The Australian Economic Review, vol. 33, no. 4, pp. 355–62  The University of Melbourne, Melbourne Institute of Applied Economic and Social Research 2000 Published by Blackwell Publishers Ltd, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA * We are indebted to Boyd Hunter, John Taylor and an anonymous referee for comments on previous drafts of this paper. 1. Introduction Noel Pearson has recently suggested that ‘to survive as a people’ his indigenous compatriots in Cape York will have to ‘get rid of passive welfare out of Aboriginal governance’ (2000, cover). Sentiments such as these are not en- tirely new. Soon after indigenous Australians were legislatively included in the social secu- rity system in 1959 and 1966, ideas about how this would undermine indigenous community life began to be voiced (see, for example, Har- ris and Turner 1976). In the 1970s, when unem- ployment payments, which became known as ‘sit down money’, were gradually spreading into remote Aboriginal communities, ideas were produced by Nugget Coombs and others for a more active alternative which would in- volve working for welfare (see Coombs 1977a, 1977b; Sanders 1985). Such an alternative, known as the Community Development Em- ployment Projects (CDEP) scheme, came to pass in a small number of remote Aboriginal communities from May 1977. The scheme proved immediately popular and led to con- certed pressure for expansion to other Aborigi- nal communities. However, administrative and budgetary problems with the scheme restricted that expansion until the mid 1980s (Sanders 1988), by which time unemployment payments had become commonplace in indigenous com- munities all over Australia (Sanders 1985). From the mid 1980s the CDEP scheme ex- panded rapidly—from 30 communities and 3000 participants in remote areas only to 250 communities and over 30 000 participants in urban as well as remote areas a decade later. The number of indigenous Australians work- ing for welfare by the mid 1990s was of consid- erable significance. It represented perhaps a quarter to a third of all indigenous employment and consumed a third of the budget of the Ab- original and Torres Strait Islander Commission (ATSIC). This budget proportion reflected the fact that although the CDEP scheme has been linked to social security entitlements, it has not formally been part of the social security sys- tem. Since its inception, it has been a program within the Commonwealth indigenous affairs portfolio. In this paper we explore what difference it makes to the socioeconomic status of indige- nous Australians to be participating in the CDEP scheme. We utilise two data sources: the 1994 National Aboriginal and Torres Strait Is- lander Survey (NATSIS) and the 1996 Census. We compare the incomes of the CDEP em- ployed with indigenous people in ‘mainstream’ employment, the indigenous unemployed and those not in the labour force. In the process, we also note major geographic differences in la- bour market shares for the CDEP employed, mainstream employed, unemployed and indig- enous people not in the labour force. We also examine hours worked for the CDEP employed compared to the mainstream employed. Finally Policy Forum: Welfare for Indigenous Australians Indigenous Australians Working for Welfare: What Difference Does It Make? J. C. Altman, M. C. Gray and W. G. Sanders* Centre for Aboriginal Economic Policy Research The Australian National University