A significant part of the literature dealing with income distribution in the U.S. suggests that the growing disparity between earnings is a consequence of the dynamics of technological change and that this behaviour is the factor explaining the increasing inequality of income between households. This explanation of the “skill-biased technological change” argues that technological change has caused a critical change in the demand of labour, producing a strong growth of the best-paid and most skilled jobs (with a higher level of education and professional training), at the expense of the less skilled and low wage jobs. This change would be the main cause of the increasing wage inequality among the employees and, consequently, of the greater gap among the households’ money income (Danzinger and Gottschalk, 1996). To the new technologies introduced in the 80s, that theory adds the complementary effect of some other factors, such as international trade, the increase of immigration and the delocalization of some production activities of U.S. firms. External trade promotes the internal specialisation in goods and services with a higher technological content and greater export orientation, which accentuates the demand of skilled labour, whereas both the growth of imports and the productive delocalization affect the low technology and low-skilled labour-intensive activities. The arrival of workers coming from other countries increases the low-skilled labour supply, widening the gap between demand and supply in that type of labour and, consequently, the gap in wages.
The indicators measuring income distribution prove clearly that during the last three decades has been a strong increase of inequality among the United States’ households, following this sequence:
-
between 1968 and 1980 the growth of households’ income decelerated, whereas inequality grew at the expense of the households’ located in the quintiles Q23;
-
between 1980 and 1992 income experienced a limited expansion, at the same time that inequality intensified at the expense of Q1234;
-
between 1993 and 2000 income growth increased again and inequality kept on accentuating against the distributive position of the quintiles Q234 and percentiles 81–95, and favouring the top-5.
The analysis reveals a greater presence of the part-time employment with low earnings, and also a growing participation of women in the labour market, obtaining considerably lower mean income than men, despite the reduction of income inequality between both genders. Simultaneously, a profound transformation of the professional and retributive structure has happened, with a strong dominance of the managers-specialists and the support personnel in sales and technical tasks; the specialised and subaltern labourers, together with the administrative personnel are the categories that have lost more presence in employment and earnings. It seems logic to argue that those changes have been partly determined by technological changes and that it exists a real relation between the technological dynamic and the formation of those earnings, and also between the mentioned earnings and households’ income distribution.
However, this analysis rejects that both relations are unidirectional, since it questions: (1) that the disparity among earnings could be explained as a consequence of labour productivity changes; and (2) that the earnings inequality was the only or the main factor influencing the growth in inequality among households.
As far as the first issue is concerned, if we consider that the disparity among earnings has been determined by productivity, then it seems inexplicable that period after period the production per hour of work has maintained a moderate rhythm of growth whereas the real wage per hour of work has declined from 1973 to 1995. It is not possible to accept that the technological change affects positively only the minority of employees that increased wages per hour of work, whereas the majority of employees experienced a standstill or even a decline in their wages per hour of work. Firstly, the factor that seems more closely linked to the growth of the employees’ mean income has been the increase of the number of working hours. This is what happened in 1968–80 and more importantly also in the two following periods: 1980–92 and 1993–2000. Therefore, this analysis reveals that the existing inequality between earnings has not been caused by differences in efficiency, but—in presence of the technological dynamics—by the change in business strategies, deregulation policies, the ‘precarisation’ of wide segments of the employed population and other socio-institutional conditions.
As regards to the second issue about the link between earnings and households’ income, indicators show that in the period 1968–80 inequality in E distribution declined, while MIH increased, whereas in the two following periods we can observe greater inequality in both. However, this connection does not express either a unidirectional relation if we take into account that inequality increased more rapidly in E than in MIH, but the Gini coefficient of E keeps on being lower than MIH's; hence, we can even recognize this fact when we consider only men's earnings, who have greater inequality among them than women's but have a lower coefficient than MIH's. Therefore, in order to explain MIH's inequality we have to take into account the remarkable influence exerted by the remaining two components of the household's income: property incomes and government transfers, since the earnings’ average income only represents 60% of the households’ mean income.
