Income inequality: a highly controversial subject

The most popular economic topic of recent months was low growth. Nowadays the focus seems to have shifted towards another hot topic: income inequality.
Last week Deputy Prime Minister Ali Babacan declared that economic growth must alleviate poverty. Umut Onaran, a deputy chairman of the Republican People's Party (CHP), published a very critical evaluation of the Justice and Development Party (AK Party) government's economic performance, claiming that the standard of living in Turkey has not improved, while income inequality has worsened. As the elections approach, I think we will be talking more about inequality and poverty. Representation of the income distribution by Lorenz curve and the Gini coefficient  What do the figures say about this controversy? Since 2006, the Turkish Statistics Institute (TurkStat) has published some measures of inequality based on Income and Living Conditions surveys, whose samples are sufficiently large to give a picture at the regional level. The 2012 figures were published recently. Thus, the period in question does not cover the entire AK Party era (2003-2012), but it allows us to observe the evolution of income inequality in the recent period. Before considering the TurkStat figures, let us recall that the relationship between economic growth and income inequality is not straightforward. Since the work of Simon Kuznets, it is largely believed that income inequality increases in the first years of economic development, while it begins to decline in the years of maturation. This inverted “U” form of the so-called “Kuznet's curve” is largely explained by the transition from an agricultural economy, where income inequality is relatively low, to an industrial economy characterized by low wages in the first years while wages increase along with labor productivity during the years of maturation. However, income inequality measures show that in a large number of developed countries income inequality continued to increase from the 1990s due to an increase in wage inequality between wage laborers, the reason being the widening wage gap between qualified and unqualified labor. Thus, it is not certain that economic growth will alleviate poverty automatically. Having said that, let's look at the TurkStat figures. TurkStat publishes two measures of inequality, the Gini coefficient and the S80/S20 ratio, plus one measure of poverty, the Relative Poverty (RP) headcount ratio. The Gini coefficient, varying from 0 to 1, is a measure taking into consideration the differences in income from each individual to others. The value of 0 means perfect equality while 1 means that one individual has the entire income. The S80/S20 measures the relationship of the 20 percent holding the greatest distribution of income to the 20 percent holding the least. It can be considered a more sensitive inequality measure than Gini in terms of changes in poverty. The RP ratio measures the percentage of individuals in the total population with an income amounting to less than 60 percent of the median income. As such, it is a measure of income inequality rather than a direct measure of poverty. What do the data say? The Gini coefficient was at 0.428 in 2006 -- note that this was the highest in Europe -- and then decreased to 0.405 in 2008, increased to 0.415 with the economic crisis of 2009, decreased again with the strong recovery to 0.402 in 2010 and has now been stagnant for two years. The same pattern is observed in the S80/S20 and RP ratios. From 9.5 in 2006, the S80/S20 ratio decreased to 8.1 in 2008 before increasing to 8.5 in 2009. Finally it decreased to 8.0 in 2010 and maintained this level for the following two years. More or less, the RP ratio followed this pattern: From 25 percent, it declined to 22.6 in 2011 and kept this level in 2012, the economic crisis having a very limited effect on RP. Without a doubt, income inequality has decreased, albeit modestly, in the last six years. However, most of the improvement occurred before the crisis, a period during which growth was based largely on gains in labor productivity (see my column “Turkey on the brink of middle income trap,” Sept. 2, 2013). I have also to note that in two regions, İstanbul and the Mediterranean area, income inequality increased in recent years while it was decreasing in others, the three best performers being Western Anatolia, Central Anatolia and the Eastern Black Sea. Although the average growth rate reached 5 percent in the last two years, the improvement in income inequality halted overall while varying a lot between different regions. Will these conflicting patterns have an impact on the AK Party's votes, particularly in İstanbul, next March? We know that voting patterns do not depend only on well-being, but we shall see.