Does how you measure inequality change how income inequality looks?

One common question that people ask about income inequality is ‘does how you measure it change how it looks’? The answer, like for many questions on statistics, is yes, and also no. The level of inequality depends on who you are measuring inequality between.

As we discuss in the About Inequality section of our website, there are several different ways of measuring inequality; the most popular are ratio measures or the Gini coefficient. In the graph below I’ve taken a few of the most popular measures of inequality and shown how they’ve changed over time[1]. It looks at the Before Housing Costs (BHC) Gini Coefficient, the ratio between the income of those at the top 10% and the bottom 10% (90/10), the ratio between the income of those in the bottom 10% and the average (50/10), the ratio between the top 10% and the average (90/50) and the After Housing Costs (AHC) Gini coefficient.  This graph doesn’t show how big each type of inequality is, but it does show how it has changed in the UK since 1961.

As you can see from the graph there are some similarities between all the measures of inequality. Inequality of every variety increased in the 1980s, the gap between the top and the bottom, the middle and the bottom and the top and the middle got wider. Other periods in the last 50 years have featured slower change than in the 1980s. After the 1980s there’s a divergence in what the different measures of inequality show. The gap between the bottom and the average income has substantially decreased over the last twenty years and is now only slightly higher than it was in 1972. The gap between the top and the bottom has also decreased but is still substantially higher than it was in 1972. However, there has been no such decrease in the gap between the top and the average.

The reduction of inequality has been mostly at the low end with large progress being made on fighting poverty in the 1990s and 2000s. The Gini, which seeks to measure inequality across the spectrum, is far higher now than it was in 1972 and has increased at a faster rate if measured after housing costs.

This tells us very little about how much inequality has increased at the very top. As data’s hard to come by for those at the very top we’re confined to using pre-tax inequality figures[2]. The following chart looks at how top income shares have changed since 1962.

It shows that whilst the income share of the top 10% has increased (most of which happened in the 1980s) the biggest increases are in the top 1% income share and the top 0.1%, both of whom have seen massive gains over the 80s, 90s and 00s; some of those gains were clearly reversed in the financial crisis but not all of them. The top 0.1%’s share of pre-tax income is now almost three times as high as it was in 1972. From these graphs you can see that there are several overlapping stories to how inequality has changed in this country but the broad picture is that it’s higher than it was 30 years ago and that trend doesn’t seem to be reversing except in the fight against poverty.

[1] Each measure has been adjusted so that it’s result in 1972 is =1 and all other results are relative to that. The Data source for this is the IFS’s Poverty, Inequality and Living Standards work using the FRS dataset.

[2] Same method as above but this time using the world top incomes database.