A more pertinent trade-based argument is that income inequality can be induced by an acceleration in the diffusion of skill based technology. For technology to raise inequality, it must be the case that the more technologically advanced industries have a higher relative use of skilled labour, so that the diffusion of these technologies increases the demand for skilled labour. This certainly appears to have been the case in the UK and US throughout the 1980s and so in this way help to explain some of the rise in inequality.
Other theories have attempted to attribute the rise in income inequality through this shift in demand away from unskilled labour towards skilled workers. In the US and the UK it is argued that this lead to a fall in the relative wage of unskilled workers and hence a rise in dispersion. This may have occurred firstly because of increased trade liberalization which increases competition from other countries where unskilled labour is abundant, reducing the relative wage of unskilled labour. Other explanations have also been advanced, notably that technological change has been biased towards skilled labour with the introduction of automation and IT. Firstly, it is argued that computer decision making has significantly substituted for human decision making in modest cognitive tasks. Secondly, computers in organisations have been complementary to large bundles of cognitive skills. Thirdly, organizational computing has been complimentary to people skills, raising their return. Fourthly, these changes are assumed to have come at a time and in the industries where organizational computing has had its largest impacts. Hence, demand for skilled labour has risen causing an increased dispersion between skilled and unskilled labour wages.
It has been argued that this demand-based argument is too narrow and fails to realize that wage determination is partly affected by interacting as members of a society. One source of social interaction in wage determination in the labour market is through collective bargaining by unions. If it is the case that unions lead to a lower earnings dispersion (and this seems reasonable to propose), then a valid explanation for the rise in dispersion in the UK and US is the decline in union power and coverage. In the UK, the proportion of union members fell from 54% in 1980 to 32% in 1995, whereas in the US the unionization rate fell from 25% to 16% between the years 1975-85. Gooling and Machin (1993) concluded that decline in unionization accounted for around 15-20% of increased earnings dispersion in the 1980s. This argument is particularly compelling with respect to the UK and US experiences, because the other explanations presumably would impinge on all countries in a similar way whereas this explanation focuses specifically on an Anglo-American phenomenon.
Another area in which the weakening of labour market institutions has had an effect on inequality is through the abolition of wage guarantee policies. For example, the abolition of Wages Council protection for lower paid workers in the UK has been shown by Dickens (1994) to have increased inequality because Wages Councils significantly compressed the distribution of earnings in Wage Council industries. Kard and Krueger (1993) also found that in the US the fall in the real value of the minimum wage had caused 20-30% of the rise in wage dispersion.
However, to fully understand why the US and UK has seen such phenomenal and typically sharp rises in income inequality, particularly in the 1980s, it is necessary to look at the political environment in these countries at the time. This is important firstly because it determines how progressive of the tax system is, which is an important mechanism in counteracting rising inequality. In the UK in 1979, the highest rate of tax on income was 83%, which was cut to 40% by 1988. The basic tax rate was also cut from 33% to 30% initially, reaching 24% by 1996. Giles and Johnson (1994) have shown that these tax changes disproportionately benefited the better off by calculating the level of taxes that would have to have been paid by various decile groups in 1995 if the 1975 tax system had still been in force They show that the 1995 system took about 30% of the income of the top decile, whilst if the 1985 system had been in place 35% of their income would have been paid in tax. Differences at the lower end are much less marked, implying that the rich have done relatively well out of the tax changes in the UK over the 1980s.
Secondly, the political environment is important because it determines attitudes, and particularly with respect to how much inequality is to be acceptable by society. Both Thatcher and Reagan in the 1980s, through their highly laissez-faire economic policies, reflected in their attitude towards the role of labour institutions and the tax system, fostered a climate in which greater levels of income inequality became socially acceptable. Akerlof (1981) has developed a model in which individuals’ utility depends not only on their own income but also on reputation and conformity with the social code. In the 1980s, it may be the case that the social code changed, setting wages at long run equilibria which were seen to be fair rather than the market clearing wage. It may be therefore concluded that the notion of fairness changes allowing for greater dispersions in wages across groups.
In conclusion, therefore, it appears as though skill-based technological change arguments and trade-based arguments may help to explain some of the underlying movements in income inequality seen across a number of countries. However, to explain why income inequality rises have been so marked in the US and the UK, factors specific to these countries must be the major cause. This occurred because of a change in the political climate, bringing about a reduction in the equalizing role of labour market institutions, a less progressive tax system and a new notion of what wage differentials society was prepared to accept as fair.
