However, there is no tax policy that is deliberately against growth or that aims to increase
unemployment. Policymakers face trade-off and the way they balance various conflicting objectives
shapes the structure of tax revenue. An efficient and neutral tax system has interesting properties but in
many cases pursuing neutrality and efficiency conflicts with redistribution. This means that defining
quality as neutrality conflicts with the view that the tax system may be an appropriate tool to achieve
some redistribution. It may also conflict with the view that some tax incentives may be welcomed, if well
targeted.
We decide not to make any reference to an “absolute optimum” and we hold decisions about the trade-off
between efficiency and equity, or neutrality and incentive to be political decisions that are not subject to
evaluation per se. We consequently consider that it is more appropriate to assess separately efficiency
losses and non-neutralities, the effect of taxes on income distribution and the effects of tax incentives.
“Quality” requires that neutrality should be properly achieved if considered as a tax policy goal, that a
progressive tax system effectively redistributes income and that incentives lead to input and/or output
additionality.
Using a large set of diversified indicators enables us to investigate if and to what extent the various
policy goals have been achieved, without imposing a specific choice between them.
Tax policymakers often consider that complexity should be avoided and that quality require simplicity.
The complexity of the tax system is however very difficult to assess and we do not account for this.
2. Indicators
2.1. Efficiency and neutrality
Efficiency requires minimising the tax rate, preferring inelastic tax base to elastic ones, or following a
“broad base, low rate” approach. Neutrality requires that price distortions induced by taxes should be
kept to a minimum, what a “broad base, low rate” approach will also ensure.
The best indicators are the marginal cost of public funds and estimates of the deadweight cost. Such
indicators are however not available on a regular basis and their computation is subject to many
assumptions, some of them being disputable.
Efficiency also refers to effective taxation and its dispersion across the tax base. This means that tax
burden indicators may give some indication about the quality of tax revenue. Another option might be to
test the broadness of the tax base.
2.1.1. Tax burden indicators
The tax/GDP ratio is well known but is not the relevant one for the assessment of the quality of public
finance. It aggregates the whole set of taxes, while tax structure is a key element of the quality of tax
revenue. Moreover, differences in the tax treatment of transfer may affect comparisons across countries
and over time^3 ; different treatments of in-work benefits may affect the indicator and result in misleading
comparisons and more broadly the choice between tax expenditures and direct expenditures made any
assessment of the quality of public finance based on the tax/GDP ratio difficult^4.
(^3) The OECD work on net social expenditures illustrates how the taxation of transfer differs across countries. See ADEMA
e.a (2005). Sweden and Denmark, that top the tax/GDP raises taxes and social security contributions from transfer that
amounts to 3.6 and 4.1% of GDP, compared with 1.7% of GDP for Belgium.
(^4) Cf. OECD (2000)