Inequality, Poverty and Growth: A First Look !

Background:

Inequality, be it of income, access to credit, health or education has become one of the most important issues in political and policy circles. In simple terms, inequality means the gap between the means of wealthy and poor. A simple way to measure inequality is to take the difference between the share of the national income earned by richest 20% of the population and the share earned by the poorest 20% of the population. For example, consider India’s case. In the last decade on an average, richest 20% of the population earned 44% of the national income as against mere 8% by poorest 20%. The difference of 36% gives a pretty good measure of the income inequality. Using this measure, the inequality in India and the USA was around 31% and 37% in 1980’s which grew to 36% and 41% respectively by 2010.  The fact that income inequality has risen in many of the advanced and developing nations over last two decades is considered by many as a major socio-economic problem.  Of course, many economists do not see inequality as a problem. Barro (2000) for example argues that inequality, especially in the developing world where the average income is low implies that there are at least certain individuals who have a bear minimum capital to innovate or be the entrepreneurs and kick-start the development process. One can find plenty of arguments on both the sides. But I feel that most of these arguments miss the fundamental point that inequality is not the same as poverty (in data as well as in principle) and that we need to start seriously thinking about the two separately and at the same time find any links between these two concepts. The purpose of this note is twofold: first, I argue that poverty and not the inequality should be our concern. Second I try to unearth some links between inequality and poverty.

Some Philosophy:

From a social perspective, do we care about the Inequality or the economic state of the poorest section? Inequality is a relative concept where we compare the rich against the poor.  But inequality tells us little about the financial condition of the poor. If the most miserable person in the country has a beautiful house, car, and a good job, do we care about how much more the richest guy in the country has? I doubt it. From a welfare perspective, we are more concerned about ensuring a reasonable lifestyle to the poorest section and not about the gap between the rich and the poor. Until the gap is an outcome of the market forces, what we care is just the absolute condition of the poorest section. If certain people as a combination of talent, efforts, and opportunities made a good fortune for themselves and at the same time also created opportunities for the underprivileged class, we should not be much concerned about the gap or the inequality.  So the first point I would like to stress is that in principle, we are more concerned about poverty reduction as against inequality reduction.

Some Empirics:

Given this view, let’s try to look at the data on inequality and poverty.

Figure 1:

Figure 1 shows inequality-poverty combination for each of the country averaged over the period 2008-2015.  Red lines indicate the global means of respective variables which divide the plot into four regions. The first striking observation is that no region is significantly relatively more populated or empty. There is at the best a weak positive relationship between the inequality and poverty. A substantial number of countries has above-average inequality but below-average poverty level, for example, Brazil.  The mean level of poverty for countries with above-average inequality is 47%, and that for the countries with below-average inequality is 33%. It means that variation in inequality explains only 14% of the poverty level difference, while the poverty ranges between 10% and 55% even leaving the bottom and top 10% of the countries. This analysis makes it clear that poverty and inequality are two different dimensions for a good deal of the countries.  

So now we know that an unequal society does not necessarily need to be a poor country. So why should we care about inequality? We must if higher inequality hampers the future economic development or the speed of poverty eradication. I will discuss the theory later, but first again let’s take a look at the data.

Figure 2:

In figure 2, I plot the initial inequality in 1980’s for each country on X-axis and the percentage of poverty change from 1980 to 2015 on Y-axis. (Note that the negative change means poverty reduction). Looking at the graph, one can see that most countries have reduced the poverty level over last three decades. But the magnitude of the reduction is not linked with the initial inequality as can be seen from the almost flat line of best fit and also from the simple observation that enough countries lie above and below the average poverty reduction line (red horizontal line) for any level of inequality.

Figure 3:

I do a similar analysis in figure 3, with the percentage change in GDP per capita on Y-axis. Because countries which started as poor in 1980’s had more scope to improve the income level, I plot the graph separately for countries which were rich in 1980’s and the countries which were poor in 1980’s. But scatter is flat for both the type of the countries, suggesting that initial inequality countries have not done poorly in terms of per capita GDP growth or in terms of eradication of poverty.

Figure 4:

But the level of growth is not all that one cares. The volatility of growth matters too as it feeds into the decision making of the consumers and producers. So I look at the link between the starting inequality level and the volatility of growth over next three decades for each country. Figure 4 plots the relationship. What we see is a flat average link between the two. If at all, the extreme long-term growth volatility countries all had low inequality to start with. Again, inequality does not seem to be associated with high volatility.

Figure 5:

Before turning to theory and economics behind the results, let’s look at the last graph. The idea is to understand if there is a notion of Inequality Trap. I plot initial inequality on X-axis and change in inequality over 1980-2015 on Y-axis in figure 5.

Figure 5:

The graph clearly indicates a negative link. That is those countries which started with high initial inequality have reduced the inequality on an average. It shows that there is no apparent evidence of inequality trap. The reduction in inequality could be due to aimed policies by the governments or the market forces. The graph tells us nothing about that, but it shows one thing that countries with high inequality to begin with have been able to reduce it more than the countries with lower levels of inequality.

    

None of these graphs presented above have got a causal interpretation in the sense that X-axis variable is not causing Y-axis variable necessarily. The graph should be seen only from correlation perspective. For example, figure 5 cannot be interpreted to mean that high inequality to begin with increases the speed of inequality reduction. A more rigorous regression analysis can shed light onto robustness of the correlation as well as the causation.

So looking at the evidence, we know four things: First poverty and inequality are different dimensions. Second, higher inequality cannot be associated with lower long-term growth rate or lower rate of poverty reduction. Third, that inequality does not increase the growth volatility and fourth that high inequality can be reduced. Recent IMF paper documents that high inequality reduces the future growth. But the horizon they consider is a five-year window as against thirty years in my analysis which leads to different results.  In summary, it seems difficult to understand the inherent negativism towards inequality. Inequality neither implies poverty nor the hindrance to the poverty reduction process, at least in the data.