Assessing the medium-term effects of an increase in VAT on living standards and inequalities using microsimulation

Mathias ANDRÉ - Anne-Lise BIOTTEAU

Documents de travail
No G2019/01
Paru le :Paru le11/02/2019
Mathias ANDRÉ - Anne-Lise BIOTTEAU
Documents de travail No G2019/01- February 2019

An increase in value added tax (VAT) leads to a price shock the same year. Then, wages and other incomes (incomes from self-employed activities, pensions, unemployment benefits...) are partially adjusted to this price increase. Besides, social benefit and income tax scales are legally indexed on inflation. We assess both the direct effects of a VAT rise and some of the delayed effects, up to three years after, on living standards (adjusted for VAT and rent expenses) and inequalities. We use the microsimulation model Ines, allowing us to simulate increases in various VAT rates. Three years after a 3-points increase in the standard VAT rate (from 20% to 23%), the real mean adjusted living standards is lower by 0.6% than what it would have been without any VAT rise. This medium-term effect amounts to 45% of the direct effect of the VAT increase. The 10% poorest individuals face a specific situation: they experience a relative loss of their mean adjusted living standards (1.8%) almost 3 times larger than the rest of the population. Indeed, indexing social benefits and incomes does not offset the rise of VAT and rent expenses. However, in the medium term, inequalities in adjusted living standards and poverty increase slightly only. In the first year, every indicator measuring living standards inequalities and poverty increase. But delayed effects almost totally compensate this direct effect for the decile ratio, Gini coefficient and at-risk of poverty rate, which are rather stable in the mid-term. The P95/95 ratio rises more after three years, since the delayed effects on incomes and tax and benefit scales barely compensate the direct effect. Only the poverty intensity increases more in the medium term than in the short-term. These results are robust relative to the main assumptions about VAT pass-through to prices and price elasticities of wages. The decrease in the mean adjusted living standards ranges from 0.3% to 0.8% (resp. 70 euros to 155 euros).