Attitudes to Welfare through Deliberative Forums – The Emergence of Reluctant

This article uses deliberative forums to examine attitudes to UK welfare futures. It makes methodological, empirical and theoretical contributions to the field. We demonstrate the value of the approach, provide insights into attitudes, in particular about priorities and how people link ideas together, and show how the UK’s neoliberal market-centredness fits with enthusiasm for state healthcare and pensions, desire to close national labour markets to immigrants and approval of government interventions to expand opportunities for those who make the effort. Findings point to the strength of the work ethic and individual responsibility alongside a regret that major and highly valued state services appear unsustainable, the construction of immigrants as simultaneously a burden on provision and unfair labour-market competitors, and backing for the development of a ‘new risk’ welfare state through social investment. The study reveals the complexity of responses to current challenges in an increasingly liberal-leaning welfare state.


Introduction
Welfare states faces severe social, political and economic pressures (Hemerijk, 2013;Kersbergen and Vis, 2014;Esping-Andersen, 1990;Taylor-Gooby et al, 2017).More specifically, indications of a decline in support for traditional state welfare services (Pearce and Taylor, 2013), anxiety about government capacity to provide adequate services in health, social care and pensions, stronger concerns about how welfare is assumed to weaken work incentives (Clery, 2016) and fears about a perceived failure of government to manage immigration (Migration Observatory, 2016) suggest pressures for change are strengthening, and indicate a shift away from the collectivist tradition towards greater individualism.In this context, better knowledge of people's attitudes to welfare and their priorities and how they are framed is of value (Svallfors, 2010;Larsen, 2008;Mewes and Mau, 2013).
This paper uses Democratic Forums (DFs) to investigate attitudes to the future of state welfare in the UK.It is part of a five-country comparative project on which we will report in due course.DF methods are rarely used in attitude research, and not previously, as far as we know, in relation to welfare state futures.We seek to contribute in three areas: to method, by demonstrating the value of the approach, empirically, by using its capacity to explore the justifications that underpin the 'headline figures' of attitudes to welfare and theoretically by developing the free market individualist characterisation of UK welfare state ideology to include the notion of 'reluctant individualism'.
Most existing research on welfare attitudes relies on structured surveys based on individual interviews with a representative population sample.These studies are valuable in charting attitudes towards social policy, but suffer limitations when academic and expert preoccupations not shared by the general public receive excessive attention (Goerres and Prinzen, 2011).
DFs adopt a different approach (Carpini et al, 2005;Steiner, 2012;Elstub and McLaverty, 2014;Chambers, 2003;Taylor-Gooby and Leruth, 2018): a group of people meets for extended discussion of a topic over a period of time.A high degree of control is retained by the participants, who frame the issue and pursue discussion as they see fit with only light-touch moderation.The method is typically used in consultations on difficult policy issues (Wakeford and Singh, 2008;Renn, 2008: ch 8;Elstub, 2015).Here we use DFs rather differently, as a research tool to gain new insights into welfare attitudes.DFs do not permit the degree of statistical representativeness often achieved by sample surveys but rather complement them.In particular they provide insight into the priorities that people recognise and the way specific themes are understood to link together.
The five sections of the article cover: pressures on the UK welfare state, in particular the trend towards greater market-centred individualism, the contribution of DFs and conventional attitude studies to understanding what people want and expect from the welfare state in the future, our method, our findings and discussion and conclusion.

Pressures on the welfare state and the attitudinal response
Many commentators agree that welfare states face severe economic, social, demographic and political pressures.These challenges raise the question of how far the future of the welfare state will resemble the past and direct our attention to understanding people's attitudes and priorities.
Social class solidarity made an important contribution to the development of welfare states (Baldwin, 1990;Marshall, 1950;Korpi, 1983;Offe, 1984).Recent developments at a structural level have fractured the traditional welfare alliances.The transition from industrial to post-industrial society, the emergence of a more post-industrial and globalised political economy and the decline of the labour movement have shifted the balance between capital and labour on which the welfare state compromise rested (Iversen and Wren, 1998;Scharpf and Schmidt, 2000).The traditional redistributive settlement has become more unstable.
Welfare states have come under strain during the past three decades from demographic change, rising aspirations (Glennerster, 2009) and labour market change, particularly as more women move into full-time employment and skills play a stronger role in determining opportunities (Taylor-Gooby, 2004).These factors generate demands for childcare and for state support as wages and opportunities at the bottom stagnate (Green, 2006).These trends are markedly insistent in the UK.Most commentators expect them to continue into the future and to grow more pressing (Pierson, 2001;Hemerijk, 2013;Van Kersbergen and Kees, 2014;Taylor-Gooby et al, 2017).The Great Recession from 2007 and subsequent stagnation exacerbate the problem of funding services (Gough, 2011).
At the level of politics, the 'fanning out' of inequalities (Atkinson, 2007) and tensions between immigrants and established populations over access to welfare state resources create further division (Alesina and Glaeser, 2004;Triandafyllidou et al, 2011;Van der Waal et al, 2013).A new populist politics erodes trust in government as welfare provider (Kriesi et al, 2012).
Interests, values and institutions are all relevant to attitudes (Chung and Meuleman, 2017) and in all three areas there is a notable shift towards individualism.The 'new politics of welfare' school argues that interest cleavages are becoming more complex as new concerns gain importance alongside class.These divisions are reinforced by the rise of welfare chauvinism and disillusion with government, particularly on the part of those who feel left behind by globalisation and economic changes (Kriesi et al, 2012;Teney et al, 2013;Hobolt, 2016).The growth of more individualist, less solidaristic values is shown by the greater suspicion of unemployed people, concerns about immigrants as labour-market competitors and the declining support for redistribution and for benefits for unemployed people and the low-waged (Larsen, 2008;Clery, 2016;Pearce and Taylor, 2013;Baumberg et al, 2012;Baumberg-Geiger et al, 2017).Concurrently, private provision is gradually expanding (Hemerijk, 2013).
The UK is often seen as a liberal-leaning welfare state (Esping-Andersen, 1990) committed to a strong role for the market and for individual rather than state responsibility (Mau, 2015), and to targeted welfare provision (second highest in the EU in means-tested proportion of spending: Eurostat, 2017).Recent policy issues include relatively low unemployment but high inequality and high poverty rates, major tax cuts and cuts on social care and benefits, while seeking a balanced budget and the differential loading of spending cuts onto the working age population (Lupton et al, 2016).These features correspond to neoliberal commitment to a free market and individual responsibility.In addition, immigration from the EU and elsewhere is high on the political agenda at the time of the fieldwork (Mewes and Mau, 2013) with a repeated government promise to reduce the rate from 'the hundreds to the tens of thousands' (Telegraph, 2010).More generally the UK faces escalating pressures on social spending.First, population ageing is expected to increase demands for pension spending (from 5.0 per cent of GDP in 2013-14 to 7.4 by 2043-44), reinforced by the triple lock on pensions and only partly offset by pension age rises and cuts in short-term and disability benefit spending (from 6.8 GDP to 6.2 per cent) (OBR, 2014, Table 3).Second, NHS spending is expected to rise from 7.3 to 7.6 and social care from 1.1 to 1.9 per cent of GDP in the period, partly offset by a 0.9 per cent GDP fall in education spending (OBR, 2015, Table 3.6).Whether these spending increases will be sufficient to sustain adequate services is unclear (King's Fund, 2017).Other concerns refer to long-term productivity within both state services and the broader economy (ONS, 2017).
In such a setting the main expectations in relation to attitudes are: an assumption that individuals should be, as far as is possible, independent, providing for themselves through work or property, that entitlement should be strictly limited to proven need, and that progressive taxation or redistribution should be rejected.The authoritative British Social Attitudes (BSA) survey (NatCen, 2017) generally endorses the work ethic and antipathy to progressive tax but shows two things: an entrenched division between support for the major-spending mass provision areas of state welfare (healthcare and pensions) and devaluing of those targeted towards the smaller groups of working age population (unemployment benefit and low-wage support); and a desire to restrict immigration (Blekesaune and Quadagno, 2003;Reeskens and Van Oorschot, 2017).At first sight, both these attitudes cut across neoliberal rejection of all state provision and approval of an open and free market in labour.In addition, a 'thermostat' theory is often applied to attitudes to unemployed benefits: as labour market conditions deteriorate the public becomes more sympathetic and vice versa (Curtice, 2010).At the time of the study however, attitudes seemed to be moving away from this logic.Despite benefit cuts, attitudes to claimants had hardened (Clery, 2016).
These comments lead to four research questions: 1. To what extent do attitudes in the UK follow an individualist market logic?2. How do attitudes to immigration fit in? 3. How does this relate to attitudes to mass services like the NHS and pensions?4. What are the expectations about welfare state futures?

Deliberative forums and other methods of investigating attitudes
Almost all research on welfare attitudes is based on quantitative analysis of structured sample surveys, the great majority on data from four surveys: the European Social Survey, the International Social Survey Programme, the World Value Survey and the European Values Study (see Svallfors, 2010), and nationally the UK BSA survey (which carries modules for the above surveys) used as a comparator for the qualitative work in this paper.There is a small body of work using focus groups (for example Burckhardt et al, 2011;Taylor-Gooby and Martin, 2010).These methods have substantial strengths and the quantitative work in particular has stimulated a growing body of academic research by generating and making available cross-nationally comparable data.
The quantitative approach finds it difficult to explore issues that are not contained within the prior assumptions of the researcher.It has a limited capacity to examine reasoning processes or the meanings individuals attach to particular concepts.Questions are typically answered through choice between pre-coded categories, with the risk of misunderstanding the language and that respondents may have no strong opinions on the issue or answers to meet what they understand as the interviewer's expectations (Goerres and Prinzen, 2011).Discursive and interactive rather than pre-structured methods are better equipped to capture attitudes about complex issues which people may not have thought through and on which their conceptual framing may differ from that of the experts who designed the questionnaire.
Focus groups allow individuals to express their opinions within a moderated group discussion of 45 minutes to one hour on a particular subject, typically according to a set of sub-themes listed in the moderators' topic guide.The procedure allows people to describe, share and debate meanings, but provides limited opportunities to develop them or to move away from the initial topic (Finch and Lewis, 2003).The DF approach seeks to re-balance control over data-gathering between participants and researchers.The topic is usually defined broadly and shaped by the participants rather than by a facilitator.Typically, there is a degree of light-touch moderation, in order to keep the discussion to a broad theme, but no predefined topic guide or schedule.The discussion is framed by participants, and researchers play a more passive role.There may be injections of relevant information, but these are typically provided in response to requests from the group and are prepared by independent experts who are available for cross-examination.The assumption is that people experiencing an issue are the best experts on their own understanding of it (Wakeford, 2007;Wakeford and Singh, 2008;Narayan, 2000).The group may be asked to agree a report to provide a focus and point to the discussion.
Political and social scientists have become interested in this method for two reasons.First, conceptions of democracy have shifted away from that of a system for managing consent from a largely passive electorate to one of democracy as an active institutional framework for promoting more widespread deliberation and citizen engagement (Mouffe, 2009;Chambers, 2003;Carpini et al, 2005;Dryzek, 2010).Second, some social psychologists have moved away from a positivist understanding of attitudes as original to an independent individual to one which sees them as social constructs, developed through interaction and expression in debate (Brown, 2011).From this perspective, interview responses in structured surveys are inevitably shaped by the interaction between a researcher and an interviewee.Conversely DF discussions are the product of a group interaction between naïve citizens.
The strengths of DFs are that they allow participants greater control over the way in which issues are defined and discussed than do structured surveys or, to some extent, than one-hour focus groups organised round a schedule of topics to which the moderator directs contributions.The approach requires that participants maintain an atmosphere of mutual respect, so that all points of view can be included and members are encouraged to participate.They allow researchers to examine the unprompted priorities of individuals in a broad topic-area.
These strengths entail corresponding weaknesses.DFs are too small to permit fully representative sampling and so offer an imprecise guide to the pattern of opinions across the population.They cannot be directed to consider specific aspects of an issue according to a researcher's system of priorities.They are best deployed in collaboration with other methods.
In this project we examine a general issue of considerable importance (attitudes to the medium-term future of the welfare state), where a number of factors interact to influence change and where popular understanding of current developments and of how they affect people's interests will have an impact.The studies referred to in the previous section show that many people feel concerned about the impact of immigration, the future of the labour market and the extent to which government policies are supporting their interests.Uncertainty is exacerbated by a growing distrust of politicians and experts (Rothstein, 2005;Ipsos-Mori, 2016).For these reasons DFs offer an appropriate methodology for the work.DFs and the similar democratic polls (Fishkin, 2011) have been used in consultation exercises (Renn, 2008;Steiner, 2012) but not, as far as we know, in welfare state attitude research.

Method
We carried out a DF exercise in Birmingham in late 2015, after the May general election had brought in an austerity-focused majority Conservative government at the time when Prime Minister Cameron was seeking to avoid a referendum on EU membership by renegotiating the UK's relationship with the EU.
Our UK research formed part of a five-country comparative study on which we will report in future work (Taylor-Gooby and Leruth, 2018).Participants were selected by a market research company with a screening questionnaire through door-knocking in contrasting areas.The group contained 34 people and achieved broad representation of the UK population by gender, age, family stage, ethnicity, work status and social class.All attended the full event and received a small financial incentive.We also carried out a brief structured survey of the participants' attitudes to welfare state issues using questions taken from the 2008 ESS questionnaire immediately before and after the DF meetings to identify areas in which opinions shifted as a result of discussion.The meetings took place over two days spaced two weeks apart.They were a mixture of plenary sessions and break-out group discussions in three groups of 11 to 12 participants to facilitate interaction.Prior to the event, participants were informed that the overarching question to be debated was the following: What should priorities of the UK government in this country be for benefits and services in 2040?
This question was phrased in general terms to facilitate broad discussion.A brief explanation of the full range of services and benefits provided by the welfare state was given.
The first one-day meeting consisted of a naïve discussion of the welfare state with no prior stimulus, so that participants could discuss and formulate themes which they considered most important.The five themes selected were (in preference order): immigration; resources and sustainability (which in practice led to discussion of inequality and redistribution); population ageing; unemployment (which led to wider discussion of labour market issues); and poor access to education and opportunity (which many participants related to immigration, labour market issues and weaknesses in government policy).We then departed from strict DF protocol to add the topic of gender issues to the discussion, because we had agreed that this would be a common theme in the cross-national study mentioned earlier.In practice, and surprisingly, most gender issues (other than issues surrounding lack of childcare and women's access to employment but including the gender pay gap) were seen as relatively unimportant by participants, despite their prominence in academic and policy debate.
The group was given the opportunity to request further information at the end of Day One.It asked for factual material on immigration, resources and public spending on welfare, unemployment, population ageing and access to educational opportunities.This was gathered from ONS sources, distributed by email between the meetings and introduced and discussed at the beginning of the second day.
The second meeting took place after a two weeks' gap, to give participants space to reflect on the issues.Day Two was initially structured around the five themes which formed the basis of the comparative study: immigration; income inequality; population ageing; unemployment and the labour market; and gender.The other issues identified by UK DF participants (resources and sustainability and education and training) emerged in the discussion of inequality and the labour market in any case.The themes of the research project corresponded fairly closely to those generated spontaneously in the first day forum, apart from gender issues, but the stress by participants on resource problems and education as a pathway to opportunity may indicate strong concern about the former and high priority for the latter.The participants were asked to formulate policy recommendations on each of these themes, to ensure the discussion remained focused.Interactions were audio and video-recorded, with additional note-taking by three observers, so that all statements could be traced to specific individuals.
We coded the data in Nvivo starting from the five issue areas identified by participants, and then extending through iteration on the basis of topics emerging in the discourse.The questions of how a particular need should be addressed (by the individual, family, community, employer or the state), the extent to which people approved or disapproved of particular policies, the sources of evidence referred to, the justification for a particular argument, the level of conflict in the group and the extent of attitude change were included, making a final 21 codes.

Findings
The DFs provide a picture of attitudes to major changes facing the UK welfare state as our participants understood them and their desired policy responses.They also show how attitudes shifted (or failed to shift) as people discussed the issues.This is brought out in the comparison of the before and after surveys which summed up people's individual views either side of the DF experience.The most noteworthy shifts are in relation to income inequality (more find it acceptable), welfare state financing (support for NHS and pensions but less confidence in sustainability) and immigration (stronger anti-immigrant sentiment).Attitudes correspond broadly to the BSA.In the first two areas broadly speaking they reinforce the neoliberal agenda: majorities emphasise the work ethic and individual responsibility and see government as unable to provide decent services as time goes on.

Welfare state and the work ethic
Spending on those of working age and especially unemployed people was perhaps the most important issue raised in relation to the future sustainability of the welfare state.In reality benefits for unemployed people account for some 2 per cent of spending, for people on low incomes 17 per cent, for children 18 per cent, for disabled people 18 per cent and for pensioners 45 per cent (IFS, 2014).Despite the fact that these statistics were presented in the stimulus material, most people stuck to their belief that unemployment benefits make up a very large proportion of welfare spending.Many of the participants expressed strong stigmatic beliefs about benefit abuse by unemployed people, following a neoliberal denigration of state welfare.
'Every time you drive past the Jobcentre on sign-on day they are outside standing with cans of beer at 10 o'clock in the morning.' (P-51) A benefits cap (at 25 per cent below average wage levels) was also strongly supported: 'We'd achieve this by saving money on benefits and giving this back to employers, who can pay this back in the form of a higher minimum wage.

' (P-81)
There is also some support for greater social equality and a lot of unease about conditions at the bottom of the labour market.Concern was expressed about zerohour contracts (although the only participant on such a contract said she valued flexibility in working hours) and about insecurity.The best way forward was seen in individual work ethic terms of a higher minimum wage and improved opportunities (further discussed in relation to education and the leading theme when the forum turned to future directions for welfare), rather than state redistribution through tax and benefits.In the final recommendations, all agreed that zero-hours contracts should be abolished: 'people on zero-hour contracts, they're not entitled to the same things as permanent full-time employees…they don't qualify for sick pay, there's no guarantee of work, they're not paying the…tax or National Insurance…they can't get loans, they can't get a mortgage, they don't know when they're working, they have really no say.'(P-45) There was extensive discussion of competition from immigrants for jobs (see the section on immigration, below) and this was the main reason most participants wanted immigration curtailed.Most believed that UK nationals should have labour market priority over immigrants, although after discussion on practicalities this was diluted to a right-to-interview for nationals.There was support for compulsory work-experience for all school-children and, following the individualist and opportunity-centred theme of the discussion, stronger regulation of trade unions.
Structured survey respondents express similar views.BSA 2014 respondents are much less keen on benefit spending for people of working age, especially unemployed people, than they are for spending on pensions or healthcare.There is strong and increasing concern that benefits reduce work incentives.Fifty-two per cent of BSA respondents believe that benefits for the unemployed are too high and discourage work while 27 per cent believe that they are too low and cause hardship, and, perhaps most important, only 18 per cent believe that the system encourages people to take paid jobs (Taylor-Gooby and Taylor, 2015).In addition, a large majority (73 per cent) support a tight benefits cap.Similarly, in our DF before-and-after surveys, half of participants believed that social benefits and services tend to make people lazy, and 56 per cent believed that large inequalities are acceptable to reward talent and efforts.What was remarkable was that these statistics increased significantly to 71 per cent and 76 per cent respectively after the forum.There is a clear shift in attitudes towards a work-ethic opportunity society with greater inequalities and strict benefit constraints rather than a tax-and-spend redistributive welfare state.Participation in the debates strengthened the individualistic viewpoints of forum members.

Welfare state finance, sustainability and population ageing
The discussion of population ageing and intergenerational issues focused chiefly on resource issues and on concerns about the sustainability of pensions, the NHS and community care, again fitting with a neoliberal mistrust in the capacity of the state to meet needs.The cost of benefits for people of working age, which bulked large in the minds of many participants (see discussion of unemployment below), was identified (counterfactually) as the most important pressure on spending.However, it is clear that participants value highly what are in fact the major-spending state services (NHS and pensions), following a welfare state agenda.Despite this, they move to a neoliberal market-centred approach.While they regard state provision as highly, they also see it as unsustainable, as this interchange, to general assent, indicates: 'That will be gone; the NHS will be gone in five years.' (P-80) 'Yes, and that to me is just awful that we'll lose that.' (P-90) 'If we keep doing what we're doing the whole thing is going to be bankrupt…It will have to change, it's got to change, but I mean, it's the biggest gift that…I mean, it's a thing of beauty, isn't it?'(P-80) 'Absolutely, the best thing England's ever introduced.' (P-87) Similarly, in another group: 'The pensions are rubbish.'(P-46) 'I don't think there will be a state pension in 2040.' (P-42) 'The state pension probably won't exist.I think we are all in agreement on that.' (P-68) Among final recommendations was an obligation for people to pay a percentage of their wage into a private pension scheme (only one participant disagreed) and support for healthier lifestyles to reduce NHS costs.
Participants also repeatedly criticised the lack of transparency in government spending and felt the authorities should be more accountable, summed up in one group as follows: 'Somebody's given £10, £15, £20 million, whether it be council or a government department, they've got to say what they're spending the money on, where they're spending the money and what are going to be the benefits to us as, as part of the society, yes, so there's got to be more accountability… and more transparency.'(P-83) In discussion of how funding shortfalls might be met there was little support for tax increases.Tax was discussed mainly in terms of tax evasion by large multinational corporations.A significant proportion of participants believed that high earners should not be taxed more: 'I feel quite strongly, if you are bettering yourself: why should you pay twice as much tax as other people?It's your money, you're earning it.'(P-44) Most participants, however, supported pay ceilings for high earners in large corporations (with only two disagreeing).
The earlier discussion of sustainability led to recommendations for pension privatisation, but the area is clearly one of tension.In relation to population ageing, the real concern about sustainability of pensions was reflected in a recommendation from one breakout group to increase the retirement age to between 70 and 75 and sharply cut funding for state pensions, provoking substantial disagreement in the final plenary.Similarly, the idea of abolishing the state pension and making private pensions compulsory was controversial, with one of the three break-out groups arguing for retention of at least some state pension provision funded by higher national insurance contributions ('it's a safety net for everyone, you know, you need that' (P-47)).
The other two groups were in favour of means-testing and privatisation to contain the pressures on the state of an ageing population: 'It's going to be a totally different system in 25 years' time than it is today… it's just going to be impossible.'(P-86) 'It's straightforward, you've earnt this amount of money…we're not going to be paying you a pension.' (P-48) 'Yes, we've got to shift that thought process…away from relying on benefit and state pension.' (P-62) Structured survey data shows a similar picture of concern about welfare state costs, with somewhat more enthusiasm for state provision in pensions, healthcare and education.Most people in the UK think the government should be responsible for and provide generous pensions and healthcare but there is real concern about the future.BSA shows about 70 per cent of respondents supporting higher spending on the NHS and pensions (Curtice and Ormston, 2015).Seventy-four per cent believed that the NHS faces a 'major' or a 'severe' funding problem.Only around a half (48 per cent) believe that it 'will still be paid for by taxes' and be 'free to all' in a decade.
Similarly, DF participants agreed that pensions and healthcare should be a government responsibility, but were more pessimistic about the sustainability of the welfare state and supportive of cuts.The experience of taking part strengthened rather than weakened these attitudes.Seventy-four per cent thought that the NHS should be entirely a government responsibility before the DF, with 68 per cent for pensions.Corresponding statistics after the DFs were 32 and 38 per cent.Most people assumed that private responsibility would dominate.The reason for pessimism was clear: 56 per cent stated that the UK would not be able to afford current levels of pensions in 2040 before the DFs, rising to 79 per cent afterwards.
A substantial group of DF participants were uncertain about spending policies before taking part in the discussions (37 per cent in favour of decreasing taxes at the cost of cutting social spending, against 23 per cent in favour of maintaining the status quo, and the plurality, 40 per cent, undecided).After the discussions more of the undecided group shifted to support more spending, but the balance remained in favour of cuts: 47 per cent against 40 per cent, with 13 per cent neutral, reflecting the acceptance of the pressures but the valuing of pensions and the NHS.It can be seen that this is a nuanced neoliberalism -pro-welfare state in principle, but mistrustful of state capacity in practice.In general, the future of the welfare state is seen as more market-centred.In relation to work, individual responsibility is valued.In relation to healthcare and pensions -the major services for those out of the labour market -the picture is more complex.
Immigration adds a further layer to the complexity of market-centred ideas.

Immigration
Immigration, selected as the most important topic by participants on Day One, is clearly a central issue, with a large majority endorsing much stricter border controls.This conflicts with an ideal typical neoliberalism that would value a free (and presumably globalised) market in labour (assuming immigrants are seen as primarily workers not benefit claimers) but fits with ideas of individual responsibility that would include ensuring that people can protect themselves against competition from those who work for lower wages.
Most participants (who included three recent immigrants) believed that immigration is too high and (mistakenly -see Dustmann and Frattini, 2013) that current rates put severe strain on job opportunities and on housing, although some also pointed to benefits from immigration in diversity and the value of skilled workers to the economy and the NHS.Most people claimed that immigrants tended to put more pressure on the welfare state than denizens.All participants agreed that there should be a points-based system limiting immigration and this was prominent in the final policy recommendations.Potential immigrants: 'must have language, a promise of a job, be able to employ people, no health issues, no criminal record, money in the bank…Incomers need to bring something to the system.' (P-70) Participants argued for harsher policing of immigrants, for example, that immigrants should be tracked through ID cards, and that any conviction should result in deportation: 'if [immigrants] come here and they are naughty, send them home.' (P-72) A minority of participants acknowledged that economic emigration should also be taken in consideration when discussing the introduction of immigration caps, especially with regard to a potential 'brain drain' and the need for skilled migrants: 'British people are going to go and follow the money abroad so, we're going to have to get other people in.' (P-43) This argument was incorporated in the final recommendation for a points-based system, but did not affect the majority conviction that competition from immigrants was highly damaging to most people's interests.
The argument that immigrants' rights to benefit should be severely curtailed corresponds to findings from the BSA Survey: 40 per cent of the 2014 BSA sample believe that immigrants from outside the EU should never receive UK benefits and 66 per cent that entitlement should be limited to six months; corresponding attitude statistics for EU immigrants are 26 and 59 per cent (Taylor-Gooby and Taylor, 2015, Table 1).The before-and-after surveys for the DF indicate that the process of discussion shifted views against immigration.A majority (50 per cent before the event, 71 after) believed that immigrants should be granted rights to social benefits and services only after having worked and paid taxes for at least a year, and that immigrants tend to receive more than they contribute (47 per cent before, 56 after).These are substantial shifts in the group and appeared to be influenced by the arguments made by some participants which were based on personal experience of immigrant workers undercutting accepted pay rates.
The forum participants linked education and training to labour market and state responsibility in a way that casts further light on neoliberalism and the individual.

Training, education and social investment
Education and training opportunities were stressed by almost all forum participants as key to addressing problems of inequality and here the state was seen as having an important role and as able to make a major contribution.These themes emerged in relation to the topic of education and also in discussion of welfare state financing, income inequality and unemployment issues.Participants also endorsed childcare but only for those who are in paid work or contribute to society through other means such as voluntary work.They took seriously the 'new risk' (Taylor-Gooby, 2004) issues of low pay and poor job opportunities, especially for young people and women with dependent children, despite current high employment.The emphasis by DF respondents fits with the focus on individual as opposed to collective advance and the fact that redistributive policies played little part in the discussion.
Childcare was not extensively discussed as a major issue, although participants in two breakout groups called for more free or subsidised provision, including company crèches.Interestingly and again following the UK commitment to the work ethic, rather than any awareness of a potential benefit in child socialisation and early education, childcare was understood as a way to ensure that more parents were in paid work.Participants were reluctant to provide childcare for those who were not in the labour market: 'I think offer childcare to people who are working, the free childcare for people who are working, and should take it away from people.' (UK-80) 'Just take away from people who don't work.' (UK-88) Education and training is endorsed because it allows those who make the effort to improve their lives as individuals within the more unequal system with weaker state provision that most participants see as inevitable.Many of the participants referred to apprenticeships as a positive policy and an expansion of apprenticeships was prominent on the final list of priorities put forward at the end of the DF, to unanimous agreement: 'if you have got the education and [do an] apprenticeship, you are getting paid, you feel like you are doing something and something is going back into society.'(P-68) And also to encourage employers to develop human capital: 'so invest in people, you know, see where their needs are, education, employment.They want to be more skilled, experienced.Like, you know big companies out there tell them to give people opportunities to develop themselves.' (UK-46) Such policies are not extensively discussed in the structured welfare state surveys, possibly because this area is often categorised by academics separately from the welfare state although policies in it influence welfare outcomes.We did not anticipate the consistent emphasis on it across almost all DF participants and it was not included in our before-and-after surveys.

Discussion and conclusion: towards 'reluctant individualism'?
This brief review of some of the most striking findings from the extended discussions in our DFs casts new light on what we know from structured surveys.Most people value the big-spending state services (especially the NHS and pensions) highly, but also stigmatise working-age benefit claimers and endorse individual responsibility.Two factors link these ideas to what might be seen as neoliberal individualism: most people misunderstand costs, exaggerating the scale of spending on workingage benefits over that on healthcare and pensions, and they mistrust the capacity of government to provide decent healthcare and pensions in the future, so that there will be little alternative to much greater individual responsibility in these areas too.
The findings also move beyond what structured surveys indicate in two ways: attitudes strengthen in both the areas mentioned above as a result of discussion.In addition, the widely-studied welfare chauvinism, that has been an important feature of UK political debate in recent years, is strengthened in a way that conflicts with pure free market neoliberalism but fits with individual determination to provide for self and dependents through paid work.
The impression of individual responsibility so far as the world of work goes is further reinforced by another finding: the fact that collective approaches to social issues are largely absent from responses in the forums.Ways forward mentioned in the more than 40 hours of discussion included some state regulation (in relation to immigration, stricter benefit regimes, enforcement of tax rules on multinational companies, requirements for employers to provide pensions, employ or interview UK nationals, and provide good affordable childcare, and banning zero-hour contracts).There were very few references to positive interventionist policies involving higher tax to expand state services or finance redistribution.Strikingly, there was no mention of other collective institutions such as trade unions or local government or even the voluntary sector as potential providers.These institutions tended to be seen as obstacles to the pursuit of individual opportunities.
In relation to the four research questions listed earlier (to do with the extent of individualism in UK welfare attitudes, the relationship with immigration attitudes, the particular issue of individualism and mass services and expectations for the future), the DFs endorse but build on existing knowledge.There is a committed and often enthusiastic individualism in labour market attitudes.People should be responsible for earning a living and too many people are seen to be work-shy, something which welfare benefits are believed to encourage.Immigrants are often an extra burden on the welfare state and in any case generate unfair competition for those who strive honestly to support themselves in the labour market.However, the individualism which sees a much greater role for personal responsibility and privatisation in relation to the NHS and pensions is less whole-hearted.Most people in fact regard state provision as highly desirable, but think that the state will not be able to sustain it.This move towards a regretful individualism we term 'reluctant individualism', a converse to Keynes's 'reluctant collectivism' (George and Wilding, 1985).
The DFs also take our understanding of neoliberal individualism in a more positive direction.Endorsement of individual responsibility extends to a demand for state engagement in helping and supporting people into decently-paid work through education and especially training summed up in the keenness with which apprenticeships were endorsed throughout the discussion, emerging as the area of strongest consensus in the final plenary session.This links up with support for childcare (limited to families who contribute back to society through paid or voluntary work) which follows the logic of 'social investment' (Morel et al, 2012).It also suggests a shift in welfare effort away from older people (the current lion's share of welfare state spending) to younger age groups, provided they are actively pursuing or are in paid work.
The DF approach contributes to research on welfare state attitudes in three ways.It adds to the range of methods available and offers opportunities to examine how people link ideas and frame issues conceptually.Second, it enhances knowledge by taking forward that understanding of frameworks and in particular by showing how individual views in areas of initial majority opinion tend to strengthen through discussion to form a larger and more consensual majority view.Third, it helps develop theory, in this case by revealing some of the complexity of broadly individualist attitudes.Alongside the market individualism that promotes a work ethic lies a more reluctant individualism to do with the perception that state welfare faces major challenges and that individual responsibility in some areas is the only way forward.In addition, the individualism extends to the rejection of labour market competition from immigrants contra the principles of a free market.Also, most people see a role for government in regulating multinational corporations and developing active programmes to enhance opportunities for those who pursue them through interventions in education and training to improve prospects for workers and especially for younger people.