Headline Numbers: How do you measure inequality?
Last week I wrote about a measure of inequality that had received much coverage in the US.
Research from the Institute for Policy Studies found that in 2014, bonuses paid to Wall Street employees had been double the total annual pay earned by all Americans who worked full-time at the federal minimum wage.
I crunched the numbers and it turned out that the same was true for the UK.
But is this actually a good measure of inequality? If a load of people earning minimum wage suddenly received a pay rise then bonuses would become an even greater multiple of minimum wage salaries.
One of the most commonly used measures of inequality is the Gini Coefficient, which gives countries a score between zero and one. A score of zero would mean that everybody in the country earned the same amount while one would indicate that all of the country's income was earned by one person. It can also be used to measure wealth inequality.
The Gini Coefficient is more than 100 years old now, and attention in inequality has been turning recently to measures that concentrate more on comparing extremes in the population - for example, looking at what proportion of wealth is held by the richest 1% of the population and what proportion is held by the bottom 50%.
Earlier in the year, Oxfam predicted that the combined wealth of the richest 1% would overtake that of the other 99% of people next year. There were problems with the way the charity extrapolated that conclusion from previous years' figures, but the conclusion was not implausible.
The figures were based on some research by Credit Suisse, which estimated the distribution of wealth across global populations.
A third way to think about inequality is in terms of poverty measured by relative incomes.
The Office for National Statistics (ONS) had figures out on Wednesday based on the definition that people were in poverty if their income was below 60% of the median level (to find the median income, line up all the people in the country in order of income and take the middle one).
It found that almost a third of the UK population had experienced poverty in at least one of the years between 2010 and 2013, which is very high by European standards.
The OECD sets out a summary of what has happened to examples of all three of these measures. Across its 34 member countries, the Gini Coefficient rose gradually from 1996, fell slightly for the financial crisis and then resumed its upward path.
Relative-income poverty grew steadily over the period and a comparison of the top 10% and bottom 10% showed growing inequality that paused briefly for the financial crisis before accelerating.
So while there are a number of different ways of calculating inequality, each of which have their own strengths and weaknesses, they seem to agree that inequality has been growing in recent years.
But these measures tell us little about other inequalities such as health, education or opportunities.
A student at university would be expected to have a low income and indeed negative wealth because of student debt, but would not necessarily be in poverty.
In health, the ONS found a gap of 18 or 19 years in the life expectancy of people in the most and least deprived areas.
And the OECD talks about how wealth and income inequality cause overall economic problems because they affect access to education for the next generation.