A Poorly Targeted Subsidy? The Fiscal and Social Costs of EU VAT Expenditures

Value-added tax (VAT) systems in EU Member States deviate substantially from a uniform-rate benchmark through reduced rates and exemptions, typically motivated by distributional concerns. This paper provides microsimulation evidence on the fiscal and distributional effects of the full set of VAT expenditures, including both reduced rates and the non-actionable exemption gap covering healthcare, education and financial services mandated by the EU VAT Directive, in five major EU economies (Germany, France, Italy, Spain and Poland) under the 2024 policy baseline. Using EUROMOD and its Indirect Tax Tool with EU-SILC 2022 microdata matched with EU-HBS 2015 expenditure data, and following a compensating variation welfare approach, we simulate a universal standard-rate VAT counterfactual and find that abolishing all VAT expenditures would reduce post-consumption-tax disposable income by 9-21% for bottom-decile households versus 3-5% for the top decile, with Italy and Spain displaying the most regressive reform-cost profiles. The difference in welfare loss between the bottom and top income deciles ranges from -6.3 percentage points in Poland to -16.3 percentage points in Italy. While VAT expenditures offer meaningful purchasing-power protection to lower-income households in proportional terms, this protection is poorly targeted in absolute terms because richer households capture a larger share of the implicit subsidy in absolute cash terms, and could be replicated more efficiently through targeted income transfers. We discuss implications for revenue-neutral VAT reform and the efficiency-equity trade-off in EU indirect taxation.