Surjit S Bhalla: Inequality - Misconceived ideas

IT DOESN'T MATTER

Somewhat surprisingly for many arm-chair experts, inequality in India has stayed broadly constant over the last 25 high-growth years.

My previous article "When will they ever learn?" documented how the pessimism about Indian growth prospects was ill-informed. This article looks at the twin side of growth""inequality. Perhaps not coincidentally, most who are pessimistic about Indian growth are also the most gloomy about trends in inequality. Are they just as wrong? A one-word precise answer: yes.

The "truth" about inequality and inequality change can only be assessed through results of household surveys, and there are several organisations conducting such surveys in India. It is quite possible that all of these surveys just do not reveal the perceived "truth" about either inequality levels, or change in such levels. If that is the case, then no one can state with any conviction what the trends in inequality have been. And we can stop the discussion here.

But we all have opinions, beliefs, and, most importantly, convictions. The common conclusion about inequality in India is that not only has it been worsening, but worsening sharply. The recent angst expressed by Prime Minister Manmohan Singh on CEO salaries is an indicator of what is believed in important policy circles, and more generally, by the chattering classes. Not only are the rich getting richer, but horribly so. For evidence, commentators point to a similarly large developing country, China. It has experienced high growth, and a large increase in inequality. And authors, even Nobel laureates like Michael Spence, head of the prestigious growth commission, are not immune to the "if it is happening in China, it is happening in India" syndrome. "Inequality often rises in the growth process ... In China, the bottom 10 per cent of the income distribution has seen its income rise by 42 per cent in the past 10 years. The middle has grown by 115 per cent and the top 10 per cent by 168 per cent. Large numbers. Everyone has benefited but not equally. Similar patterns can be seen in other rapidly growing countries such as India" (Leipziger-Spence, Financial Times, May 14, 2007).

Note the gratuitous add-on phrase "such as India". Obviously, it would be appropriate if it were true.

The most widely used, and quoted, results on inequality have been conducted by the National Sample Survey Organisation (NSSO) of India. This organisation has been conducting consumption inequality surveys since 1951. The most popular measure of inequality is the Gini coefficient. If all individuals have the same consumption (or income), then the Gini has a value of 0, perfect equality. If one individual has all the consumption, the Gini has a value of 100""perfect inequality. 

CONSUMPTION INEQUALITY IN INDIA, 1951-2005
Year	Nominal	Real	Nominal	Real
1951		35.5
1977		32.0
1983	32.5	31.9	30.3	29.4
1987/88	32.9	31.3	30.2	28.4
1993/94	32.5	30.3	30.1	27.7
1999/00			32.0	28.8
2004/05	36.3	32.5	34.6	30.5
Source: NSSO consumer expenditure surveys Notes: The difference between the old and new definition of consumption is that the new definition accounts for some consumer durables on a 365-day recall basis rather than a 30-day basis

The table reports the consumption Gini for selected years prior to 1983, and all large sample NSSO surveys conducted since then, including the most recent 2004-05 survey. Two Gini measures are reported""a nominal Gini (inequality unadjusted for price differences), and a real Gini (inequality adjusted for cost of living differences between states and between rural and urban areas). Clearly and obviously, inequality should be measured in real terms; a person living in Mumbai having a 10 per cent higher income is most likely poorer than a person living in rural Bihar.

The pattern of real inequality in India is as follows: a large decline in inequality between 1951 and 1983, and then a flattening. For the high-growth period 1983-2005, the pattern is a V shaped one""a small 5 per cent decline in inequality between 1983 and 1994, and a corresponding equal increase over the next decade. The inequality change for the entire post-1980 high growth period: zilch, actually 1.8 per cent. In contrast, the real income Gini in China is estimated to have increased by at least 40 per cent between 1983 and 2003 (nominal Gini increase from 28 to 45).

So what is going on? How come in India there has not been any change in inequality, but in China there has been a large change? One simple possibility is that inequality in both countries has behaved "as expected". China, a formerly communist economy, went through a large structural change. Inequality change there has closely paralleled the inequality change in other formerly communist countries like Russia, Vietnam, Cambodia, Ukraine, etc. The average inequality change (between pre-and post transformation years) in the formerly Soviet Union and Eastern Europe countries is 43 per cent, almost exactly the same as that observed for China (39 per cent). Fast-growing countries like Japan, Korea, Taiwan, Indonesia, etc. all show steady to declining inequality. India shows a similar pattern.

There is supposed to be the Kuznets curve effect in developing countries. According to this hypothesis, because of faster productivity growth in urban industrial areas, inequality first worsens and only after this rural-urban differential growth is over does inequality stabilise. This hypothesis ignores a powerful, and opposing force observed during the development process""the spread of education. In poor countries, labour is a major asset for at least 90 per cent of the population, if not more. And educational attainment is a major component of this "labour asset". The rich always had education, and good-quality education. Ditto for the children of the rich. Their educational attainment shows no change over time. In contrast, the bottom 90 per cent increases educational attainment by leaps and bounds. And this can, and does, cause inequality levels to stabilise, if not improve.

No matter what definition of inequality is used, or what survey, the fact remains that inequality in India has not changed over the last 25 years. Nor is this pattern unique to India. It is observed for several other fast-growing economies, i.e. fast economic growth is neither necessary, nor sufficient, for inequality to worsen. An equally strong pattern is observed for formerly socialist economies: inequality there has worsened considerably. The two patterns are different and should not be confused to be the same. Just because India and China are both fast-growing, large developing economies, it does not at all follow that inequality trends are even broadly comparable.

The views expressed are my own and should not be attributed to the National Statistical Commission, of which I am a member. Details about inequality in India and China are contained in Second Among Equals: The Middle Class Kingdoms of India and China, forthcoming 2007, Peterson Institute of International Economics, Washington, DC

ssbhalla@gmail.com