Parading Utopia on the road to nowhere? An introduction to the special issue on the policy impact of the European basic income experiments

Basic income experiments have emerged across Europe in recent years, but until now analysis has focused on their design and the scientific interpretation of their results, rather than the subsequent policy impact of these projects. This special issue addresses this gap. The papers all focus on whether and how the European basic income experiments have made an observable impact on the basic income debate and social security reform more generally. The special issue includes country case studies of the three countries in Europe that have completed their experiments, Finland, the Netherlands and Spain, as well as a case study of Scotland, where a feasibility study did not result in a field experiment, and of Ireland, which is in the process of planning at least one experiment. Two papers then also examine the effect of these experiments on the debate at EU level and outside Europe, in Australia. The special issue provides a novel contribution that advances both the scholarly and policy debates surrounding basic income at a time when COVID-19 appears to have increased interest in the policy and equally seems to have propelled the idea of experimenting with basic income even further into the mainstream.

It is common in historical accounts of basic income advocacy to reach for epidemiological metaphors when describing the rise and fall of interest in the idea of a basic income. 2015 marked a year of 'exponential' growth in interest and support for basic income and signalled the beginning of a new 'wave' of mobilisation (Campa, 2019;Widerquist, 2018). Two of the most notable triggers for the sudden rise in (social) media and political interest were the steps taken in Finland and the Netherlands to prepare for basic income experiments in their respective countries. Despite notable differences in their design and context, both projects captured the public imagination around Europe and news of cutting-edge experimentation spread across media outlets. Yet, by 2019, whether due to poor public relations management or more objective policy failures, the notion that these experiments had failed or fizzled out was widespread in media reporting (e.g. Bendix, 2019).
Interestingly though, rather than the white heat of the moment precipitating an equally rapid descent in interest across Europe as this wave peaked, an apparent conveyor belt of basic income experiments and projects has emerged with no signs of abating. In 2020, the Irish government committed to a basic income pilot in its Programme for Government and recently announced a Basic Income for the Arts pilot scheme 1 , while the Welsh government has just started up a basic income pilot of its own, targeted at care leavers. 2 This follows a feasibility study carried out jointly by the Scottish government and four local authorities in Scotland, as well as the conclusion of another experiment, B-Mincome, conducted by Barcelona city authorities with financial support from the European Union. Outside of Europe, projects in South Korea, Brazil, Canada and across US cities further strengthen the notion of a constant churn of basic income experimentation and policy interest, as opposed to a fading away of the energy associated with the initial experiments in Finland and the Netherlands. For better or for worse, it appears that basic income trials are here to stay. This in turn raises an interesting set of questions for policy researchers. Have the European basic income experiments had any impact on social security policy development? What factors have affected the relative success or failure of the various experiments? And what lessons can we apply to future basic income experiments and basic income policy development as well as to social security reform in general? As it stands, these questions are not yet being seriously addressed or even raisedin academic and policy debates surrounding basic income experimentation. Ongoing discussion largely focuses on questioning the role of basic income experiments, their optimal design and implementation and the 'correct' scientific interpretation of the results (e.g. Castro and West, 2022;De Wispelaere et al., 2019;Kangas et al., 2021;Laín, 2022;Merrill et al., 2021;Widerquist, 2018). A systematic discussion of subsequent policy impact is missing.
This special issue is focused on addressing this gap by answering the specific questions about policy impact mentioned before. The different contributions all focus on either an individual country case or the European Union as a whole, and explore whether and how the basic income experiments have made an observable impact on the basic income debate and the ongoing debate around social security reform more generally. The special issue includes country case studies of the three countries in Europe that have completed their experiments: Finland, the Netherlands and Spain. It also includes a case study of the feasibility study in Scotland, which did not result in a field experiment, and a case study of Ireland, which is in the process of planning at least one experiment. The motivation for including countries where experiments have not (yet) taken place is that we surmise that even serious political discussion of basic income experimentation may have an impact on policy development. In addition, it is interesting to explore what lessons the later cases may have learned from the earlier ones. These European country-level cases are complemented by a paper examining the effects of European basic income experiments on social security debates at the EU level, where basic income has also entered policy and political debates. Finally, the special issue includes a paper on how the Finnish experiment influenced policy development in Australia, as a case study of policy impact through global policy transfer. We discuss each paper in more detail in the final section of this introduction.
Overall, these papers offer a novel contribution to the growing scholarly basic income debate, through an innovative assessment of the role that experiments and pilots can play and their value in basic income policy development, an area that so far remains underdeveloped. We believe this contribution to be important for advancing both the scholarly and policy debates at a time when COVID-19 appears to have cast basic income even more to the centre of social security debates, and also seems to have propelled the idea of experimenting with basic income even further into the mainstream.

(Re)thinking the policy impact of basic income experiments
There were many motivations for this special issue on the policy impact of basic income experiments. Crucially, recent academic publications and policy reports have all focused on the empirical results of the three completed experiments (Kangas et al., 2021;Muffels, 2021), with little attention devoted to either examining the political and policy factors leading up to the experiments or systematically considering their policy impact on the ongoing social security debates in each of these countries. Meanwhile, at the EU level the basic income debate has been growing, in part inspired by media reports or political discussions of the national experiments, but this too has not been systematically examined.
However, two other important substantive factors motivated the desire to offer a different perspective. The first was a wish to contest the notion that these experiments are inherently scientific endeavours motivated solely by the desire to find out if basic income 'works' and evaluate the evidence. Given that these experiments are initiated, funded and in many cases run by governments, agencies and parties with electoral and ideological goals, we suggest that they should be treated as political events with accompanying analysis of the achievement of policy goals, rather than only examining the 'neutral' effects of the intervention on labour market participation, health and so on. Basic income experiments, like most large-scale social experiments, operate at what might be called the 'science-policy interface', and this means they need to be assessed from both a scientific and a policy perspective (Rogers-Dillon, 2004).
The second factor was a narrative of comprehensive failure that has surrounded the experiments, which we contend is primarily due to either overinflated expectations or insufficient nuance about the outcomes. It is certainly the case that none of the experiments have led to the implementation of a universal basic income. Given some of the media coverage of the experiments that implied this was nigh, some observers may deem the outcomes a disappointment. Yet, the experiments were never likely to result in the full implementation of a basic incomeat least not in the short run and so their impact should be viewed with more realistic expectations. This in turn requires the development of a framework that allows us to conceptualise and measure variability and nuance in the policy impact of basic income experiments. 3 What do we mean by the policy impact of basic income experiments and by what metrics should we judge this impact? We propose an adaptation of the work of Marsh and McConnell (2010) and Checkland et al. (2021), who identify multiple dimensions for assessing the success of policy or pilots, such as the 'process' of policymaking, the 'programmatic' dimensions of policymaking and the 'politics' of policymaking. We similarly locate three dimensions of policy impact that can be matched to three key 'good-faith' reasons why policymakers would seek to initiate a basic income experiment and therefore can be used to ex post evaluate its impact. The first, which we liken to 'process', is to produce and disseminate high-quality scientific or procedural knowledge about the effects and operation of basic income-like schemes. The second broad goal, which we label 'programmes', is the desire for the experiment to lead to the implementation of basic income-like policy, whether through amending existing or initiating new programmes. Finally, the third goal of basic income advocates, which falls into the category of 'politics', would be for the experiment to increase support among the public, politicians and stakeholders and provide an avenue for coalition-building around basic income policy development.
Acknowledging the multi-dimensionality of policy impact in this way helps us to recognise varying patterns of policy impact across the basic income experiments. While the Scottish feasibility study clearly achieved little to no impact in terms of process due to the failure to launch an experiment, the project did have a considerable impact on political support for basic income and similar schemes, and arguably also sat alongside programmatic change that reduced sanctions in devolved benefits and thus moved social security in a more basic income-like direction. On the other hand, the Finnish experiment arguably had significant success in terms of process, as it produced unique evidence on the policy intervention examined and disseminated it widely, but it appeared to have had a dampening effect on political support for basic income within the country and coincided with measures to increase sanctions for individuals receiving unemployment benefits. Thus, we contend that there is not singular success or failure but different experiences across countries on different dimensions of policy impact. Our heuristic device for assessing the policy impact of basic income experiments must explicitly acknowledge both multi-dimensionality and variation across country cases.
The case studies also help to inform why the basic income experiments have had the (lack of) impact we observe. In so doing, we can start identifying distinctive drivers that are useful for structuring our understanding of why experiments led to specific policy outcomes. It helps to separate out three distinct 'layers', each comprising a cluster of distinctive drivers that interact with each other in predictably complex ways. First, the details of experimental design differed considerably across the experiments and must be examined as an important determinant of what happened afterwards in terms of process, programme or policy impact. Factors such as the scale and ambition of the experimental project, whether it was set up as an evaluative pilot or a 'mechanism' experiment of an intervention that was not strictly feasible for that level of government, or whether the experiment was explicitly framed as a basic income experiment, are likely to have an important influence on subsequent policy impact.
Second, there are political or actor-centred factors that may matter considerably. These include factors such as whether the experiment was initiated by national, regional or local governments (top-down vs. bottom-up), whether it was primarily driven by a single political individual or a coalition of actors, whether there was an existing basic income community within the country and the existence of veto players such as strong unions, civil service and coalition partners. Finally, institutional and structural factors have an importantperhaps the most important --bearing on the policy impact of the experiments. As is familiar from the comparative politics of social security and welfare states, existing social security institutions determine the possible reform trajectories after the experiment has finished. Equally, political institutions, such as the electoral system, federal structure or the role of social partners, influence the actors which are involved and which can wield decision-making power. Two main challenges in exploring the drivers underlying basic income experiments are to disentangle the interaction of individual factors determining policy impact and to specify the relationship between the layers of drivers, and how this in turn determines the manoeuvrability for basic income advocates, enabling them to boost the policy impact of experiments and pilots. This is in addition to common problems associated with social science research, such as the small-N problem and identifying the causal impact of experiments, given the non-randomness of countries that instigate these trials.

The cases
The Finnish experiment, which was the first to get underway in January 2017, is perhaps the best known and has been an inspiration for many other experiments across the world. The Finnish project still stands out as the world's only nation-wide RCT of a basic income scheme, albeit one only targeted at those receiving unemployment benefits, and for that reason is widely regarded as the leading example in the growing field of basic income experimentation. However, its impact domestically has been far less impressive than its international reputation might suggest. In his article, Heikki Hiilamo describes the experiment as a 'missed opportunity' and offers a pessimistic analysis of the impact the experiment has had on policy debates and especially on the prospects of implementing basic income in Finland. He identifies four distinct 'denial' moments where policymakers failed to act on the evidence from the basic income experiment, which show that the direction of travel in Finland is not towards the implementation of a basic income. Hiilamo suggests that 'cheap support' from political parties, the prior existence of generous and relatively unconditional social security and an ongoing focus on activation offer the best explanations for why the experiment fell flat. The future of basic income in Finland remains uncertain although, as Hiilamo notes, a (small) window of opportunity may occur again in 2023, when a recently appointed Parliamentary Social Security Committee publishes its report on alternative models for social security.
Unlike the linear development of the highly centralised national Finnish experiment, the trajectory of the multitude of small-scale municipal Dutch basic income experiments was bumpy and faced many obstacles. Also distinct from the Finnish experience, and key to understanding the politics underlying the Dutch experiments, is the continuous political battle concerning how to view the experiments and how to interpret their results. Femke Roosma, in her article, deftly charts the complexity of assessing the political landscape surrounding the basic income experiments in the Netherlands. Many municipalities adopted a deliberate strategy of not labelling these experiments as having anything to do with basic income, instead preferring to frame them as 'trust experiments' or 'experiments with low regulation'. The experiments proceeded under a complicated regulatory frameworkor rather under two different sets of regulationsand, as Roosma outlines, even after evaluation, the battle over how to present and interpret the results did not abate. In terms of policy impact, Roosma argues that while on the surface the experiments did not seem to inform the immediate policy debate, several indirect political and policy effects can be credited to them. A recent coalition agreement of all governing parties includes a point on allowing social assistance beneficiaries to keep a larger share of their wages when working part time. Meanwhile, since the reporting of the experiments' findings, a number of progressive parties have adopted more radical basic income proposals as part of their election manifestos. For Roosma, the future of basic income in the Netherlands is perhaps not quite rosy but certainly not gloomy either.
In her article on B-Mincome, the third case of a completed basic income experiment in Europe, Leire Rincón claims that the basic income experiment undertaken by the Barcelona city council in 2017-2019 had a peculiar political boomerang effect. B-Mincome, perhaps more so than either the Finnish or Dutch experiments, was very much driven by a coalition of basic income advocates, many from within the city council. The experiment has become rather (in)famous for adopting a very complicated design that in addition to the 'standard' basic income dimensions also ended up testing various forms of activation and even the use of local currency. Rincón argues that several political, legal and institutional barriers first shaped the complicated design of the B-Mincome experiment and, second, prevented subsequent policy and political impact on the area of most interest to basic income proponentscash transfers. However, Rincón does list a number of more nuanced policy impacts at the municipal level, including improvement of social community policies, which she thinks can be attributed to the experiment. Rincón's article also points out one of the key challenges when assessing the policy impact of basic income experiments: the existence of non-experiment factors as mediating factors or alternative drivers or barriers to policy development. Rincón discusses how both COVID-19 and the introduction of Spain's first nation-wide minimum income policy (Ingreso Minimo Vital) at around the same time as the experiment complicate any assessment of the impact of B-Mincome considerably. One of the main impacts of B-Mincome may end up being that it sparked a new, much larger and less convoluted basic income pilot in the Catalan region, which is in itself an interesting type of policy impact that in turn may prove consequential for the future of basic income in Catalunya or even Spain.
The Scottish case is fascinating as an example of how a great deal of preparatory work and collaboration over numerous years, culminating in the release of the final report of the feasibility study in June 2020, can still have policy impact even when the field experiment itself is effectively aborted. 4 Sara Cantillon and Francis O'Toole take up the challenge of recounting the events that led up to the commitment of the Scottish Government to support four local authorities (Fife, North Ayrshire, Glasgow and Edinburgh) in their collaboration to examine the feasibility of a basic income pilot in Scotland. The paper outlines the unique and evolving devolution settlement in Scotland as an important context for the impetus to experiment with social security. Cantillon and O'Toole helpfully unpack the research produced during the feasibility study, which raised many issues and concerns about basic income, as well as providing a case for the experiment to go ahead. Crucially, the legislative and administrative limits facing the Scottish Government meant that it was not possible to proceed with its initial plans without the buy-in of the UK Government. Nevertheless, the final report outlined its proposal for an extremely ambitious pilot of a high-level basic income (£213.59 a week) for at least 2500 individuals in one area and a lowlevel basic income (£57.90 a week) for at least 14,600 individuals in another area. Given the Conservative government's hostility to basic income in general (whatever the level), their lack of cooperation was unsurprising and no subsequent steps were taken to progress with the proposals.
Was it then just a waste of time? Cantillon and O'Toole argue not: as a process of clarification and learning about the policy options, the feasibility study served an important purpose and has built a coalition of political actors and stakeholders keen on working towards a new unconditional minimum income guarantee for Scotland.
The special issue then moves from the cases where projects have concluded to cases where discussions are ongoing or projects are in development. The motivation for including each of the remaining cases is to explore the policy impact of experiments on basic income development in outside jurisdictions. First in line is Ireland, where at least one pilot scheme is due to begin this year. As in Finland and the Netherlands, political interest in basic income has a much longer history in Ireland than in most other European countries. In her case study, Helen Johnston places the Irish Government's commitment to a basic income pilot in a context of over 30 years of government commissions, a Green Paper on basic income in 2002 and campaigns from civil society, notably from Social Justice Ireland. The interest has ebbed and flowed, but in 2020 the formation of a coalition government that included both the Green Party, a longstanding supporter, and Fianna Fáil, which cited a commitment to investigate basic income in its election manifesto, led to the commitment to a universal basic income pilot 'in the lifetime of the Government'. Interestingly, Johnston notes the two-track process by which events have unfolded, with a Basic Income for the Arts launched in 2022 while the Low Pay Commission is simultaneously exploring a more universal pilot. However, although the full details of the government's plans for basic income are not yet apparent, it is an opportune moment to take stock of how things have reached this point as well as being aware of how this brazen twin-track approach may both offer opportunities as well as present unique challenges for basic income policy development, going forward. The Irish case also serves as a striking exception to the trials that have mushroomed since 2015, as it is the only example, since Finland, of a national government driving the process. 5 Dominic Afscharian, Viktoriia Muliavka, Marius Ostrowski and Lukáš Siegel provide a very different perspective by examining debates about basic income within the European Union. They provide their own conceptual framework of basic income trials as a buffer against the risks for political actors that wish to capitalise on the popularity of basic income in the abstract without suffering the political backlash that may occur if trying to implement policy outright. Afscharian et al. use contributions from the Conference on the Future of Europe, election manifestos from European elections and European Parliament debates from 2009 to identify the tendency for political actors to propose trials rather than policies, and offer their framework as an explanation for why this is likely to be the case. Moving from the empirical to the normative, Afscharian et al. also make a case for basic income trials as a tool to influence policy agendas and potentially break the stalemate that currently exists in the attempt to construct a 'Social Europe'. While this paper is inevitably firmly grounded in the specific political, legal and institutional environment of EU social policy, it affirms a central message of this special issue, namely that basic income experiments can indeed have policy impact beyond the implementation of a full-scale universal basic income and that this impact needs to be acknowledged if we want to better understand basic income as part of the policy process. In the last paper, we leave the continent of Europe and head down under to Australia, where Troy Henderson and Ben Spies-Butcher offer reflections on how the European basic income experiments were received there. By tracing the historical tendency of the Australian Left to identify with Scandinavian social democracy, Henderson and Spies-Butcher argue that this 'Nordic Connection' explains why the Finnish experiment had such a prominent effect in raising the salience of basic income in Australian policy debates. By contrast, the other planned, ongoing or completed experiments in Europe seem to have been largely ignored in the Australian debate. The authors chart the ties and scholarly networks between Australia and Finland and provide a convincing case for the existence of policy transfer and the diffusion of ideas. However, the effects should not be overstated, as there is still not widespread support among policymakers in the country. Attempts to push for a basic income trial in Australia have remained confined to the margins of political debate.

Conclusion
This special issue has three distinctive but related aims. In the first place we aim to broker a new research agenda into a long-neglected and long-overdue aspect of basic income experiments: their subsequent consequences on basic income and further social security policy development. As outlined in this introduction, we believe this neglect is rooted in either a misunderstanding/misrepresentation of basic income experiments as 'pure' scientific ventures or else in an exceedingly implausible view of how the policy process works. The former fail to appreciate the inherent political nature of large-scale social experiments, while the latter fail to understand the slow, gradual and stepwise nature of policy design and implementation. The different contributions in this special issue are tasked with identifying some of the gaps in these respects in current basic income research and with starting to fill them.
Second, this special issue illustrates through a variety of case studies just how variegated and complicated assessing policy impact of basic income experiments really is. Policy impact turns out to be multi-dimensional as well as multi-layered, with different experiments adopting different trajectories in terms of when, how or indeed what precise effects they have on basic income policy development. Qua case studies, basic income experiments often appear rather idiosyncratic, with each experiment apparently pursuing its own objectives, guided by distinctive designs and adopting implementation modalities catering for the specifics of each case. On the surface, the basic income experiment landscape soon starts to look like a patchwork quilt from a bygone era. But peeling away this first layer we rapidly begin to encounter a series of factors that directly or indirectly determine the prospects of conducting an experiment, its particular design and implementation modalities, but also its likely subsequent impact. It is here we can start to identify certain factors, in an attempt to explain both similarities as well as differences across very distinct cases. Interestingly, these drivers are familiar enough to social security and welfare state scholars, and as such one of our aims is also to bring this aspect of basic income research firmly into the fold of 'mainstream' comparative social policy.
Third, we contend that in order to pursue this novel research agenda, we need to start developing a framework for the policy impact of basic income experiments, a heuristic device that allows us to conceptualise the multiple dimensions of impact ('process', 'programmes' and 'politics'), the range of impacts to be observed along each of these dimensions, and the variability in what we call 'impact profiles'. Such a framework needs to be built in close collaboration with empirical case study research, both in-depth exploration of single case studies and comparative research of the variation in underlying determinants of the available cases. This is a challenging research agendanot least because cases (especially in Europe) are few and the time frame is too short to establish any firm conclusion. It is, however, work that, in our view, needs to be undertaken now while the policy interest in basic income experimentation remains high and the political will exists to improve the design of trials.