Rationale and definition:
Concerns about inequality focus on the top and bottom ends of the income distribution. Indicator 68, on “relative poverty,” tracks the bottom end of the income distribution, while this indicator monitors changes at the top end of the distribution. We see two options for such an indicator. First, countries may track the share of incomes generated by the richest 10% of the population. An alternative indicator is the increasingly popular Palma ratio, defined as the ratio of richest 10% of the population’s share of gross national income (GNI) divided by the poorest 40% of the population’s share.
The Palma ratio seeks to overcome some of the limitations of the widely used Gini Coefficient, which fails to take into account changing demographic structure (e.g. the effects of a baby boom or an aging population) and is insensitive to changes in the tails (top and bottom) of the income distribution, which is where most movement occurs.1 Furthermore, using a simple ratio, as opposed to the more complex Gini Coefficient measurement, is more intuitive for policy makers and citizens. For example, for a given, high Palma value it is clear what needs to change: to narrow the gap you raise the share of income of the poorest 40% and/or you reduce the share of the top 10%.
The income share of the top decile and the Palma ratio are formulated using household survey data relating to income and consumption (usually from World Bank PovCal / World Development Indicators). Such data can be disaggregated by income deciles in countries, allowing for comparative analyses between countries and regions. Further disaggregation by centiles, regions or groups would require complex analysis of the original household survey data, which at present may not be feasible on a national/ global scale.
Comments and limitations:
An important limitation of the income share of the top decile and the Palma ratio (as well as the Gini Coefficient) is that the indicators cannot be decomposed (i.e. overall inequality is related consistently to inequality among sub-groups). Furthermore, data is based on household surveys, some of which measure income and some consumption. The mix makes international comparison quite challenging, as the distribution of consumption tends to be less unequal than that of income. But since no means of adjustment (income vs. consumption) is readily acceptable, it is common practice not to adjust the surveys. To improve the quality of this data we recommend expanding the collection of pure income-based data, for example via the Luxembourg Income Study, which currently has micro-data for 40 countries.2
Preliminary assessment of current data availability by Friends of the Chair:
Primary data source:
Potential lead agency or agencies:
UNSD, World Bank, OECD (with Luxembourg Income Study).
Palma, G (2011). Homogeneous middles vs. heterogeneous tails, and the end of the ‘Inverted-U’: The share of the rich is what it’s all about. Cambridge Working Papers in Economics.
See a list of LIS available datasets.