In Defence of Liberty

Driven by data; ridden with liberty.

Statistics and Lampposts XXV: The Perceived Chasm

the-perceived-chasm-natalie-tapsom.jpg

Is the gap between rich and poor really getting wider? (Edited: Natalie Tapson)

On May Day’s broadcast of The Andrew Marr Show, Nigel Farage, the leader of the United Kingdom Independence Party (UKIP), was talking to its host about the upcoming local elections and the European Union referendum.

In particular, the UKIP leader claimed to Andrew Marr:

The gap between the rich and poor is getting bigger and bigger. We have been through a decade where the people earning average salaries: they are 10% worse off than in 2007. That isn’t right.

This claim about the chasm between “the rich and poor” is commonly asserted, but often without evidence [1].

Household income inequality

The Office for National Statistics produces a report on household income inequality, and the effect of taxes and benefits [2].

There are multiple ways in which the inequality of household income can be presented. The Gini coefficient receives wide usage, and is either expressed as a value between 0 and 100, or as a percentage between 0% and 100%. The zero value represents complete equality, where all households have the same income, and the maximal value is the case where one household has all income, and all other households have nothing.

The Gini coefficient is calculated as half of the relative mean absolute difference. This calculation can be applied to original household income (income before any taxes and benefits), gross income (following the addition of cash benefits), disposable income (after the addition of cash benefits, and the subtraction of direct taxes), and post-tax income (after the addition of cash benefits, and the subtraction of both direct and indrect taxes).

the-perceived-chasm-ons-2

Gini coefficients for different income measures, 1977 to 2014-15. (Source: Office for National Statistics)

Following the starting point in 1977, the Gini coefficients for these four measures rose until 1990. Household income inequality dipped between 1990 and 2001-02, before falling irregularly — with a spike around 2006-07. Equivalised post-tax income is not yet available for the year 2014-15.

Other indicators

There are limitations with every indicator. The Gini coefficient cannot distinguish between different shapes in the income distribution, and so other measures should be considered, in order to obtain a fuller view of British income inequality.

The S80/S20 ratio is the ratio of the total income received by the top 20% of households with the highest income, to that of the 20% of households with the lowest income.

A point comparison is the P90/P10 ratio, where the ratio of the income of the household at the bottom of the top 10% to that of of the household at the top of the bottom 10%. In other words, this measure is based on two households: the income of the one household represented by the 90th percentile and the income of the one household represented by the 10th percentile.

The Palma ratio takes the ratio of the income share of the richest 10% of households to that of the poorest 40% of households. The intermediate 50% of households are likely to have a stable share of income, so they can excluded from the calculation without a substantial loss of information.

the-perceived-chasm-ons-1.png

Change in Gini coefficient, S80/S20 ratio, P90/P10 ratio and Palma ratio for equivalised household income, 1977 to 2014/15. (Source: Office for National Statistics)

 

These measures all show a similar pattern: there were twin peaks in 1990 and 2001-02, and household income inequality has dropped irregularly since — spiking in 2006-07.

Since that year, income inequality has fallen in most years, on most measures.

The report also finds that the gross income shares for the top 10% and 20% have remained remarkably stable since 1990.

the-perceived-chasm-ons-3

Share of gross income, 1990 to 2014/15. (Source: Office for National Statistics)

Furthermore, the report also concludes that, as a whole, indirect taxes are regressive on household income, because they widen the Gini coefficient, unlike direct taxes and cash benefits.

the-perceived-chasm-ons-4.png

Change in Gini coefficients because of cash benefits and taxes, 1977 to 2014/15. (Source: Office for National Statistics)

“The people earning average salaries”

Considering Mr Farage’s other, less vague, claim — “the people earning average salaries: they are 10% worse off than in 2007” — this does not appear to be accurate. All three standard measures are still lower than their peaks around 2007, but the figure is not close to 10% [3].

the-perceived-chasm-ons-5.png

Growth of median (and mean) household income and gross domestic product (GDP) per person, 1977 to 2014/15. (Source: Office for National Statistics)

Mean equivalised disposable household income was 2.3% lower in 2014-15 than its zenith in 2006-07. In 2014-15, median equivalised disposable household income was just 0.1% beneath the figure in 2007-08. GDP per capita in 2015 Q1 was 0.3% under its peak in 2008 Q1.

The gap between rich and poor does not appear to be widening, and the drop in average incomes is not as large as Mr Farage claims.

References

[1] Masters, A., 2016. The Rich Get Richer. In Defence of Liberty. Available from: https://anthonymasters.wordpress.com/2016/02/05/the-rich-get-richer/ [Accessed: 2nd May 2016]

[2] ONS, 2016. The effect of taxes and benefits on income inequality: 1977 to financial year ending 2015. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfinances/incomeandwealth/bulletins/theeffectsoftaxesandbenefitsonincomeinequality/1977tofinancialyearending2015 [Accessed: 2nd May 2016]

[3] ONS, 2016. Household disposable income and inequality: financial year ending 2015. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfinances/incomeandwealth/bulletins/householddisposableincomeandinequality/financialyearending2015 [Accessed: 2nd May 2016]

Advertisements

Information

This entry was posted on May 3, 2016 by in Statistics and tagged , , , .
%d bloggers like this: