'Faith-based' organisations and contemporary welfare.

Melville, Rose ; McDonald, Catherine

Introduction

The recent electoral success of the conservative and Christian Family First Party in the Australian Senate suggests that this is an appropriate juncture at which to review the reconfiguration of the institutional relations of welfare. Developments in welfare policy--loosely encapsulated in the notion of welfare reform--have been accompanied by calls to further devolve service delivery to the local community level (or, at least, away from the state). Increasingly, these calls are accompanied by references to 'faith-based' organisations as desirable sites for service delivery (McClure, 2000). Our principal claim is that much of the local utilisation of the notions of 'faith-based' organisations and 'faith-based' service delivery represent uncritical adoption of what is essentially the language of American welfare reform--terminology that is inappropriate and, therefore, not particularly useful in the Australian context. Further, we suggest that its continued use, particularly as an omnibus term for religious non-profit service delivery, perpetuates convenient myths and obscures our capacity to develop a finely grained and critical analysis of emerging patterns of welfare delivery and their consequences. In reality, 'faith-based' service provision is, we propose, code for a mode of American welfare radically different to that which its users would purport to create in this country and to that which it replaces.

To advance this proposition, we first discuss the notion of 'faith-based' organisations and its links to American welfare reform, particularly via a controversial clause of the American Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) known as Charitable Choice. We do this to highlight the intensely moral and, to a large extent, culturally singular agenda promoted by Charitable Choice and the discourse of 'faith-based' service delivery in that country. Secondly, we discuss some key differences between the role of religion in welfare service delivery and in the institutional relations between the churches and the state in Australia and the United States of America (US). Thirdly, we illustrate some of the problems with the terminology and practices of 'faith-based' service delivery in the context of American welfare reform, and consider the actual and likely consequences of its uncritical adoption in Australia. Finally, we propose an agenda for research and analysis around the contemporary role of religion and religious non-profits and welfare--a topic made more salient by recent political developments that signal the intensification of Australian welfare reform.

Charitable Choice and Welfare Reform

Religion, particularly Christianity, plays a significant (and from the Australian perspective, extraordinary) role in American society (Belcher, Fandetti & Cole, 2004). It is not surprising then that when the US set about dismantling its New Deal-inspired Assistance for Families with Dependent Children (AFDC) program through the introduction of Temporary Assistance for Needy Families (TANF), it introduced a specifically religious element. This was not a particularly unexpected development because, during the 1980s, the Reagan administration stridently championed religious organisations as more effective than state agencies and secular non-profit organisations in the delivery of welfare. Supported by conservative scholars (Olasky, 1992), key Republicans such as Newt Gingrich forcefully asserted a privileged role for religious welfare service delivery. Further, prominent conservative think-tanks such as the Brookings Institute and the Manhattan Institute promoted public debate by hosting a series of influential conferences about the role of religion in social policy and welfare reform (Chaves, 2001). Support was also evident on the other side of American politics, as former Democrat President Clinton and Vice-President Al Gore also endorsed the involvement of religious groups in welfare service delivery (Cnaan and Bodie, 2002). The role of religion in welfare reform was cemented in 1999 when the US Department of Health and Human Services sponsored a national conference on Welfare Reform and the Faith Community (Chaves, 2001).

In effect, Charitable Choice represents the successful mobilisation of an influential religious cadre within the welfare reform movement--an influence that has permeated other associated administrative and political developments at both federal and state levels. In 1996, when PRWORA was going through the legislative process, Republican Senator John Ashcroft introduced the Charitable Choice provision, its primary purpose being:</p> <pre> ... to allow States to contract with religious organizations, or to allow religious organizations to accept certificates, vouchers, or other forms of disbursement ... on the same basis as any other non-governmental provider without impairing the religious character of such organizations, and without diminishing the religious freedom of beneficiaries of assistance funded under such program (Section 104 (b) cited in Cnaan and Brodie, 2002: 224). </pre> <p>However, it went further than this in that Charitable Choice enjoined the states to actively privilege and promote faith-based service delivery. One of the (paradoxical) outcomes of TANF is that it has managed to overcome long-standing problems of fragmentation in local service delivery systems (Walter & Petr, 2000). Under TANF, previous seemingly intractable problems with service integration were reversed as a complex array of state and county health, welfare, and employment services were drawn into the welfare reform system by the use of fiscal incentives. Accordingly, Charitable Choice applies to a wide range of programs and services--all of the state and county services provided through the auspices of TANF--food stamps, Medicaid, Supplementary Security Income, food services (meals, community pantries, household budgeting counselling, soup kitchens), employment services (such as job search, training, vocational education, and ESL programs), and a range of community services (domestic violence services, drug and alcohol services, health clinics, residential care, and supervised community housing). In 2001, early in his presidency, President Bush, by executive order, established the White House Office of Faith-Based and Community Initiatives with the explicit brief of involving faith-based organisations in all aspects of social and welfare policy.

Charitable Choice is not only unique in terms of its ubiquity. Whereas previously religious organisations in receipt of federal funds were required to remove any symbols or reference to their faith in public documents, spaces, and forums, under Charitable Choice this manifestation of the separation of church and state was overturned. In effect, this represents a fundamental reversion from the principles of modern publicly-funded service provision to a form of governance more attuned to the conditions and institutional relations of the 19th century. As suggested earlier, Charitable Choice represents the political victory of religious (largely Christian) activists in the realm of social policy. And while the changes wrought by PRWORA were extensive, they have, according to Belcher et al (2004: 269) been 'overtaken by a growing movement among people who argue for the total devolution of the welfare state in favour of faith-based organisations replacing government as the provider of social welfare.'

While the American churches have always played a role in the provision of welfare, Charitable Choice directs attention beyond the increasingly secularised bureaucracies of the mainstream churches towards religious congregations. Prior to 1996, religious congregations played an important yet residual role largely in emergency welfare service delivery in the form of, for example, food and clothing pantries, and limited financial aid (Chaves, 1999; Cnaan, 1999, 1997). Some congregations accepted state funding but, in the main, the bulk of their support came from financial, in-kind, and labour contributions from church members (Pipes & Ebaugh, 2002). Chaves (1999), in his 1998 National Congregations Study, for example, found that while 57% of congregations were engaged in welfare service delivery, only 3% were in receipt of government funds. Charitable Choice directly targets these groups--congregations that, for the most part, reside outside of the institutions and structures of the post-New Deal liberal welfare state.

Further, and as Hopkins and Cupaiuolo (2001) note, sponsors of the faith-based initiatives brought to fruition by Charitable Choice rely heavily for support on the Christian conservative movement. For some, this does not represent a problem (Sherman, 2001; Skillen, 2000). Other commentators, however, argue that much faith-based welfare in the US, captured abroad within the domain of public policy by Charitable Choice, relies on a theology and set of values that stand at odds with the liberal welfare state (Belcher et al, 2004; Hopkins & Cupaiulolo, 2001). At the same time, mainstream Protestant churches (which provided the bulk of formal welfare services along with the Catholic Church in the secularised pre-TANF welfare state) represent a shrinking proportion of US church attendees. As a result, the ability of this group to participate at the congregational level in comprehensive faith-based programs is compromised (Belcher, Fandetti, & Cole, 2004). It should be clear from the preceding discussion that the term 'faith-based' in the context of American Charitable Choice and welfare reform is employed to signal a particular mode of service delivery in a reconfigured set of institutional relations with the state. In summary, it represents a fundamental moralisation of welfare that significantly reconfigures the nature and trajectory of the American liberal welfare state.

The Church and Welfare

Unlike in America, in Australia the term 'faith-based' tends not to be used to differentiate between particular types of religious organisations involved in the delivery of welfare services. Rather, the Australian social policy literature commonly conflates the term 'churches' and 'charities' and does not distinguish between the two institutional forms, their beliefs systems, and practices. According to Wards and Humphreys (1995:17), one of the main reasons for this deficit is the lack of a general history of religion in Australia. While there are several partial accounts of the activities of religious bodies in the provision of social welfare (Lyons, 2001; Gleeson, 1996, 2000; Challen, 1996; Cleary, 1994; Swain, 1992, Kennedy, 1985), there is no comprehensive social history on the role of churches, whether large institutions or single congregations (Garton, 1994:24). Dickey (1987) is one of the few social historians who does provide a wide-ranging account of the establishment and subsequent role of early charitable religious institutions to care for the sick, poor, widowed, orphaned, and imprisoned. His work investigates the religious and moral underpinnings of religious welfare, the relationship between the organisations and the state, their functions, funding source, and the nature of services provided.

In contrast, there is an extensive established body of literature on the involvement of the church and religious organisations in the provision of social welfare in the US (see, for example, Wuthnow, Hodgkinson, & Associates, 1990). This literature indicates that there are interesting and significant differences in the evolution and role of religious bodies in the provision and role of welfare services in that country, differences that help to explain the breadth and depth of the American academic literature on the welfare state and non-profit sector. Indeed, reference to this literature also demonstrates that the American non-profit academic community has engaged extensively in ongoing analyses of Charitable Choice and the developing role of 'faith-based' service delivery (Jeavons, 2004; Austin, 2003; Smith & Sosin, 2002; Chaves & Tsitsos, 2001; Farnsley, 2001; Cambell, 2000; Wittberg, 2000).

In the US, large groups of early settlers were drawn from people fleeing religious persecution in Europe, who brought with them strong religious commitments and a belief in religious freedom. There has always been a stronger religious identification in American society than in Australian society, evidenced by their higher levels of church attendance and membership (Wuthnow, 1988). Australia, on the other hand, was founded as a British colony populated by convicts rather than free settlers, and religion is seldom mentioned in official histories of early settlement. On the 'insistence of Evangelicals within the English Church,' one Anglican chaplain accompanied the first fleet of 1,000 people (Challen, 1999:26-27,). While the representatives of the early church and military authorities (the precursor to the state) were closely intertwined in the business of managing the settlement (as were most non-convict early settlers), this relationship was nevertheless ambivalent (Challen, 1996:27). In Tasmania, for example, the convict chaplains and religious instructors were public servants not subject to church governance or doctrine and, as such, acted as overt agents of social control (Challen, 1996:27).

The influence of religion in the establishment of 19th century social welfare in Australia was significant, particularly in the establishment of British-style benevolent societies. Transplanting middle class British dislike of the Poor Laws with its (crude) system of entitlement, early charitable welfare was constituted within a strong moralistic and religious discourse, often not the official church doctrine but a curious mixture of doctrinal or theological language practices that combined with and complemented the controlling functions of the state. The emphasis was on saving the 'soul', 'reforming the character', 'making the poor more work ready', distinguishing between the 'deserving and non-deserving poor', and so forth. Early 19th century welfare in Australia constituted a clear example of governing through 'pastoral care' (Dean, 1999) in which religion informed practices undertaken by charities and, in doing so, promoted the interests of the state. One effect of these origins was the early engagement of the non-profit sector more broadly in the business of the state.

During the 20th century, a more indigenous and unique model of welfare evolved that involved considerable state intervention in the economy in the form of centralised wage fixing and tariff protection (Castles, 1983). In this model, charitable provision of welfare services developed alongside state-based provision, eventually evolving into a mixed economy of care with relatively cooperative and stable institutional relations between the sectors. As Australia became a federated nation binding the former colonies together, it evolved into a society considerably more secular than the US. Within that, and as Challen (1996:26) notes, Australians developed an expectation that the church charities will provide social services much like the state, regardless of whether people presenting to them are members of those churches or not.

In many ways, the role of religion in welfare service delivery was normalised to the point where it was taken for granted. Indeed, Australian charitable service delivery (be it religious or otherwise non-profit) has really earned the sobriquet of the 'Shadow State' as developed by Wolch (1989) in respect of the British context. Much of the historical analysis of Australian social welfare, for example, focuses on the development of the income maintenance system (Jones, 1996) and industrial arbitration system (Castles, 1983). Although social policy analysts and non-profit scholars in the 1990s increasingly acknowledged the importance of the non-profit sector more generally in the provision of social welfare, there has been very little analysis of the role and (social and economic) contribution of church institutions in the provision of non-government services (McDonald, 2000; Garton, 1994).

The largest inquiry into the non-governmental sector in Australia (IC, 1995: XVIII), while noting the 'economic contribution' of large church-sponsored organisations, contains little analysis about the religious character of service delivery, governance, or policy initiatives. This is especially interesting given that the top six social welfare providers (and a total of nine of the top fifteen providers) were church organisations. Cleary's (1994) landmark study of Catholic human service agencies generated some of the most original data about the contribution of churches to social welfare, estimating that their annual expenditure (including health) exceeded $5 billion. Similarly, the workforce was estimated at 130,000 paid staff, making it equivalent to the number of public servants employed by the two largest states (Cleary, 1994-52).

These two lone examples aside, the absence of critical discussion about the church sector in Australian social welfare discourse is quite striking, particularly noticeable when examining key Australian policy textbooks of the past two decades. The terms 'churches', 'charities', and 'non-governmental welfare' are largely absent from the main indexes, and there is little to no substantive content in the works of influential writers such as Dalton et al (1996) or Jamrozik (2001). In an earlier book, Graycar and Jamrokiz (1994) devote a mere three pages to a discussion of the work on non-government organisations, including a brief overview of the history of the Sydney Benevolent Society in NSW, drawing substantially on the work of Kewley (1965). In their analysis, they (like most) conflate the term charities, churches, and non-government welfare. An exception is the recent work of Winkworth and Cameilleri (2004) on Catholic Social Welfare Services.

Several large-scale studies have documented the contribution of the non-government welfare sector (including religious organisations) in the provision of Australian welfare services (Milligan, Hardwick & Graycar, 1984; Industry Commission, 1995; Lyons, 1995). The sector as a whole expanded in Australia post-World War II, with substantial encouragement in the form of state funding complimented by private contributions and church-based support. It is important to note, however, that the development of the non-government sector was accompanied by the development of state provision of services up until the mid-1980s, when the state began to progressively withdraw from direct service delivery in response to economic liberalism. In other words, religious-based non-profit organisations were part of a broader non-profit sector drawn into a set of institutional relations with the state as part of a highly normalised state project. As such, the uniqueness or singularity of parts of the sector (such as religious organisations) was institutionally compromised in a manner wholly unlike the situation in the US.

There are also two other major differences between the US and Australia in terms of extent of state support of religious provisions of social welfare. While both countries have provided significant direct subsidies to the non-profit sector, the respective taxation systems reflect a major difference in the treatment of funding of the activities of religious organisations. In the US, congregation members who pay tithes (usually at least 10% of their income) to the church to support the staff and pastoral care activities can claim the full amount as a tax deduction. The American tax benefit to congregations provides institutional recognition of the congregational pastoral care function that is also, in effect, state subsidisation of those functions. In this way, the structural condition for the role of faith-based congregational welfare provision under Charitable Choice was in place before welfare reform came along. In Australia, no such provision exists. Members of congregations can only claim tax deductions on items that pertain to educational activities in which the church may be involved. Donations to registered charities (which may or may not be run by churches) are also claimable as tax deductions. This demonstrates a quite different relationship with the state, with considerably less institutional recognition of congregations.

Faith-based organisations and practices in the United States of America

Assessments of the impact of charitable choice on the American welfare system are confounded by the lack of clarity and agreement about the term 'faith-based' among researchers, state, and private sector funders (Sider & Unruh, 2004:109-110). For example, it is used to describe various institutional forms (i.e. small congregations vis-a-vis large-scale social service agencies), or to distinguish between the level of religiosity involved in service delivery (Smith & Sosin, 2001; Working Group, 2002). This has a number of serious implications. Comparative research is limited, as the entities labelled as 'faith-based' can be institutionally and culturally-religiously very different to each other. It makes it difficult for public servants implementing charitable choice programs to allocate money if they cannot clearly distinguish between different types of faith-based organisations. Lack of consensus among those employing the terminology of 'faith-based' under Charitable Choice, for example, can result in some public sector funders continuing old practices of allocating money to large traditional church charities, whereas others will seek out small congregations providing social services (Sider & Unruh, 2004:109-110). When there is little agreement about what constitutes a 'faith-based' organisation, let alone 'faith-based' programs, the resulting confusion means that many miss out on funding opportunities. It is also hard to distinguish between congregations that sponsor social service programs with little religious content and programs that possess significant religious content but little formal affiliation to faith-based organisations (Sider & Unruh, 2004:109-110). In effect, American funders, policy makers, and researchers are using 'faith-based' terminology to make distinctions between organisations and religious social services programs that are not the same. The kind of 'religious content and affiliation' now permitted to receive public funding (but previously excluded on legal and constitutional grounds) is open to considerable levels of interpretation with high degrees of ambiguity because of the lack of uniform language surrounding this class of organisation. Until this is worked through, policy makers and researchers will not be able to demonstrate any qualitative or quantitative differences between services offered by secular or faith-based services (Sider & Unruh, 2004:110). Charitable Choice is premised on the belief that faith-based organisations are more suited to providing a range of personal (life-transforming) and social services than either secular or state institutions (Campbell, 2002). This belief will remain intact unless clear distinctions can be made between faith-based organisations, and secular and state institutions and, more importantly, between different kinds of faith-based organisations.

The ambiguity surrounding 'faith-based' language is not a new problem in the American context. A number of scholars have developed a series of typologies to address this deficit. They fall roughly into two categories. Some researchers focus on the institutional characteristics employing concepts derived from organisational theory (Jeavons, 1997; Smith & Sosin, 2001). Others focus on the level of integration of religion icons, symbols, practices, and content in social service programs (Monsma, 1996). Groggin & Orth (2002) and Unruh (in press cited in Sider & Unruh, 2004) have attempted to integrate both aspects into multidimensional typologies of faith-based organisations. They argue these typologies more accurately reflect the range of faith-based organisations, as they enable more finely grained analysis between overt (and covert) forms of religious expression and different organisational entities to emerge. Unruh (in press cited in Sider & Unruh, 2004:116) has developed a classificatory taxonomy of faith-based organisations based an analysis of institutional arrangements and program characteristics. The five types of faith-based organisations proposed are: faith-permeated organisations, faith-centred organisations, faith-affiliated organisations, faith-background organisations, and faith-secular partnerships.

Nevertheless, such typologies are inherently problematic. For instance, any 'ideal type' is hard to apply in practice, as there is room for considerable variation in classifying organisations (Sider & Unruh, 2004:129). More importantly, they are based on Anglo-centric models of protestant religion and Christianity (Cameron, 2004) and, as such, do not provide appropriate models for other religions (such as Buddhist, Jewish, or Islamic organisations or faith practices (Jeavons, 2004)). A definition of faith-based organisations must be inclusive of the diverse range of faith, institutional, and religious practices in contemporary American society.

Advocates of the Charitable Choice initiatives frequently make a series of 'claims' about the unique characteristics of faith-based organisations (especially small congregations) to deliver a particularistic type of social service, which is substantially different to secular non-profits and the state. It is often asserted that they are capable of providing more 'personalised and friendship-based' intensive care capable of promoting long-term change in lives. It is also asserted that they are more trustworthy than other kinds of organisations, as they are based in local networks and 'grass-root' communities, and that they are more efficient ways to deliver face-to- face social services than large, impersonal, secular bureaucracies (Glenn, 2000). Empirical data, however, tends not support such claims (Farnsley, 2001, Chaves & Tsitsos, 2001, Campbell, 2003).

Research to date indicates that small-congregation faith-based organisations are not doing very well under the Charitable Choice regime. Farnsley's (2001) study found overwhelming evidence that small congregations were disadvantaged in applying for government funds. This was so even with the provision of special assistance and advice in the grant application process. The kinds of disadvantage they experienced are common to all small non-profits. For example, lack of grant writing, budgetary, accountability, and program evaluation skills detracted from the quality of the applications. In addition, poor literacy and program development also contributed to applicants receiving a low rating in the grant process. One of the most surprising findings was the lack of religious activities or references to religion in all but two of the applications. Given that this is what is supposed to distinguish 'faith-based' organisations from secular service providers, this is a major contradictory trend (Farnsley, 2001:106).

Further analysis of empirical research points toward a class and ethnic bias in the uptake of Charitable Choice funding opportunities. In Farnsley's study (2001:108), 95% of churches did not apply for any funding. Furthermore, they may not have been aware they were eligible to do so despite intense efforts by the local Mayor's office to publicise it among local churches. The largest number of applications came from poor and black congregations, with some middle class professionalised congregations entering the fray. Furthermore, upper class and white congregations do not apply for funding, nor do they provide 'congregation-based' social services (Farnsley, 2001: 108).

Other researchers note similar trends. Chaves (1999) found few churches knew about the faith-based provisions. Those who did were predominantly from black churches, new to the funding game. White churches were less likely to engage with government agencies regardless of whether they knew about the funding opportunities or not. This is especially so for conservative churches with an Evangelical background who, by comparison with more liberal mainstream churches, were more proactive in providing social services.

This body of research illustrates that the assumptions made by advocates of American faith-based funding initiates are deeply flawed. Campbell's study (2002) of contracting bids across five Californian counties reveals that small congregations are not responding to funding opportunities. Instead, the take up is from large church agencies and newly formed coalitions of churches and secular organisations, which predate the welfare reform process. The impetus pushing these organisations into the welfare arena is the emergence of a 'revival' of a 'reformulated community development' and 'civic-engagement leadership' training to suit the needs of low income Afro-American and Hispanic neighbourhoods, and new refugee and migrant groups (Campbell, 2002:215-218). It is not welfare reform.

Implications of adopting the discourse

Adopting the discourse in the Australian context is, we suggest, counterproductive for several reasons. First, Anglo-Christian faith practices have been eroded by changing migration patterns over the past two decades. Although most overseas born migrants still come from Western Europe (7.5%--with Great Britain comprising 5.7% of this group), the number of people from the 'three Asian regions' is increasing. They comprised 5.7% of the overseas born population (ABS, 2004a). The growth of these latter groups, and the ethnic composition of the humanitarian and refugee programs, is serving to diversify religious and faith-based practices within Australia. Accompanying this migration is the development of faith-based (ethnic) organisations, which operate uniquely if not idiosyncratically within local (welfare) institutional arrangements.

More important to the debate about the relevance of Charitable Choice initiatives in the Australian context is the changing trends in religious affiliation and practice in this country. ABS data shows (see Table 1 below) that there was a decline in most mainstream churches between the 2001 and 1996 census, but a significant growth among non-mainstream churches such as the Church of the Latter Day Saints, the Oriental Christians (16%), and the Pentecostals (11%). Catholics and Baptist denominations grew approximately 4% but did not keep pace with the overall population growth (6%). Growth has also occurred among the Orthodox religions. The largest growth (relative to population size) was in the Buddhist (2%), Hindu, Islamic and Jewish communities. The Islamic and Hindu communities grew by 40% since the last census (ABS, 2004b; Hughes, 2004--see Table 1). While there are some methodological problems with the figures, as noted by Hughes (2004), these general trends are indisputable and indicate a change in institutional religious arrangements and practices.

Accompanying such demographic shifts is the increasing secularisation of society. The overall pattern indicates a decline in the importance of religious identity to Australians. Table 2 (below) shows the shifts in religious affiliation in Australia since 1901 (ABS, 2004b). This claim is supported by the findings of the 1998 Community Attitudes Survey (Hughes, 2004:4). In Hughes's analysis (2004:4-5), approximately 30% of people indicated that religion played a role in varying degree in their sense of social and personal identity. A more refined examination of the data reveals some subtle differences in it.</p> <pre> Only (10%) of people indicated that religion was the single most important aspect of their identity. It is seen as more significant than gender, education, country of birth, their occupation, social class, family income or personality. Eleven percent (11%) said that their religious identify was extremely important. A further 12% said it was important, but not in comparison to other attributes (Hughes, 2004:4). </pre> <p>What is important to this current debate is that approximately half of the respondents (45%) said religion played no role in their sense of identity at all (Hughes, 2004:4).

Age is also an important dimension driving changes in religious practices and beliefs. There have been significant shifts in young people's perceptions of spirituality and religions affiliation (Powell, 2004), in the 18-24 year old age group (Powell, 2004). In the 2001 ABS census, 8% of young people (18-24 year olds) claimed to be Christian. In contrast, 13% claimed Buddhist affiliation, while 20% claimed they had no religion. Furthermore, the largest group of Hindu and Muslim faith affiliates are younger people (ABS, 2004).

The decline in religious affiliation among mainstream churches has serious implications for the provision of contemporary Australian church-based welfare provision. As Wittberg (2000) notes, these trends point to the increasing secularisation of society, accompanied by (and contributing to) a decreased capacity by the mainstream churches to provide services. It also points to the growth of a new cohort of 'faith-practising' people whose future engagement with mainstream church welfare is, at best, unknown. Furthermore, large-scale charities, such as the Benevolent Society, and church-based institutions, such as Mission Australia, are in the process of 're-inventing' themselves, attempting to discard any vestiges of the 19th century charitable and paternalistic model of welfare, and embracing the rhetoric and practice of social entrepreneurialism. These organisations are responding to an external environment that is demanding that they obtain a larger share of funds from non-state sources. Further, the largest growth in church-based (or congregation-based) services in Australia is in education (independent schools) and in aged care, both of which are areas that (at present) fall outside of the welfare reform agenda. For all of these reasons, we suggest that the language of American religious-based welfare reform is at odds with the contemporary conditions of 21st century Australia.

Contemporary agenda for research

As indicated above, research into the role of religious organisations in contemporary Australian welfare is still minor despite the large economic contribution they make to its provision in the non-profit sector. The insertion of the faith-based discourse into the Australian context immediately raises the debate about whether religious organisations can, and should, run welfare reform. Attendant to this are questions about the capacity of small-congregation based organisations to provide comprehensive welfare programs. The American evidence shows that they tend to provide small, localised, and short-term emergency relief and not large-scale projects and programs (for example, by providing temporary shelters for the homeless rather than longer-term housing options for low income and poor families). As Harris's (1995) work indicates, the potential to change this focus is largely untested, and where change does occur, it is sporadic.

An ongoing area of interest is the extent to which church-based organisations will try to insert their religious beliefs and practices into the delivery of welfare services. The American data provides a mixed picture. Farnsley's (2001) study shows that many newcomers tend to leach out religious content, whereas Campbell's (2002) study reveals a range of responses to this issue. The Catholic Church, for example, interpreted religious practice to mean social justice advocacy for the poor in government and county forums. By comparison the Methodists used Charitable Choice to expose congregational members to 'poor people', but not as an opportunity to proselytise. The Salvation Army continued to emphasise self-help based on its long-standing evangelical approach to its work. One of the few Australian studies of the impact of religious ideology and practices on the delivery of emergency relief services by church-based organisations found that the punitive and deserving culture of the 19th century was the dominant discourse in the latter decades of the 20th century (Wearing, 1998). The diversity of approaches once again underscores the importance of not classifying church or faith-based organisations into one homogenous whole. They are diverse and eclectic in their approaches and practices. Given that major growth is occurring in the non-mainstream and non-Christian religions in Australia, new forms of welfare assistance based on a range of religious practices and tenets may emerge in the future. It will be important to track these changes, as they could signal either a return to 19th century welfare models or a more radical, community--communitarian model combined with a hybrid form of social entrepreneurialism without overt religious practices or ideologies.

Conclusion

In summary, we suggest that the casual use of the language of American welfare reform--for this is what the term 'faith-based' refers too--is not particularly appropriate or helpful in the Australian context and at this juncture unless considerably more thought goes into exactly what we mean. Uncritical adoption of American 'faith-based' welfare language unwittingly imports American-style welfare reform and, more importantly, a model of service delivery that has not proved itself particularly effective in terms of the policy goals it proposed to pursue. At a minimum, it predisposes Australian users to unreflective engagement with a mode of welfare reform that sits at odds with Australian institutional relations and social conditions. At worst, it has the potential to promote beliefs and myths about the role of the non-government sector that distort our capacity to comprehend the complexities of the Australian mixed economy of welfare. That said, there is nothing wrong with explicitly developing and articulating localised and indigenous meanings of such terminology, particularly if it will contribute to ongoing debates about the institutional model of 'welfare' we wish to progress in the 21st century. Such a meaning should, however, acknowledge both the local past and the local present if it is to have any real intellectual purchase in Australia debates. Finally, we suggest that Australian versions of 'faith-based' service delivery should be developed from and grounded in sound empirical research.

References

Australian Bureau of Statistics (2004a) Migration, Australia, 2004, Catalogue no. 3412.0, Canberra.

Australian Bureau of Statistics (2004b), 'Culture and Recreation, Religion', Year Book Australia, Catalogue no. 1310.0, Canberra.

Austin, M.J. 2003, "The Changing Relationship Between Nonprofit Organizations and Public Social Services Agencies in the Era of Welfare Reform', Nonprofit and Voluntary Sector Quarterly, 32, 1, p 97-114.

Belcher, J.R. Fandetti, D. and Cole, D. 2004, 'Is Christian Religious Conservatism Compatible with the Liberal Social Welfare State?' Social Work 49 (2): 269-276.

Border, R. 1962, Church and the State in Australia 1788-1872, London, SPcK.

Cage, R. A. 1980, 'The Origins of Poor Relief in New South Wales: An Account of the Benevolent Society', 1813-62', Australian Economic History Review, Vol 20, pp 153-69.

Campbell, D. 2000, 'Beyond Charitable Choice: The Diverse Service Delivery Approaches of Local Faith-Related Organizations', Nonprofit and Voluntary Sector Quarterly, Vol. 31, No. 2, pp 207-230.

Castles, Francis. 1983, The Working Class and Welfare, Sydney, Allen and Unwin.

Challen, M. B. 1996, 'The Changing Roles of Church and State in Australian Welfare Provision', Social Security Journal, June: 26-31.

Chaves, M. 2001, "Religious Congregations and Welfare Reform' Society, 38, 2, p 21-27.

Chaves, M. and Tsitsos, M. 2001, 'Congregations and Social Services: What They Do, How They Do It, and With Whom', Nonprofit and Voluntary Sector Quarterly, Vol, 30. No. 4, pp 660-683.

Cleary, M. 1994, 'Management Dilemmas Facing Catholic Human Service (Health, Education and Welfare) Organisations in Australia', in Melville, R. (ed) Third Sector Welfare: The Industry Commission's Inquiry and the Future of Non-Government Organisations, Discussion Paper No. 5, UNIYA, Sydney.

Cnaan, R.A. 1999, The Newer Deal: Social Work and Religion in Partnership, New York, Columbia University Press,

Cnaan, R.A. 1997, Social and Community Involvement of Religious Congregations Housed in Historic Religious Properties: Findings from a Six City Study, Philadelphia, Program for the Study of Organized Religion and Social Work, University of Philadelphia.

Cnaan. R.A. and Boddie, S.C. 2002, 'Charitable Choice and Faith-Based Welfare: A Call for social Work', Social Work, 47 (2):224-235.

Dalton, T. et al 1996, Making Social Policy in Australia: An Introduction, Sydney, Allen and Unwin.

Dean, M. 1999, Governmentality: Power and Rule in Modern Society, London, Sage.

Dickey, B. 1987, No Charity There: A Short History of Social Welfare in Australia, Sydney, Allen and Unwin.

Farnsley, II A. E. 2001, 'Can Faith-Based Organizations Compete?' Nonprofit and Voluntary Sector Quarterly, Vol. 30, No. 1, pp 99-111.

Garton, S. 1994, 'Rights and Duties: Arguing charity and welfare 1880-1920' in Wearing, M. and Berreen, J. (Eds) Welfare and Social Policy in Australia, Sydney, Harcourt Brace.

Glenn, C. L. 2000, The ambiguous embrace: Government and faith-based schools and social agencies, Princeton, NJ, Princeton University Press.

Gleeson, D. J. 1996, 'Catholic charity during the 1930s Great Depression', Australasian Cathlolic Record, Vol, 73, No. l, pp 68-80.

Gleeson, D. J. 2000, 'Professional Social Work and Welfare Bureaus: The Origins of Australia Catholic Social Work', The Australian Catholic Record, Vol. 77, No. 2, April.

Graycar, A. & Jamrozik, A. 1994, How Australians Live: Social Policy in Theory and Practice, 2nd edition, South Melbourne, Macmillian.

Harris, M. (1995) 'Quiet care: Welfare work and religious congregations', Journal of Social Policy, Vol. 24, No. 1, pp 53-71.

Hiscock J. and Pearson, M. 1999, 'Looking Inwards, Looking. Outwards: Dismantling the "Berlin Wall" between health and social services', Social Policy and Administration, V33, 2: 150-163.

Hopkins, J. and Cupaiulolo, A. 2001, 'For Better or For Worse? Faith-based social work', Policy and Practice, 65: 24027.

Hughes, P. (2004) Trends in Religious Identification: Details from 2001 Census, Christian Research Association, August, [available online] http://www.cra. org.au

Industry Commission. 1995, Charitable Organisations in Australia, Report No. 45, June 16, Melbourne, AGPS.

Jeavons, T. H. 2004, 'Religious and Faith-Based Organizations: Do We Know One When We See One? Nonprofit and Voluntary Sector Quarterly, Vol, 33. No, 1. pp 140-145.

Kennedy, R. 1985, Charity Warfare: The Charity Organisation Society in Colonial Melbourne, Melbourne, Hyland House.

Lyons, M. 1995, Third Sector: Contributions of nonprofit and cooperative enterprises in Australia, Sydney, Allen & Unwin.

McDonald, C. 2000, 'The Third Sector in the Human Services: Rethinking its Role' in I. O'Connor, P. Smyth and J. Warburton (eds). Contemporary Perspectives on Social Work and the Human Services: Challenges and Changes Melbourne, Longman, 84-99.

Milligan, V., Hardwick, J. and Graycar, A. 1984, Non-Government Welfare Organizations in Australia: A National Classification, SWRC Papers and Proceedings, No. 51, Sydney, University of New South Wales.

Olasky, M.M. 1992, The Tragedy of the American Community, Washington D.C., Regnery Gateway.

Pipes, P.F. and Ebaugh, H.R. 2002, 'Faith-based coalitions, social services and government funding', Sociology of Religion, 63 (1): 49-69.

Powell, R. (2004) Why People Don't Go to Church ... And What Churches Can Do About It, Christian Research Association, July, [available online] http:// www.cra.org.au

Sherman, A. 2001, 'Faith-based organisations and welfare reform', Policy and Practice, 59: 16-19.

Skillen, J.W. 2000, 'Welfare Reform and Faith-Based Organisations', Journal of Religion, 80, 555.

Smith, S. R. & Sosin, M. R. 2002, 'The Varieties of faith-based agencies', Public Administration Review, Vol, 61. No, 6. pp 651-670.

Swain, S. 1992, 'Besmirching our reputation: sectarianism and charity in Geelong', Victorian Historical Record, Vol. 63, No. 1, p 51-60.

Walter, U.M. and Petr, C.G. 2000, 'Template for family-centred interagency collaboration', Families in Society: The Journal of Contemporary Human Services, V81, 5: 494-503.

Wards, R. & Humphreys, R. 1995, Religious Bodies in Australia: A Comprehensive Guide, 3rd revised edition, Wantirna, New Melbourne Press.

Wearing, M. (1998) Working in Community Services: Management and Practice, Sydney, Allen & Unwin.

Winkworth, G. & Camilleri, P. 2004. 'Keeping the Faith: The Impact of Human Services Restructuring on Catholic Social Welfare Services', Australian Journal of Social Issues, Vol 39, No. 3, pp315-328.

Wittberg, P. 2000, 'Called to Service: The Changing Institutional Identities of American Denominations', Nonprofit and Voluntary Sector Quarterly, Vol. 29, No. 3, pp 357-376

Wuthnow, R. 1988, The Restructuring of American Religion, Princeton, Princeton University Press.

Wuthnow, R. Hodgkinson, V. and Associates. 1990, Faith and Philanthropy in America: Exploring the Role of Religion in America's Voluntary Sector, New York, Jossey Bass. Table 1: Religious Affiliation 1996 2001 Change '000 % '000 % % Christianity Anglican 3,903.3 22.0 3,881.2 20.7 -0.6 Baptist 295.2 1.7 309.2 1.6 4.8 Catholic 4,799.0 27.0 5,001.6 26.6 4.2 Churches of Christ 75.0 0.4 61.3 0.3 -18.2 Jehovah's Witness 83.4 0.5 81.1 0.4 -2.8 Lutheran 250.0 1.4 250.4 1.3 0.2 Orthodox 497.0 2.8 529.4 2.8 6.5 Pentecostal 174.7 1.0 194.6 1.0 11.4 Presbyterian and 675.5 3.8 637.5 3.4 -5.6 Reformed Salvation Army 74.1 0.4 71.4 0.4 -3.7 Uniting Church 1,334.9 7.5 1,248.7 6.7 -6.5 Other Christian 420.6 2.4 497.9 2.7 18.4 Buddhism 199.8 1.1 357.8 1.9 79.1 Hinduism 67.3 0.4 95.5 0.5 41.9 Islam 200.9 1.1 281.6 1.5 40.2 Judaism 79.8 0.4 84.0 0.4 5.2 Other religions 68.6 0.4 92.4 0.5 34.6 No religion 2,948.9 16.6 2,906.0 15.5 -1.5 Not stated/inadequately 1,604.7 9.0 2,187.7 11.7 36.3 described Total 17,752.8 100.0 18,769.2 100.0 5.7 Source: ABS data, 1996 and 2001 Censuses of Population and Housing (Year Book Australia, 2004). Table 2: Major Religious Affiliations Christianity Anglican Catholic Other Total Census year % % % % 1901 39.7 22.7 33.7 96.1 1911 38.4 22.4 35.1 95.9 1921 43.7 21.7 31.6 96.9 1933 38.7 19.6 28.1 86.4 1947 39.0 20.9 28.1 88.0 1954 37.9 22.9 28.5 89.4 1961 34.9 24.9 28.4 88.3 1966 33.5 26.2 28.5 88.2 1971 31.0 27.0 28.2 86.2 1976 27.7 25.7 25.2 78.6 1981 26.1 26.0 24.3 76.4 1986 23.9 26.0 23.0 73.0 1991 23.8 27.3 22.9 74.0 1996 22.0 27.0 21.9 70.9 2001 20.7 26.6 20.7 68.0 Not stated/ Other No inadequately religions religion described Total Census year % % % '000 1901 1.4 0.4 (a) 2.0 3,773.8 1911 0.8 0.4 (a) 2.9 4,455.0 1921 0.7 0.5 (a) 1.9 5,435.7 1933 0.4 0.2 12.9 6,629.8 1947 0.5 0.3 11.1 7,579.4 1954 0.6 0.3 9.7 8,986.5 1961 0.7 0.4 10.7 10,508.2 1966 0.7 0.8 10.3 11,599.5 1971 0.8 6.7 6.2 12,755.6 1976 1.0 8.3 11.4 13,548.4 1981 1.4 10.8 11.4 14,576.3 1986 2.0 12.7 12.4 15,602.2 1991 2.6 12.9 10.5 16,850.3 1996 3.5 16.6 9.0 17,752.8 2001 4.9 15.5 11.7 18,769.2 (a) Includes 'object to state'. Source: ABS data, Census of Population and Housing (Year Book Australia, 2004).