Abstract
The period since 2008 has seen high-profile debates over radical policy ideas such as universal basic income. As Jurgen De Wispelaere has noted, basic income has attracted interest from a wide range of political actors, but parties have struggled to sustain a firm commitment to the idea, especially as they have moved closer to power. This article tests and refines De Wispelaere’s analysis of ‘cheap political support’ for basic income through a case study of the UK Liberal Democrats, who have twice endorsed the idea (in 1990 and 2020) but subsequently dropped it. Detailed analysis of the party’s decision-making under Paddy Ashdown and Sir Ed Davey highlights the importance of party leaders, changing strategic goals, and substantive ‘policy reasoning’ by working groups. The article shows that the Liberal Democrats’ retreats from basic income were driven not only by electoral calculations, but also by critical scrutiny of specific policy proposals.
Keywords
One of the defining features of western politics since the 2008 financial crisis has been the renewed appetite for ‘big ideas’. Alongside proposals for a ‘Green New Deal’ (Green, 2024) and the campaign for ‘Medicare for All’ in the United States (Oberlander, 2023), this trend has been typified by the global wave of interest in Universal Basic Income (UBI). Basic income has drawn strong normative support from political philosophers such as Philippe Van Parijs (1995), who see it as a way of achieving ‘real freedom for all’: eliminating poverty in a way that gives citizens maximum freedom to make choices about work and consumption (Bidadanure, 2019). In the face of widespread disillusionment with mainstream politics, radical ideas such as UBI serve as mobilising demands which allow politicians to build new coalitions of support around alternative futures (Béland and Cox, 2016; Sculos, 2019). Yet parties often find it difficult to sustain support for these ideas, either in opposition or when moving into government. Partly for this reason, most developed countries have only implemented UBI in highly attenuated forms – for instance, through pilot schemes and temporary pandemic-era cash transfers (De Wispelaere, Chrisp, and Morales, 2025).
Political scientists have long seen parties as a crucial part of the ‘chain of democratic linkage’ between voters and public policy (Dalton, Farrell, and McAllister, 2011). In parliamentary systems, in particular, the prospects for implementing radical ideas depend heavily on whether parties include them in their manifestos, win electoral support, and prioritise them in the formation of governing programmes (Schmidt, 1996; Zohlnhöfer, 2009). Despite a vast literature on party strategy, however, the ways in which parties view ideas such as UBI are not well understood, since most of the comparative work on party policy change focusses on left-right shifts in parties’ spatial positions (Böhmelt et al., 2016; Fagerholm, 2016) rather than the adoption or rejection of specific proposals. ‘Big ideas’ such as UBI have powerful symbolic resonances and far-reaching implications for state activity, which are unlikely to be fully captured by the Manifesto Project dataset that underpins much existing work (Gemenis, 2013). Understanding why parties support or reject these ideas thus requires us to open up the ‘black box’ of intra-party debate (Wenzelburger and Zohlnhöfer, 2021) and examine ‘the processes of policy reasoning through which political parties adopt and change policy’ (Jacobs and Hindmoor, 2024: 2).
This article seeks to advance our understanding of radical policy innovation through a case study of the UK Liberal Democrats’ engagement with UBI. Since the party’s creation in 1988, the Liberal Democrats have twice endorsed basic income as party policy (in 1990 and 2020) before dropping the idea (in 1994 and 2023) in favour of more conventional social policies. On the face of it, the Liberal Democrats’ behaviour fits closely with the phenomenon of ‘cheap political support’ identified by Jurgen De Wispelaere (2016), in which relatively marginal political actors take up UBI ‘without either the commitment or the capacity. . . to build a sustainable coalition’ around it, and then become ‘reluctant to spend political capital on a policy that remains highly divisive’ (De Wispelaere, 2016: 132–133, 134). A detailed analysis of the Liberal Democrats’ policy reasoning allows us to test De Wispelaere’s thesis that parties’ support for UBI is often ‘cheap’ (and thus of little value to UBI campaigners), to examine why the party retreated from the idea, and to specify the causal mechanisms involved more precisely.
This case study contributes to our understanding of the politics of ‘big ideas’ in three main ways. First, it provides confirmatory support for De Wispelaere’s thesis that parties are most commonly drawn to UBI at moments of political weakness. The Liberal Democrats’ engagement with UBI in 1988–94 and 2015–23 suggests two reasons why this might be the case: because electoral failure opens up a ‘policy window’ within the party’s internal processes, and because party elites see big ideas as a way of achieving greater visibility and distinctiveness. Once the party’s future prospects seem more secure, concerns about the risks of radical policy commitments are likely to become more salient.
Second, the case highlights the importance of policy reasoning to party policy development. As Michael Jacobs and Andrew Hindmoor (2024: 13) have argued, parties engage in ‘substantive debates about the merits of alternative policies and the likelihood of their being successful in achieving the party’s goals’, which intersect with (but are conceptually distinct from) judgements about electoral strategy. In both the early 1990s and the early 2020s, we can see Liberal Democrat working groups ‘puzzling through’ the practical implications of UBI and comparing it against other policy options. The evidence of these working groups’ discussions echoes findings from public opinion research that ‘high support for the abstract idea of a basic income is fragile and susceptible to “wilting”’ once concrete decisions about policy design come into focus (Chrisp, Pulkka, and Rincón Garcia, 2020: 224). This has important implications for our understanding of UBI debates more generally.
Third, the case reminds us that the ways in which parties evaluate policy options depend on the analytical frames adopted, on parties’ different (and sometimes changing) strategic goals, and on institutional incentives. As we shall see, the Liberal Democrats’ decisions to drop UBI in 1994 and 2023 partly reflected the way in which the practice of manifesto ‘costings’ focussed attention on the tax cost of UBI and the ‘opportunity cost’ of drawing resources away from other spending commitments. This caution about tax rises was shaped in turn by Paddy Ashdown and Sir Ed Davey’s focus on winning over ‘target voters in target seats’, particularly in rural and suburban areas which traditionally elected Conservative MPs (Cutts, Russell, and Townsley, 2023). The high salience of costings and the challenges which the ‘first-past-the-post’ system poses for smaller parties are distinctive features of the UK political landscape. UBI might be a more politically sustainable policy option for parties which operate in less tax-sensitive political cultures, or under proportional representation (PR).
The article is structured as follows. The first section draws on John Kingdon’s influential multiple streams framework to identify the conditions that must be met for a political party to support UBI, outlines some theoretical expectations for how we might expect parties to evaluate the idea, and considers how these expectations might apply to a party like the Liberal Democrats. The second section explains the research methods. The following four sections trace the Liberal Democrats’ engagement with UBI in the two periods studied, working backwards from the key decisions to adopt or abandon the idea to identify the relevant actors, institutions, and motivations. A final section discusses the research findings and draws some conclusions.
Political parties and UBI
Although the idea of a UBI can be traced back more than a century, it has received a surge of support in recent years in response to concerns about precarity, inequality, and the impact of automation (Sloman, 2018). High-profile pilots and an international network of campaign groups helped to push the idea up the global agenda during the 2010s (Caputo and Liu, 2020), and the COVID-19 pandemic appeared to open a ‘policy window’ for universal cash transfers in developed countries such as the United Kingdom (De Wispelaere, Chrisp, and Morales, 2025: 167). Political science research on UBI has expanded in parallel. In particular, Jurgen De Wispelaere has pioneered a growing body of work on ‘the conditions under which basic income proposals enter the policy process’ and the barriers to UBI implementation (De Wispelaere, 2016: 627; see also De Wispelaere and Noguera, 2012). Several scholars have also used survey data to examine public opinion on UBI and identify the social coalitions which might form around different schemes (Chrisp, Pulkka, and Rincón Garcia 2020; Laenen, 2023; Roosma and van Oorschot, 2020; Vlandas, 2021).
Economists and political philosophers have long used basic income as a teaching aid: a device for thinking through questions about redistribution (Atkinson, 1995) and the rights and obligations of citizenship (White, 2003). Analysis of the practical politics of UBI is complicated by the fact that the term can cover a wide variety of policy designs. In particular, the Basic Income Earth Network’s standard definition – ‘a periodic cash payment unconditionally delivered to all on an individual basis, without means-test or work requirement’ – rules out income-tested payments (such as a negative income tax), but says nothing about the level of the basic income or the way it is funded (Basic Income Earth Network, 2025). Nevertheless, UBI proposals generally share four characteristics which mark the idea out from other social policies:
1. UBI involves recurring cash payments from government to individuals, and thus represents a permanent addition to current spending.
2. UBI is non-targeted, since it creates a universal benefit entitlement (perhaps with age variations). Parties which want to shape the pattern of winners and losers in a detailed way must do so by adjusting the tax system or other policies.
3. UBI is likely to be relatively expensive, even if interactions with other policies mean that the net cost is lower than the gross cost.
4. UBI would involve significant change to the tax and benefit systems, with potentially disruptive implications for people’s incomes and work incentives.
It is not difficult to understand why UBI might be attractive to politicians. In addition to its substantive appeal, support for the idea sends out a powerful signal about a party or candidate’s commitment to poverty relief, social rights, and personal freedom; indeed, the very cost and ambition of the idea may amplify this signal (Clark and Lee, 2016). At the same time, putting UBI in a party manifesto runs the risk of provoking intense normative objections and triggering the psychological tendency towards risk aversion (Jordan, Ferguson, and Haglin, 2022; Kahneman, 2011). Proposals for UBI trials can be seen as a way of squaring the circle by allowing parties to reap some of the benefits of support without paying the political costs of a firm commitment.
Partisan interest in UBI has ebbed and flowed over the years. When the modern European basic income movement took shape in the 1980s and 1990s, support tended to be strongest among environmentalists (such as GroenLinks in the Netherlands) and some left-liberals, and weakest among orthodox social democrats and conservatives, who were more firmly committed to traditional productivist norms around work and welfare (Van Parijs and Vanderborght, 2017: 189–203). As UBI gained a higher profile during the 2010s, it won backing from a wider range of sources, including radical left-populist groups such as Podemos in Spain (Noguera, 2019), pragmatic liberals such as the Centre Party in Finland (Perkiö, 2020), and senior figures in mainstream left parties, such as Benoît Hamon and John McDonnell. In the UK, the Green Party and the Scottish National Party have been particularly vocal supporters (Cantillon and O’Toole, 2022). However, parties which have endorsed UBI have often struggled to maintain their commitment to it, and basic income trials (most notably in Finland in 2017–18) have not been followed by a wider roll-out (De Wispelaere, Doussard, and Chrisp, 2025).
Writing in the early stages of the most recent wave, De Wispelaere (2016) pointed out that public support for UBI was concentrated among individuals and groups that were ‘marginally positioned in terms of their ability to influence [government] policy’. For these actors, support was generally ‘cheap’ in the sense that ‘there are few political costs associated with supporting BI’ when ‘one is never put in the position of having to defend one’s support against a sceptical – at times even hostile – political base’ (De Wispelaere, 2016: 133). De Wispelaere suggested that such parties were likely to drop or water down their commitment to UBI as they moved closer to ‘policy responsibility’, either as a ‘Downsian electoral move’ or because the ‘opportunity cost’ of trying to introduce a basic income became ‘too high to be politically sustainable’ (De Wispelaere, 2016: 133–134). He also noted that the apparent breadth of political support for UBI partly reflected a tendency to ‘hide policy detail’ under a ‘veil of vagueness’ (De Wispelaere, 2016: 136). Once decisions about the level and financing of universal payments came into view, coalitions formed around the principle were liable to disintegrate.
De Wispelaere’s analysis of the problems of cheap support is an intuitively plausible one, sketched out by reference to Podemos’ experience with UBI in Spain in 2014–15 and brief comments on other examples. However, not all of the causal mechanisms are clearly specified, and with the partial exception of a fuller account of Podemos’ trajectory by José Noguera (2019), De Wispelaere’s argument has not been tested in detail.
Drawing on John Kingdon’s influential multiple streams framework (Kingdon, 1984), a more formal model of parties’ engagement with UBI might be structured as follows. The adoption of UBI as party policy requires four conditions to be met: (1) awareness of basic income as a potential solution to a policy problem, (2) some form of ideological congruence between UBI and the party’s principles and objectives, (3) an agent or ‘policy entrepreneur’ who mobilises support for the idea, and (4) the opening of a policy window within the party’s internal processes which makes it possible to break with previous policy commitments. The decision to support or drop UBI could be motivated by a number of factors: substantive policy analysis, electoral strategy, internal party management, activist recruitment and identity-building, or a desire to become ‘coalitionable’ or achieve policy influence (Schofield, 1993).
Within this framework, De Wispealere’s theory of ‘cheap support’ can be set out more precisely. His argument suggests that a party supports UBI when the normative appeal of the idea is not constrained by detailed policy analysis or strategic considerations. Over time, and especially as a party grows or moves closer to power, (1) substantive policy work highlights the trade-offs involved and weakens internal support for UBI; (2) the party becomes more aware of the political costs involved in seeking to win support for UBI from voters, the media, interest groups, and potential coalition partners; and (3) the party becomes more sensitive to the ways in which the pursuit of UBI might hamper its ability to achieve other objectives. When the substantive and strategic costs of the commitment to UBI outweigh the perceived benefits, the party is likely to drop it.
Existing bodies of literature on UBI and political parties also allow us to form some broader theoretical expectations about which parties are most likely to support the idea, as set out in Table 1. Ideologically, the case for UBI has generally been rooted in arguments about individual autonomy and social justice, with claims about the scope for reducing bureaucracy, advancing gender equality, and improving labour-market flexibility also featuring in particular contexts (Bidadanure, 2019). Structurally, we would expect UBI to attract more support from new parties than older ones, all things being equal, since older parties are more likely to have invested political capital in existing welfare programmes. Small parties may find it easier to reach internal agreement on an idea such as UBI (Vanderborght, 2004: 19) and may see it as a way of carving out a ‘niche’ in the electoral market (Meyer and Miller, 2015); both internal party democracy and ‘personalization’ around the leader can also allow new ideas to rise quickly up a party’s policy agenda (Schumacher and Giger, 2018). Conversely, strong ties to organised interest groups (such as trade unions) can be expected to act as a barrier to UBI support, since the material benefits are likely to be relatively diffuse, and accrue disproportionately to labour-market outsiders (Rincon and Vlandas, 2023). The need to fight elections under majoritarian institutions may also be a barrier, since research suggests that ‘broad, non-targeted programs are. . . systematically smaller under majoritarian elections’ (Persson and Tabellini, 2001: 28) and that ‘more proportional systems. . . have higher transfers and lower public good spending, ceteris paribus’ (Milesi-Ferretti, Perotti, and Rostagno, 2002: 637). Finally, we can expect a fully fledged UBI (though not necessarily UBI trials) to be more attractive to a policy-seeking or vote-seeking party than to an office-seeking one, since it can be difficult to fit radical and costly policy ideas into a negotiated coalition programme.
Theoretical expectations about UBI support.
Taken together, these criteria suggest that we should expect the Liberal Democrats to be more receptive to UBI than large established parties such as Labour and the Conservatives, but more cautious than green and radical-left parties in the UK and elsewhere. The ideological fit with UBI is relatively promising, since the Liberal Democrats have generally defined their goals in ‘social-liberal’ terms (Howarth, 2007), placing heavy normative weight on individual autonomy and highlighting the ways in which ‘economic deprivation, lack of education, disadvantage or discrimination’ can limit personal freedom (Liberal Democrats, 2002: 7). However, the party also includes some market-liberal elements, and a major thrust of its electoral strategy involves competing with the Conservatives for relatively affluent swing voters in ‘first-past-the-post’ Westminster elections (Cutts, Russell, and Townsley, 2023). As a result, the party is likely to be more sensitive than green or radical-left parties to the tax cost of UBI, especially in a ‘maximalist’ form where the payment is set at a high enough level to cover a wide range of needs and significantly reduce economic inequality.
Organisationally, the Liberal Democrats have ‘traditionally been viewed as a “bottom-up” party with a relatively high degree of influence open to grass-roots members and party activists’ (Evans and Sanderson-Nash, 2011: 459). Party leaders are elected by ‘One Member One Vote’, and the constitution puts formal policy-making in the hands of the Federal Policy Committee (FPC), ad hoc policy working groups, and the biannual party conference (Brack, 1996). 1 This system of internal democracy gives ordinary members a more direct role in policy development than is the case in most ‘governmental’ parties, and provides considerable scope for the kind of ‘policy entrepreneurship’ that has pushed UBI up the political agenda in other contexts (Frankel, 2020; Sloman, 2019). As several observers have pointed out, however, both the party leader and the parliamentary party tend to play a bigger role in policy-making than the constitution suggests (Evans and Sanderson-Nash, 2011; Russell, Fieldhouse, and Cutts, 2007). This makes it difficult to characterise the party as either ‘activist-dominated’ or ‘leadership-dominated’ (Schumacher, De Vries, and Vis, 2013). Alongside its organisational structure, the party’s relative newness and weak ties to interest groups might be expected to make it more receptive to UBI, but a desire to be ‘coalitionable’ may push it the other way, towards policies that can be accommodated more easily in a coalition programme.
Methods
Historical case studies have an important role to play in strengthening our understanding of the drivers of party change, as Tim Bale (2012: 3) has noted. Close analysis of the Liberal Democrats’ debates over UBI provides an opportunity to trace a party’s ‘policy reasoning’ around the idea in detail, in line with classic justifications for case study research (Gerring, 2004). The choice of the Liberal Democrats as a case study can be justified in two ways. First, the party is one of the largest and most ‘mainstream’ political parties to have firmly endorsed UBI in a major western democracy. Since their founding in 1988 through the merger of the Liberal Party and Social Democratic Party (SDP), the Liberal Democrats have won between 7% and 23% of the vote in UK general elections; they have usually been the third party in the House of Commons, and have formed part of coalition governments both at UK level (in 2010–15) and in the Scottish Parliament and Welsh Senedd (Wager and Bale, 2019). Second, the source base for qualitative analysis of Liberal Democrat decision-making processes is unusually strong. For the 1988–94 period, the party’s central archive and the private papers of former leader Paddy Ashdown are open to researchers at the London School of Economics. For the later period, many policy documents have been published, and videos of party conference debates from 2020 and 2023 can be found online on the party’s YouTube channel.
For the purposes of this research, the author conducted a systematic survey of relevant policy papers and of documents in the Liberal Democrat papers and the Ashdown papers at the LSE, including the minutes of the FPC, the minutes and papers of the 1988–90 and 1992–94 tax and benefit working groups, and Ashdown’s unpublished diaries and correspondence. The party conference debates from 2020 and 2023 were transcribed from YouTube, and the author was also able to draw on nine interviews with past and present Liberal Democrat activists, party staff, and basic income campaigners. 2
The sources were used to carry out a process-tracing analysis of the Liberal Democrats’ internal debates over UBI, seeking to identify the relevant actors and their motivations in order ‘to provide a narrative explanation of a causal path that leads to a specific outcome’ (Vennesson, 2008: 235). The focus was on explaining the party’s decision-making at ‘moments of collective indeterminacy’ through ‘temporal tracing’ (Ermahoff, 2019: 595), rather than applying a coding scheme or using formal tests to establish causal inference. Within each period, the central question was whether the Liberal Democrats’ decision to endorse and then drop UBI reflected the drivers suggested by De Wispelaere’s ‘cheap support’ theory.
The evidence base was well suited to this research strategy. As Armando Lara-Millán, Brian Sargent, and Sunmin Kim have noted, contemporaneous ‘documents that give us information about moments of positive contingency’, ‘capture the processes through which historical actors learn by mistake’, and ‘provide a clue to plausible alternatives’ are among the most valuable forms of evidence for process tracing (Lara-Millán, Sargent, and Kim, 2020: 346, 347). Nevertheless, both archives and interviews can be biased and misleading. The analysis thus required a reflexive and critical approach to the sources, informed by the author’s historical training: ‘triangulating different types of documents’, reflecting on their provenance and purpose, and ‘probing consistency and constancy’ (Ermahoff, 2019: 596). The interviews were used to build contextual understanding of the party and UBI campaigns in Britain, rather than to obtain direct evidence about policy choices. Although the risk of bias and error cannot be wholly eliminated, this critical approach to the sources helps to build our confidence in the findings.
The first endorsement: Paddy Ashdown and ‘Citizen’s Income’, 1988–90
The Liberal Democrats’ first engagement with UBI came towards the end of a wider wave of interest in the idea, which ran from the 1970s to the 1990s (Sloman, 2018). In the face of deindustrialisation, mass unemployment, and changing gender norms, social security systems geared towards male-breadwinner family structures and Keynesian full-employment policies came under strain around the world. In Britain, the Thatcher governments pursued labour-market deregulation and regressive tax and benefit reforms, which contributed to sharp rises in poverty and inequality. Basic income enthusiasts such as Charles Handy (1984) and Bill Jordan (1987) presented it as part of an alternative strategy for smoothing the transition to a post-industrial economic model, reconciling labour-market flexibility with the need for income security. The creation of the Basic Income Research Group (BIRG) in 1984 and the Basic Income European Network (BIEN) in 1986 provided institutional focal points for UBI research and advocacy.
British Liberals’ interest in basic income was not entirely new. Juliet Rhys-Williams had put the idea of universal cash payments on the Liberal Party’s internal agenda as early as the 1940s, when she developed plans for ‘a new social contract’ as a more comprehensive, redistributive, and gender-neutral alternative to William Beveridge’s National Insurance scheme (Sloman, 2019: 74–83). Two party committees endorsed versions of Rhys-Williams’ proposal, and in the 1970s the Liberal activist Philip Vince devised a Credit Income Tax scheme that would replace both tax allowances and benefits with refundable weekly tax credits (Vince, 1979). However, the party never made this a central campaigning theme, and when the SDP-Liberal Alliance was formed in 1981, it adopted Dick Taverne’s alternative plan for tackling poverty through a means-tested Basic Benefit (SDP-Liberal Alliance, 1983).
After the 1988 merger, basic income came on to the Liberal Democrats’ agenda from two directions. First, as part of a wide-ranging exercise to develop policies for the new party, the FPC set up a Working Group on Tax and Benefits and asked the veteran Liberal peer Nancy Seear to chair it (Tattersfield, 1988). Seear had been a member of a Liberal Party committee which worked on Juliet Rhys-Williams’ scheme in 1949–50, and was still enthusiastic about the idea. Led by Seear and Vince, the group developed plans for a radical overhaul of the tax and benefit systems which would include the replacement of tax allowances with refundable personal credits or a UBI, the integration of means-tested benefits into a new Low Income Benefit, and the abolition of the State Earnings-Related Pension Scheme (SERPS) (Vince, 1989).
Second, the new party elected a UBI sympathiser as its leader. Paddy Ashdown seems to have come across the idea during the 1970s or 1980s, possibly through reading Handy and Jordan’s work, and had signalled his interest before the leadership election (Ashdown, 1987: 18). After he was elected leader, he endorsed ‘a partial Basic Income’ in his book Citizen’s Britain (1989), which he wrote with help from Bill Jordan (interview with Bill Jordan). Ashdown characterised UBI as one of the central institutions of a post-Fordist liberal society in which citizens moved in and out of different forms of work, and combined paid work or self-employment with household duties and care (Ashdown, 1989: 20). Though it might ‘take several decades of growing prosperity’ before the payment could be high enough to meet subsistence needs, ‘every step we take towards it will diminish dependency, liberate economic power in the hands of the citizen and cut back on the oppressive power of the state bureaucracies over people’s lives’ (Ashdown, 1989: 83).
Ashdown’s enthusiasm for UBI seems to have been sincere, but it also served two important political purposes. Most obviously, it signalled the Liberal Democrats’ willingness to engage with radical ideas at a time when the newly-merged party was struggling to establish itself, especially after a disastrous showing in the 1989 European Parliament elections, where it came fourth behind the Green Party with 5.9% of the vote. As Ashdown noted in his diaries, Citizen’s Britain was part of a ‘strategy for survival’ at a time when the party was in danger of ‘being squeezed out by the Labour Party moving into the centre ground and by the Greens capturing the radical agenda’ (Ashdown, 2000: 51). In early planning for the 1992 election, he told colleagues that he wanted the Liberal Democrat campaign to be ‘radical, fresh, and risky’, and argued that the party ‘wouldn’t capture the public imagination unless we had new things to say’ (Ashdown, 2000: 90).
As well as giving the Liberal Democrats some policy definition, basic income also served as a form of ‘flanking’ for Ashdown’s efforts to ‘recaptur[e] the political ground of choice and individual liberty from Mrs Thatcher’ (Ashdown, 1988: 12). As he explained in a subsequent interview, he ‘shifted the economic policy deliberately quite strongly to the right’ and sought to underline the party’s commitment to ‘competition, small business and enterprise’ (Rawnsley et al., 2001: 5). By framing UBI as a ‘Citizen’s Income’, Ashdown presented it as a concrete embodiment of the new politics of citizenship championed by Charter 88, and sought to reassure activists and voters that the Liberal Democrats were not embracing Thatcherite values.
The political challenge – as so often for UBI enthusiasts – was to devise a workable scheme. This was complicated by the high salience of income tax rates in the late 1980s and the way in which the Conservatives had repeatedly attacked Labour and the Alliance over tax. As Philip Vince (1988) pointed out to the working group, radical tax-benefit restructuring was likely to create ‘a great many losers’, and the party needed to be clear about what tax increases it was ‘prepared to campaign [for] . . .as a socially caring party’. Hermione Parker of BIRG had calculated that a basic income ‘sufficient to permit the abolition of all existing cash benefits, without people on low incomes losing out’ would require income tax rates of at least 70%, which she thought was ‘not feasible and probably never will be’ (Parker, 1988: 5, 7). Seear’s working group alighted on the idea of turning the personal tax allowance into a Citizen’s Income of just over £10 per week for 16–64 year olds, to be paid for largely by broadening the income tax base and phasing out the married couple’s allowance, and then raising the Citizen’s Income over time as resources became available. Child Benefit would also be raised significantly. This package would avoid the need for a politically sensitive increase in marginal rates for most taxpayers, but would still bring about a significant redistribution towards the very poor and families with children (Liberal Democrats, 1990).
The Liberal Democrats’ embrace of UBI thus took place in a policy window in which the launch of a new party coincided with a wider public debate about economic change and tax-benefit reform, in which basic income campaigners featured prominently. The policy was ideologically congenial to the party leader (Ashdown) and senior activists (Seear and Vince), who acted as agents to put it on the party’s internal agenda. Identity-building and party management motives can be detected alongside substantive policy analysis: at a low political ebb, Ashdown’s primary focus was on establishing the Liberal Democrats’ visibility and distinctiveness.
The first retreat, 1990–94
The launch of the Common Benefit report produced by the working group passed off smoothly, and the package was endorsed (with small modifications) by the March 1990 spring conference. The journalist John Pienaar (1990) reported that Liberal Democrat strategists saw the document as a key step towards establishing ‘a distinctive identity’ and that the policy would be ‘a central plank in the manifesto for the next general election’. However, Citizen’s Income was much less prominent in the 1992 election campaign than Pienaar suggested. It did not feature in the ‘five Es’ which Ashdown used to structure his campaign: education, the economy, the environment, Europe, and electoral reform (Holme, 1995). The social security section of the manifesto promised to ‘work towards the eventual creation of a new “Citizen’s Income”, payable to all irrespective of sex or status’ and worth about £12.80 per week, but even here, the focus was on more immediate benefit changes (Liberal Democrats, 1992a). Bill Jordan (1992: 15) felt that ‘the Liberal Democrats suppressed any mention’ of Citizen’s Income during the campaign, and ‘systematically missed opportunities’ to talk about it. After the election, FPC set up a new Tax and Benefits Working Group (TBWG), and its report led to the party abandoning Citizen’s Income at its September 1994 conference (Liberal Democrats, 1994a).
Why did the Liberal Democrats retreat from UBI so quickly? The story is a complex one, which shows how policy analysis and political judgements intersect to shape a party’s policy reasoning. Ashdown and his colleagues had worried from the outset that the Common Benefit scheme was politically risky, especially because the base-broadening tax reforms would hit some households with modest incomes (Kirkwood, 1989). The proposals were modified to remove the most obvious points of criticism (Liberal Democrats, 1991), but the party’s economic researcher Ed Davey (1992) thought the ‘cumulative impact’ of the tax changes in the 1992 manifesto was still ‘very difficult to justify’. Davey also observed that MPs and candidates struggled to explain ‘the distinction and differences between Citizen’s Income and minimum income’ (Davey, 1992). By contrast, the commitment to put 1p on the basic rate of income tax in order to ‘invest’ an extra £2 billion a year in education was widely felt to have ‘captured the voters’ imagination’ (Liberal Democrats, 1992b). The Tory attack on ‘Labour’s tax bombshell’ during the 1992 campaign made the tax issue seem particularly salient (private interview), and Ashdown concluded that the Liberal Democrats should focus on ‘re-distributing opportunities’ rather than ‘re-distributing wealth for its own sake’ (Ashdown, 1992). FPC also created a standing ‘Costings Group’ to monitor spending proposals and ‘provide the FPC and the Parliamentary Team with a fiscal framework for debates and policy initiatives’ (Liberal Democrats, 1994b).
In this changing political context, the logic of a partial basic income faced more critical scrutiny from the new TBWG. The barrister Sir William Goodhart, who chaired the new working group, pointed out that replacing the personal allowance with a UBI would only benefit ‘those who neither have significant incomes nor get benefits’: ‘mainly non-working married women’, plus some students and trainees, many of whom were not especially poor (Goodhart, 1992, 1994a). For UBI supporters, this was beside the point, because the payment of just over £10 a week proposed in Common Benefit – equivalent to about £16.50 in the 1994 tax system – was only ever intended as the first step towards creating a more adequate basic income (Vince, 1994). Goodhart’s retort was that raising Citizen’s Income in such a way would be enormously costly: for instance, doubling the rate to £33 a week would ‘require the equivalent of an extra 10 pence on income tax’, which ‘would not only be politically unacceptable but would also pre-empt other spending increases that we might wish to adopt in fields such as education’ (Goodhart, 1994b).
Ashdown regretted that abandoning Citizen’s Income would deprive the party of ‘an element of distinction & radicalism’ (Liberal Democrats, 1994c), but reluctantly accepted the force of Goodhart’s argument (Ashdown, 1994). Many activists seem to have been similarly ambivalent (White, 1994). When the final report came to the September 1994 conference, Goodhart argued that UBI ‘should be sent back to Utopia, where it came from’, and delegates agreed to drop the policy (Wintour, 1994). This retreat proved less contentious than might have been expected, perhaps because delegates’ energies were consumed by debates over a minimum wage, cannabis decriminalisation, and the future of the monarchy. By the time of the 1997 election, the ‘penny for education’ was firmly entrenched as the Liberal Democrats’ ‘flagship’ policy (Russell and Fieldhouse, 2005: 38) and UBI had largely been forgotten.
On the basis of the archival record, Goodhart stands out as the driving force behind the retreat from UBI, though it could not have happened without Ashdown’s acquiescence. Critical analysis of the case for a partial UBI was clearly a central motivation, but this analysis took place within a particular frame of reference, in which the electoral risks of tax rises and the opportunity cost of UBI featured more prominently than in 1989–90. After the qualified success of the 1992 campaign, in which the Liberal Democrats held their ground but failed to make hoped-for gains, earlier concerns about policy distinctiveness gave way to a focus on ‘credibility’ and efforts to establish a clearer set of priorities. 3 The stronger result in 1997 – when the Liberal Democrats more than doubled their seats to 46 MPs – seemed to vindicate Ashdown’s disciplined focus on winning over ‘target voters in target seats’ and encouraging tactical voting (Holme and Holmes, 1998: 19).
The second endorsement: Covid as a ‘policy window’, 2015–20
In September 2020, the Liberal Democrat party conference voted 715–250 to commit the party to ‘campaign for a Universal Basic Income’, ‘funded in a socially just and equitable manner to create a fairer social security system for all’ and ‘implemented based on the best available international evidence’ (Liberal Democrats, 2020a). As the party’s newly-elected leader, Sir Ed Davey declared that UBI would be ‘a huge step towards the fairer society we, as liberals, should champion’ (House of Commons Library, 2020: 10). Less than three years later, however, the party’s 2023 spring conference dropped the idea in favour of a means-tested guaranteed income scheme. Work and Pensions spokesperson Wendy Chamberlain, who had backed UBI in 2020, explained that she had ‘gone on a journey’ and concluded that a means-tested solution ‘would tackle poverty in a targeted way and at a fraction of the cost of a UBI’ (Liberal Democrats, 2023a). Older activists could have been forgiven for feeling a sense of déjà vu: thirty years on from the debate over Common Benefit, history seemed to have repeated itself. Why did the Liberal Democrats return to UBI and then retreat from it again?
For two decades after 1994, UBI disappeared from the Liberal Democrats’ internal policy agenda. The party’s social policy initiatives were dominated by education, health, pensions, and social care, which featured prominently in the 1997, 2001, 2005, and 2010 campaigns (Finn, 2020). When Nick Clegg took the party into coalition in 2010, his failure to deliver on a promise to abolish university tuition fees came to symbolise the perceived ‘betrayal’ of Liberal Democrat voters, as the party leadership signed up to a Conservative-led austerity programme (Atkins, 2020). At the same time, the Liberal Democrats’ neglect of working-age social security allowed Iain Duncan Smith and George Osborne to set the agenda by introducing Universal Credit (UC) and imposing real-terms benefit cuts on lower-income households.
The Liberal Democrats’ collapse from 57 seats to 8 seats in the 2015 general election coincided with a new wave of interest in basic income in the UK, which was driven partly by the reaction against austerity (Sloman, 2018). The Green Party made a £72 per week UBI one of its flagship policies in the run-up to the 2015 campaign, though it later watered down its commitment after party leader Natalie Bennett struggled to explain how it would be paid for in a BBC interview (Dennison, 2017: 84). Later in the decade, the idea of a UBI pilot attracted wide interest in Scotland, and the Scottish government under Nicola Sturgeon funded a feasibility study (Cantillon and O’Toole, 2022). Labour’s Shadow Chancellor John McDonnell also indicated his support for pilots, and the basic income enthusiast Geoff Crocker worked with the Compass pressure group to set up a campaign hub, known as the ‘Basic Income Conversation’.
Until the Covid pandemic hit, however, the Liberal Democrats continued to reject the idea. A 2015–16 Working Group on Mending the Safety Net was ‘initially strongly attracted to the thinking’ behind UBI and negative income tax, but concluded that ‘these policies would be unable to cope with the diverse and complex variations in benefit needs existing in the UK, in particular the huge variation in housing costs. . . and support for people with disabilities’ (Liberal Democrats, 2016: 11). UBI ‘would entail increases in expenditure, potentially very significant’, and was ‘not the best way to spend scarce resources’ (Liberal Democrats, 2016: 12). Both socially and politically, it made more sense to focus on targeted changes to UC and legacy benefits, like scrapping the Conservative government’s benefit cap and two-child limit. 4
The Liberal Democrats’ embrace of UBI in 2020 was thus a sharp break from the position which the party had taken since the mid-1990s, and is best understood as a product of two contextual shifts. First, the failure of the party’s ‘Stop Brexit’ campaign in the 2019 election threw the party into a strategic crisis, since pro-European positioning had helped to double party membership (to more than 126,000 in December 2019) but had not translated into extra parliamentary seats. Jo Swinson’s defeat in East Dunbartonshire also triggered a leadership election, fought by Davey and the Oxford West MP Layla Moran. Comparative research has found that poor election results commonly stimulate policy shifts (Somer-Topcu, 2009), and this effect is often mediated through changes in a party’s leader or ‘dominant faction’ (Budge, Ezrow, and McDonald, 2010).
Second, the Covid pandemic opened up a policy window in which UBI could be framed as a necessary crisis response. In the UK, 100 MPs (including 9 Liberal Democrats) signed an Early Day Motion calling for ‘a temporary universal basic income or an emergency measure to help freelancers and the self-employed [a]ffected by the COVID-19 outbreak’ (UK Parliament, 2020). As acting leader, Davey took a particular interest in highlighting gaps in the UK government’s support for the self-employed, but also argued that ‘we should look more seriously at proposals for a universal basic income’ as a way of tackling wider social inequalities (Davey, 2020: 156).
The combination of the Covid pandemic with a party leadership race created an unusually permissive environment for UBI within the Liberal Democrats. Layla Moran declared that she wanted to ‘take the Lib Dems back to our radical roots’, and argued that support for UBI could help establish the party as ‘more radical than Labour’ (Payne, 2020). The Cardiff councillor Rhys Taylor (2020) developed a case for UBI in an essay collection edited by Moran, Build Back Better, and drew attention to modelling which suggested that a basic income of £60 a week was realistic and affordable (Taylor, 2020). Davey was more cautious, but pointed to his experience of working on Common Benefit with Paddy Ashdown, and suggested that the Treasury could part-finance a UBI by taking stakes in firms which could not repay business-support loans (Johnston, 2020). When the Welsh Liberal Democrats’ leader Jane Dodds convened an informal group of UBI supporters to prepare a conference motion, the group received encouraging signals from both candidates. Davey was eventually elected leader in August with 64% of the vote – a result widely seen as a victory for the party’s centrist establishment.
The motion debated by the September 2020 conference – which was held online due to Covid – committed the Liberal Democrats to the principle of UBI while asking FPC to work out the details. 5 Its proposer, Adam Bernard, opened the conference debate by making the case for UBI in distinctively liberal terms, as a policy which would break with ‘conditionality’ by giving people ‘the tools to make their own choices about their own lives’. The London-based activist Richard Flowers echoed this point, arguing that Covid had shown ‘that government can make radical changes to how we support each other’ and that UBI would help to build ‘a generous, caring society’. Other speakers warned that the motion was a ‘blank cheque’ and that UBI would become ‘a greedy cuckoo in our policy nest’, absorbing money that would be better spent on other priorities (Liberal Democrats, 2020b). Nevertheless, members voted overwhelmingly to adopt UBI as party policy.
The 2020 conference vote can be seen as an attempt to short-circuit the Liberal Democrats’ usual policy-making process at a time when the party was searching for direction. This time, the agents were mainly activists motivated by sincere policy views, but the attitude of the leadership candidates and party elites (including the Federal Conference Committee) was important, and identity-building and electoral considerations were clearly present. Support for UBI was useful as a way of demonstrating the party’s relevance and openness to new ideas after the failure of its anti-Brexit campaign, and Scottish and Welsh Liberal Democrats – preparing to fight devolved elections in 2021 – were particularly prominent.
The second retreat, 2020–23
After the September 2020 motion had passed, the FPC faced the challenging task of turning UBI from a principle into detailed policy. By the time a specific UBI proposal came to the March 2023 spring conference, the economic and political situation had been transformed by the post-Covid economic recovery, rapid inflation, rising interest rates, and the collapse of Conservative support in the wake of the ‘partygate’ scandal. Winning over former Tory voters in so-called ‘Blue Wall’ seats quickly became the dominant strategic focus of Sir Ed Davey’s leadership, as a string of by-election victories highlighted the scope for the Liberal Democrats to make gains from the Conservatives in southern England (Cutts, Russell, and Townsley, 2023: 163–166). From a low of 5% in some opinion polls in mid-2020, the party was routinely polling at 10% or above by the end of 2021.
As in the Ashdown era, the development of Liberal Democrat policy on UBI passed through two successive working groups. First, from January to June 2021, the former Southwark councillor Paul Noblet chaired a Working Group on UBI, which took evidence and drew up a costed scheme. Drawing on Malcolm Torry’s work for the Citizen’s Basic Income Trust, the working group focussed on the idea of replacing the personal tax allowance and its National Insurance equivalent with a UBI of £71 per week, at a net cost of about £30 billion (Liberal Democrats, 2021). In structural terms, this was similar to the Common Benefit proposal, though the post-2010 increase in the personal allowance meant that the UBI would be worth much more in real terms – about 12%, rather than 6%, of average weekly earnings. 6 The working group pointed out that £71 per week was a sweet spot, above which the cost would rise rapidly as the UBI started to give net benefits to basic-rate taxpayers. If UBI was treated like earned income, UC claimants would derive some benefit from the scheme, gaining about £26 a week. Modelling by the US-based UBI Center suggested that the policy would reduce absolute poverty by about one-quarter and deep poverty (which it defined as an income below half the poverty line) by about 70% (Liberal Democrats, 2021: 20). As in Common Benefit, however, the biggest gains would go to people who had no taxable income and no means-tested benefit entitlement.
The £30 billion cost of the working group scheme was not totally unrealistic, since the 2019 Liberal Democrat manifesto had set out plans for almost £63 billion a year of new spending, including £14 billion for universal free childcare and over £10 billion for schools. On the other hand, introducing UBI would make it harder to finance these other policies, and some activists wondered whether the £71 scheme was attractive enough to justify the cost. The FPC thus referred UBI to a new Fairer Society Working Group, chaired by the former MP Julia Goldsworthy, which was asked to consider how it would relate to the party’s other plans for ‘tackling economic unfairness’ (Liberal Democrats, 2023b: 52), with a particular eye on the need to ensure that policies could be ‘communicated effectively on the ground’ (Goldsworthy in Liberal Democrats, 2023a).
Goldsworthy’s group made two analytical moves which undercut the rationale for UBI. First, the group took up the concept of ‘deep poverty’ – which it redefined, in line with the Joseph Rowntree Foundation’s work, as less than 50% of median income – and suggested that the Liberal Democrats should set a goal of ending deep poverty within a decade (Liberal Democrats, 2023b: 15, 26). Once the policy problem was reframed in this way, raising UC parameters and removing work requirements seemed to be much more cost-efficient than introducing UBI. Second, by labelling this alternative policy as a ‘Guaranteed Basic Income’, the working group blurred the distinction between UBI and means-tested income support. This allowed it to make a virtue of the potential for means-testing to ‘maximise poverty reduction for any given level of state expenditure’ (Liberal Democrats, 2023b: 25).
At the March 2023 spring conference, a succession of senior party figures lined up to back GBI and condemn UBI as unrealistic. For instance, Tim Farron warned that UBI would be ‘electoral Kryptonite’, ‘hardly helping anyone who needs it’ but giving ‘our opponents and the media a big flipping stick to beat us with’, while the former MP Andrew Stunell invoked the spectre of a Liberal Democrat MP floundering on television as they struggled to justify the pattern of gains from a £71 UBI scheme. 7 By contrast – Stunell argued – ‘Guaranteed Basic Income debugs UBI, avoids stuffing the pockets of the rich with extra cash, and so can provide even more help for the poorest families’ (Liberal Democrats, 2023a).
Faced with these appeals to urgency, progressivity, and electoral pragmatism, the conference voted decisively in favour of ‘GBI’ (Liberal Democrats, 2023a). Davey’s party thus reverted to the focus on raising and liberalising UC which featured in its 2017 and 2019 manifestos, reinforced by the new goal of ending deep poverty. As in 1994, a working group provided a vehicle for the retreat from UBI, reframing the issue in a way that made the rationale for basic income seem less compelling and more conventional benefit changes more attractive. Party elites justified the shift in policy through a mixture of substantive and electoral arguments about the limitations of a ‘partial’ UBI scheme. Though the party’s formal policy-making processes were respected, the decision to shelve UBI went with the grain of Davey’s focus on ‘Blue Wall’ seats. Dropping UBI reduced the need for politically contentious tax rises and allowed the party to prioritise other spending commitments – particularly on the National Health Service and social care – in an increasingly constrained economic environment. As in the 1990s, this pragmatism seemed to pay off as the Liberal Democrats made sweeping gains in the 2024 general election, winning a record 72 seats.
Discussion and conclusions
‘Big ideas’ such as UBI have an important part to play in contemporary political debate, helping politicians and activists to envision alternative futures (Wright, 2010) and stimulating policy innovation (Dumont, 2022). There are good theoretical reasons why small parties might be attracted to radical ideas as a way of building political identity and raising their profile. Nevertheless, experience suggests that parties’ support for UBI is often shallow and fleeting. By using process tracing to identify the drivers of party policy change, this case study of the Liberal Democrats’ engagement with basic income helps us understand why.
For the most part, the Liberal Democrats’ decision-making on UBI under Paddy Ashdown and Ed Davey fits closely with De Wispelaere’s analysis of ‘cheap support’. The parallels between the two periods are certainly striking. In the aftermath of both the 1988 merger and the 2019 election, leadership change and a sense of political weakness coincided with a wider public wave of interest in UBI to open an internal policy window for the idea. In 1989–90, the leader and senior activists were key policy entrepreneurs, and detailed work was carried out before UBI was endorsed by a party conference; in 2020, grassroots activists played a more significant role. Normative arguments for UBI were prominent in both cases, and the COVID-19 pandemic was especially important as a stimulus to action, highlighting gaps in the social safety-net and temporarily reshaping perceptions of what was ‘affordable’. The Liberal Democrats’ subsequent retreats were driven partly by electoral calculations, and especially by growing awareness of the political costs involved in trying to justify basic income proposals. De Wispelaere’s warning that support for UBI becomes harder to sustain once attention turns to policy details is also borne out by analysis of the internal debates within the party.
The case study of the Liberal Democrats thus supports De Wispelaere’s basic analysis of why parties endorse and then abandon UBI. At the same time, by attempting to delineate internal actors and their motivations more carefully, this article also suggests the need for a number of refinements. First, the Liberal Democrats’ experience highlights the need to pay close attention to parties’ goals, which De Wispelaere does not consider systematically. Classic literature on party objectives focusses on the pursuit of ‘policy’, ‘office’, and ‘votes’ (Müller and Strøm, 1990), but small and medium-sized parties like the Liberal Democrats may have more limited intermediate goals, such as party survival and parliamentary representation (Russell and Fieldhouse, 2005: 199, 200). Moreover, the case suggests that these two intermediate goals may have different implications for policy. ‘Big ideas’ may help a small party to gain visibility, build its political identity, and mobilise activists, but it is less clear that they help to win seats under ‘first-past-the-post’. Under both Ashdown and Davey, Liberal Democrat strategy was oriented first towards survival and then towards legislative seat maximisation, in the belief that a large bloc of Liberal Democrat MPs was a prerequisite for achieving policy objectives. In the run-up to the 1997 and 2024 elections, strategists seem to have believed that policy controversy was liable to undermine efforts to attract tactical support and protest votes at constituency level. Of course, the Liberal Democrats’ focus on target seats reflects the incentives created by the UK electoral system, as Thomas Quinn (2017) has shown. Radical policies such as UBI might be a more straightforward asset for smaller parties in PR systems, which can win both votes and seats through ‘niche’ strategies (Bischof, 2017).
Second, the case study highlights the importance of policy costings as a frame through which parties evaluate and aggregate policy proposals. The salience of tax and spending in UK election campaigns shapes policy development within parties, as well as public debate (Sloman, 2024): the Liberal Democrats’ commitment to producing a ‘fully costed’ manifesto imposes discipline on the policy-making process and forces the FPC to engage with budgetary trade-offs. 8 Though the culture of costings is particularly strong in the UK, similar practices can be found in several other countries, including Australia, Canada, the Republic of Ireland, and the Netherlands (Robson and Jarvis, 2020), and Podemos in Spain also faced pressure to show how it would pay for its UBI policy (Noguera, 2019: 292). The issue here is not just that basic income is expensive, but that the shadow-budgeting process places it in competition with other policy commitments and thus complicates internal coalition-building. Opinion polling suggests that many left-leaning ‘welfare state supporters’ are attracted to UBI in principle, as David Weisstanner (2022: 101) has shown. As the cost of even a limited UBI scheme becomes clear, however, those who are strongly committed to improving public services such as health, education, and childcare are liable to see the proposal as a threat to their own spending priorities. This problem may also affect the prospects of other costly ideas, such as ‘Green New Deal’ policies (Robertson, 2025).
Third, the importance of policy analysis in the ‘wilting’ of support for UBI should not be under-rated. Since UBI campaigners are mostly convinced by the normative arguments for the idea, there can sometimes be a tendency to see retreat as a failure of nerve. Whatever its normative merits, however, UBI is not easy for politicians to justify as a practical policy proposal. As Luke Martinelli (2020) has noted, an ‘adequate’ version – set high enough to cover basic needs – would be extremely expensive and could lead to significant economic and social disruption, while an ‘affordable’ version would not by itself eliminate poverty. 9 In an analytical landscape structured by perceived budgetary constraints and static distributional modelling, the opportunity cost of putting UBI in a manifesto looks very high, and any specific UBI scheme is likely to be vulnerable to pairwise comparisons with alternative policy options (Aerts, Marx, and Verbist, 2025).
Further research is needed to assess how far these findings apply to other cases. Parties have distinctive structures and cultures, shaped by their own experiences and national contexts, which mean that the process of policy reasoning is likely to play out in different ways; the existing tax and benefit structures which act as the ‘baseline’ for reform also vary markedly between countries. It may be that a comparative study would identify multiple pathways by which parties move towards and away from UBI. Some of the most high-profile proposals have been closely linked to presidential candidates – such as George McGovern in the United States in 1972, Benoît Hamon in France in 2017, and Lee Jae-myung in South Korea in 2022 – and lost momentum after these candidates were defeated. Other parties, such as D66 in the Netherlands in 2021, have achieved strong election results after campaigning for UBI-type policies, but then settled for more limited reforms in coalition negotiations (Roosma, 2022). At the time of writing in early 2025, the latest wave of enthusiasm for the idea seems to be fading. Even so, as policy-makers continue to search for ways of tackling economic insecurity and managing disruptive challenges such as the rise of AI, the role of parties in the UBI debate is likely to repay continued attention.
Footnotes
Acknowledgements
I am grateful to interviewees for their assistance, and to Jurgen De Wispelaere, José Tomás Labarca, and the journal’s anonymous reviewers for their comments on earlier versions of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This article draws on research supported by the Keynes Fund for Applied Economics at the University of Cambridge.
