ESRC RESEARCH GROUP ON CARE, VALUES AND THE FUTURE OF WELFARE

University of Leeds

Workshop Paper No 11
Prepared for Workshop Three
Analysing Policy Change and Implementation
Friday 11 February 2000

Simon Duncan

NEW LABOUR'S 'RATIONALITY AND MORALITY MISTAKES' - AND SOME ALTERNATIVES

1. Introduction: family law, social behaviour and government intervention

Following its landslide election victory in 1997, the 'New Labour' government in Britain has taken the express aim of changing social behaviour as part of its project of 'modernisation'. A major part of this drive is to use legislation to sustain and induce particular types of partnership and parenting, and to discourage other, less favoured, forms. This is because, as Tony Blair put it in a 1996 speech, while 'family values' are the key to a 'decent society', there is a 'moral deficit' which leads to an 'indifference to the undermining of family life' (Blair, 1996a). In this view government should take action to inculcate appropriate values and so 'rebuild social order and stability' (Blair, 1996b). The New Labour version of what this means in policy terms has now been codified in its 1998 Green Paper Supporting Families.(1)

I examine the validity of this enterprise in terms of its underlying assumptions about social behaviour and economic decision making, and how this might be influenced through legislation. I argue that in Supporting Families the government implicitly assumes a universal model of 'rational economic man' and his close relative the 'rational legal subject'. In this view, people take individualistic, cost-benefit type decisions about how to maximise their own personal gain. Change the financial and legal structure of costs and benefits in the appropriate way and then people will modify their social behaviour in the desired direction. Alternatively, people may make sub-optimal decisions where they lack information about this cost-benefit structure. In this case simply providing better information, or educating them so that they can access it and act upon it more effectively, will have the desired social effects.

But what if this is not a correct version of how people make important decisions about their moral economies - about how partnerships should be formed, sustained and dissolved; how parenting should be carried out; how this might be combined with paid work; and about who does what sort of paid and unpaid work? Certainly the assumption of rational economic man, and the theories of neo-classical economics which underlie it, have been vigorously challenged both in general and in their application to complex social decision-making (See Workshop 1 paper by myself - 'Challenging rational action theory'). Similarly, recent empirical research suggests that people do not act like rational economic man in making decisions about their moral economy (See workshop 2 paper by Bren Neale ' The proper thing to do? Theorising family, kinship and social change: a qualitative approach'.) Rather, people seem to take such decisions with reference to moral and socially negotiated views about what behaviour is expected as right and proper, and that this negotiation, and the views that result, varies between particular social groups, neighbourhoods and welfare states. These decisions are not simply individual, therefore, but are negotiated in a collective way. Calculations about individual utility maximisation, and in particular perceived economic or legal costs and benefits, may be important once these understandings are established, but are essentially secondary to such social and moral questions. Decisions are still made rationally, but with a different sort of rationality to that assumed by the conventional economic and legal model.

My point is, therefore, that people may take decisions on parenting, partnering and work on quite different grounds to that assumed by the New Labour government. If people do not act according to the model of rational economic man and the rational legal subject, then legislation based on such assumptions might well be ineffectual. This is what I have labelled the 'rationality mistake'. The proposals in Supporting Families may be one example. At worst, for instance if a response to this weak effect was to introduce compulsion (as in fact appears to have happened with the New Deal for Lone Parents), then such policies could force large numbers of people to do what they consider morally wrong. Quite apart from the ethical implications of such policy, it would probably still be inefficient. This is what I call the 'morality mistake'. The fate of the Child Support Act introduced by the Conservatives in 1991 gives a good example. Will New Labour - despite a different political vocabulary - fall into the same trap?

Section 2 of this paper examines Supporting Families in this light, Section 3 reviews recent empirical research on how people do make decisions about their partnering and parenting, and - concluding that New Labour is in danger of compounding a 'rationality mistake with a 'morality mistake' - Section 4 goes on to sketch out some alternative policy directions. For more detail see Barlow and Duncan 2000a, b (sections 2 and 3) and Duncan and Edwards 1999, ch. 9 (section 4).

2.The 1998 Green Paper: supporting families or a 'rationality mistake'

Communitarianism offers New Labour a 'third way' between the neo-liberalism of the new right and the supposedly outmoded social democracy and Marxism of 'old Labour'. Individuals are not seen as asocial creatures but as being and acting through their contextual social experiences and relations - where the communities in which individuals are embedded are especially important.. Markets, therefore, are not the natural basis of individual behaviour but are just one more social institution. This perspective has a strong normative tone, where community is regarded as a 'good thing'. Individuals who are 'socially excluded' will not behave in the 'right way', and communities that are fragmented or dysfunctional will produce such individuals. Hence the community as a set of institutions should be fostered and supported (see Driver and Martell 1997).

Figure one around here

Economic efficiency is interwoven with both social cohesion and social morality in this model of society - rational economic actors need a community base (see figure 1). For example, launching the Social Exclusion Unit in his first major speech outside parliament since the 1997 election, Tony Blair (1997) talked of a growing underclass of unemployed young men and young single mothers, and the need to bring this 'new workless class back into society and into useful work'. It was, he went on, 'an offence against decency that work should be allowed to disappear .... to be replaced by an economy built on benefits, crime, petty thieving and drugs'. 'Work' in this speech simply means paid work, and 'society' seems limited to the employed. Receipt of benefits is not only set alongside criminality, but is also held to mean both economic inactivity and personal idleness. Family practice, or rather 'the family' as an ideal form, therefore comes to play a pivotal role (see Duncan and Edwards, 1999). But proper family values, the key to a 'decent society' are in doubt given all the changes in families where there is increasingly a 'parenting deficit' - or so communitarian discourse goes. Mothers are out at work. Fathers may be absent. Hence children are left without moral guidance or emotional support. This is exacerbated by a 'moral deficit' where, as Blair (1996a) claims, there is 'indifference to the undermining of family life' . Moreover, it is increasingly difficult for government to manage family life, where partnering, and even parenting, increasingly takes place outside formal marriage. Some commentators even see New Labour's interest in parenting as approaching the status of a 'moral reform crusade' (Coward 1998); certainly the House of Commons Parliamentary Group on Parenting sees a national strategy for parenting as laying the foundations for 'social responsibility and self-discipline' which would 'promote important social objectives' (The Guardian 9.6.98).

The Conservative government's 'back to basics' campaign during the 1990s became quite open in proclaiming a moral agenda in favour of the traditional family, and it contained a strong element of vilification of other family forms, most notably lone motherhood (see Smart and Neale 1998, Duncan and Edwards 1999). In contrast New Labour proclaims moral tolerance. Nevertheless, it still firmly states that marriage is the ideal state and that living with two biological and preferably married parents is the best for children. This perspective has again been evident in the furore over repealing the infamous 'section 28'.

This somewhat contradictory position reflects a paradox set up by New Labour's version of communitarianism. On the one hand there is a supposed parenting deficit, but on the other hand all adults below pensionable age have the ascribed duty to take on paid work. Traditional marriage, with two parent married families, seems to offer the best way of dealing with the contradiction, for this is the family form that best facilitates the combination of parenting with paid work. Lone motherhood, in contrast, epitomises the contradiction between paid work and parenting - there is less disposable time for one parent to achieve either at adequate levels. The intricacies and ambivalences of step-parenting, cohabitation and all the other 'new family forms' just complicate matters and in any case are seen as more likely to lead to family breakdown. In the terms of Figure 1, parenting by both biological parents who are also married is therefore the best and most efficient family form in linking social morality, social cohesion and economic efficiency.

Supporting Families makes proposals to operationalises this preference, and this is to be achieved through changing the financial and legal parameters under which parents, as rational actors, are assumed to operate. All in all, the Green Paper asserts, these proposals will be good for parents, children, business, the economy and society (para 3.19). The Green Paper itself is particularly significant as the first cross-government social policy wholly conceived and developed under Tony Blair's leadership, and as such lies at the core of New Labour's values (cf.Travis, 1998: Wintour 1998). For all these reasons it can be taken as a good indication of the government's intentions, and I will therefore use Supporting Families to allow an examination of government assumptions about social rationality and morality in family practices, and how these should be linked through use of the law .

I will not comment on here on all the proposals in the Green Paper (see Barlow and Duncan, 1999), except to note that it is permeated by a neo-classical 'carrots and sticks' view of family decision making . Supposedly sub-optimal behaviour is the result of a lack of information, or a lack of ability to use it properly. Change this ('change the culture' as the Green Paper puts it in para 1.20) and more optimal behaviour will result - in this case desired parenting or labour market behaviour. However, particular groups, such as the more disadvantaged in problem council housing estates or the sons of lone mothers, need to be educate, instructed - at at times coerced - as to what the right information is. At times this seems to lead to a 'blame the victim' approach to social problems. The activities of employers, for example, in providing low paid or insecure jobs are not questioned. Rather, employers are simply encouraged, in rather unspecific terms, to introduce 'family-friendly employment', based on voluntary cooperation. No coercion here for those with anti-social behaviour! (2) The point for us here is not only that this policy emphasis can be seen as resulting from New Labour's prescriptive and moralistic version of communitarianism, one which emphasises individual responsibility at the expense of socio-economic reform. Rather, this emphasis becomes naturalised where the sovereignty of individual preferences and behaviour is an axiom of the neo-classical version of social behaviour. This is, after all, the very foundation of rational economic man and, from this starting point, it does indeed make little sense to see the origins and causes of social problems lying in wider social conditions, still less in the actions of employers and firms.

The guts of the whole project lie in chapter 2 of the Green paper, on taxes, benefits and employment, and in chapter 4, on strengthening the institution of marriage. Where the latter maps out a basic response to family change, and is the section of the Green Paper with most novel ideas, the former uses the financial levers available to the government for implementation.

And in chapter 2 it is the The New Deal for Lone Parents, already piloted in test areas and indeed already implemented nationally from October 1998, which is most significant. I will therefore take this part of chapter 2, as well as the proposals in chapter 4, as case studies of New Labour's family policy.

The New Deal for Lone Parents

From October 1998 lone parents on income support, whose youngest child is over school age, have been invited to an interview with a personal advisor in their local job centre in order to discuss finding paid work. This is to be backed up by improved daycare and information services. A basic premise - as cited in para 2.16 of the Green Paper - is the received wisdom that 85 per cent of unemployed lone mothers (the vast majority of lone parents) say they would like paid work if practical problems could be overcome.

The New Deal for Lone Parents has already been piloted in eight labour exchange areas, starting in the summer of 1997 . By December 1997 ministers were claiming it as an outstanding success, in the furore following the cutting of lone parent benefit, when as many as a quarter of those participating had found jobs (Marsh, 1997). This claim is repeated in the Green Paper (para 2.18), while the latest figures for the pilot areas suggest that almost a third of participants had found jobs by April 1998. The trouble is that these figures refer only to those lone parents who participated. In fact little more than half the target group was successfully contacted and, of this half, only a quarter actually participated and attended an interview. In other words, as Table 1 reveals, fully 74 per cent of contacted lone parents did not respond. This leaves just 4 per cent of the target group actually finding work through the New Deal, a figure that is well within the range for those who would have found jobs anyway, These uptake figures are not only low, they are very much lower than the 85 per cent of lone mothers quoted in the Green Paper as wanting paid work.

Table 1 The New Deal for Lone Parents: take up in the pilot areas by 24 April 1998

Source: calculated from Hansard Written Answers, 16.6.98.

What are the reasons for this extremely low uptake rate? It could be argued that lone parents face severe practical problems in finding day care. In part answer to this sort of reasoning, the government has increased investment in after school clubs and also the 'disregard' for daycare expenses in calculating the amount of family credit to be paid - although early evidence suggests that few lone parents have gained (Daycare Trust, 1997; Roberts 1998). This is to be replaced by a daycare tax credit when WFTC comes into operation. This daycare constraint, however, would mostly affect the uptake by those lone parents actually participating in the scheme for whom, as we have seen, employment uptake is 32.1 per cent (see Table 1) -perhaps reasonable enough if still far below the 85 per cent supposed potential. But as Table 1 shows the greatest drop-out rate resulted from the non-response of lone parents - 74 per cent of those contacted did not attend an interview.

Why have most lone parents simply not responded to the apparently reasonable offer of help in finding jobs? For this non-responding majority, it could be that their knowledge of likely wages, usually low (because of low hours worked and/or low wage rates) and often insecure, coupled with the extra costs of travel to work, school meals, and day care, mean that they see the interviews as just a waste of their scarce time (cf Duncan and Edwards 1999).

This explanation (as with the previous point about the scarcity of day-care) would fit into rational economic man type explanations for lone mothers' behaviour. The problem is rather that the government has underestimated the level of constraints and overestimated the quality, rewards and availability of jobs. Lone mothers' own cost-benefit analyses, therefore, would usually have a different outcome to that imagined by the government - it would remain most rational, in this neo-classical sense, for them to remain on benefits.

Government spokespersons tend to assume that it is lone mothers who have got their equations wrong, and this is why they need persuasion to access better information. This government response seems to neglect, however, the fact that most jobs on offer will not provide an adequate household income. Thus what seems in government eyes to be a move away from 'dependency' (depending on income support) may, in lone mothers' eyes, offer little improvement in income levels or security. In fact this situation will also have a limited effect on lone mothers benefit dependency. For many lone mothers getting jobs through the scheme will simply swap dependency on income support (the out of work benefit) to dependency on family credit (the in-work benefit). According to the House of Commons Select Committee for Social Security this transfer may even increase social security spending (The Guardian, 17.2.98). In total, an evaluation study estimates that the New Deal pilots in fact helped only 1-2 per cent of lone parents in the pilot areas to move off Income Support by March 1998 (Hales et al 1999, for a follow up evaluation see Finch et al 1999).

There is another, and more fundamental, possible problem with the government's assumptions, however. It may well be that lone mothers are employing a different sort of rationality to the neo-classical model implicitly assumed. According to recent research, practically all lone mothers lone mothers see their moral and practical responsibility for their children as their primary duty and that for many (although not all) this responsibility to be a 'good mother' is seen as largely incompatible with significant paid work (Duncan and Edwards 1999, Standing, K 1999 ,Von Drenth et al 1999). Interviews with job advisors would then be at best an irrelevance, and at worst a threat. However, it is paid work that New Labour erects as a moral duty, not the unpaid caring which most lone mothers place first. This, I contend, is a 'rationality mistake'. Lone mothers make decisions about taking up paid work on quite different grounds to that assumed in the Green Paper.

By February 1999 Tony Blair was quoting unattributed figures of 80 per cent employment take-up for lone parents attending New Deal interviews, but fully 94 per cent of those contacted did not attend an interview in the first place (Daily Mail 10.2.99). (We might imagine that the 4.8 per cent of the total taking jobs might have done so anyway,). This massive non-response has not brought about, it seems, any reexamination of the behavioural or economic assumptions in the New Deal for Lone Parents. Rather, the government is to force its own assumptions upon lone parents. Attendance at job interviews is to be made compulsory, as part of the Welfare Reform and Pensions Bill, with complete loss of benefits as penalty for non-compliance. According to Alistair Darling as Social Security Secretary, this 'harsh but justifiable measure' is necessary to confront head-on lone parents' 'poverty of ambition and poverty of expectation' (The Guardian 11.2.99) . More generally, as Tony Blair wrote in a pre-publication 'softener' in the Daily Mail (10.2.99), the message to claimants is that 'If you can work, you should work'. However, as recent research shows, lone parents in fact have considerable ambition and expectation for their children, and they undertake considerable work in trying to achieve this, albeit unpaid caring work which leaves them formally 'unemployed'. In forcing its own version of rationality upon lone parents the government risks making large numbers to do what they consider morally wrong. Rationality mistake is compounded by a 'morality mistake'.

Strengthening Marriage

Chapter 4 of the Green Paper maps out a basic response to family change. On one level, it claims that intervention aims to help the parenting relationship - whether married or not - to succeed. In any case, government competence is limited where 'families do not want to be lectured about their behaviour or what kind of relationship they are in' (para 4.2). Yet at another level, the Green Paper states that the government's preferred parenting structure is marriage. As the preamble makes clear:

'marriage does provide a strong foundation for stability for the care of children. It also sets out rights and responsibilities for all concerned. It remains the choice of the majority of people in Britain. For all these reasons, it makes sense for the Government to do what it can to strengthen marriage.' (para 4.8)

What is more, the vast bulk of the chapter - which indeed is entitled 'Strengthening Marriage' - is concerned with how marriage can be supported and encouraged. Other possible partnership and parenting forms are hardly mentioned - despite the fact that in 1996 21 per cent of children were born to cohabiting parents, with another 14 per cent born to lone mothers. Both figures are increasing, and cohabitation is predicted to double by 2020. Yet only about half a dozen of the 49 paragraphs could have much relevance to such parents, and only three consider cohabitants. Nothing at all is said about same-sex parenting. What can the Green Paper say to all these parents other than 'get married'?

The chapter goes on to propose a number of measures to strengthen marriage. These include better preparation for marriage, including a clear statement of rights and responsibilities, prenuptial agreements about the distribution of money and property, an enhanced role for marriage registrars in providing premarital counselling, modernisation and personalisation of the civil marriage service, access to mediation and counselling to support marriages in difficulty, and better information before divorce so as 'to increase the chance of saving more marriages' (para 4.12). Clearer rules on property division on marital breakdown are proposed to reduce conflict between married couples.

In contrast proposals affecting cohabiting families are limited to just two suggestions. First, is the introduction of a non-religious and public child-naming ceremony which may also be used to stage the public signing of a parental responsibility sharing agreement, where parents are unmarried. This is designed to encourage public assertion of both parents' commitment to a child, whether or not they are living together. Second, the Green Paper rather grudgingly suggests that 'it might therefore be worthwhile' to produce a guide for cohabitants setting out their legal rights in relations to income, property, tax, welfare benefits, and responsibility towards their children, to be made available in Citizens Advice Bureaux and libraries (para 4.15). This does nothing to address the complexity and inadequacies of the law relating to cohabitation. While enforceable prenuptial contracts for those intending to marry are proposed, the Green Paper is silent on the issue of legally enforceable cohabitation agreements. Nor is there counselling to save cohabitation relationships, in sharp contrast to the proposed efforts to be invested in marriage-saving. The Green Paper therefore fails to acknowledge, yet alone address, the need for better family law-based regulation of cohabitation relationships. And for lone parents, as we have seen, the discussion is hardly about parenting at all (save for the negative assumption that their sons will lack male role models). The Green Paper simply sees the issue as getting lone parents into paid work. In terms of policy discourses (Figure 2) lone mothers and cohabitees are placed as social threat, and social problem. Only if cohabitees or lone parents marry will they be rewarded with the legal protection and government support they and their children need.

Figure 2 around here

The means of implementing this discourse, of strengthening marriage and reducing the importance of other family forms - or at least the threat they pose, is seen in terms of rational economic man and the 'rational legal subject'. The government appears to believe that changing financial and legal parameters, as in the Green Paper, will thereby alter the calculus for people's decision making about partnering and parenting, and therefore in turn lead to the desired changes in behaviour. More lone parents will take up paid work, more couples will marry, less will cohabit and fewer will divorce. The problem is that the basic assumption about how people do make decisions about their moral economies - about how partnerships should be formed, sustained and dissolved; how parenting should be carried out; how this might be combined with paid work; and who does what sort of paid and unpaid work - might be incorrect. The whole enterprise might then become irrelevant - or even oppressive - because of this 'rationality mistake'.

In the next section I will review recent empirical research on (1) how lone mothers make decisions about combining care for their children with paid work and, (2) how couples decide whether to cohabit or marry.

3. Moral rationalities and family decision-making

Family sociology has become more interested in what families actually do, rather than - as before - what they ought to do, or are assumed to do, and how deviant families and family members can then be seen in this light. As David Morgan (1996) has put it, the focus is now on 'family practices' rather than 'family problems'. Among other things, this shift has meant new research and knowledge about how people make decisions on how to conduct their family life. Janet Finch's and colleagues' work on family obligations has become something of a formative classic in this area ( Finch 1989, Finch and Mason 1993), and work by several CAVA members has also been formative. The empirical focus was how notions of moral obligation and responsibility between kin might be changing in the context of rapid changes in family structures. The question of moral values, of how they are formed and how they inform action, was then brought back into empirical family sociology. At around the same time, theorists pointed out that 'ordinary people' do not have to be versed in the intricacies of moral philosophy in order to act morally or form moral judgments, and that moral and ethical reasoning are 'everyday social and textual practices' (eg Bauman 1993, Sevenhuijsen 1998). Moral decisions are not just the preserve of philosophers, religious leaders or politicians, but form the basis of everyday life.

It then follows that how people actually make these moral decisions become crucial in understanding social behaviour and change. As research has shown, such decisions will also vary according to the different contexts of social groups, social places and social histories. Certainly this has important implications for the construction of social policy, at least in terms of its efficiency if not ethically. Put baldly, and with hindsight, this might seem obvious. It was just that much social research on families and decision making assumed people were either the passive respondents to external stimuli, or rational economic men simply making decisions in terms of personal costs and benefits. In either case, they were fairly uniform. Exceptions were in some way deviant. The problem is, New Labour still seems to assume this.

What, then, does this sort of research say about the social behaviour problematised in the Green Paper? What sort of moral decisions do people make about partnering and paid work? I draw here on separate studies on (i) lone mothers' decision making about paid work (Duncan and Edwards, 1999) and (ii) the decision to cohabit or marry (Barlow, 1998).

A major result of the first study was that lone mother's decisions about whether to try for a job or not is primarily influenced by 'gendered moral rationalities' - that is, their socially negotiated understandings about the proper relationship between good motherhood and paid work. These rationalities provide answers to, or guidance on, questions such as ' Is it right that I, as a mother bringing up children by myself, should try for a full time job? What are my responsibilities, how will my behaviour affect my children? What do others expect of me, what do they see as right and how will they treat me as a consequence?' All the interviewees in the study (65 in Britain, and 30 in Germany, Sweden, and the USA) experienced lone motherhood in terms of responsibility towards their children but - crucially -their gendered moral rationalities differed in terms of how this basic responsibility was best discharged .

Some lone mothers, who subscribed to a 'primarily mother' gendered moral rationality, gave primacy to the moral benefits of physically caring for their children themselves, based at home. Children's and mother's needs overlapped, and these mothers saw paid work as contradictory or even inimical to this moral responsibility to be good mothers. Others, with a 'primarily worker' gendered moral rationality, gave primacy to a right to paid work for themselves as separate to, and autonomous from, their motherhood. Finally, another group of lone mothers held a mother / worker integral gendered moral rationality. They saw paid employment as part of their moral responsibility to their children as good mothers, providing them with both financial security and a good role model. For all the mothers, these moral decisions were not simply established individually, but rested on on moral, emotional and practical support - and pressure - from others. Impressions about the financial costs and benefits of taking up paid work or not were still important, but remained secondary to this essentially moral and social evaluation of what was the right and proper thing to do for them and their children.

These different gendered moral rationalities were associated with different social groups living in different types of neighbourhood, as figure 3 shows (4). In Britain, both White middle class lone mothers living in suburban areas, and White working class lone mothers living in peripheral social housing, were likely to hold a 'primarily mother' gendered moral rationality - despite the considerable class differences between the two groups. Those lone mothers holding 'alternative' views of families and gender roles, usually informed by feminism and living in gentrifying areas, were most likely to hold a 'primarily worker' gendered moral rationality. Finally, it was Black lone mothers, mostly living in large council estates in inner city areas, who tended towards a 'mother / worker integral view'.

Figure 3 around here

The generality of these gendered moral rationalities was backed up by analysis of the 1991 household Sample of Annonymised Records (SARs) from the census. This showed that for any given level of constraints or human capital, Black lone mothers usually showed higher employment rates and, in particular, higher rates of full-time work. Indeed, the least resourced and most constrained Black lone mothers (those living in council housing, aged under 30, with more or younger children, with lower educational qualifications and so on) often held employment positions similar to the most resourced and least constrained White lone mothers. The most resourced and least constrained Black lone mothers moved towards the (White) 'male norm' of continuous full-time work. Analysis of the longitudinal study (LS) census data for 1981-91 showed that these differences were also stable over time. For example, Black lone mothers who had (re)partnered by 1991 were even more likely to be in full-time paid work than in 1981. For (re)partnering White lone mothers, the converse was true; they were even more likely to be unemployed or to be in part-time work in 1991. This is all the more remarkable when we remember that most Black lone mothers live in labour market areas, particularly inner cities and industrial towns, that have generally performed worst in terms of providing jobs.

There are a number of important implications for public policy from this study. Not least of these is that those lone mothers who chose not to take up paid work, far from falling into a deviant sub-culture that abhors self-reliance and social responsibility, did so on the basis of what they believed to be the morally proper thing to do as a mother. Given this strong moral underpinning, then their decisions about their non-involvement in the labour market were perfectly rational. There views are aligned with dominant conventional views about mothering and family life, and in this way unemployed lone mothers are not 'socially excluded' or in some sense 'outside society'. In contrast those lone mothers who did prioritise paid work, or saw it as integral to being a good mother, held views that are alternative to - or even 'deviant' from - dominant conceptions, but are equally moral and rational. The mother / worker integral gendered moral rationality in particular challenges the simple mother versus worker dichotomy which has dominated social policy up to now. Furthermore, not only are each of these views both moral and rational, but they were deeply social, mediated through lone mothers experiences of being members of particular social groups in particular areas.

The limited effect to date of the New Deal for lone parents, where the vast majority of lone parents do not respond, seems a good example of the 'rationality mistake' in the light of this research. Interviews with job advisors may be at best an irrelevance, and at worst a threat. Unfortunately, the response to this weak policy effect has not been to reassess the social assumptions made, but to make interviews compulsory for those lone mothers receiving benefits. This threatens to force large numbers of people to do what they consider morally wrong.

I can call this response to policy inadequacy the 'morality mistake' and it appears on two levels. A first stage is to assume that people are not behaving 'rationally' (in terms or rational economic man and the rational legal subject) because of lack of information or, more pejoratively, ignorance. Hence, for example, the need for lone parents to have daycare and labour market situations explained to them. If policy is still ineffectual - as seems quite likely - then this 'morality mistake' can move to a second, more authoritarian, stage. People are not behaving 'rationally' because of their own moral or cultural deviancy. Compulsion can be justified and rationalised as the unfortunate effect of their 'poverty of ambition and poverty of expectation' or, more pejoratively, as resulting from the 'dependency culture'. Currently, government spokespersons seem to mix the two. The idea that people take what they consider to be morally appropriate decisions in their situation, and that they have worked hard in reaching these decisions in particular situations, is not considered. Ironically, in view of the theoretical claims of communitarianism, such legislation also rides roughshod over the varying community contexts in which people reach such decisions. This response seems based on an equally mistaken moral view - the 'morality mistake' - that lone mothers are wilfully irresponsible or morally inadequate. Ironically, in view of the claims of communitarianism, it also rides roughshod over 'community' norms about parenting and paid work. economy, and may go badly wrong.

The second study focusses on beliefs about cohabitation compared to marriage, and is based on a small pilot survey of 30 mothers (eleven married, eleven cohabiting and eight lone mothers) in the contrasting social and labour market areas of Great Yarmouth in Norfolk and Merthyr Tydfil in South Wales. (Current research funded by the Nuffield Foundation will build on this by means of both a national survey undertaken by the National Centre for Social Research, and by in-depth interview. See also Smart and Stevens 2000 for research on cohabitation breakdown).

Almost two-thirds of the interviewees saw marriage as an ideal family form, in that it symbolised stability and commitment . Interestingly, this included the majority of cohabitees as well as half the lone mothers. This ideal view parallels, therefore, the view of marriage taken in the Green Paper. However, and crucially, respondents took a different view of the moral reality of their own situations. Thus all the cohabitants had considered marriage, and all indicated that most people assumed they were married and that no stigma was attached to cohabiting. But they had rejected marriage largely because they thought it made no difference to the success of their relationships and / or they had previous bad experiences of marriage. (They also -inaccurately - believed this rejection had no legal implications). Indeed, around half of these respondents actively saw marriage as in some way threatening to their relationship, because it would change their partner's behaviour for the worst. (Lone mothers saw marriage more of a source of unhappiness and disappointment). A smaller group of cohabitants (four of the eleven) did want to marry, but saw cohabitation as a trial marriage. While these mothers saw the cost of a 'proper marriage' as a disincentive, they did not doubt the validity of cohabitation as a partnering and parenting form. Marriage was again more of an ideal rather than some superior family form in practice. At the same time, few of the married mothers had actually got married because of its ideal characteristics, and around half had done so because of their wider social position in terms of religious beliefs or pressure from partners or parents.

In this way most respondents saw the ideals of marriage as just that, an ideal not obtainable in practice. For many mothers therefore, and particularly the cohabiting and lone mothers, cohabitation was seen as equal to, or even superior to, marriage. These views are not acknowledged in New Labour's 'social threat' / 'social problem' view of unmarried families as replicated in the Green Paper (section 2).

The practical advantages of marriage given by respondents, whether married or not, are particularly illuminating. These do not refer to the superiority of marriage for partnering and parenting as supposed in the Green Paper. Rather, they referred to marriage as a social symbol. This symbolism was to be achieved in two major ways - through a change of name and through a full-blown 'white wedding' in church.

The desire by cohabitants to have the same surname as their children and partner was cited as a major reason for marriage, and this had been a major reason for marrying given by four of the five mothers who had previously cohabited. It was the birth of children which commonly predicated this move. Conversely, most of the cohabiting mothers saw having a different surnames to their partner and children as the greatest disadvantage of not marrying. (Two had formally changed their surname to their partner's and another two families had all adopted double-barrelled names). Female name-changing is of course not a legal requirement, but is rather a powerful tradition. Presumably, this is taken as a social signifier of a 'proper family', one which follows accepted gender norms about roles and responsibilities - this is the very same reason why name changing is actually rejected by many professional married women and by those with 'alternative' feminist views.

It was also clear that the cohabiting mothers were not prepared to marry in a simple register office wedding. If they were to marry, it was on condition that they had a full-blown white church wedding. It was the wedding as a social display and symbol, not the institution or ideal of marriage as a partnership or parenting form, which was endowed with significance in the context of their lives. This is dramatically underscored by the fact that eight these mothers had actually refused their partner's offer of marriage in a Register Office! Those cohabitees in the 'trial marriage' group fully accepted that this might mean that they never married. These were the only unmarried respondents who indicated that financial incentives would have a decisive effect on their decision to marry - but only if this enabled them to obtain the highly desired 'white wedding' in church.

This essentially social signifying role of marriage was buttressed by the 'common law marriage myth'. Nearly all respondents firmly believed that the law treated cohabitants with children of the relationship in all respects as if they were married. This allowed marriage to be dismissed as 'only a piece of paper'. Although the law has not recognised common law marriage since the Clandestine Marriages Act of 1753, both married and unmarried cohabiting couples volunteered this as an acknowledged legal status. Yet this is far from being an accurate reflection of the legal position. This ignorance of the law can have severe results (see Barlow and Duncan 1999, 2000a,b) but in a wider sense the ignorance of the law it displays is quite rational. This is because couples generally see their partnership - and its strength or weakness - in terms of a relationship, not in terms of an institution

So marriage was often seen as an ideal state, but in terms of everyday moral adequacy few respondents saw marriage as a superior partnering or parenting form. It was the strength of mothers' relationship with their partner that was decisive, and this was unaffected by whether marriage had taken place or not. Similarly, marriage was seen as largely irrelevant to the welfare of children. Respondents, unlike government spokespersons, did not easily confuse partnering and parenting forms (married, cohabiting etc) with those processes (love, support, communication etc) that lead to the success or failure of these relationships. In this sense, the respondents took rather more sophisticated moral judgments than the government. People do not decide upon their moral economies according to the model of rational economic man and the rational legal subject. In this way chapter 4 of the Green Paper on 'Strengthening Marriage' perpetuates the 'rationality' and 'morality' mistakes identified in the New Deal for Lone Parents.

4. Some alternative policy directions

It is, of course, far easier to oppose than propose. What, government spokespeople might retort, are the alternatives? An overall starting point is that uniform prescriptive policies are likely to be both inefficient and oppressive. There is also an overemphasis on social forms (such as paid work, marriage) rather than social relations (working including caring, partnering, parenting) The alternative is to try to develop supportive and flexible legislative frameworks which do recognise the varying ways in which people take moral decisions. Ironically, communitarian thinking usually asserts that questions of values and morality are essentially local just because they are embedded in, and relative to, particular 'communities' (Driver and Martell, 1997). Unfortunately, New Labour's particular adaption of communitarianism is in danger of prescribing moral values that ride roughshod over this plurality. Policies for partnering, parenting and paid work need to be locally and socially sensitive.

A first corrective to the 'rationality mistake' is that welfare to work type policies need to respect local definitions of good motherhood and responsibility towards children. The issue of motherhood and paid work cannot sensibly be addressed without considering children's needs, and these needs are appreciated differently by different social groups in different places. At the same time mothers themselves, the recipients of policy, should be brought on board to allow them to express their needs and understandings, and to find some degree of control over policies that affect their lives. As research on urban regeneration has also shown (Allen, 1998), the opportunity to find some control over change is vital, otherwise policy intervention risks becoming just something that is done to you by more powerful outsiders.

Improved daycare provision is an example. Different groups of lone mothers have different ideas about what constitutes good daycare, and about how much day care is morally appropriate. The African-Caribbean lone mothers in the interview sample from inner London usually preferred formal nursery provision for the whole day. In contrast the White working class lone mothers living on the peripheral public housing estate in Brighton were more likely to see relatives as the best carers, probably part-time, if they were willing to leave their children with anyone else at all. One national response might be to extend the daycare 'disregard' to informal carers in calculating family credit. On a local scale it may be better in some areas to concentrate investment in day nurseries with professional staff, but in others to develop initiatives to support care by relatives through training, back-up professional support, and communal facilities. In other areas there may already be quite extensive community and voluntary day care facilities, which some mothers prefer as being more in tune with their own biographies and expectations, as in the gentrifying area in Brighton, and these would need different forms of support again. Furthermore, all these local daycare initiatives should be targeted at all mothers in an area, not just lone mothers. It is not only that restricted access to day care affects partnered mothers just as much as lone mothers. There is also a danger that in singling out lone mothers they will become stigmatised as bad mothers, especially in areas where the dominant conception of good mothering is to be at home, or with minimal paid work when children reach school age - as with lone mothers interviewed in the 'suburban' middle class area of Brighton (see Duncan and Edwards 1999, ch 4 for details).

There are some problems with this strategy, which also bring up more general, and quite crucial, issues about the distribution and definitions of work, caring and time. In the short term, this sort of 'bottom up' welfare to work could risk the consolidation of social and gender inequalities. Part-time informal daycare might not be as socially or educationally advantageous as full-time professionally run nursery care, and in this way class and neighbourhood inequalities might be perpetuated or exacerbated. At the same time, however, if we take moral and cultural differences seriously, then we have to give different life models equal recognition. Top-down social engineering, as expressed in the past for example by 'building away poverty' in demolishing old localities and constructing mammoth estates, is unlikely to be self-sustaining whatever the immediate material benefits. A possible way forward out of this dilemma is to emphasise that day care is for the benefit of children and in children's best interests not, as has become dominant in the British policy discourse, a device for getting mothers into the labour market. As research has shown, mothers see their responsibility to their children as of prime importance. Valuing day care as positive for children gives the best long term strategy for allowing mothers to balance their beliefs about good mothering with labour market opportunities. All too often, the much higher day care rates in other west European countries like France or Sweden have simply been equated with gender equality and women's labour market participation - forgetting that this development was also seen very much as an educational and social benefit for children. As research also shows, British lone mothers do not simply want more day care, it is the type and quality of day care, including who would be looking after their children, that is crucial. Quality and purpose are just as important as quantity.

A 'bottom-up' welfare to work strategy also risks a perpetuation of gender inequalities. For informal day care would be provided overwhelmingly by female relatives or friends, and thereby consolidate the distribution of caring work to the 'private realm' of women and families. (Even when it reaches the public sphere, formal care for young children is also a mainly female task). This brings up the question of the definitions of work and welfare. If we were to redefine 'work' as not just paid work, but also work to care for children and the elderly, and at the same time redefine welfare to include the receipt of care, then we can see that 'welfare to work' is a misnomer: 'Welfare' is not opposed to 'work', because most people receive and carry out both; welfare and work are mixed. Nor is care simply a barrier to paid work, or something that is simply necessary - both care-giving and care-receiving are meaningful and valuable in their own right. This qualification provides a powerful challenge to the whole political philosophy underlying welfare to work, that of paid work providing the basis of both economic rationality and citizenship. As Selma Sevenhjuisen (1998) points out, this latter view invokes a moral norm of self-sufficiency, and a related liberal view of human nature, which assume that people are, or should be, detached individuals whose aim is autonomous and separate behaviour. We should all be Robinson Crusoes or, if we are not, welfare to work will help us become one. In contrast, the morality arising out of care giving and receiving accounts, both philosophically and practically, for the relational and interdependent state of the human condition. Politically, this challenges the primacy given to paid work in the communitarianism of New Labour which is manifest in the New Deal for lone mothers. Instead the question becomes 'what do lone mothers need to care well for their children without this conflicting with their economic independence?'.

If we take this on board an implication is to remove the gendered and unequal valuations of full-time paid work, part-time paid work and unpaid work. These valuations do not just lead to income or resource inequalities, they also produce inequalities of time and status. Seen in this light, there is little advantage in lone mothers swapping daycare for low pay, short-time work at unsocial hours. Rather than erecting paid work - any paid work - as a moral duty it would be better to move towards moral and financial neutrality towards all forms of socially necessary work. An overall alternative, therefore, is that national 'welfare to work' should be redesigned as locally sensitive 'welfare and work'.

The argument is similar with respect to cohabitation and marriage. Rather than prescriptive policies focussing on form (marriage versus cohabitation), a more socially and communally aware communitarianism could develop policies to support partnering and parenting relations. If the government so chose, they could endow cohabitation with similar rights and responsibilities to those held by married partners. This would provide legal security and state support for both partners and children within unmarried relationships. I can briefly note here how this is in contrast to some other European countries. In Scandinavia, for instance, cohabitation and marriage have long held equality before the law, and same sex cohabitation, and more recently same sex marriage, have been drawn into the same orbit. Lone mothers are just another type of 'worker citizen' where all adults below pensionable age are treated as autonomous, and supported in taking up paid work. The great advantage is that a large proportion of parents (up to 50 per cent) are not legally and policy marginalised. In France the Jospin government has equalised the legal rights of cohabiting and married couples by introducing 'PACS' - civil union or civil solidarity contracts available to all unmarried cohabitants whether heterosexual or same-sex . The French legislation will allow such couples to opt in to legal rights akin to marriage for all purposes including social security benefits, inheritance, maintenance and property division on relationship breakdown. In contrast, the British government's Green Paper proposals demonstrate a blindness to the social significance of unmarried cohabitation.

It is essential that New Labour redefines its communitarianism along these more pluralistic, voluntaristic and redistributive lines if lone mothers are to successfully balance their responsibility to their children with the opportunities of paid work, and if cohabitants are to become more successful partners and parents..

Notes
1. Supporting Families: A Consultative Document, The Stationary Office, 4 November 1998.
2. A basic framework of maximum working hours and extended maternity and parental leave has been enacted under the Employment Relations Act 1999 , partly under the pressure of EU directives. However, the former are riven by exceptions while the latter, especially where maternity leave over 18 weeks and all paternity leave is unpaid, will effectively exclude low earners who cannot afford to take them up. These low earners will include many lone parents or those living on inner-cities that the government sees as most likely to display aberrant parenting behaviour.
3. Cambridge, Cardiff, Croydon, Halesowen, Hamilton, Sheffield, Warrington, Warwick.
4. The gendered moral rationalities were identified by taking from the interview transcripts all statements about motherhood and paid work, and children's needs, including contradictory statements within one account. Similar statements were grouped together and in this way the three main forms of relationship between motherhood and paid work were distinguished, and the triangular model in figure 2 was constructed on this basis. The position of each interviewee was then plotted as appropriate.

References
Allen, T. (1998) 'Housing renewal: doesn't it make you sick' Housing Studies,

Barlow, A. (1998) 'Family structuring, legal regulation and gendered moral rationalities; some empirical findings' paper given to Socio-legal Studies Association Annual Conference, Manchester Metropolitan University, 16 April (available from author at Law Department, University of Wales, Aberystwyth).

Barlow, A. and Duncan, S. (1999) 'New Labour's communitarianism, supporting families, and the 'rationality mistake' Centre for Research on Family, Kinship and Gender, University of Leeds, Working Paper 10.

Barlow, A. and Duncan, S. (2000a/b) 'Family law, moral rationalities and New Labour's communitarianism', Journal of Social Welfare and Family Law Parts I & II.

Bauman, Z. (1993) Postmodern Ethics, Cambridge, Polity Press.

Blair, T. (1996a) Speech to the CPU conference, Cape Town, 14 October,

Blair,T. (1996b) in Radice,G. (ed.) What Needs to Change: New Visions for Britain , London, Harper Collins.

Blair, T. (1997) 'The will to win', speech to launch the Social Exclusion Unit at the Aylesbury Estate, Southwark, 2 June.

Coward, R. (1998) 'Busybody's charter,' The Guardian, 9 June.

Daycare Trust (1997) Childcare Disregard in Family Credit, Who Gains? London, Daycare Trust.

Driver, S. and Martell, L. (1997) 'New Labour's communitarianisms' Critical Social Policy, 17, 3, 27-46.

Duncan, S. and Edwards, R. (1999) Lone Mothers, Paid Work and Gendered Moral Rationalities, Macmillian, London.

Finch, H. and O'Connor W, with Millar, J, Hales, J, Shaw, A and Roth, W The New Deal for Lone Parents: Learning from the Prototype Areas DSS Research Report 92, Leeds.

Finch, J. (1989) Family Obligations and Social Change, Cambridge, Polity Press.

Finch, J. and Mason, J. (1993) Negotiating Family Responsibilities, London, Routledge.

Hales, J Shaw, A. and Roth, W. (1999) Evaluation of the New Deal for Lone Parents: a Preliminary Assessment of the Counterfactual . In house report 42, Social Research Branch, Department of Social Security.

Marsh, A. (1997) 'Making it work for lone parents' The Guardian, 31 December.

Morgan, D. (1996) Family Connections, Cambridge, Polity Press.

Roberts, Y. (1998) 'Left holding the baby' The Guardian, 27 January.

Sevenhuijsen, S. (1998) Citizenship and the Ethics of Care, London, Routledge.

Smart, C. and Neale, B. (1998) Family Fragments? London, Polity.

Smart, C. and Stevens, P. (2000) Cohabitation Breakdown Family Policy Studies Centre, London and Joseph Rowntree Foundation, York.

Standing, K. (1999 )'Lone mothers and 'parental' involvement' Journal Social Policy 28, 3, 479-95.

Travis, A. (1998) 'Straw plays Spock' The Guardian 3 November.

Von Drenth, A, Knijn, T and Lewis J (1999) 'Sources of income for lone mother families: policy changes in Britain and The Netherlands and the experiences of divorced women' Journal Social Policy 28, 4, 619-641.

Wintour, P. (1998) 'Nannying gives Labour pains' The Observer, 25 October.

BACK