Pareto models, top incomes, and recent trends in UK income inequality


By: Jenkins, Stephen P.

Statistical agencies and other researchers typically estimate income inequality levels and trends from either household survey data or tax return data, but rarely combine the information in the two types of data source. The result is that very different impressions about how inequality is changing over time may arise. Estimates from tax return data show a substantial rise in inequality over the last two decades in both the UK and USA, whereas survey-based estimates of inequality show much less change. For the UK, for example, the share of total income held by the richest 1% increased by 29% between fiscal years 1996/97 and 2007/08 whereas the Gini coefficient for household income increased by 7%. For the USA, the corresponding increases over the same period are 30% and 2%. Research users may reasonably ask what the ‘true’ picture of inequality trends is. There is a good case for providing them with answers using methods that combine information from survey and tax data in order to take advantage of the strengths of each source, and this is what I do. Tax return data provide better coverage of top incomes than do survey data, and survey data provide the ability to create income variables with the same definitions, so that combination is done on a like-for-like basis. I analyse income inequality levels and trends for the UK by combining inequality estimates from survey and tax data. As part of this analysis, I also provided new findings about survey under-coverage of top incomes in UK survey data: the problem becomes apparent at around the 99th percentile in the 1990s but at around the 95th percentile in the 2000s. In addition, I provide new results about how to summarise the distribution of top incomes using Pareto models, arguing in favour of a Pareto II model rather than the Pareto I and for using modelling thresholds rather higher than often employed. My conclusions about aggregate UK inequality trends since the mid-1990s are broadly robust to the way in which I employ the information about top incomes in the tax data. For example, the Gini coefficient for gross individual income rose by around 7% to 8% between 1996/97 and 2007/08, with most of the increase occurring after 2003/04. When I use only survey data, with tax data not exploited at all, the Gini coefficient is estimated to decrease by around 5% over the same period.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: