Social Problems and Public Policy

Tim Blackman & Roberta Woods. Handbook of Social Problems: A Comparative International Perspective. Editor: George Ritzer. Sage Publication. 2004.

The Decline of Public Policy

An experiment is under way in Vancouver, where residents are being invited to play a computer game exploring different policy options for the development of their region. Called QUEST and inspired by the video game SimCity, it is part of the Georgia Basin Futures Project aimed at helping the public to understand alternative futures for complex and interacting ecological, social, and economic systems (see The game enables players to generate scenarios for the region based on options chosen from a menu of goals such as full employment, more land set aside for parks, lower housing densities, or fewer deaths from vehicle accidents. The computer works with the player’s chosen options and shows how they interact, conflict, or reinforce each other when the scenario is run forward. The idea behind QUEST is to demonstrate the systemwide consequences of individual decisions, particularly the consequences for the sustainability of the system. The game’s designers do not see the main challenge for sustainable development to be a lack of policy or technical tools. Instead they see the challenge to be a lack of public understanding of how choices about single issues interrelate to produce surprising outcomes, including undesired outcomes that, if known, would alter the choices previously made.

This is a dilemma that goes to the heart of public policy. Public policy often involves conflicting policy goals, short-term and long-term trade-offs, and unforeseen consequences. Despite the advance of individualism in both political and economic spheres over the past few decades, the persistence of “social problems” continues to demand a macroperspective through which the relational and emergent nature of these problems can be understood (Emirbayer 1997). Giddens (1998) has argued that there is a moral transition involved in this: “The new individualism, in short, is associated with the retreat of tradition and custom…. We have to make our lives in a more active way than was true of previous generations, and we need more actively to accept responsibilities for the consequences of what we do and the lifestyle habits we adopt” (pp. 36-7). Social problems such as crime, unemployment, and ill health both affect private lives and are emergent social outcomes of private choices. They are, however, beyond private action alone to solve.

Robert Reich (2001) takes up this theme in his book on American economic and social trends, The Future of Success, arguing that there is an increasing divergence between the public interest and the private actions of individuals. A particular problem is the sorting mechanisms fuelled by the dynamics of the new economy. Reich argues that without strong public policies these mechanisms will drive income inequalities, school segregation, and neighbourhood polarisation to points where the resulting divided and insecure society emerges as an unintended consequence on a huge and intractable scale. His concern is that the emergent nature of society needs to be more widely understood as a consequence of decisions made in everyday life by large numbers of people. The question he puts is whether the big choices about American society are made as societal choices, through democratic institutions, or as private choices through markets. He asks us to

[i]magine that several decades ago a giant genie appeared in the American sky, offering the nation a big choice: “Either you keep the economic arrangements you have now, stay working as you are, or—do I have a deal for you! By the start of the next century, some of you will be extraordinarily rich, most of you will be better off in terms of what you can buy, and the economy will balloon. But that’s not all…. The other part of the deal I’m offering is this: Your jobs will be less secure, your incomes less predictable; there will be wider disparities of earnings and wealth; and your society will fragment.” (P. 230)

Today there seems to be less optimism that societal choices, based on a public interest that is arrived at democratically, can be made and politicians trusted with them. In the optimism of the 1960s it was possible for the Kennedy and Johnson presidencies to commit to the Great Society welfare reforms as a means, it was claimed, of achieving social integration and a better life for all. This optimism faded with the onset of world recession after the postwar boom. Political elites and the mass media in high-income Western societies asked the question, Can we afford welfare? and the New Right claimed that “excessive” and “untargeted” welfare spending was crowding out individual initiative and private enterprise (Clarke and Piven 2001).

The public interest rationale for public policy is thus under pressure from the spread of individualism and social fragmentation. This chapter discusses how both public policy and public management have become recast by these trends: the emphasis on the responsibility to find employment, rather than the right to work; on local social policy rather than national welfare states; and on containing government expenditure rather than welfare expansion. These newer individualistic, local, and managerialist approaches have in fact been shown to work in particular circumstances, and some useful initiatives have been developed, such as better targeted support for individuals, greater consultation, and improved management of large projects. But they are too often “technical fixes” and by no means always work even within their own terms. Above all, they fail to engage the public in choices about social futures and are insufficiently systemic to change course in a sustainable way from the societal trajectory of division and insecurity identified by Reich.

The chapter concludes by considering social problems as emergent phenomena that demand an understanding of both the key parameters that produce them and the local complexity of their dynamics, dynamics in which conventional command-and-control public bureaucracies are ill equipped to intervene. First, however, the chapter examines the state of public policy in the United States, Europe, and Japan, discussing a series of examples that progress through a consideration of the individual, the local, and the national as spheres of policy action. By considering these spheres of action in terms of possibilities for achieving change, the chapter aims to understand how policy measures can address complex social problems, drawing on the ideas of complexity theory. Complexity theory does not treat these spheres as separate. Instead, it focuses attention on how different spheres of action interact, producing outcomes that emerge and follow trajectories of change and development that gravitate towards important general qualities. Understanding these processes is crucial to the success of policy interventions.

The Individual Sphere: Welfare-To-Work

In America during the 1980s, the New Right critique of social welfare focused on the image of a demoralised “underclass” of people overly dependent on state handouts. This saw an intensification of policy measures aimed at moving people into work and reducing dependency on income maintenance while maintaining the deregulated economic environment considered to be the engine of growth (Clarke and Piven 2001; Martinson and Holcomb 2002; Stoesz 2002). The underclass concept was also influential in Europe and Japan, although it was often used as a warning about pursuing U.S.-style deregulation. In Japan especially, the concept has been interpreted in terms of a shameful dependency on public assistance to be avoided if at all possible, alongside a Scandinavian-like policy commitment to maximising employment (J. Campbell 2002; Garon 2002). In Europe, there are certainly strong elements of these American and Japanese perspectives, but the less judgemental idea of “social exclusion” has come to dominate official social policy discourse in most countries.

Originating from the French concept of solidarity and the role of the state in furthering social integration, social exclusion describes the idea of a rupturing of the social bond between an individual and his or her society that can be tackled by appropriate inclusive measures that are part of the mainstream of public policy making and implementation. It is a multifaceted idea spanning a range of social problems from truancy and school nonattendance to long-term unemployment. The policy responses have been equally multidimensional, from individual counselling to improved parenting or employability to cross-sectoral area-based initiatives to connect unemployed people with job opportunities or improve health status. Across all of these responses there is a common aim of “reinserting” individuals, families, or groups into a cohesive civic culture. In contrast, the underclass idea justifies policies that emphasise the “special needs” of the personally inadequate, sometimes punitively. America’s post-1996 welfare reforms, for example, place considerable emphasis on benefit claimants needing to be personally responsible for finding employment, with public financial support conditional on time limits and various compliance, enforcement, and sanctioning arrangements (Martinson and Holcomb 2002). This regime coexists with the incarceration of American young men on a huge scale, disproportionately ethnic minorities, unable to sustain jobs in a deregulated labour market that generalises insecurity and division (Downes 2001). Policies of social inclusion and reinsertion stand in opposition to this growth of “a post-Keynesian state that replaces the social-welfare treatment of poverty by its penal management” (Wacquant 2001:95).

Reinsertion policies, however, have had significant lacunae, such as the inadequacy of social security benefits for those excluded from employment. Indeed, the main focus of European social inclusion policies has been on training and job subsidies aimed at reconnecting unemployed people with labour markets, albeit still generally more regulated ones than in the United States. Although the language is less harsh, the concept of social exclusion has in common with the idea of an underclass a concern with individual responsibility, especially to find employment. Targets to get more and more welfare benefit recipients into employment have been widely adopted by European states. The necessary programmes have not been cheap, and rely on various types of wage subsidy as well as some direct employment creation, mainly aimed at people whose employability is affected by disadvantages such as low skills, age, or disability. Detailed empirical studies of the European experience are now appearing, with a particularly interesting evaluation by Van Oorschot (2002) of the so-called Dutch miracle: the recent decrease in unemployment that has accompanied welfare-to-work measures during a period of strong employment growth.

Clearly the success of welfare-to-work programmes is strongly influenced by the level of job creation, as well as the willingness and feasibility of unemployed people moving to areas with better job opportunities—a particular issue in Europe where there are relatively low levels of labour mobility (Boeri, Hanson, and McCormick 2002). Van Oorschot’s study considers the efficacy of welfare-to-work in the apparently favourable conditions of high employment growth. Dutch policies during the 1990s sought to change the behaviour of both unemployed people, by making benefits conditional on them improving their employability and taking up work if offered, and employers, who are often reluctant to hire them. Wage supplements, interviews, and counselling are among the measures aimed at getting the unemployed into work, including disabled claimants, while employers also receive tax reductions. In addition, it was recognised that nearly half of all social assistance claimants had very low employability due to factors such as age, caring responsibilities, or language problems. They have been targeted for “social activation” to tackle social isolation and apathy through voluntary work that exempted them from having to find employment.

Although unemployment has fallen during implementation of these measures, and many people found employment through them, Van Oorschot shows that the number of unemployed beneficiaries has been significantly lower than the growth in jobs. What has happened is that full-time jobs held by men in the industrial sector have been replaced by a larger number of part-time jobs often taken up by new labour market entrants, especially women. As a result, the total hours worked annually in the Dutch economy actually changed very little during the 1990s. Furthermore, a claimant moving onto benefits at the end of the 1990s was no more likely to move out of benefits than at the beginning of the decade. Some policy measures conflicted with the welfare-to-work policies, such as replacing short-term disability benefits with an obligation on employers to continue paying wages, which has discouraged some employers from hiring people suspected as having health problems.

Van Oorschot concludes that the welfare-to-work programme has not cured long-term unemployment in the Netherlands but has stigmatised those who remain on benefits. In particular, he argues that it has reduced the level of social protection in Dutch society because employment growth has enabled governments to cut back the generosity of insurance benefits. If unemployment rises again, much larger numbers of Dutch employees will be dependent on inferior means-tested public assistance. The programme has therefore created the potential for a future social problem to emerge, depending on future employment scenarios.

Waddan (2003) reaches a similar conclusion about America’s post-1996 welfare-to-work reforms. These have seen a dramatic fall in benefit-dependent households, again during a sustained period of economic growth. An interesting feature of the U.S. case is the variability in how the reforms have been implemented at state level so that, for example, it is apparent that there is no necessary connection between punitive measures against benefit recipients pursued in some states and reductions in benefit dependency seen in all states (Waddan 2003:26). A feature of the reforms that appears to have worked is that the time-limiting of benefits encourages people to seek work rather than eat into their time-limited welfare entitlement. However, what really matters is whether people move out of poverty. The studies that Waddan reviews have so far shown that most of those who move into work have not achieved earnings that lift them out of poverty, and some, who are neither any longer on welfare nor working, have seen their poverty deepen.

Modern social policies often individualise social problems despite their public interest rationale, in contrast to the big picture that the QUEST project is working with and that is an obvious concern for environmental sustainability. Welfare-to-work policies, for example, focus on individual employability. It is true that at an individual level, factors such as low education, ethnicity, age, or interview technique are likely to predict the risk of unemployment, regardless of the actual level of unemployment in a society, and therefore programmes that target individual needs seem sensible. But this neglects the big picture. These individual characteristics are only relevant insofar as they interact with a wider parameter value, the unemployment level. At this scale, individual factors are not likely to be important determinants of unemployment because what matters is the level of unemployment (Davey-Smith, Ebrahim, and Frankel 2001). Adjust this parameter value and the behaviour of individuals, and the barriers and opportunities they face, appear quite different. Thus, while welfare-to-work programmes can give a high priority to providing minimum levels of qualifications, especially for young people, and are therefore aimed at being both responsive and preventative, a key incentive for young people to gain qualifications is the prospect of good employment. These prospects still vary widely because of geographical differences in the availability of jobs.

Even in low unemployment societies, problems of poverty and inequality persist. Mackenbach et al. (2002) show that in the Nordic countries of Europe, with their long histories of relatively egalitarian social and economic policies, there are substantial inequalities in self-reported morbidity, and these are being accentuated by the general trend towards better health status being greater among higher income groups. Although they continue to comprise a distinctive social-democratic welfare regime, typified by high rates of female employment, a more equal distribution of incomes, and more universal provision of public services across all social classes, these countries have not eliminated difficult social problems (Salonen 2001). The extremes of America’s racialised urban ghettoes and gated communities, or even the United Kingdom’s inner-city deprivation and outer “sink” estates, are absent, but those at the bottom of the income and education scales still live more segregated lives in the most deprived housing areas, often ethnically concentrated. Social exclusion as a concept, therefore, still has relevance because social problems are not just issues of poverty.

Japan is a significant example in this respect. Its combination of high employment, generous but often unclaimed social security benefits, and the duty of family members to support each other has created a society with one of the lowest rates of poverty and welfare dependency among the world’s more developed economies (Garon 2002). However, this is also now a system with considerable tensions. The increasing participation of women in employment is necessitating the replacement of their unpaid care work with funded services, and the principle of the deserving poor that underpins its public welfare provision is being challenged by the rise of social problems such as homelessness. Homelessness in particular has revealed the extent to which Japanese social policy depends on the family: much of the problem has been caused by a combination of recession since the early 1990s and the vulnerability to unemployment of able-bodied male labourers who have no family support (Ezawa 2002). In addition, the centrality of the family has been shaken by the “discovery” of child abuse and, excluding sexual abuse, of natural mothers as the main perpetrators (Goodman 2002a). A society that has prided itself on the virtual elimination of poverty is having to come to terms with social problems as public issues.

There are two aspects of the concept of social exclusion that are particularly important. Firstly, it recognises how exclusion from rights and customary standards of material comfort can be a result not just of low income but also of age, gender, geographical location, or ethnicity, conditions that often interact together and with key parameters such as the unemployment level (Burchardt, Le Grand, and Piachaud 2002). Secondly, social exclusion is a dynamic as well as a multidimensional concept. It goes beyond the statistics of poverty to capture the sense in which exclusion not only happens to some people but is done to some people by other people, whether wittingly or not. These observations—that social problems are produced by interactions between initial conditions and key parameters, and that they are consequences of actions taken by all of us—imply adopting a wider perspective than the individual with their special needs. One-to-one support is always valuable but rarely sustainable; the issue is one of a supportive social context, which has encouraged much recent attention to the state of neighbourhoods.

The Local Sphere: Polarising Neighbourhoods

For Byrne (1999), one of the most visible manifestations of social exclusion—deprived neighbourhoods—are products of policies that have had the effect of “promoting exclusion through space” (p. 122). In the United Kingdom, he argues, there has been an interaction between spatially targeted urban regeneration programmes and national social policy changes that, far from reintegrating poor people in these locales, has trapped them in low-paid, insecure employment with minimal social protection. This is not a process that is unique to the United Kingdom or even other liberal welfare regimes:

The evidence is now overwhelming. It does not show that all post-industrial cities are polarized. It shows they are polarizing, and that those which are fully subject to liberalizing post-industrial capitalism are polarized…. This is as much a matter of ideological hegemony as of inevitable systemic tendency. (Pp.112-13)

The hegemony of liberal economics has on the whole been facilitated by policy measures rather than contained or reversed by public policies deployed as alternative instruments of control over the societal trajectories otherwise determined by liberalisation. For example, in the United Kingdom the redistributive effects of regional policies have not countered the spatial impact of cutting higher-level taxes; public policy planning and management often mimic the principles of liberal economics; and urban regeneration rarely achieves more than marginal citizen participation given the priority accorded to the needs of development capital (Hamnett 1997; Hill 2000; Newman 2001). The underlying cause of mobile capital seeking the highest return is paradoxically treated as a solution by regeneration programmes aimed at attracting private investment to “underper-forming” areas. An alternative public policy response would be instead to tax this mobility and distribute the revenue to abandoned areas as an ongoing process of resource equalisation aimed at achieving comparable living standards across the space economy. Reade (1987), for example, has advocated taxation of development value as a logical source of resources for repairing the damage caused by capital and population migration, a type of displaced feedback that could move spatial economies to much more even patterns of development. On a much larger scale, the taxation of international financial transactions has been proposed as an important tool of poverty reduction on a global scale (Townsend 2002). The common counterargument, of course, is that these interventions would impair wealth creation, but this is a further example of not modelling the social effects of policy choices.

The migration of population and capital has meant that labour market problems in deprived neighbourhoods are compounded by “concentration effects.” The people left behind are increasingly coping not just with poverty but a rise in crime, drugs, antisocial behaviour, dereliction, and stigma that are widely perceived as problems for everyone, not just the residents trapped in these areas. Often there are also problems with failing public and private sector services that find it difficult to recruit staff, are insufficiently compensated for the extra costs of meeting needs in these areas, and have problems of viability due to low demand. Local job creation has often not reduced unemployment because new jobs are taken by commuters, in-migrants, or new labour market entrants (M. Campbell 2002). Social problems are diverse but largely emanate from the lack of access to employment that can sustain even average standards of living. Lone parenthood is one example. Webster (2002) shows how the dramatic rise in lone parenthood in the United Kingdom since the early 1980s, and earlier in the United States, has followed the equally large rise in male unemployment. In the main, men without jobs are not wanted as long-term partners. Echoing Van Oorschot’s (2002) study, Webster also shows that British welfare-to-work policies aimed at lone parents and using in-work tax credits have had a very modest impact on lone parent employment in the worst unemployment areas, with the take-up of tax credits proportionately much higher in low-unemployment areas. This again emphasises the need not just for supply-side measures but demand-side measures that explicitly aim to close geographical inequalities in access to employment.

Other social problems closely connected with deprived neighbourhoods include drug use, smoking, and poor health. These have all been linked with stresses caused by living in these neighbourhoods (Boardman et al. 2001; Duncan, Jones, and Moon 1999; Kleinschmidt, Hills, and Elliott 1995; Marsh et al. 1999; Reijneveld 1998; Thomson, Petticrew, and Morrison 2002). Such findings have led to policy prescriptions that emphasise tackling the diminution of social capital by combining improvements to the local environment and public services with community development, individual support with accessing opportunities and building personal confidence, resident involvement, and concerted multiagency work (Lupton and Power 2002). There has been mixed success with these local interventions, and it is often unclear why some succeed while others fail. Conditions external to the neigh-bourhood are critical, so local action to tackle decline is likely to be necessary but not sufficient. For example, Nevin et al. (2001) show how the risk of neighbourhood abandonment in the urban centres of northwest England is associated with a combination of neighbourhood conditions and external factors. The key external factors are the employment level and interest rates, and these affect the rate at which residents exit to more popular areas. Large amounts of money have been spent on renewing many deprived areas only for these external factors to operate in a way that leads to continuing abandonment and ultimately demolition.

In the United States, a key goal of the federal HOPE VI program has been to spatially deconcentrate very low-income households. The program has funded the redevelopment of rundown public housing areas to create more mixed income neighbour-hoods and provided either public rehousing or vouchers for private renting to many residents so that they can move to other areas. A tracking study by Buron et al. (2002) found that for most residents, their housing environment improved as a result of the program, but a substantial proportion experienced continuing problems with drug trafficking and violent crime, and about half of those who moved into private renting using the vouchers found it difficult to pay the rent. Also, while the intention was to encourage rehoused residents to participate in new, more supportive social networks in mixed income areas, relatively little interaction with new neighbours occurred.

While neighbourhood context clearly has some effect on quality of life—particularly crime, the quality of local services, and the quality of the environment—there are few reports in the literature of neighbourhood renewal achieving any significant reduction in social inequalities, and its role as an instrument of social policy appears quite modest. Health, and especially mental health, can benefit to some extent from neighbourhood improvements, but health impact assessments have also warned against the disruptive and potentially health-damaging consequences of renewal programmes, particularly if accompanied by rent increases, and the negative health effects that may occur for residents in adjoining unimproved neighbourhoods (Hirschfield et al. 2001). Area-based programmes such as neighbour-hood renewal also inevitably exclude many areas in an attempt to target resources, and in the United Kingdom this has been identified as a cause of racial and social unrest as some areas and groups are perceived to benefit while others lose out (Cantle 2001). In addition, unemployment and other types of deprivation may affect far more individuals living outside the most deprived neighbourhoods than within them. In Scotland, McLoone (2001) shows that if area-based initiatives targeted 20 percent of the most deprived postcode sectors nationally (equivalent to the U.S. zip code), only 41 percent of unemployed people and 34 percent of low-income households would be included. Over half of all post-code sectors would have to be targeted to include 80 percent of unemployed people or 74 percent of low-income households. McLoone concludes that there needs to be greater emphasis on national strategies rather than neighbourhood targeting.

Individual support regardless of neighbourhood has been shown to be effective in reducing the incidence of a range of social problems. For example, Olds et al. (1997) report results from a longitudinal study of the effect of a programme of prenatal and early childhood home visits by nurses to first-time mothers in the United States. A follow-up study when the children were 15 years old found that the intervention group of mothers was almost half as likely as a comparison group to have been identified as perpetrators of child abuse and neglect. Among women who were unmarried and on low incomes, the intervention was associated with fewer subsequent births, less use of welfare, fewer arrests, and lower levels of drug and alcohol problems compared with mothers in the nonintervention group. Clearly, individual interventions can make a difference and therefore may deny many potential beneficiaries their better outcomes if restricted to geographically defined deprived areas.

However, the model of intervention at this individual level remains one based on personal deficits. While this can appear effective, it largely works within the structures of exclusion that generate the problems that are then individualised as inadequate people unable to cope with the pressures around them. For example, given the choice between receiving home visits or a higher income, it is likely that the vast majority of mothers in the Olds et al. (1997) study would choose the latter. They are likely to make that choice out of a clear recognition that their problems are poverty-related. But a problem with this argument that it is structural causes rather than symptoms that should be tackled by public policy has been a paucity of work modelling and demonstrating the tangible effects of a structural approach.

In this respect, a recent report by Mitchell, Dorling, and Shaw (2000) is very important: it takes up this issue by deploying statistical modelling to simulate what Britain’s map of the public’s health would look like under different national policy scenarios. They simulate changes at a national level in income inequality, in unemployment, and in rates of childhood poverty, and model the resulting health outcomes for each of Britain’s 723 local parliamentary constituencies, modelling these outcomes according to the values of these key parameters. If national taxation reduced income inequality back to its level in 1983—a fairly modest target—an estimated 7,500 lives of people aged under 65 would be saved each year because of the statistical link between premature mortality and income inequality. Further lives could be saved by ending unemployment and eradicating childhood poverty. Rather than local policies trying to moderate some of the effects of national macroeconomic and social policies, this research shows that these national policies themselves are both causes of, and therefore keys to tackling, poor health in deprived areas.

The National Sphere: Ageing Societies

If we turn to what is often presented as the social problem of ageing, and the position of people for whom employment may not be an option in order to avoid social exclusion, what is the significance of public policy for their welfare? In the United States, the population is ageing but continuing to grow due to immigration; in Europe and Japan, ageing is coinciding with population decline, and this is causing particular concern about financing pensions and care services (Bermingham 2001). Clearly, this is a sphere where public policy responses need to be on a national scale. Initial conditions vary greatly across countries, and Blackman, Brodhurst, and Convery (2001) explore whether national “welfare regimes” are a useful way of understanding cross-national differences in care systems at a time when demographic change is putting all systems under pressure. State responsibility for providing health and social care for older people is greatest in the Scandinavian social-democratic welfare regimes, where political action has established universal publicly funded care as part of women’s right to paid work. High levels of taxation in these high-income countries provide the resources for provision on this scale, and there is widespread political support for the services because beneficiaries span the social classes and there is no significant opportunity to exit to a private market in care. In contrast, publicly funded care services for older people are currently least developed in countries such as Italy, Greece, Ireland, and Japan, where relatives, primarily daughters and female spouses, are expected to provide this care. The role of the state has been very limited in these “familist” welfare regimes, with dependency on public or voluntary services a last resort and associated with stigma.

Both the social democratic and familist regimes are under pressure due to a declining 15-to 64-year-old population. In the social democratic regimes, this is causing labour shortages and cost inflation. In the familist regimes, it is the shrinking pool of family carers that is critical and compounded by economic and social changes associated with greater individualism, such as the eclipse of the male “family wage” by work-based wages, the growth of female education and employment aspirations, and, among older people, a trend towards living independently rather than coresiding with their adult children (Blackman et al. 2001; Izuhara 2002; Watanabe 2000). Both regime types face sharply rising government spending on care, although factors such as morbidity and disability trends, the shifting balance between formal and informal care, and future changes in unit costs of care mean there is a very substantial funnel of doubt about the extent of this future increase in demand for services and their costs (Bermingham 2001).

The key to understanding differences between welfare regime types is welfare culture, especially the very different attitudes to the respective roles of the state and the family. However, while welfare culture helps to explain cross-national differences, from a social policy perspective there is a particular concern with how well these different welfare mixes foster the inclusion of older people, a group that has tended to be neglected by the employment-focused concept of social exclusion. Giarchi (1996), for example, reveals the enclosed nature of family life for some older Italians, whom he describes as “captive within households” (p. 374), and whose disengagement from social life is actively fostered, especially in the south of Italy. Hugman (1994) cites studies that have found high levels of self-reported loneliness among older people in Greece and Poland, both countries with an apparently high degree of family centredness. Eurobarometer sample data have shown considerable cross-national disparity in the life satisfaction of older people, ranging from 68 percent in Denmark stating that they are very satisfied with their lives, to just 6 percent in Greece (Walker and Maltby 1997). Despite the family orientation of Greek society, in which older people traditionally have a central position, 36 percent in Greece said they often felt lonely. This compared with less than 5 percent in Denmark, where the great majority of older people live independently and have less contact with relatives.

Cultural factors are important in interpreting these findings, as older people in Greece are likely to have high expectations about the range and frequency of contacts they should have with their families. In general, however, older people prefer to receive the support they need to live as independently as possible in their own homes. While social care is increasingly recognised as a significant factor in achieving this independence—so that older people receive assistance in carrying out activities of daily living—pension income and health care are of critical importance. Older people are of course one group for whom in general welfare-to-work programmes are not appropriate, although the pressure of ageing populations on pensions has caused a slowing down or reversal of earlier trends to early retirement, and recently some countries have raised their statutory retirement ages. Older people are also a group that has been relatively neglected by programmes to tackle deprived neighbourhoods, despite being vulnerable to the problems of these areas, such as high crime rates and poor facilities (Scharf et al. 2002).

The significance of ageing as a social problem is associated with the fact that across the countries of the Organisation for Economic Cooperation and Development (OECD), 40 percent to 60 percent of public expenditure is sensitive to age structure (OECD 2001). As well as pensions, this includes health care and social care spending where demand is increasing sharply as more people survive into their 80s and 90s. The OECD estimates that age-related public spending will rise by 6 percent to 7 percent of GDP by 2050, a forecast that takes into account countervailing trends such as pension reform. The impact will be greatest in countries where relatively generous earnings-related pension schemes occur together with a rapid ageing trend.

The old-age dependency ratio, or the number of retired people as a proportion of the employed population, is often presented as the key to the affordability of age-related spending. Italy faces the most acute problems of all the OECD countries as its ratio increases from 29 percent in 2000 to a forecast 67 percent by 2030 (OECD 2001). By contrast, the United States—which is facing less pressure from population ageing than the OECD average—is expected to see its old-age dependency ratio rise from 22 per cent to 38 per cent over this period. While the ratio is deteriorating in all countries, countervailing trends include increasing female participation in employment, falls in unemployment, improved productivity of those in employment, and immigration. Countries have also addressed affordability by reforming pensions. As with the social security system, generally these have taken two forms: price reforms have reduced the level of benefits while volume reforms have reduced access to insurance schemes. Companies are also increasingly replacing final salary pension schemes with money purchase schemes that do not guarantee benefits. As a result, many countries now face pensioner incomes falling well behind wages and pressure to save privately for retirement.

Age-related public spending on health and social care varies considerably across countries but it is not the high-spending social-democratic welfare regimes that appear to be least sustainable. The greatest pressure is being experienced by the South European and Japanese familist regimes where the increasing participation of women in employment and smaller family sizes are limiting the scope for family care on which frail and disabled older people have overwhelmingly depended. In addition, these countries tend to have many older people receiving relatively generous pensions that have helped to provide cash for care, and the generosity of these pension systems is having to be cut back due to deteriorating dependency ratios. The response has been to start expanding the provision of home help and day care services, most dramatically in the case of Japan, where a series of three Gold Plans between 1990 and 2000 has greatly increased spending on these services and short-stay community centres and nursing homes, although provision still lags behind the United Kingdom, for example (Goodman 2002b). In Greece and Italy, many families cope by hiring an immigrant worker to live in, part of a general increase in demand for immigrant workers to take on household tasks as young educated women turn away from traditional mother and homeworker roles. Indeed, immigration is sometimes suggested as a solution to deteriorating dependency ratios in OECD countries to expand the supply of labour, but it is unlikely that it can have any marked effect (Bermingham 2001; Boeri et al. 2002; OECD 2000). The issue is discussed further later in the chapter when considering scenario planning.

Two policy responses to the challenge of providing care in ageing societies are apparent, reflecting wider issues about affordability of welfare arising from political resistance to tax rises and concerns about the efficiency and performance of public services. The first is prevention, whereby governments are trying to identify how early intervention can prevent higher costs later on. The focus of welfare-to-work programmes on young people and the home nurse visiting programme described above are typical examples of preventative public policy. In the field of ageing, it is not just growth in the number of older people that is a significant spending driver but also rates of disability and institutionalisation. Costs can be moderated downwards substantially by measures to promote healthy ageing, prevent disability, and prevent admission to high-cost acute care and nursing institutions.

The second policy response is to increase the efficiency and effectiveness of public policies and services by more closely matching services to needs, achieving better integration between services, and making much more use of research evidence about “what works.” In the field of older people’s care, initiatives such as the United Kingdom’s introduction of care management have been shown to achieve significant improvements in service productivity (Davies, Fernandez, and Nomer 2001). In many countries, this type of response has also involved making greater use of the private sector in a variety of ways: as sources of investment through private-public partnerships, of spare capacity to deal with demand pressures, of management expertise, and of alternative provision. Thus, new ways of working are emerging that emphasise efficiency and effectiveness in tackling social problems, especially target-setting—a new paradigm for national social policies.

Social Policy, Targets, and Complexity

Policy decisions are made by combining three factors: resources, evidence, and values. Currently, most decisions are based mainly on values and resources, although there is increasing attention paid to “evidence-based practice.” It is particularly the issue of values that causes most concern when considering the efficiency and effectiveness reforms of governments attempting to contain the cost of rising welfare demands. This is currently the subject of considerable debate in the United Kingdom. At one extreme it is argued that, say, if food production can be left to private enterprise, then there is no reason why education or health care cannot be provided through private enterprise as long as this “works.” At the other extreme, it is claimed that public services should be motivated not by profit but by meeting needs, and that the ethical basis of public services is compromised by commercial criteria.

These arguments are complicated by observations that private enterprise cannot function without public policy to provide a regulatory framework and tackle market failures, and that no profit-making enterprise is likely to succeed without meeting the needs of its customers. In fact, there has been an increasing blurring of any distinction between private and public services in many areas. Private providers are often commissioned to deliver public services according to the needs-based criteria, and public services are subjected to new performance management regimes that have imposed private sector practices such as cost centres and targets. There are many examples from around the world of the implications of contracting in the public services. Martinson and Holcomb (2002), for instance, show how in the case of American welfare reform, local welfare agencies now contract with numerous private and not-for-profit organisations that provide the employment services and case management that benefit payments are conditional upon. These performance-based arrangements require considerable management and monitoring, although the improved accountability achieved has generally been welcomed. But characteristic problems with performance management are also evident, such as the perverse incentive for organisations to meet their job placement targets by “creaming,” or focusing on the job-ready. The response, common elsewhere, has been to expand the number of performance measures.

However, the spread of an “audit culture,” with its panoply of measures and targets, is further undermining public policy and the distinctive public services ethos on which the delivery of social policies depends. This is because social problems are complex and tackling them requires flexibility and learning, which are discouraged by the command-and-control tendencies of audit cultures. While apparently a debate about management practice, this issue is in fact far more fundamental because governments are increasingly framing social problems as needing to be managed rather than solved.

Few managers of public services are now without responsibility for a set of quantitative targets that they are held accountable for achieving, although the possible dysfunctional effects of targets are well documented (Smith 1995). These include gaming, when teachers focus their efforts on students at the borderline of an exam performance measure while neglecting other students, and biasing, such as when hospitals give a lower priority to clinical need to meet waiting time targets. Consequently, success in meeting targets can lose credibility as a method for judging performance, especially if a large number of centrally dictated targets conflict with meeting local priorities. The response is often to seek better targets, and a great deal of effort has been invested in this type of work (Boyne 2002). Attempts to improve the validity of targets as measures of performance tend to lead their to overproliferation, but there are more fundamental problems with management by targets than the search for valid and reliable measures of performance. These follow from three problems connected with the inadequate theory on which the approach is conventionally based.

The first problem is that in order to hold an identifiable management accountable for performance, the system that delivers the outcome a policy objective aims for is often misspecified. Thus, a school is specified as the system for delivering examination results when the school actually comprises many systems, some achieving better results than others. Schools also have inputs from other systems, such as students’ achievements in feeder schools and their material and home circumstances (Fitz-Gibbon 1996; Byrne and Rogers 1996). These all substantially influence outcomes at school level. Examination performance is likely to vary more across subjects within schools than for all subjects across schools, and the characteristics of a school’s intake are much more important than the effect of the school alone in determining its results.

Attempts to address this issue of system misspecification include “whole system” approaches. Rather than focusing on specific interventions or services, this approach looks at all the people in a defined population and implements strategies that intervene in the wider system parameters and the local context, both of which affect outcomes for the population. Whole system intervention is also based on “theories of change” (Pawson and Tilley 1997). Theories of change are working ideas about how change will occur as a result of the intervention, as well as what the expected outcomes will be. This contrasts with the “black box” approach to how change occurs that typifies target-setting. Indeed, O’Neill (2002) identifies how new audit regimes monitor requirements about both processes and targets, yet there is no necessary connection between conforming to process requirements and meeting the organisation’s targets.

The wider system parameters that influence local outcomes demand national and increasingly international policy measures such as managing the economy at full employment or benefit/tax credit levels that lift all children out of poverty. The local context is the level of organisational interventions such as training or housing programmes tailored to local conditions. It is at this level where much effort is being expended on “joining up” initiatives and services, with whole-systems thinking influential in working towards a “postbureaucratic” local social policy (Netherlands Institute for Care and Welfare 2002). A third intermediate level of the region or city-region is also increasingly advocated as a space of strategic action for mobilising collective effort by agencies and integrating programmes within urban systems defined by networks such as travel-to-work patterns (Healey 2002).

A good account of the case for whole-systems public policy is set out in the final report of the Committee on the Quality of Health Care in America. In an important appendix, Plsek (2001) argues that public policy does not achieve its aims in many circumstances because it is delivered by organisations that apply “mechanical systems thinking” well beyond the situations where it is appropriate. Such thinking may work where there is a high degree of certainty about outcomes from actions and a high degree of agreement among the people taking the actions. Plsek gives routine surgery as an example—in this type of situation it can make sense to fully specify appropriate behaviour and reduce variation. However, there are many other situations that lie in a “zone of complexity,” where there are only limited levels of certainty and agreement, and where detailed plans and controls do not make sense. Plsek’s example is delivering primary care services, where there are many different models that have worked in some situations but not in others. Taking this perspective, it is clear that most public policy is in the zone of complexity.

The zone of complexity, Plsek argues, calls for interventions not by mechanical organisational systems with detailed plans and controls but by complex adaptive systems (CASs). A CAS comprises individual agents whose actions are interconnected and who can learn. The responses of these agents and of the environments in which they act are not mechanical and predictable but nonlinear and unpredictable in detail. Policy frameworks for complex adaptive systems need to create the conditions under which desirable outcomes are possible rather than try to prescribe agents’ behaviour to achieve these outcomes. Prescription has its place, but in terms of the policy framework defining a few rules that are locally applied. Detailed action is then decentralised and this produces local variation, ideally producing the desired outcomes in a number of possible ways that are suited to local context, but also generating knowledge about good and not-so-good interventions.

As Plsek points out, these ideas are not new, although the synthesis offered by this account is new and has been identified as a new field of “complexity theory.” Complexity theory focuses on interactions, especially the whole system as the causal system that produces the outcome a policy measure aims for. By only focusing on part of this system, there is a danger that important causal factors will be ignored (Örtendahl 2002). Cross-agency targets will not help if there is no mechanism that brings agencies together as part of an interacting system with inclusive thinking and learning. There is increasing recognition of whole systems in health and social care and attempts to find system solutions by, for example, creating new multiprofessional agencies and programmes (Shaw 2002). But it is still often the separate subsystems that are held accountable for performance, with their own targets.

A second theoretical problem with management by targets is that the approach assumes that the delivery system, even if treated as a whole system, is in equilibrium, so that a given input achieves a proportionate output. There seem to be few examples of this input-output equilibrium and many that suggest there is no such direct relationship. Pawson and Tilley (1997) review extensive evidence that there is no direct relationship between money spent on crime reduction initiatives and successful outcomes because local contexts vary in ways that produce quite different interactions between the extra resources and local behaviour. While there will be circumstances when additional resources produce proportionate outcomes, it seems more common for additional resources to produce disproportionately large or small outcomes, or for the outcomes to vary unpredictably. Governments largely ignore this complexity in their tendency to make better outcomes—defined, measurable, and accountable—a condition of any extra public spending, together with the associated apparatus of performance management and audit. However, the systems into which extra resources are injected are too complex for the relationship to be so direct. Governments then turn to bad theory when the systems fail: it must be poor management that is to blame, and this can be exposed using performance indicators and league tables of performance, even though evidence is beginning to accumulate that this cannot be the sole or even main reason. For instance, White (2002) shows that the local level of deprivation almost perfectly explains differences in performance between many local authorities in London; where it does not, this seems to be a consequence of imperfections in the deprivation measure. In general, performance appears to be reflecting local conditions rather than significant variations in management competence.

A third and related theoretical problem with targets when addressing complex social problems is that the state of a system is regarded as amenable to command-and-control. An intervention in the initial conditions of a system produces an outcome that depends on both these initial conditions and the subsequent interaction between the intervention and the system’s self-organisation. The potential for nonlinear change is huge under these conditions, and command-and-control will not bridge the implementation gap that can result from this complexity. In some organisations, management practice has adapted by striving for alignment between the values of the organisation and those of its employees and customers so that it can work on the basis of high levels of trust and adaptability rather than command-and-control and prescription. Fitz-Gibbon (1996) concludes from a study of school effectiveness that the key to effectiveness is to feed back information on valued outcomes to those responsible, so they can make adjustments that “close in” on the outcomes. The fact that outcomes are coproduced also means that the delivery organisation is not the only part of the system. Its customers, clients, and users coproduce the outcome by bringing their own resources and behaviour into an interaction with the service provider.

Complexity and Learning in Public Policy

Intervening in complex social problems needs less prescription of detail and more intelligent accountability through reports and inspections for a smaller number of dimensions. As Plsek (2001) identifies, paradoxically the best way to achieve complex outcomes is to work with the self-organising capacities of local agents within a set of a few simple rules, rather than adopt the mechanistic approach of designing complex policy and procedural machinery that has led to the proliferation of targets considered earlier. Plsek and Wilson (2001) suggest that these dimensions should be limited to general direction pointing, boundary setting, resource allocation, and permissions. An example would be a policy on health inequalities. The policy approach could be a general direction pointing that all policy areas contribute on an ongoing basis to health improvement actions that narrow inequalities. The boundaries could be that actions must have a user focus (such as user participation in programme design and evaluation), that each service shares a common aim and pools resources with at least one other (for example, housing and health promotion), and that demonstrable learning occurs from information feedbacks that prioritise evaluation by users (consultation, surveys, operational data). Actions would have to keep their spending within common formula-based budgets (to take account of local conditions) with multiagency access to them. Accountability would be based on open access to information and intelligent audit with a focus on theories of change rather than black box evaluation.

If performance management of target achievement is based on bad theory, it might be expected not to work. A leading figure in the U.K.’s New Labour government, David Blunkett, was recently quoted as saying, “There is mounting frustration in government that, after nearly five years in power, the promised transformation in public services has yet to be seen. Centrally set targets have proved elusive, and even the modest pledges of the 1997 election were unexpectedly difficult to achieve” (Perkins 2002). The problem has been an approach based on the introduction of targets that have little relationship to internal local system states and little regard for the influence on performance of wider system parameters.

Organisations that attend to their internal system state, especially the knowledge of their personnel, trust, and alignment, and scan and respond to their environments, especially with scenario modelling and anticipatory planning, are learning organisations. This is a term now frequently encountered in the management literature, but there has been a surprising lack of engagement in this literature with pedagogic research, which might be expected to provide some useful clues for learning organisations. Issues of trust and blame, which now feature prominently in debates about targets and performance, are also prominent in debates about good pedagogy. For example, Biggs (1999) contrasts “theory X” and “theory Y” approaches to pedagogy, with the former assuming that students cannot be trusted and the latter supporting student autonomy and, in particular, student learning. Theory X is reflected in practices such as negative reinforcement using anxiety to “motivate,” time stress, blame-the-student explanations of failure, cynicism towards lecturers perceived as not believing in what they are doing, and little input by students to decisions that affect them. The parallel with bad management practices is clear. Management by targets can engender a culture of “Who is to blame and who is to be rewarded?” while management by learning aims for a culture of “What do we know and how do we find out what works?”

An increasingly widespread approach to the learning organisation in public policy is the paradigm of evidence-based policy and practice (Trinder and Reynolds 2000). This is the approach taken by Fitz-Gibbon and Tymms (2002) to school improvement. They do not reject using indicators but emphasise the need for good “tracking” systems as well as controlled experimentation to find out what practices work best: performance management then becomes a collective research enterprise. For example, Fitz-Gibbon (1996) advocates the use of confidence intervals in monitoring indicators of exam performance that compare “value added” (residuals) with predictions of exam achievement derived from regression analysis using data on prior achievement. But although Fitz-Gibbon and Tymms (2002) argue for collecting a good range of indicators, there is an issue with their methodology due to the nature of public services as complex and interacting systems. Their approach implies that the general relationship between inputs and outputs is explainable in terms of linear regression, with residuals representing exceptions—either very good or very poor performance. In fact, typically more than half of such variation remains unexplained and the relationship between inputs and outputs is nonlinear and discontinuous. Interactions complicate any direct, uniform relationships between independent and dependent variables, with outcomes emerging from these interactions in ways that are very difficult to predict. This has led Byrne (2002) to argue that focusing on relationships between variables is the wrong approach, because what matters is cases (or systems) and how their condition is determined by their interactions with environmental parameters that make possible a number of system states (such as good performance and bad performance).

The problem has seen both management theory and policy evaluation methodology converge on complexity theory as a new type of systems thinking (Pawson and Tilley 1997; Stacey 2000). Complex phenomena have been the subject of this chapter, as most social problems are complex. They are characterised by nonlinear interactive processes, the emergence of new phenomena, patterns of both continuous and discontinuous change, and outcomes that cannot be predicted from past trends. The organisations that intervene in complex social problems are themselves complex adaptive systems, yet, as discussed above, policy and management approaches often do not enhance this adaptability but seek instead to regulate it (Zimmerman, Lindberg, and Plsek 2001).

Complexity reveals itself over time not in terms of uniform relationships but of transitions in system behaviour due to interactions with either internal or external changes. A steady state represents convergence on what is called an “attractor” when, say, all the schools in a regional education system return much the same outcomes year-on-year. But a system may shift from a steady state if a key parameter changes, such as a cut in budgets or a new policy on selection. The system might begin transforming and outcomes flip between different values, with local systems—the schools—starting to diverge into different states (a phenomenon identified by Byrne and Rogers 1996). Finally, the parameter value may change further and shift the system into chaotic behaviour so that local systems start returning different values in wholly unpredictable ways. Although a conceptual sketch, this scenario might well capture the effects on delivery organisations of increasingly cutting back their budgets. It could, for example, explain the observation made by Blackman et al. (2001) that standards of social care for older people show increasing internal variability as the national social care system becomes less well resourced.

The greatest significance of an ability to learn is in dealing with uncertainty. Uncertainty is a feature of many social problems existing in the zone of complexity. OECD governments have responded by using scenario planning. This has focused on the effects of key parameter changes on system behaviour so that possible futures can be anticipated and planned for in the long term. In scenario planning language, the key parameters are known as “drivers” and the attractors are called “outcomes” or “worlds” (S. Davies et al. 2001). The phase transitions seen in the behaviour of complex systems are modelled in scenario planning by “wild cards” that hypothesise unexpected changes with significant impacts.

Returning to the example of national policy responses to ageing discussed above, the OECD (2000) has modelled possible ageing scenarios for nine countries using two key parameters, persons aged 65 and over and the labour market participation rate, with the dependency ratio as the outcome. The importance of the second parameter, the labour market participation rate, is important not only because of its effect on dependency ratios but also because it is much more susceptible to policy action than ageing. Three scenarios are modelled, all assuming that women’s labour market participation continues to increase in line with recent trends. In the first “benchmark scenario,” men’s participation rate is assumed to continue to fall in line with recent trends. In the second “constant scenario,” men’s participation rate is kept stable, and in the third “reversal scenario,” it gradually returns to that of 1960. The reversal scenario is therefore likely to counteract the effect of ageing on the dependency ratio. Modelling this effect shows that it has only a modest influence, but, if instead of the dependency ratio, outcomes are modelled for the share of the total population that is employed, a considerable easing of the effects of ageing is evident. The effect depends on the level of the existing participation rate; in Japan, where existing rates are high, the effect is far less pronounced. The reversal scenario is used to justify the authors’ conclusion that there is nothing inevitable about the effects of population ageing. For most countries, policy measures that increase the participation of older workers in the labour market should substantially contain the economic effects of ageing. This presents a rather different picture to the “burden of care” perspective that has driven attempts to improve the targeting of services for older people.

This conclusion, though, should be qualified by noting the possibility of phase transitions caused by wild cards. Three demographic wild cards are identified in the scenario planning literature: massive migration from the developing world into the developed world; Europe and Japan failing to take the necessary policy measures to cope with the challenges of their ageing populations; and a collapse of the sperm count, causing an even more substantial decline in population replacement rates than forecast (S. Davies et al. 2001). Although immigration has been proposed as a way of addressing Europe’s and Japan’s decline in working age populations, the scale of immigration that would be necessary to offset declines in the ratio of working age populations to older people is huge: for Japan, Bermingham (2001) estimates that it would need to be 9.7 million immigrants per annum! Nevertheless, large-scale immigration remains a possibility, and the implications for the culture and society of the receiving countries are also huge, so much so that Byrne (2002) uses this scenario as an example of likely phase transition, in contrast to the OECD’s reversal scenario, which envisages the use of policy measures to steer countries away from critical falls in the proportion of the existing population that is employed, with no qualitative transition in the nature of their societies. Which trajectory occurs depends on how the key parameters move and the policy decisions taken. The modelling presents a policy choice between increasing the employment participation of older people and large-scale immigration, with the latter likely to have least effect but most social and political impact.


This chapter began with a discussion of public policy as representing the public interest in relation to social problems. Recent developments such as welfare-to-work, neighbourhood targeting, and efficiency measures to cope with the affordability “crisis” of ageing have eroded universalist conceptions of welfare as social protection and the moderation of inequalities through redistributive mechanisms. While there are features of these developments that “work,” there is a great danger that, as the QUEST designers recognise, “society” will become less and less meaningful because the extent of individualisation in policies, practices, discourses, and information means that people can no longer “see” the society that is emerging.

Management by targets is one aspect of this problem. The alternative to command-and-control is to create conditions in an organisation that are learning-orientated and discursive, and this is what is needed in order to intervene in complex social problems. It involves reformulating targets as feedbacks, the monitoring information needed to track change. If this information is fed back continuously, it encourages constant dialogue about what it means, what the organisation is achieving, and what it should be achieving.

Table 4.2 summarises how managing complexity looks quite different to managing by targets. In these examples, the systems are defined differently and the key variables and parameters are different. Public policy based on complexity should work better than policy based on targets and the theory X tools of audit culture because it is based on developing and testing theories about why things actually happen—in pedagogic terms, deep rather than surface learning. But there is also a need to establish where things are going and where we want them to go so that rather than just responding to trends, there is an explicit effort to steer the trajectory of social and economic change. What is needed is more QUESTs that involve people in public debate about social problems and social trends. People can then engage with these problems and trends as active and informed agents taking action to achieve desired social futures. Without this, action is likely to reflect individual interests because these are tangible, rather than the collective interest that is less tangible in modern and complex market economies.

Social problems create opportunities for system transformation if they become framed as turning points (or “crises,” from the Greek krinein, to decide) inviting choices between alternative future scenarios. For example, the response of Japanese governments to the ageing issue during the 1980s was to promote more of the same: Japan’s “welfare society” ideology, based in practical terms on the role of women as unpaid carers at home. However, in the face of women’s political mobilisation against the extent of this imposed role and a broad-based popular campaign, the Japanese state has been forced to introduce long-term care insurance. This has expanded socialised care from its previous focus on low-income older people without family support to the middle class. Eto (2001) observes that the untenable situation experienced by many Japanese women meant that the “strong sense of crisis they felt often transformed ordinary housewives into feminists” (p. 21), but what is also significant is that, while the familial principle remains strong in the government reforms that have been made, progressive municipal mayors in Japan have used the introduction of long-term care insurance as an opportunity to build local welfare systems. These have developed as collective responses to the social and economic trends that the crisis of family care brought into the arena of public debate and political action.

The issue, then, is the availability of means that enable people to link their private concerns and preferences with an appreciation of, and engagement with, alternative outcomes at system or societal levels. Scenario planning, for example, should not be the preserve of experts and political elites but a popular practice. There is potential for this in the growth of computer modelling in schools and colleges as a tool for students to better understand the world around them and, in particular, to discover system behaviours that are not evident at an individual level (see, for example, Colella, Klopfer, and Resnick 2001). Using models that show how social problems and trends originate and develop should help to reestablish a public sphere for public policy. Byrne (2002) calls such computer-based models “macro-scopes”; these increase the scale of our perspective from the traditional community level of observation and action, which has become detached from real economic and social processes, to the system level at which markets and governments operate. The significance of initiatives such as QUEST is that they take the macroscope concept further by showing how existing policy tools are available to take social trajectories in new and different directions: futures that are chosen, rather than futures that just arrive. Reich (2001) used the image of the giant genie in the quotation at the beginning of this chapter. With the computing power, modelling expertise, and Web access now available, perhaps the scenario he describes can move beyond fantasy to a reality of reflective societies.