Working Paper: CEPR ID: DP17135
Authors: Facundo Alvaredo; Mauricio De Rosa; Ignacio Flores Beale; Marc Morgan
Abstract: Large gaps exist between income estimates from inequality studies and macroeconomic statistics, questioning our representation of flows and the relevance of economic growth. We take stock of these gaps by confronting multiple datasets in Latin America, finding that surveys account for around half of macroeconomic income over the past twenty years. Less than half of this gap is due to conceptual differences, the remainder coming from growing measurement issues, which mainly concern capital incomes. Top tails in administrative data and surveys present diverging averages, especially for non-wage incomes, and different shapes. We discuss implications for both inequality levels and trends.
Keywords: income distribution; data gaps; surveys; national accounts; administrative data; latin america
JEL Codes: D3; E01; N36
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| measurement gaps (C52) | accuracy of income distribution representation (D31) |
| discrepancies (D80) | assessment of economic growth distribution (F62) |
| measurement issues (C52) | gap between surveys and national accounts (E01) |
| undercoverage of capital income (H24) | growing measurement gaps (I24) |
| divergence in averages (C29) | higher and more heterogeneous inequality (I24) |
| discrepancies between income statistics (D31) | implications for inequality analysis (I14) |