The recent increase in earnings inequality in a number of industrialised countries is by now a wellrdocumented event. Countries have differed in their experiences, with the most pronounced increases taking place in the UK and the USA. The ratio of the 90th to the 10th percentile of the male wage distribution rose from 2.51 to 3.11 in the UK and from 3.26 to 4.35 in the US over the period 1980-1995.
These changes have been decomposed into three main elements. First, an increase in betweenrgroup inequality, that is, rising educational and agerrelated wage differentials. Second, the dispersion of wages of workers with the same observable characteristics, withinrgroup inequality, has augmented. Third, the lifetime earnings of each individual have become more volatile.
Work on the first two aspects is abundant and well discussed. For examr ple, Acemoglu (2000) reports that between 1979 and 1995, the wage ratio of university to high school graduates increased by some 25% in the USA. His estimates of withinrgroup inequality also display a sharp increase: the 90-10 log wage residual differential rose by 17% over 1980r1995. In contrast, the increase in the volatility of individual lifetime earnings has received much less attention. In a seminal paper, Gottschalk and Moffitt (1994) use individual data to examine changes in transitory? movements in white males earnings in the US.
They decompose each worker/s earnings into permanent and transitory income, and compare their behaviour in the 1970s and the 1980s. Their results indicate that a third of the overall increase in the disr persion of earnings is due to a rise in the instability of earnings, and that a substantial part of the widening in the distribution of earnings is due to a greater volatility of the wages received by each particular individual staying in the same job. An increase in the variance of the transitory component of income is also found for the UK by Blundell and Prescott (1998).
