The politics of piloting. The case of minimum income schemes in European cities
Abstract
enEU member states have adopted Minimum Income Schemes (MIS) to prevent destitution and ensure a minimum standard of living through means-tested income support combined with Active Labor Market Policies (ALMPs). However, the effectiveness of MIS has been hindered by limited coverage, low take-up rates, inadequate cash transfers, strict conditionalities, and the limited impact of ALMPs. Public opinion is polarized, leading to potential policy changes. Pilot projects have emerged as a strategy to address implementation barriers, facilitate evidence-based policy making, and improve stakeholder relationships. This paper investigates the political conditions under which pilots are promoted and the effects these policy decisions have on scaling up through a qualitative analysis and comparison, using two exemplary cases—B-MINCOME in Barcelona and Weten Wat Werkt in Utrecht.
摘要
zh欧盟成员国已采用最低收入计划(MIS)，通过将经济状况调查的收入支持与积极劳动力市场政策(ALMPs)相结合，以防止贫困并确保最低生活水平。然而，MIS的有效性受到“覆盖范围有限、使用率低、现金转移不足、条件严格、以及ALMPs影响有限”等因素的阻碍。舆论两极分化，导致潜在的政策变革。试点项目已成为一项用于应对实施障碍、促进循证决策和改善利益攸关方关系的战略。本文使用巴塞罗那的B-MINCOME计划和乌得勒支的Weten Wat Werkt计划这两个典型案例，通过定性分析和比较，研究了促进试点计划的一系列政治条件以及这些决策对扩大试点规模一事的影响。
Resumen
esLos estados miembros de la UE han adoptado Planes de Ingreso Mínimo (MIS) para prevenir la indigencia y garantizar un nivel de vida mínimo a través de un apoyo a los ingresos sujeto a verificación de recursos combinado con Políticas Activas del Mercado Laboral (ALMP). Sin embargo, la eficacia de los MIS se ha visto obstaculizada por una cobertura limitada, bajas tasas de utilización, transferencias de efectivo inadecuadas, condicionalidades estrictas y el impacto limitado de las PAMT. La opinión pública está polarizada, lo que lleva a posibles cambios de políticas. Los proyectos piloto han surgido como una estrategia para abordar las barreras a la implementación, facilitar la formulación de políticas basadas en evidencia y mejorar las relaciones con las partes interesadas. Este artículo investiga las condiciones políticas bajo las cuales se promueven los pilotos y los efectos que estas decisiones políticas tienen en la ampliación a través de un análisis cualitativo y una comparación, utilizando dos casos ejemplares: B-MINCOME en Barcelona y Weten Wat Werkt en Utrecht.
INTRODUCTION
All the member states of the European Union have Minimum Income Schemes (MIS) to cope with the new challenges brought by globalization, automation, and the economic crisis (Ayala et al., 2021; Coady et al., 2021; Figari et al., 2013; Frazer & Marlier, 2009, 2016; Immervoll, 2010; Peña-Casas, 2005; Peña-Casas & Ghailani, 2013). MIS are ‘non-categorical, anti-poverty schemes providing rights-based means-tested income support, that are typically flat-rate and tax financed. MIS thus constitute a universal yet selective last-resort safety net for the working-age population’ (Natili, 2019, p. 1). Their main goal is ‘to prevent destitution and to ensure a decent minimum standard of living for individuals and their dependants when they have no other or insufficient means of financial support’ (Crepaldi et al., 2017, p. 6). MIS usually include cash transfers combined with Active Labor Market Policies (ALMPs), which are programs supporting people entering the labor market through, for instance, training in job search activities or programs to promote entrepreneurship. ALMPs usually imply the monitoring of job search efforts and their sanction in case of inactivity.
Literature assessing the effectiveness of MIS' goals usually agrees that these schemes are extremely useful in alleviating poverty, reducing deprivation, and in producing benefits for individuals and households—for instance by increasing their expenditure for food, education, and health services (Almeida et al., 2022; Bastagli et al., 2019; Leventi et al., 2019). Nevertheless, the implementation of MIS has proven to be difficult and problematic on at least two levels. First, its adoption varies significantly across member states in terms of coverage of the target population, take-up rates, adequacy of cash transfers, and conditionalities (Crepaldi et al., 2017; De La Rica & Gorjón, 2019; Frazer & Marlier, 2016; Immervoll, 2010; Konle-Seidl, 2021; Marchal & Van Mechelen, 2013; Peña-Casas & Bouget, 2013). Second, the institutional and organizational settings where MIS are embedded challenge their implementation, particularly when they are associated with ALMPs. Partial results obtained through activation policies are generally attributed to administrative burdens faced by beneficiaries, low capacities, skills, and resources in public employment services and social assistance institutions, lack of coordination, and cooperation between services (Champion & Bonoli, 2011; Crepaldi et al., 2017; Heidenreich et al., 2014; Heidenreich & Rice, 2016), and a ‘tendency to prioritize different groups in need of support who may be easier to reintegrate into the labor market’ (Frazer & Marlier, 2016, p. 9). Furthermore, the discretion of street-level bureaucrats and creaming practices in delivering services can also hinder fair and equal access to ALMPs and, therefore, their effectiveness (McGann et al., 2020).
Against this background, minimum income experiments and pilot projects emerged as a possible strategy to cope with complexity and to try to overcome barriers to implementation. Pilots, if properly designed and implemented, may play a useful legitimizing function and may provide support for evidence-based policy making. A policy pilot is ‘a governance tool that contributes simultaneously to policy formulation and policy implementation, aimed to generate learning, usually through a linked evaluation and to reduce the risk of implementation failure’ (Ettelt et al., 2022, pp. 385–386).
Piloting has raised interest among scholars and policy makers as part of the ‘experimentalist turn’ in social science (Huitema et al., 2018). Pilots can address several problems affecting policy implementation ranging from anticipating conflicts to pre-assessing impacts and promoting adaptation and learning. In the case of MIS, they can serve to test new ideas, but they can also provide an opportunity to figure out what the policy will look like and what characteristics implementation would assume. For these reasons, they have a strong political component (Rogers-Dillon, 2004). Pilots are useful as a policy tool, but they can also be used as a governance tool by political actors with the aim of improving their relationships with other actors and as a strategy to promote ideational change (Ettelt et al., 2022). Accordingly, pilots can also be defined as ‘experimental implementation schemes’ (Sager & Gofen, 2022, p. 355) or ‘meta-implementation arrangements’, which pre-allocate power and roles and therefore shape the institutional and organizational environment where the process of implementation will take place.
Political dynamics behind experimentation affect the way pilots are designed, implemented, and possibly scaled up since they could support them but, at the same time, they can also be a barrier to their development. Political actors can strategically use pilots to inhibit conflicts on specific issues or to confirm existing courses of action but they can also simply reverse them at the end of the experimentation (Bailey et al., 2017; Huitema et al., 2018; Nair & Howlett, 2016). Moreover, pilots may hinder the development of policy innovation, constrained by their own nature: limited scale, representativeness, and short-term focus (Nair & Howlett, 2016). Pilot programs typically operate on a small scale, often in specific geographic areas or with a limited number of participants. This limited scale can make it difficult to assess how a policy will perform when implemented at a larger scale, if the pilot is not well designed. The participants or areas chosen for a pilot may not be fully representative of the broader population or context in which the policy will eventually be implemented. This can lead to skewed results and challenges in generalizing findings. Finally, pilots often have a relatively short duration, which may not allow for a comprehensive assessment of long-term impacts and unintended consequences of a policy.
Drawing on these premises, the paper aims to achieve two goals. On the one hand, it addresses the political conditions under which pilots are designed and implemented and possibly scaled up into mainstream policy making. The analysis will focus on two case studies of MIS, a policy area where experimentations have flourished in recent years as a means to pre-test adequacy, coverage, and potential drawbacks of ALMPs' implementation. On the other hand, the paper is aimed at analyzing the effects produced by political dynamics surrounding pilots on ‘scaling up’ and the implementation of MIS at a wider level.
The research is based on a qualitative critical examination and comparison following the Most Different Systems Design (MDSD) of two recently implemented and internationally renowned pilot projects: the B-MINCOME of Barcelona and the Weten Wat Werkt of Utrecht. Although the two pilots reach similar conclusions, they have some different characteristics within an analogous experimental design—aimed at assessing different combinations of approaches across testing groups and a control group—and a similar goal—that is to acquire knowledge on the most suitable and effective method of implementation of MIS.
This paper is organized in the following way. First, we position our article in the literature on experimentation and implementation arrangements highlighting both benefits and limitations of pilots from a political point of view. Then, we explain case selection and we expose the methodology used to carry out our qualitative research. In the fourth section, we present and discuss the political determinants of the design of the two minimum income pilots and the effects produced by them on the subsequent implementation of the policy they have tested. Finally, the conclusive section includes a discussion of the implications of the findings for future research in this area.
PILOTS, POLITICS, AND META-IMPLEMENTATION ARRANGEMENTS
Political benefits and challenges of experimentation
The implementation of MIS has proven to be challenging in all countries where they have been adopted due to the impact that institutional, organizational, and normative aspects have on the process. Implementation is not necessarily linear and coherent. Rather, it often requires ‘difficult changes in the supporting stakeholder coalition, changes in the structures and rules of familiar institutions, and new patterns of interaction’ (Crosby, 1996, p. 1405). Implementation poses both political and organizational challenges such as legitimizing the process, building a constituency to support it, gathering and mobilizing adequate resources, redesigning organizations, managing process coordination, and monitoring policy changes (Crosby, 1996). In the case of complex issues such as MIS, implementation would be more uncertain and riskier due to lack of knowledge of results, and potential high political and financial costs. To cope with these problems, several governments, at different institutional levels, have opted to pre-test MIS using an experimental approach based on piloting.
Policy pilots are ‘rigorous early evaluations of a policy (or some of its elements) before that policy has been rolled out nationally and while is still open to adjustment in the light of the evidence compiled’ (Jowell, 2003, p. 11). They usually take the form of small-scale tests designed to assess the feasibility of methods and procedures for later use on a large scale (Everitt, 2021). Pilots and trials gained the attention of policy scholars as tools for policy design in the 1960s when they started to be adopted in the UK and USA to manage ambiguous and conflictual issues (Bailey et al., 2017), but they became increasingly popular at the beginning of the 2000s as part of the ‘experimental turn’ in political science aimed at collecting evidence-based information on policies, stimulating creative solutions to complex problems, and encouraging social, political, and policy learning (Ansell & Bartenberger, 2016; Huitema et al., 2018). According to Matland, experimental implementation usually takes place when policy ‘goals are clear and widely supported’ but they have ‘unclear means of implementation’ so ‘implementing policies of this type can be technology-forcing and can lead to the development of entirely new capabilities’ (1995, p. 167).
Pilot projects can support policy makers in producing knowledge about the policy context and the potential solutions that can be implemented within it (Nair & Howlett, 2016). Pilots fulfill, therefore, several goals: to help frame problems (Peters, 2005; Rochefort & Cobb, 1994); to foster early policy evaluation; and to review the policy process and adapt it to circumstances and tackle unanticipated conditions through learning (Ettelt et al., 2015; Kivima & Rogge, 2022; Swanson et al., 2010). Lastly, they can be used rhetorically to portray possible policy scenarios and to address political conflicts associated with controversial issues (Bailey et al., 2017). All of these goals are inherently political. As Rogers-Dillon clearly points out, experimental pilots are ‘institutional channels’ that ‘demonstrate to the public and policy elites whether or not a policy will work. Defining what works is a powerful political tool’ (2004, p. 2). In the case of MIS, this implies conflictual redistributive dynamics, controversial party support, complex bureaucratic implementation, and unclear results, especially as when associated with employment and social activation measures, the adoption of pilots can be politically strategic. In this regard, pilots are also providing room for a better understanding of policy implementation by creating a shortcut. Pilots allow for anticipating potential consequences of policy implementation, creating a temporary overlap between the formulation, the implementation, and the evaluation stages of policy making. Thus, pilots can be used to pre-design and test implementation arrangements to be adopted later and to assess opportunities, constraints, and political strategies available to actors (Brodkin & Kaufman, 2000). Moreover, they offer the advantage of testing implementation arrangements at a small-scale level, with limited risks, and a lower level of political contestation (Nair & Howlett, 2016; Natili, 2019).
At the political level, piloting nevertheless also poses, several challenges, especially with reference to their application to MIS. As clearly highlighted by Brodkin and Kaufman (2000) experimentation in welfare policies can be strategically used by politicians to promote new ideas but, paradoxically, it can also be used to confirm existing processes and paradigms by postponing conflicts. This mechanism of reinforcing existing courses of action is enhanced through a process of design and data interpretation that can either disregard or selectively reconstruct information (Brodkin & Kaufman, 2000, p. 523). Thus, politicians seem to support experimentations only if they perceive that they will confirm a decision already taken and not because they represent ‘opportunities for wider experimentation that might produce results that could challenge the general direction of policy’ (Ettelt et al., 2015, p. 301).
A final relevant political problem relates to pilots' ‘political sustainability’ and the already mentioned problem of scaling up their results. In the case of MIS, existing experiences of piloting suggest that governments can easily interrupt experiments if they were introduced by the opposition, if they conveyed a conflicting policy paradigm and/or diverted resources to opposing political constituencies (Jessoula et al., 2014; Jessoula & Natili, 2020; Natili, 2019; Sacchi & Bastagli, 2005).
Pilots as meta-implementation arrangements
According to Sager and Gofen (2022), implementation arrangements are the institutional settings and organizational structures through which implementation takes place and that shape the context within which actors behave, influencing, therefore, policy outputs. The political dynamics through which implementation arrangements are chosen are still under scrutinization (Casula, 2022; Steinebach, 2022). An exception is the article by Sager and Gofen (2022) that traces the variables influencing this choice back to institutionalism, multi-level governance, and performance management. The study of pilots can add further insights on the topic due to the increasing relevance this experimental approach has gained in policy-making processes. In fact, piloting offers the opportunity to analyze the political dynamics driving their design and the definition of the implementation arrangements to be adopted. The ‘politics of policy experiments’ (Nair & Howlett, 2016 p. 70) can enable or hamper the design and implementation of pilots. They can have an impact on established power equilibria among actors, on power relationships between central and local institutions, on enabling or hampering controversial policy reforms, on allocating resources, and on legislative changes (Checkland et al., 2021; Ettelt et al., 2022; Nair & Howlett, 2016). Most importantly, they can be used as ‘shadow institutions’—or alternative informal channels—to manage political conflicts over issues in a less visible and controversial environment than legislative arenas (Brodkin & Kaufman, 2000).
METHODOLOGY
Drawing on these premises, the present article analyses two pilots, the B-MINCOME of the Municipality of Barcelona and the Weten Wat Werkt of the Municipality of Utrecht, specifically aimed at testing MIS with different combinations of activation policies. The two case studies are part of a broader project called POLBIS-The Politics and Policy of Basic Income Schemes: Lessons from Italian and European Cases (Nesti et al. 2023). Both are based on Randomized Control Trials (RCT), that is, experiments aimed at testing their hypotheses through the creation of randomly assigned groups and a control group. B-MINCOME took place between 2017 and 2019 and Weten Wat Werkt between 2018 and 2019. It is important to note that, even if the two pilots are extremely similar in their design strategy, they are set apart by the two distinct socio-economic environments. The labor markets in Utrecht and Barcelona present distinct opportunities and challenges for job seekers. Both cities are a prominent economic center in their respective countries, but the job markets are different. Utrecht offers a robust job market with a particular focus on knowledge-based industries. Multinational corporations and research institutions are prevalent, providing a wide array of opportunities in fields such as IT, life sciences, and engineering. Barcelona, meanwhile, is marked by a dynamic job market that emphasizes tourism, creative industries, and technology startups. In terms of unemployment rates, the Netherlands, including Utrecht, generally records lower figures compared to Spain. The latter has historically faced higher unemployment rates, although these have been gradually improving. In conclusion, both Utrecht and Barcelona offer opportunities above the national average but with a noteworthy distance between the two. The in-depth analysis of the two case studies will shed light on the institutional context and the politics of experimentations and on the politics of implementation. We follow the guiding principles of Ashworth et al. (2019) to present our methods and focus on the rationale for qualitative choice, sampling frame, data collection, and analysis. Our research consists of an interpretive comparative study based on interviews (Tavory & Timmermans, 2014) and secondary data.
The choice of the adoption of this method together with a deductive approach stems from the fact that it allows the collection of a rich amount of data that contribute to providing an in-depth insight, interpretation, and comparative analysis of the role played by political actors and their strategies in shaping the design of the pilots, which is a key point for the purpose of this research.
The chosen cases represent a purposive theoretically-driven sample. The sampling frame satisfies two theory-driven criteria. On the one hand, we were looking for MIS pilots implemented at the local level following the Most Different Systems Design (MDSD): same output (perceived success), different context, and, partially, design. On the other hand, the pilots should have been at the center of a public debate and under closer scrutiny of the managing authority.
In order to study the two pilots, we used two complementary methods: secondary documentation analysis and semi-structured interviews. Desk research was carried out through the analysis of 64 documents (scholarly articles, institutional documentation, project-related documents, and other gray literature). We also conducted 16 semi-structured interviews with selected key informants that participated in the whole cycle of the development of the two projects, 10 from the B-MINCOME pilot and 6 from the Weten Wat Werkt pilot. The list of the key informants interviewed is given in Annex 1. The sample of interviewees was drawn following the non-probability sampling technique, that allowed us to reach the highest number possible of participants, and that is compatible with the maximum variation sampling strategy used to select the case studies.
To undertake the analysis of the semi-structured interviews, we utilized thematic analysis that provided us with the flexibility needed to interpret the data and to identify broader patterns and categories of meaning across the different interviews and projects. The thematic analysis helped us to understand the similarities and differences of the two projects in relation to the recurrent and fundamental themes in minimum income design. The process of data analysis started following the six-step approach developed by Braun and Clarke (2006). Firstly, we transcribed the interviews using the verbatim transcript technique using computer-assisted qualitative data management and analysis software, NVivo. Once an interview was transcribed, it was manually checked to verify the software did not commit an error. The second step consisted of coding the interviews using the same software. This process consisted of highlighting phrases or expressions that seemed important to the aim of the research. Finally, during the writing stage and the final revision of the paper, a final iteration of the data analysis was implemented.
DESCRIPTION OF CASE STUDIES
B-MINCOME
The pilot B-MINCOME was implemented between 2017 and 2019 by the Municipality of Barcelona and was co-financed by the local government and the European Union under the Urban Innovative Action (UIA) with 5 million and 12 million euros, respectively, for a total of 17 million euros (Colini, 2017; Laìn et al., 2019). The project was led by the Area for Social Rights, Global Justice, Feminism, and LGBTI Affairs of the Barcelona City Council and managed along with a consortium made up of several partners involved in the design and evaluation of the pilot: The Young Foundation, Novact (International Institute for Non-violent Action), Ivàlua (Catalan Institute of Public Policy Evaluation), the IGOP-UAB (Institute of Government and Public Policies at the Autonomous University of Barcelona), ICTA (Institute of Environmental Science and Technology of the Autonomous University of Barcelona), and DAMA (Data Management Group—UPC). The design and implementation of B-MINCOME, along with the activation policies, were carried out through collaborative efforts involving multiple municipal bodies, under the direction of the Social Rights Department. These bodies were: the Municipal Institute of Social Services (IMSS), the local development agency Barcelona Activa, the Direction of Community Action, the Barcelona Education Consortium, the Barcelona Municipal Housing Trust and the districts of Nou Barris, Sant Andreu and Sant Martí.
The main objective of B-MINCOME was to test what would be the most effective way to reduce inequalities, fight social exclusion, and avoid the reproduction of poverty through the combination of a passive policy, a guaranteed income named Support municipal d'inclusiò (SMI), active policies and programs for socio-occupational integration (Lucas Martín, 2018). The municipality wanted to test the introduction of the SMI alongside the existing Catalan PIRMI (Programa Interdepartamental de Renta Mínima de Reinserción—Interdepartmental Program of Minimum Insertion Income) and to assess which new strategy to combat poverty could be adopted alongside them (Colini, 2019).
The pilot involved one of the most economically deprived areas of Barcelona, Eix Besòs, and included ten neighborhoods (Ciutat Meridiana, Vallbona, Torre Baró, Roquetes, Trinitat Nova, Trinitat Vella, Baró de Viver, Bon Pastor, Verneda-La Pau, Besòs–Maresme) placed across three districts (Sant Andreu, Sant Martí and Nou Barris). The project was designed in 2016, implemented from October 2017 to October 2019, and evaluated from October 2019 to December 2019. Overall, it lasted 36 months.
To test the pilot's goals, 1.383 households were selected among voluntary applicants through a stratified random draw. Two groups were then randomly created: the control group consisted of 383 households, and the treatment group consisted of other 1.000 households. The latter received the SMI (463 euros) and was further divided into two groups, one only receiving the minimum income, and the other receiving the minimum income and forced to participate in an activation program. The activation program further divided the group into four subgroups, one for each active policy: training and employment, social entrepreneurship, aid for the refurbishment and rental of rooms, and participation in community networks. All participants were assigned to groups using a randomized control trial methodology.1
In a very concise way, the results highlight that the MIS introduced with the pilot contributed significantly to reduce beneficiaries' severe material deprivation, housing, and food insecurity and that it improved participants' financial and material well-being, especially with reference to the lives of children. Other positive impacts were an increased satisfaction with personal life and happiness, the strengthening of familiar and neighborhood relationships, and the reduction of individual isolation through the community participation policy. Lastly, the pilot helped improve the relationship between social workers and participants due to the development of ‘more horizontal and less assistance-based’ bonds and a general perception of increased proximity between the neighborhood and public administration (Riutort et al., 2021, p. 18). On the other hand, there was no strong evidence that the activation measures included in the pilot succeeded in increasing employment and the conditionalities did not significantly impact on beneficiaries. Rather, participation in active measures appeared to have a negative impact on beneficiaries' ability to find a job due to the time required to participate in them. Other limits of the pilot related to its design. As Hill-Dixon et al. (2019) reported, the way groups were created provoked disagreement among participants and many of them ‘felt that they had been randomly assigned to a program which did not align with their skills and interests, or that the training or work opportunity was not aligned with the real labour market’ (Hill-Dixon et al., 2019, p. 6). Furthermore, information on the ‘rules of engagement’ of the pilot was not clearly communicated to the participants, so confusion and misunderstanding and a sense of anxiety and uncertainty emerged among them.
Turning to the focus of our research, the design of the B-MINCOME project was shaped by several political choices. The first stemmed from party dynamics within the City Council. The introduction of a local economic support was part of the electoral program of Barcelona en Comù (BeC)—the party led by Ada Colau—during the 2015 municipal elections. Colau's agenda focused specifically on fighting inequalities and poverty, and the creation of an economic support for inclusion aimed at meeting the basic needs of people in poverty was one of the issues BeC promised to introduce. The original idea was to include the financial support in the list of municipal social services and to finance it with municipal resources. SMI was initially intended by a group of BeC councilors as a Universal Basic Income (UBI), that is ‘a periodic cash payment unconditionally delivered to all on an individual basis, without means test or work requirement’.2 At that time there was a huge debate in Spain about the need to introduce a UBI to combat poverty (Noguera, 2019). The SMI should have complemented the PIRMI, a minimum income scheme introduced in Catalonia in 1997. The Catalan Parliament was not satisfied by PIRMI (Fernández, 2020) and was reviewing it in 2017. Indeed, the PIRMI was replaced by Citizenship Guaranteed Income (Renda Garantida de Ciutadania, RGC) in July 2017. However, a municipal income was not introduced due to institutional and political barriers (Rincón García, 2022). The creation of a local monetary benefit would be outside the city's competences since Spanish municipalities are not allowed to introduce welfare cash transfers (see below). Moreover, although Colau won the election, BeC gained 11 seats in the City Council out of 41, and without a majority of votes, it was unable to create wide political support for the UBI initiative.
The B-MINCOME project, the way it was presented or had the favorable vote of all the parties, that is, we didn't have, we didn't have any… ehm no votes against it. What's more, in the way it was presented, we tried to present the project so that there were no opposing parties, because in the end what we were doing was approving different combinations of minimum income, right? With active inclusion policies, which in some way could reflect the different ideas that the different parties had about what a minimum income or a guaranteed income had to be in, in real operation, right? And in the end, it was a pilot funded by the European Union. Therefore, we did not have, we did not have political opposition to do the project. (Interview B-4)
Thus, the choice to transform the municipal UBI into a MIS subjected to activation measures financed by the EU was a more acceptable solution ideologically and economically for all the parties involved.
It was also somewhat the ambition of all ‘municipalist’ cities to deal with issues that are generally the responsibility of other administrations, even at a more supra-municipal scale, to deal with them at a local level. (Interview B-8)
There is a very complex issue in Spain and that is… who, which level of public institution is in charge of social policies. So, there is a multi-layer political problem in Spain. …The town councils, according to the Spanish Constitution and the Statute of Catalonia, which is like the Political Constitution of Catalonia, in theory it is the town councils that are in charge of social policies. But at the same time, it is also the autonomous communities, in this case the Government of Catalonia, that are responsible for social policies. (Interview B-4)
The coordination of the project, therefore, was extremely difficult not only for the high number of local actors involved and the numerous activities carried out but also due to the different institutional local competences that had to be carefully integrated in order to avoid deadlocks in each stage. On the contrary, the strategy pursued by the municipality did not clash with the central and regional governments' competencies. Rather, they positively welcomed the project as an opportunity to learn and gain more information about the function of MIS.
So the B-MINCOME has served precisely because it coincides in time with these two public aids, the Spanish and the Catalan. It has served to demonstrate that conditionality does not work, that withdrawal policies, limited policies, do not work either and that we have to tend towards the unconditional universal basic income. […] I think it is an important element to take into account to understand why the idea of the unconditional UBI is now much more famous than it was three years ago in Spain and in Catalonia in general. (Interview B-4)
Therefore, a ‘side effect’ of B-MINCOME was to increase the public debate on MIS and UBI in Spain just in a period (2021) when two new pilots, one at the national level to test a new MIS measure, the Ingreso Minimo Vital (Minimum Vital Income) and one at the regional level to test a UBI scheme, the Catalan Renta Bàsica Universal, were launched (Interview B-2; Rincón García, 2022). In addition, two experts involved in the B-MINCOME experimentation also participated in the other two pilots.
B-MINCOME ended in 2019 without policy consequences. The municipal income was not introduced and there was an unsuccessful attempt to extend the pilot for two years (B-MINCOME II).
Weten Wat Werkt
In 2015, the Dutch government issued the Participatory Act (PA), a social assistance law replacing the Work and Social Assistance Act of 2003, based on the principle that everyone should have the possibility to participate in society through a regular job (Hillamo, 2022). Under the PA, any person who legally resides in the Netherlands without sufficient means is entitled to receive a minimum income to support himself/herself. The benefit is regulated by strict conditionalities such as mandatory participation in activation programs to search for a job and to improve labor skills. Moreover, beneficiaries must perform social services to give back something in return for the assistance they receive—a workfare measure called Tegenprestatie or ‘service in return’. Therefore, the PA tries to prevent moral hazard by imposing conditionalities that incentivize beneficiaries to be reintegrated in the labor market and earn an income as soon as possible, and by monitoring and sanctioning compliance (Verlaat et al., 2020).
The implementation of the PA relies on a multi-level architecture where the central level determines economic benefits, while municipalities have a certain leeway to adopt activation programs. This approach led to differences throughout the territory, with some municipalities adopting strict control over conditionalities and others leaving more freedom to beneficiaries to choose their own activation pattern (Hillamo, 2022; van der Veen, 2019). The PA was, therefore, criticized for the unequal treatment of beneficiaries (Knijn & Hiah, 2019), for the administrative burdens imposed on them and on caseworkers, and for inflicting harsh conditions and sanctions (Verlaat & de Kruijk, 2019), thus creating ‘a climate of distrust between local government and its vulnerable citizens’ (van der Veen, 2019, p. 3). Moreover, the complex decentralized implementation and the obligations for municipalities to include workfare provisions in their legislation generated contrasting attitudes and criticism from local institutions toward the PA.
Remarkably, the PA contains an article on innovation (Art. 83), which allows municipalities designated by the Ministry to experiment with alternative solutions in order to find more effective applications of the law. This provision, from 2015 onwards, opened up the opportunity for Dutch municipalities to launch pilots in collaboration with universities (Verlaat & de Kruijk, 2019). The front-runners were the municipalities of Utrecht, Groningen, Tilburg, and Wageningen whose objective was to examine how eliminating job-seeking obligations and mandatory training, along with reducing the rate at which labor market earnings affect assistance grants, would impact upon the motivation to work. Furthermore, the research teams in the four cities proposed to investigate the consequences of these modifications not only in terms of the number of people transitioning to paid employment but also in terms of the health and well-being of beneficiaries, as well as of job satisfaction of caseworkers (van der Veen, 2019).
The Municipality of Utrecht organized the pilot Weten Wat Werkt in collaboration with the city of Zeist and the University of Utrecht in 2017 to investigate the best way to guide people on social assistance back to paid work or other forms of social participation (Verlaat et al., 2020). The pilot was co-funded by the European Social Fund and the Research Talent program of the Dutch Scientific Organization (NWO). In 2018 researchers began the recruitment of volunteers among the 11.110 recipients of social assistance. Among them, 752 persons were selected and randomly distributed into four different groups: one control group and three treatment groups. The latter were composed of people exposed to different activation policies: (1) those exempted from the requirement to actively search for and accept employment, (2) those with additional assistance and guidance from caseworkers, and (3) those losing less benefit as their earnings from employment increased.
The research team followed the four groups for sixteen months to examine the impact of the measure provided in each group on labor participation, social engagement, health and well-being, client satisfaction, and the financial status of beneficiaries. The results of the three groups were compared with those of the control group. The implementation phase lasted from June 2018 until October 2019, including extension for an additional three months to allow a linear transition of participants toward the return to the regular workfare approach (Verlaat et al., 2020).
Concerning the results of the pilot (Verlaat et al., 2020, 2021), all the three group interventions demonstrated that granting more autonomy to claimants leads to increasing labor market participation and to more chances of obtaining a permanent contract. Findings related to the second treatment group (extra support) were generally positive for people with low education and employability, but they were not statistically relevant for all participants. Finally, researchers did not find significant results in terms of increased social participation, health and well-being, client satisfaction, and financial situation of all participants in the pilots.
And many people were dissatisfied with the Participation Act. Politicians, but also people who are on social assistance and interest groups, because they were dissatisfied with the act being too much… too much governed by distrust, giving very little autonomy to the people on social assistance, being strict and heavy fines was, yes, the paradigm of distrust that they really dislike. […] So even before the Participation Act became official, as of January 2015, people were thinking, OK, we want something else. We want a social assistance and welfare paradigm one of trust and one where you give people opportunities and trust that it will take them. (Interview U-5)
That was possible because here, municipalities are responsible for that scheme. So, it's decentralized. The rules and regulations are national, but the execution of the scheme like paying the benefits, counselling people, monitoring people that receive the benefits that is done on the local level by the municipalities. So, they had kind of they felt the urge being responsible for this scheme to experiment with it. And the new act also provided room for that because it included an article that said you kind of experiment with the policy to improve it. (Interview U-1)
Well, the original thought about the whole research project was that different municipalities, they wanted to experiment with the basic income […] The national government didn't want it. So that's why we didn't get the possibility to do that […]. We're far too limited to experiment in that way. (Interview U-5)
And in time, these ideas of doing experiments come together and form that have boiled up to the Ministry of Social Affairs. And when it went and eventually came to the Ministry of Social Affairs, they say, well, okay, now we have so many different experiments. Now we have to regulate and we have to narrow down what we actually want to do. Yeah, and that is where the restriction on inventions came in place. (Interview U-4)
So, and it took a lot of discussion also between our municipality and the national government about how we would conduct research and what we had to do to be allowed to do the research project. […] I don't know if you heard about it, but part of the Participation Act is Tegenprestatie in Dutch. […] It is just compensation for them getting benefits, and politically there's not much support for that in Utrecht. So we don't do that. And our national government said, well, you don't have to exactly do it, but at least you have to mention it in your local laws. So that's what we did. (Interview U-5)
Certain people are forgotten in the Participation Act and the incentive for municipalities is this is very financial to help those people or you can get out of the Participation Act fast. You can get out of the benefits quickly. […] And we need more, we need more money and more attention and more possibilities to help those people. […] So many municipalities had financial problems due to changes in debt. And there is a tendency I think that municipalities are more critical towards the national government. There is an ongoing debate as to whether municipalities are adequately funded for the task that they have. (Interview U-5)
It went wrong already with the term universal basic income because […] here the universal basic income it's very polarizing. […] It's really difficult to compromise. So it started as a universal basic income experiment. But of course, again, that isn't when you see what we really did. But also the rhetoric has changed pretty quickly from a basic income experiment to an experiment with trust or an experiment with autonomy. (Interview U-5)
The pilot ended in 2019 and the results were published in the following months. Due to their inconsistencies, the national government did not change the PA and the workfare approach of the law (Roosma, 2022).
DISCUSSION AND CONCLUSIONS
The pilots analyzed in the previous paragraphs represent two interesting case studies of MIS policy experiments implemented at the local level. Both experimentations started to test innovative solutions to the problem of effective activation and control over excessive administrative burdens. They were, indeed, framed as an opportunity to carefully re-design MIS already in place and to find alternative implementation arrangements. These findings are quite consistent with literature on experimentation that views pilots as a strategy to pre-test solutions and to anticipate the potential consequences of policy implementation.
In policy terms, the results B-MINCOME and Weten Wat Werkt achieved were supportive: in Barcelona, beneficiaries' well-being was improved while in Utrecht labor participation increased. Notably the two cases were based on different assumptions but achieved the same results. B-MINCOME was aimed at finding the most effective and efficient strategy to tackle poverty while Weten Wat Werkt was more focused on finding an alternative way to implement social assistance based on trust in beneficiaries of welfare provisions.
Albeit some results of the pilots were disputable, they both demonstrated the ineffective impact of activation measures on increasing employability of beneficiaries of MIS. This was an important result since the usefulness of activation measures is quite controversial (Heidenreich et al., 2014) but strongly supported by the majority of parties.
Concerning the political side of piloting—the politics of pilots—the two experiences prove to be quite interesting, particularly in terms of interinstitutional relations. The two pilots were initiated by local governments to contest and/or compensate for national weaknesses thanks to the opportunity granted by social regulation and constitutional relations. Local ruling parties in opposition to national and regional governments promoted experimentations to demonstrate central inefficiencies, thus searching for electoral gains. But this strategy intersected with multi-level governance dynamics generating different results. In Barcelona, conflicts with the other institutional levels did not occur. Rather, experimentation became an opportunity to capitalize the local experience for other initiatives and to promote mutual learning. Nonetheless, multi-level governance affected the organization of the pilot and made coordination difficult. In the Dutch context, center-periphery relations were more complex and party dynamics led to strenuous negotiations between institutional actors. In the end, the central government was able to constrain local ambitions and to lock in the national policy.
Pilots were also strategically adopted to advance and to legitimize the debate on UBI in both countries. At the heart of both pilot proposals was the desire to simplify activation processes and to reduce administrative burdens on claimants. The two strategies followed similar patterns but achieved different results in this respect. Both cities were forced to reframe their UBI proposals in order to build a winning supporting coalition. Since UBI still represents a controversial issue, B-MINCOME and Weten Wat Werkt were turned into MIS experimentations. In any case, pilots achieved re-ignition of the debate on UBI, and in Catalonia, a new experimentation based on the UBI approach was subsequently launched. In the Netherlands, although the PA was not changed, Weten Wat Werkt contributed to raising a discussion on workfare and social assistance and to reinvigorating the debate on UBI.
In the Netherlands, pilots drew attention to the power of local policy makers. As stated by Roosma, ‘the support of local policy makers was not only crucial for the national Government's approval to the (official) experiments in the first place but also mattered in struggle for the framing and in the interpretation of the experiments’ (2022, p. 209). On the contrary, in Spain the pilot showed that ‘The local government does not have the regulatory competence to introduce non-emergency income support schemes. Aside from its limited legal powers, the local government is also subject to key gatekeepers of its functions and competence, preventing policy change and hindering alternative reforms (Rincón García, 2022, pp. 222–223).
Overall, our research highlights that pilots can be understood as meta-implementation arrangements, aimed at testing innovative courses of actions and interinstitutional relations. Following Sager and Gofen (2022) the pilots analyzed here were designed by multi-level institutional and regulatory dynamics that open opportunities for the local level to try to redistribute power from the center to the periphery. Additionally, pilots were used to reframe policy paradigms and to promote policy change at the normative level.
These findings shed new light on the role of local authorities in pilot projects. The limited research on the topic highlights that pilots can be used by central governments to ‘steer at distance’ local actors, that is, to control and to tame them (Ettelt et al., 2022). The Catalan and the Dutch pilots partially confirm this perspective, demonstrating that in multi-level institutional settings pilots can be adopted by municipalities to try to reframe policies against the central government, especially as a strategy of party opposition. Nevertheless, this does not imply that the municipalities completely succeed. Both Barcelona and the Dutch cities, in fact, were capable of initiating the pilots and testing their own policy solutions. But even if they challenged the dominant paradigm for the period of experimentation, the central government did not play a passive role in this process: it tried to weaken the position of local actors and to exploit its institutional powers to reverse experimentations, to prevent the scaling up of the pilot, and, ultimately, to re-confirm already existing policies. The nature of the policy, the control over the allocation of the resources to implement it and the distribution of policy competencies among institutional levels to design and implement it also play a crucial role in this dialectic relationship.
ACKNOWLEDGEMENTS
Authors are grateful to Jesper Labardi for his valuable research support, to the anonymous reviewers for their insightful critiques and suggestions, which have greatly improved the manuscript, and to the participants of the IWPP3 workshop at Corvinus University, Budapest, for their stimulating discussions and feedback.
FUNDING INFORMATION
Cariparo Foundation—Excellence Research Projects, Grant/Award Number ID52032.
CONFLICT OF INTEREST STATEMENT
No potential competing interest was reported by the authors.
Endnotes
ANNEX 1: LIST OF INTERVIEWS
B-1. Civil servant, IERMB & Municipality of Barcelona, 11/05/2021.
B-2. Project manager, UIA, 14/05/2021.
B-3. Civil servant (Coordinating activity), Municipality of Barcelona, 17/05/2021.
B-4. Senior researcher, IERMB & Municipality of Barcelona, 18/05/2021.
B-5. Researcher, IGOP Barcelona, 02/06/2021.
B-6. Researcher, The Young Foundation, 04/06/2021.
B-7. Civil Servant (Coordinating activity), Municipality of Barcelona, 09/06/2021.
B-8. Researcher, IGOP Barcelona, 09/06/2021.
B-9. Case worker, Municipality of Barcelona, 18/06/2021.
B-10. Case worker, Municipality of Barcelona, 18/06/2021.
U-1. Researcher, University of Utrecht, 31/03/2021.
U-2. Researcher, University of Utrecht, 19/04/2021.
U-3. Policy advisor, Municipality of Nijmegen, 21/05/2021.
U-4. Professor, University of Groningen, 03/06/2021.
U-5. Policy advisor, Municipality of Utrecht, 08/06/2021.
U-6. Policy advisor, Municipality of Wageningen, 09/06/2021.
Biographies
Giorgia Nesti is currently Associate Professor of Political Science at the University of Padova and co-coordinator of the Observatory on Urban Policies (URBES); she undertakes research on public policy and public administration, with a specific focus on innovation, digitalization, and their social impacts. Her recent research projects are on digital and just transition, minimum income, policy implementation and street-level bureaucracy.
Matilde Cittadini is currently ar doctoral student at the Institute of Government and Public Policy (IGOP) at the Universitat Autònoma de Barcelona. Her doctoral research focuses on the study of minimum income schemes, employing an intersectionality theory approach to explore the complexities and nuances of policy impacts across different social group. She started her research project during an internship with the POLBIS project, under the auspices of an agreement between the University of Padua and the Universidad de Barcelona (UB).
Matteo Bassoli is currently Associate Professor of Political Science at the University of Padova and co-coordinator of the Observatory on Urban Policies (URBES); he undertakes research on public policy and social inclusion, with a specific focus on the functioning of private-public partnerships and their network structure. His recent research projects are on street level bureaucracy, minimum income, gambling and migration policy.
Open Research
DATA AVAILABILITY STATEMENT
The data that support the findings of this study is not available in accordance with GDPR and the sensitive content.
