tion may be justified. Barr's argument is useful, but the dividing line between scale and structure is not always clear, as illustrated, for instance, by the case of top-up vouchers (Glennerster, 1996). The argument also fails to deal with issues such as the techniques used in providing social benefits, which may matter a great deal.
Of course, changing conditions demand social-sector reform everywhere. Although the reasons for this may seem self-evident, they should be recapitulated briefly here because of their relevance to transition countries:
Globalization. Globalization demands (1) open and competitive economies, which can initiate an upward spiral in efficiency and a downward spiral in labor costs, and (2) basic changes in labor markets, the effects of which range from deindustrialization to large-scale, long-term unemployment. In countries undergoing transformation, the challenges posed by globalization are especially great, because they are undertaken in conditions of sharply reduced gross domestic products (GDPs) and abrupt increases in prices, unemployment, and black market activity (Standing, 1996).
Demographic trends. Demographic trends include (1) increasing longevity; (2) declines in fertility; and (3) changing family patterns, particularly the multiplication of single-parent families. In many transition countries, these problems are being aggravated by a general health crisis, including increasing mortality and the spread of infectious diseases (Cornia et al., 1996; Nauck and Joos, 1995).
Tendencies that weaken solidarity and increase individualism. Many of the relatively better off now espouse "postmodern" attitudes toward former social policy arrangements. They demand better value for their money, including more individually tailored services and more choices. They are also less willing to share the collective costs of change. All these phenomena are appearing in transition countries, perhaps in an accentuated form as a socio-psychological reaction to the sudden lifting of former oppressive controls.
The modern welfare state has not coped well with the consequences of these trends, even in the West. Hence we have seen the emergence, even in many of the most developed and relatively affluent societies, of social fragmentation and threats of massive social exclusion and disaffiliation.
These developments may not be inevitable, however. Resources in developed countries remain plentiful; it is only their distribution that has become more unequal (World Bank, 1996a; Atkinson et al., 1995). Many analysts are searching for new insights, arguments, and instruments to address these trends. In an interesting sign of this new thinking, the eminent capitalist and financier George Soros (1997) recently warned about the potentially dangerous social consequences of unconstrained market domination.
Civil society's increasing resistance to attempts to scale back the welfare state-as indicated by demonstrations in several Western European countries