Abstract

Over the past 15 years, producing distributional measures of household income and wealth that are consistent with national accounts aggregates has been a major area of innovation in economic statistics. The first G20 Data Gaps Initiative, launched in 2009, highlighted household distributional information as a priority, and this emphasis has continued in subsequent initiatives. The forthcoming update to the System of National Accounts is likewise expected to provide guidance on statistics relating to the distribution of household income, consumption and wealth.

The European Central Bank (ECB) has developed Distributional Wealth Accounts (DWA). Since 2024 the ECB, all euro‑area countries and Hungary have published these accounts on an experimental basis. Compilation has followed a largely centralised approach: for all countries the DWA draw on the Household Finance and Consumption Survey (HFCS), financial accounts balance sheets and non‑financial balance sheets. Because these data sources are harmonised, a common methodology has been applied. While the accounts were initially compiled centrally by the ECB, national compilers are increasingly taking responsibility for producing their own DWA.

The principal advantage of this approach is methodological harmonisation across countries, which improves comparability. Its drawback is that country‑specific data sources are not always fully exploited. Countries outside the European Union that are developing distributional national accounts typically make fuller use of national data. Moreover, the current DWA omit certain monetary items that form part of the SNA wealth concept, such as currency; items such as pension entitlements and non‑life entitlements are currently reported only as memo items.

This paper discusses how remaining reporting gaps might be addressed and examines further challenges. At present, the compilation of wealth accounts is largely separate from income and consumption accounts and is not always consistent with them. There is growing demand for consistent, preferably three‑dimensional, statistics that jointly cover income, consumption and wealth. Finally, the paper considers considerable communication challenges: multiple data sources publish distributional wealth information — HFCS, national micro wealth statistics, the DWA, the World Inequality Database’s DINA and the UBS Global Wealth Databook — yet they yield different results and sometimes measure different concepts. Explaining these differences, their origins and their implications for users is a central task for producers of distributional statistics.