Inequality: the Kuznets inverted U
A second important structural change that occurs with development is that of
inequality. In a 1955 paper, Simon Kuznets asked whether inequality in the
distribution of income increased or declined as a country experienced
economic growth. The likelihood of increasing inequality (a rising share of
profits in national income and stagnant wages) was implied if not discussed in
detail in the Lewis model. Kuznets used data for the US, England and Germany
showing that rising inequality during the early stages of the Industrial
Revolution was reversed after the 1920s. For example, the income share of the
bottom quintile in the US increased from 13.5 per cent in 1929 to 18 per cent
in the mid-1940s. With a small sample and unreliable data Kuznets admits that
the attempt to discern any trend ‘comes perilously close to pure guesswork’
(1955:6), but he nevertheless drew some plausible and influential hypotheses
from his findings. The early tendency to widening inequality, Kuznets argues,
can be explained by the concentration of savings among the wealthiest house-
holds. In 1955 the top 5 per cent of income units in the US accounted for nearly
two-thirds of savings and so accumulated the bulk of income-yielding assets.
The shift from agriculture to industry will permit households that migrate early
to take advantage of high incomes in the industrial sector. Kuznets suggested
that incomes from services, professional and entrepreneurial sources (rather
than accumulated property income) would tend to increase their share of total
incomes over time. Continued migration from the agricultural sector would
reduce returns in urban industry and gradually equalize them across the whole
economy (there are parallels with the Lewis model). Any remaining differ-
ences in incomes would then be determined by individual excellence rather
than inherited position, or as Kuznets put it, inequality would be undermined
by the ‘the dynamism of a growing and free economic society’ (1955:11).
Kuznets’proposition concerning the rise then fall of inequality, which has
since become known as ‘the inverted U-shaped hypothesis’, has been subject
to much subsequent testing, some scholars finding that sustained growth will
ev entually reduce inequality, others that a decline in inequality will either take
decades or require supporting policies (Randolph and Lott,1993). The model
cannot account for some recent inequality stories such as the dramatic lurch
into inequality in Russia, where the range of incomes between the top and
bottom decile rose from a multiple of 4.5 in December 1991 to 13 in mid-1994
(Nolan,1995:295). There is a heated contemporary debate about whether
government policy can master this process and enforce ‘pro-poor’ economic
growth such that economic growth disproportionately benefits the poorest.
The role of agriculture in structural change
The share of GDP or employment accounted for by agriculture typically
declines relative to other sectors during the course of growth and development.
168 Patterns and Determinants of Economic Growth