However, economic growth and poverty have not been strongly related within the OECD in the past generation. There is little evidence of a relationship between poverty and household income growth in either a positive or negative direction. For example, Ireland has had very rapid income growth over the period and a large rise in poverty, while income growth has stagnated in Belgium in combination with a considerable reduction in poverty.
There are a few problems here. The first is the puzzling assertion that Belgium has “stagnated” in the last generation. The OECD’s own figures (XLS file) say that Belgium’s per capita GDP has grown by about 50% since 1980, adjusted for inflation and purchasing power parity ($19,000 approx in 1980 to $30,000 approx in 2010, in 2000 prices). Ireland’s has grown by about 200% ($11,000 to $31,000), but it started from a base about a third smaller than Belgium’s. Belgium’s growth rate was certainly a lot lower than Ireland’s, but they’ve converged on roughly the same point.
A bigger problem is the definition of poverty, which is relative and considered within single countries. It’s quite misleading to claim that Irish economic growth didn’t reduce poverty. The OECD uses a relative definition of poverty – the "percentage of persons living with less than 50% of median equivalised household income". Poor people in Ireland (and Belgium) are a lot less poor than they were thirty years ago with regard to the options they have available to them. The gap between them and the rich might be wider, but this matters less to most people than their life expectancies, economic security levels, and other absolute values. Saying that economic growth made poor people a lot richer but rich people even more rich is quite different to saying that “economic growth and poverty have not been strongly related within the OECD in the past generation”.
But let’s set aside debate over absolute vs relative definitions of poverty and take the relative definition for granted. Why is relative poverty only considered within one country and not globally? Comparing a poor Irish person’s wealth level to a poor Bulgarian person’s wealth level over the last thirty years would give quite a different picture. Maybe, say, Ireland’s economic growth attracted large numbers of immigrants who were still earning far more in Irish “poverty” than in “wealth” in their own country. The figures would show an increase in Ireland’s relative poverty level, even though by any measure everybody was better off than before.
The standard measure of inequality, the Gini coefficient, gives Tanzania and Malawi a more “equal” score than New Zealand and Japan. I know where I’d rather be, rich or poor. The measure of inequality itself is not worthless, but defining poverty as inequality within one country certainly is. By any measure of actual outcomes, economic growth is good for the poor. And if a measure of poverty doesn’t reflect that, it’s a bad measure.