3.104	1.051***	.404	2.860	1.112**	.473	3.039
No party change (ref. cat. Party change)				.407	.281	1.502							.373	.361	1.452
Radical left continuity/arrival (ref. cat. Right continuity/arrival)							.893*	.478	2.443				.735	.500	2.086
Social democracy continuity/arrival (ref. cat. right continuity/arrival)							.899***	.334	2.457				.777**	.353	2.174
No stable majority (ref. cat. stable majority)										.278	.361	1.321	.292	.411	1.339
Constant	−.611	.249	.543	−.950	.343	.387	−1.533	.406	.216	−.788	.333	.455	−1.887	.525	.151
Nagelkerke	.055			.075			.114			.068			.120
Number of cases (PB)	295			274			229			274			229

Source: Own elaboration based on primary data.

Note. *p < .1; **p < .05; ***p < .01. Bold values are significative differences between categories.

Party change does not have a statistically significant effect, either on its own (M2) or in the full model (M5). These findings confirm Hypothesis 1, suggesting that a change in governing party does not necessarily increase the likelihood of PB discontinuation. In other words, government continuity alone does not ensure PB stability.

More interestingly, the ideological direction of the mayor’s party in office after elections shows a significant effect (M3). Specifically, when radical left and social democratic parties remain in–or come to–power, the probability of PB continuity increases. This finding if further supported in the full model (M5), although the coefficient for the radical left falls slightly below conventional significance thresholds. These results lend support to Hypothesis 2: compared to right-wing parties, both social democratic (M3 OR = 2,457, p = .007; M5 OR = 2,174, p = .028) and radical left parties (M3 OR = 2,443, p = .062) continuing in power o coming to power are approximately twice as likely to sustain PBs.

In contrast, the government strength–measured as having held an absolute majority across two consecutive terms– does not show a statistically effect on PB continuity (M4 and M5). Thus, no evidence is found to support Hypothesis 3.

Turning to the control variables, municipal wealth emerges as a consistent predictor: higher income per capita is significantly associated with greater likelihood of PB continuity, with p-values ranging from 0.002 to 0.019 across models. In contrast, population size does not have a significant impact. Both control variables remain stable across all models.

Discussion

The aim of this article was to analyze the extent to which changes in local party governments following municipal elections affect the stability of Participatory Budgeting (PB) processes. Drawing on an original dataset covering 295 Spanish municipalities that implemented PBs between 2015 and 2023, the study finds that a change in the ruling party does not significantly increase the likelihood of PB abandonment. Rather, the key explanatory factor is the ideological orientation of the party in office. Specifically, the arrival or continuation of left-wing parties–both radical left and social democratic–increases the probability of PB continuity, compared to scenarios in which right-wing parties either remain in power or take office. These findings support Hypotheses 1 and 2: PB continuity is less about government turnover per se and more about the ideological priorities of the governing party. Government changes only become relevant when they imply and ideological shift, while left-wing incumbency tends to favour the maintenance of PB practices.

These results confirm—but also extend—previous findings. In line with earlier research (Alves and Allegretti, 2012; Spada, 2014), our study reinforces the idea that a change in government does not, by itself, necessarily lead to the interruption of PB. However, our analysis adds an important dimension by emphasizing the role of party ideology, not only in cases of political turnover but also in instances of government continuity. This is the central contribution of our work: demonstrating that right-wing governments are significantly less likely to maintain PB processes, whether they are newly elected or continue in power.

This insight also enriches ongoing debates on the role of party ideology in shaping participatory governance. While much of the existing literature focuses on the initiation and implementation of participatory mechanisms (Donaghy, 2020; Gherghina, 2024), our findings resonate with Wampler and Goldfrank (2022) argument that different sets of factors drive the adoption, diffusion, and disappearance of democratic innovations. Our contribution shows that ideological party differences persist–not only in the decision to adopt participatory policies, but also, albeit to a lesser extent, in their continuation or abandonment.

This finding is particularly relevant the broader field of democratic innovations. PBs are designed to be permanent institutional mechanisms with a cyclical participatory process, expected to improve through iterative learning. Yet, despite its widespread diffusion, PB frequently struggles to become a stable feature of local governance. As Curato and Calamba (2024) argue, understanding the political drivers of its institutional fragility— such as party ideology and municipal wealth, as highlighted by our results—can inform the design of more resilient participatory mechanisms.

Additionally, our analysis finds no consistent evidence that the repetition of absolute majorities by the same party (Hypotheses 3) significantly affects PB continuity. Finally, in line with previous research, our study confirms that higher levels of municipal wealth increase the likelihood of PB continuity (Bezerra and Junqueira, 2022; Kukučková and Bakoš, 2023; Spada, 2014). While larger municipalities are more likely to adopt PB (Ebdon and Franklin, 2006), our results show no evidence between population size and PB continuity.

This set of conclusions is not without limitations. One of the main challenges in analyzing the continuity of PBs in Spain lies in the absence of a comprehensive census or longitudinal panel. Despite recent advances at the regional level (Peiruza et al., 2024; Pineda and Abellán, 2021), this gap remains unresolved. As a result, any generalization of our findings should be made with caution. Our research strategy sought to combine multiple data collection approaches to approximate as diverse a reality as possible across municipalities. Nonetheless, an important direction for future research would be the development of a national PB census. Such an effort would facilitate more robust comparative analyses– both over time and across regions–regarding the contexts in which PBs emerge, the institutional designs employed, and the policy impacts achieved.

Alternative research strategies could also help address the research questions raised here. In our study, party change was operationalized as a change in the party holding the mayor’s office. However, prior case studies suggest that the promotion of participatory mechanisms is often more closely associated with changes in individual political leadership (Fernández-Martínez et al., 2023). In this regard, it is possible that the arrival of a new mayor or participation councillor–regardless of party continuity– can significantly alter a municipality’s participatory agenda. Qualitative approaches (e.g., Becerril-Viera et al., 2024; Röcke, 2014) could therefore be particularly useful in differentiating between party turnover and changes in political leadership. Moreover, future research could explore whether shifts in the ideology of the promoting actor correspond to changes in the democratic vision that underpins the PB process. This would help differentiate types of PBs according to their underlying goals and extend the focus to broader local participatory agendas (Asenbaum, 2025).

Regarding ideology, our classification follows the same three-category schema used in previous studies (Bezerra and Junqueira, 2022; Font and Rico-Motos, 2023), partly due to limitations in sample size. While the parties included in the ‘social democratic’ and ‘right’ categories are internally consistent, the ‘radical left’ category is more heterogeneous. It includes traditional post-communist parties as well as newer, local bottom-up citizen coalitions. A more fine-grained analysis of these internal distinctions could yield greater nuance in future research. Likewise, the variable used to capture government strength does not reflect the full complexity of local coalitions dynamics. Rather, it is intended to isolate an ideal condition for PB continuity–namely, the repetition of a single-party absolute majority–against other, more unstable scenarios. It is likely that the considerable heterogeneity within the ‘non-repetition of absolute majorities’ category may have diluted potential effects.

Finally, regarding external validity, our findings are likely shaped by the broader political and institutional context in Spain. The general level of polarization, the degree of politicization of participatory policies, and the national style of participatory governance all influence the relationship between ideology and PB continuity. In contexts where participatory mechanisms are less polarized and more technocratically oriented, one might expect the effect of party ideology to be weaker. Similarly, in countries where local participatory policies are tightly regulated by national legislation—as is the case in Poland or Peru, which have state-led PB programs—political variables may play a less decisive role in shaping PB trajectories.

Footnotes

Declaration of conflicting interests

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

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Spanish Ministry of Science, Innovation and Universities under grant number; PID2019-106731GB-I00.

José Luis Fernández-Martínez

Isabel Becerril-Viera

Joan Font

Notes

Appendix

Table A1.

Codebook of dependent and independent variables.

Original variables	Original categories	Recode variables	Recode categories
Existence of PB (2015_2019)	1 = yes 0 = no	PB stability	1 = continuity case 0 = discontinuity case
Existence of PB (2019-2023)	1 = yes 0 = no	PB stability	1 = continuity case 0 = discontinuity case
Party of the Mayor (2015-2019)*	1 = social democracy 2 = right 3 = radical left	Party change	0 = no change (same mayor’s party) 1 = change (different mayor’s party)
Party of the Mayor (2019-2023)	1 = social democracy 2 = right 3 = radical left	Party change
Party of the Mayor (2015-2019)	1 = social democracy 2 = right 3 = radical left	Ideological change/continuity	0 = radical left continuity/arrival 1 = social democracy continuity/arrival 2 = right continuity/arrival
Party of the Mayor (2019-2023)	1 = social democracy 2 = right 3 = radical left	Ideological change/continuity
Type of government majority (2015-2019)	0 = no majority 1 = majority	Stable single party majority	0 = not stable majorities (the rest: new majority, loosed majority, never majority) 1 = stable majorities (majority preserved)
Type of government majority (2019-2023)	0 = no majority 1 = majority	Stable single party majority
Number of inhabitants (2021)	Continuous	Population size	0 = up to 5000 0.25 = between 5001 and 10000 0.5 = between 10001 and 20000 0.75 = between 20001 and 50000 1 = above 50000
Average net income per inhabitant (2018)	Continuous	Municipal wealth	0 = up to 8748.79 0.25 = from 8748,8 to 9896.39 0.5 = from 9896,4 to 10987,59 0.75 = from 10987,6 to 12764,99 1 = more than 12765

Source: Own elaboration based on primary data.

*This note applies for variable Party of the Mayor of both legislatures 2015-2019 and 2019-2023: regionalists and independents Mayors are excluded from our analysis.

Table A2.

Descriptive statistics of the independent variables according to continuity or discontinuity PB cases.

Variables	Categories	% (n)		Total % (n)
Variables	Categories	PB discontinuity	PB continuity	Total % (n)
Inhabitants (standarized from 0 to 1)	Up to 5000	18.8 (27)	10.6 (16)	14.6 (43)
	5001 to 10000	16.7 (24)	16.6 (25)	16.6 (49)
	10001 to 20000	20.1 (29)	17.2 (26)	18.6 (55)
	20001 to 50000	19.4 (28)	25.2 (38)	22.4 (66)
	>50000	25.0 (36)	30.5 (46)	27.8 (82)
	Total	100 (144)	100 (151)	100 (295)
Average net income per person 2018 (standarized from 0 to 1)^a	Up to 8748.79 Є	25.7 (37)	14.6 (22)	20.0 (59)
	From 8748,8 to 9896.39 Є	23.6 (34)	16.6 (25)	20.0 (59)
	From 9896,4 to 10987,59 Є	18.8 (27)	21.2 (32)	20.0 (59)
	From 10987,6 to 12764,99 Є	18.1 (26)	21.9 (33)	20.0 (59)
	More than 12765 Є	13.9 (20)	25.8 (39)	20.0 (59)
	Total	100 (144)	100 (151)	100 (295)
Party change^b	Yes	30.8 (40)	25.0 (36)	27.7 (76)
	No	69.2 (90)	75.0 (108)	72.3 (198)
	Total	100 (130)	100 (144)	100 (274)
Ideological change/continuity^c	Radical left continuity/arrival	14.5 (17)	13.4 (15)	14.0 (32)
	Social democracy continuity/arrival	52.1 (61)	68.8 (77)	60.3 (138)
	Right continuity/arrival	33.3 (39)	17.9 (20)	25.8 (59)
	Total	100 (117)	100 (112)	100 (229)
Stable single party majority^d	Yes	22.3 (29)	12.5 (18)	17.2 (47)
	No	77.7 (101)	87.5 (126)	82.8 (227)
	Total	100 (130)	100 (144)	100 (274)

Source: Own elaboration based on primary data.

^aWe use the Average net income per person 2018 variable to measure the municipalities’ wealth.

^bTotals are lower because we excluded from the analysis the municipalities that experienced a mayor party change during one of the legislatures analysed.

^cTotals are lower because we excluded from the analysis party change which involved parties not belonging to the three main categories.

^dTotals are lower because we excluded the municipalities for which we have no information.

Author biographies

José Luis Fernández Martínez is a Tenured Researcher at the Spanish National Research Council (CSIC), based at IESA-CSIC. He holds a PhD in Social Sciences from the University of Granada and has been a postdoctoral researcher at other CSIC centers. He has taught at several Spanish universities and has been a visiting researcher at the University of Westminster. His research examines citizens’ expectations of participatory and deliberative democracy and their impact on public policies. He currently also studies the challenges posed by the use of Artificial Intelligence in public decision-making.

Isabel Becerril Viera is a PhD candidate at the Institute of Advanced Social Studies (IESA), a centre attached to the Spanish National Research Council (CSIC), and at the Pablo de Olavide University (UPO). She has undertaken a research stay in Germany at the Institut für Politikwissenschaft at the University of Münster, and another at the Instituto de Investigaciones Políticas (IIP), affiliated with the Universidad Nacional de San Martín (UNSAM) and the Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) in Argentina. Her research focuses on democratic innovations and how they are influenced by party ideology.

Joan Font Fàbregas holds a PhD in Political Science from the Autonomous University of Barcelona (1993) and an MA from the University of Michigan. He has been Director of IESA (CSIC), professor at the UAB, visiting professor at several universities, and Research Director at CIS. His research focuses on mechanisms to incorporate citizens’ preferences into public policies through referendums, elections, advisory councils, surveys, and citizens’ juries. He has collaborated with organizations such as Oxfam Intermón, Open Society Foundations, IDB, OIDP, and Fundació Jaume Bofill. He has authored several books and published in leading international journals.

References

Alarcón

Galais

Font

, et al. (2019) The effects of economic crises on participatory democracy. Policy & Politics 47(2): 265–286. https://doi.org/10.1332/030557318X15407316045688

Alves

Allegretti

(2012) (In) stability, a key element to understand participatory budgeting: discussing Portuguese cases. Journal of Public Deliberation 8(2): Art. 3. Available at: https://doi.org/10.16997/jdd.147

Anduiza

Font

Mas

, et al. (2008) The electoral impact of direct-democratic practices. International Journal of Urban and Regional Research 32: 473–491. https://doi.org/10.1111/j.1468-2427.2007.00764.x

Asenbaum

(2025) A multiperspectival approach to democratic theory: five lessons for democratic innovations. Politics 0(0): 1–20. https://doi.org/10.1177/02633957251320041

Avritzer

Wampler

(2008) The Expansion of Participatory Budgeting in Brazil: An Analysis of the Successful Cases Based upon Design and socio-economic Indicators. World Bank.

Baiocchi

Ganuza

(2016) Popular Democracy. the Paradox of Participation. Stanford University Press.

Becerril Viera

Ganuza

Rico Motos

(2024) Birds of a feather flock together: influence of ideology in the implementation of participation. Democratization 31(8): 1719–1739. https://doi.org/10.1080/13510347.2024.2340638

Bezerra

CDP

Junqueira

MDO

(2022) Why has participatory budgeting declined in Brazil? Brazilian Political Science Review 16(2): 1–33. https://doi.org/10.1590/1981-3821202200020001

Blanco

Ballester

(2011) ¿Participar para transformar? La experiencia de los Presupuestos Participativos en la provincia de Barcelona. Gestión y Análisis de Políticas Públicas 5(5): 117–144. https://doi.org/10.24965/gapp.v0i5.448

10.

Blanco

Lowndes

Salazar

(2021) Understanding institutional dynamics in participatory governance: how rules, practices and narratives combine to produce stability or diverge to create conditions for change. Critical Policy Studies 16(2): 204–223. https://doi.org/10.1080/19460171.2021.1984265

11.

Curato

Calamba

(2024) Deliberative forums in fragile contexts: challenges from the field. Politics 0(0): 1–14. https://doi.org/10.1177/02633957241259090

12.

De Vries

Nemec

Špaček

(2021) International Trends in Participatory Budgeting: Between Trivial Pursuits and Best Practices. Springer Nature.

13.

Dias

Enríquez

Cardita

, et al. (2021). Participatory Budgeting World Atlas 2020-2021. Epopeia Books. Available in: https://www.oficina.org.pt/publicacoes/participatory-budgeting-world-atlas-2020

14.

Donaghy

(2020) Participation meets politics: political shifts and the longevity of participatory governance institutions. Political Science & Politics 53(1): 6–9. https://doi.org/10.1017/S1049096519001331

15.

Ebdon

Franklin

(2006) Citizen participation in budgeting theory. Public Administration Review 66(3): 437–447. https://doi.org/10.1111/j.1540-6210.2006.00600.x

16.

Elstub

Escobar

(2019) Defining and typologising democratic innovations. In: Elstub

Escobar

(eds) Handbook of Democratic Innovation and Governance. Edward Elgar Publishing. https://doi.org/10.4337/9781786433862.00009

17.

Falanga

Lühmann

Nicoletti

, et al. (2020) Participatory budgets in Canoas (Brazil) and Cascais (Portugal). A comparative analysis of the drivers of success. Journal of Civil Society 16(3): 273–293. https://doi.org/10.1080/17448689.2020.1788246

18.

Falanga

Verheij

Bina

(2021) Green(er) cities and their citizens: insights from the participatory budget of Lisbon. Sustainability 13(15): 8243. https://doi.org/10.3390/su13158243

19.

Feenstra

Tormey

(2021) From social mobilisation to institutional politics: reflecting on the impact of municipalism in Madrid and Barcelona. Social Movement Studies 22(1): 80–98. https://doi.org/10.1080/14742837.2021.1993181

20.

Fernández-Martínez

García-Espín

Jiménez-Sánchez

(2020) Participatory frustration: the unintended cultural effect of local democratic innovations. Administration & Society 52(5): 718–748. https://doi.org/10.1177/0095399719833628

21.

Fernández-Martínez

García-Espín

Alarcón

(2023) Assessing the new Municipalism reform of advisory councils: the cases of Madrid and Barcelona (2015- 2019). Urban Affairs Review 59(5): 1567–1600. https://doi.org/10.1177/10780874221109444

22.

Font

Rico Motos

(2023) Participatory institutions and political ideologies: how and why they matter? Political Studies Review 0(0): 700–717. https://doi.org/10.1177/14789299221148355

23.

Francés

Carratalá

Ganuza

(2018) 20 years of participatory budgeting in Spain. Hope for democracy 30: 275–287.

24.

Ganuza

Baiocchi

(2012) The power of ambiguity: how participatory budgeting travels the globe. Journal of Public Deliberation 8(2): Art. 8. Available at: https://doi.org/10.16997/jdd.142

25.

Ganuza

Baiocchi

(2019) The long journey of participatory budgeting. In: Elstub

Escobar

(eds) Handbook of Democratic Innovation and Governance. Edward Elgar Publishing Ltd, pp. 77–89. https://doi.org/10.4337/9781786433862.00013

26.

Ganuza

Gómez

(2008) Control político y participación en democracia: los presupuestos participativos. Estudios de progreso. Fundación Alternativas 38: 1–54.

27.

Gherghina

(2024) Political Parties and Deliberative Democracy in Europe. A Convenient Relationship? Routledge.

28.

Gherghina

Geissel

(2020) Support for direct and deliberative models of democracy in the UK: understanding the difference. Political Research Exchange 2(1): 1809474. https://doi.org/10.1080/2474736X.2020.1809474

29.

Gherghina

Tap

Soare

(2023) Participatory budgeting and the perception of collective empowerment: institutional design and limited political interference. Acta Politica 58: 573–590. https://doi.org/10.1057/s41269-022-00273-4

30.

Gómez del Peso

(2014) Los presupuestos participativos: dimensiones analíticas y líneas de debate en la experiencia española. (Doctoral thesis). Universidad Carlos III de Madrid.

31.

Hibbs

(1977) Political parties and macroeconomic policy. American Political Science Review 71(4): 1467–1487. https://doi.org/10.2307/1961490

32.

Högström

Lidén

(2023) Do party politics still matter? Examining the effect of parties, governments and government changes on the local tax rate in Sweden. European Political Science Review 15(2): 235–253. https://doi.org/10.1017/S1755773922000388

33.

Jacquet

Minsart

Dodeigne

(2024) The spread of participatory budgeting: procedural diversity, municipal context, and electoral drivers in the Belgian context. International Review of Administrative Sciences 90(4): 773–792. https://doi.org/10.1177/00208523241238083

34.

Kukučková

Bakoš

(2023) Key criteria influencing stakeholders’ decision-making about PB continuation: the case of the Czech Republic. Journal of Public Administration and Policy 16(2): 101–121. https://doi.org/10.2478/nispa-2023-0015

35.

López

Gil-Jaurena

(2021) Transformaciones del Presupuesto Participativo en España: de la aplicación del modelo de Porto Alegre a la instrumentalización de las nuevas experiencias. OBETS Revista de Ciencias Sociales 16(1): 151–174. https://doi.org/10.14198/OBETS2021.16.1.10

36.

Mayor

(2017) Presupuestos Participativos en la Región de Murcia: una visión crítica [Conference session]. In V Congreso Internacional Sobre Innovación Tecnológica y Administración Pública. La reforma de la Ley estatal de Transparencia: Retos y posibilidades, Universidad de Castilla-La Mancha, Toledo. https://eventos.uclm.es/14455/detail/v-congreso-internacional-sobre-innovacion-tecnologica-y-administracion-publica.html

37.

Mcnulty

(2019) Democracy from Above? the Unfulfilled Promise of Nationally Mandated Participatory Reforms. Stanford University Press.

38.

Nez

(2022) Democratie reelle: L'Héritage des Indignés espagnols. Editions du Croquant.

39.

Peiruza

Borge

Márquez

, et al. (2024). Political consequences, levels of participation and continuity in participatory budgeting in Catalan municipalities [Conference session]. In European Consortium for Political Research General Conference, University College Dublin, Ireland, August 12–15, 2024. https://ecpr.eu/Events/Event/PaperDetails/75503

40.

Pineda

Abellán

(2021) Análisis de las experiencias de presupuesto participativo de la Comunitat Valenciana a través de una taxonomía. Arxius De Ciències Socials 43: 165–174. https://doi.org/10.7203/acs.43.29084

41.

Premat

(2009) The implementation of participatory democracy in French communes. French Politics 7(1): 1–18. https://doi.org/10.1057/fp.2009.5

42.

Ramis-Moyano

(2024) Political ideology and Mini-Publics’ initiation: different frameworks, different participation? In: Gherghina

(ed) Political Parties and Deliberative Democracy in Europe. A Convenient Relationship? Routledge, pp. 35–49.

43.

Ramis-Moyano

(2025) Democratic innovations and political parties: reevaluating the influence of ideology. Contemporary Politics 1–20. Available at: https://doi.org/10.1080/13569775.2025.2502027

44.

Röcke

(2014) Framing Citizen Participation. Participatory Budgeting in France, Germany and the United Kingdom. Palgrave-Macmillan.

45.

Schneider

Busse

(2018) Participatory budgeting in Germany. A review of empirical findings. International Journal of Public Administration 42(3): 259–273. https://doi.org/10.1080/01900692.2018.1426601

46.

Seawright

Gerring

(2008) Case selection techniques in case study research. A menu of qualitative and quantitative options. Political Research Quarterly 61(2): 294–308. https://doi.org/10.1177/1065912907313077

47.

Sintomer

Herzberg

Allegretti

, et al. (2013) Participatory budgeting worldwide. Dialog Global 25: 1–93.

48.

Smith

(2009) Democratic Innovations. Cambridge University Press.

49.

Spada

(2014) The diffusion of participatory governance innovations: a panel data analysis of the adoption and survival of participatory budgeting in Brazil. Latin American Studies Association 32: 1–53.

50.

Svidroňová

Baláž

Klimovský

, et al. (2024) Determinants of sustainability of participatory budgeting: slovak perspective. Journal of Public Budgeting, Accounting and Financial Management 36(1): 60–80. https://doi.org/10.1108/jpbafm-04-2022-0082

51.

Van der Does

(2022) Citizen involvement in public policy: does it matter how much is at stake? Public Administration 101(3): 772–787. https://doi.org/10.1111/padm.12846

52.

Van der Meer

Van Deth

Scheepers

(2009) The politicized participant: ideology and political action in 20 democracies. Comparative Political Studies 42(11): 1426–1457. https://doi.org/10.1177/0010414009332136

53.

Wampler

(2008) A difusão do Orçamento Participativo brasileiro: “boas práticas” devem ser promovidas? Opinião Pública 14(1): 65–95. https://doi.org/10.1590/s0104-62762008000100003

54.

Wampler

Goldfrank

(2022) The Rise, Spread, and Decline of Brazil’s Participatory Budgeting: The Arc of a Democratic Innovation. Palgrave-Macmillan.

55.

Welp

(2022) The Will of the People. Populism and Citizen Participation in Latin America. De Gruyter.