Launch of World Wealth and Income Database

Readers might be interested in the new World Wealth and Income Database, which was just launched at the American Economic Association (AEA) annual meeting in Chicago.

It is coordinated by a small core team located at the World Inequality Lab at the Paris School of Economics.

The presentation slides from the AEA are available here and the corresponding explanatory paper is visible here.

The database aims to offer open access to the most extensive available database on the historical evolution of the global distribution of income and wealth, both within and between countries.

From an Irish perspective, what’s most notable is the paucity of data on the distribution of income and wealth, something that Patrick Honohon commented upon as governor of the central bank in 2014.

However, there does seem to be updated data (most likely from Brian Nolan), on the top 1% income share from 1938-2009.

Screen Shot 2017-01-18 at 14.25.03

By Aidan Regan

I'm an Assistant Professor at the School of Politics and International Relations, University College Dublin (UCD), and Director of the Dublin European Research Institute (DEI). My research is primarily focused on comparative and international political economy.

3 replies on “Launch of World Wealth and Income Database”

The data for Ireland are sourced from tax returns.

This is likely to bias the 1% share upwards for two reasons:

-Social transfers are about €20bn in Ireland, maybe a fifth of household gross income Most people whose sole income is social transfers (the unemployed, disabled and pensioners) are not obliged to file a tax return. Child benefit does not appear on income tax returns either. So you are essentially taking a big chunk of low-income people out of the analysis
-The category variable is tax units, not individuals. Married couples comprise one tax unit. Married people tend to have higher incomes than non-married people

This effect is likely dynamic as well as static. Social transfers comprise a higher share of household income than they did 20 or 40 years ago.

The SILC is of course a more comprehensive measure of household income.

It shows a very stable upper quintile share and Gini coefficient over time:

@examiner: You say that a lot of transfer incomes do not figure in Revenue data, and I presume that a lot of very low income households are also missing from the data (no individual household or employer returns). In that case the Revenue statistics are not much use for income distribution purposes: you need household income per equivalised adult or some such measure.

You say the category variable is tax units: do you mean by this “tax cases” which I think is the term used by Revenue. You say “Married couples comprise one tax unit.”. I don’t think this is necessarily the case. A married couple which submits separate rather than joint returns, I think counts as 2 units. In some cases there are advantages in making a joint return, but I wonder if this is universally true (especially for very high incomes)?.

I wonder how reliable are the older Irish data with a persistent fall in the share of the 1% from 1938? how reliable are other data?

There was massive tax avoidance or in the case of farmers, they were exempt from income tax.

Who would have logged private company stakes including De Valera’s control of the Irish Press that was established with seed money that had been contributed by Irish Americans for the nascent state?

Owners of serviced land in County Dublin, with no capital gains tax, benefited when the average price rocketed by 530% in 1963-1971 compared with a rise in consumer prices of 64% in the period.

When a CGT and wealth tax was announced in 1974, the 5 shareholders of the Irish Times bailed out and the resultant jump in company debt almost bankrupted the firm during the subsequent recession.

Besides, the wealthy held funds offshore.

Most self-employed were dodging taxes and in subsequent years the Revenue Commissioners sent teams to towns across to country to build up records of shops and pubs.

Comments are closed.