There is a certain irony in the fact that at a time when Western industrial economies are entering their eighth year of solid economic growth, attention has focused on the limitations of the social security system. Governments, for long prepared to tolerate rising income inequality as the price to be paid for increased reliance on market flexibility and market forces, see joblessness and the unequal distribution of (paid) work as requiring a policy response.

Although unemployment has fallen markedly in some Organisation for Economic Co-operation and Development (OECD) countries, including Canada, Denmark, Ireland, New Zealand, the UK and the USA, elsewhere it has either fallen only modestly or actually increased. For the OECD as a whole, the unemployment rate is projected to fall by less than one percentage point between 1997 and 2001 despite average growth over the period of around 3 per cent per annum (OECD 2000). The decline of unemployment in countries such as the UK, USA and New Zealand has been accompanied by substantial rises in income inequality (Easton 2000; Smeeding 2000).

Considerable attention has focused on increasing inequality (particularly in the USA), yet as Nobel Prize winner Amartya Sen has pointed out, inequality in the US income distribution must be seen in the context of its low unemployment rate relative to Europe (Sen 2000). In Australia, this debate has been promoted by the recent report of the Reference Group on Welfare Reform (McClure 2000), which cites inequity in the distribution of work as an issue to be addressed, but makes no explicit reference to the need to reduce inequality in the distribution of income.

Has the combination of globalisation and technological change given rise to a new trade-off between inequality and unemployment? Or is it simply that an ex ante rise in inequality is the trigger for falling unemployment and declining ex post inequality? What has happened to the traditional argument that low unemployment is not only inherently desirable, but is also an effective way to reduce poverty and inequality? Is the only realistic path to lower unemployment for a country like Australia to follow the USA and accept a dose of rising inequality during the transition—and possibly also beyond it?

I think not. A recent study of the rise in US income inequality in the 1990s reveals the extent of the problem and highlights the factors that are driving it (Mishel, Bernstein & Schmitt 1999). Across the middle and lower sections of the US income distribution, the main culprit has been declining wages. Real hourly wage rates fell for the bottom 60 per cent of American workers between 1989 and 1997, resulting in a decline in real median earnings and a rise in median family income of less than US$300. In contrast, executive salaries have soared, increasing on average from 20 times those of the average production worker in 1989 to 116 times as high by 1997. In other words, it now takes the average US executive around two days to earn what the average production worker earns in a year.

Taxation has also played a major role, with the wealthiest 1 per cent of US families experiencing a decline in their annual tax bill of almost US$37 000 between 1977 and 1998. Even the inequities of Australia’s Goods and Services Tax (GST) reforms pale in comparison against these figures. Much of the increase in American inequality has occurred (as it has in Australia) at the top of the distribution. Changes in tax policy are needed to bring the rich back within the sight of the rest of society, though there is little sign of this occurring in the foreseeable future in either country.

However, it is what has been happening at the bottom of the income distribution that has far greater implications for social security. As Australia follows US trends (though to a lesser degree), wages and working conditions are declining at the bottom of the distribution. This is leading inevitably to a situation where the attractiveness of work relative to social security is declining, making it more difficult to induce those on social security into the workforce.

The persistence of high unemployment is thus in part a direct consequence of the rise in earnings inequality and job insecurity generally. Strategies that attempt to coerce the unemployed and other jobless social security recipients back into work will fail (or, at best, achieve only limited success) unless they also address the issues of low wages, job security and unemployment.

At the same time, the changing nature of risk has undermined the support role of social security. One of the initial goals of the welfare state—as Giddens (2000) has emphasised—was to institutionalise the management of risk so as to offer protection against hazards such as disability or job loss that were seen as outside the control of individuals. Over time, however, moral hazard emerged as an issue, particularly as mass unemployment made it increasingly difficult to police eligibility by supplying job offers to welfare clients. We are now in a situation where, far from protecting the individual against uncontrollable external risks, social security allows individuals to avoid the internal risks associated with participating in an increasingly flexible (and thus precarious) labour market.

Thus, while it is entirely rational from the short-term perspective of individual welfare clients to avoid risk by minimising the effort put into job search activity, the longer-term consequence is to entrench them in a cycle of dependency and social exclusion. Over time, their human capital erodes and they become less attractive to employers, apparently confirming their own assessment of poor labour market prospects. Meanwhile, the welfare budget remains high, preventing government from cutting middle-class taxes. These kinds of arguments have caused the focus of social security policy for people of working age to shift away from meeting the needs of people when they are without work, to developing strategies for integrating them (back) into the world of paid work.

In reviewing some of the policy developments that are taking place in Australia and elsewhere, it is important to emphasise that changes are required both in the social security system and in the labour market. An important factor behind the difficulties facing the social security system relates to what is happening in the labour market. Without addressing these labour market issues, reform of the social security system will achieve only limited success.