Author:Dalamagas, Basil

    It has long been argued that (direct-indirect) tax policy must resolve the problem of finding a suitable trade-off between equity and efficiency. An equity-efficiency tradeoff is claimed to result when the maximization of the productive capacity of an economy (incentives to work and invest, growth prospects etc.) leads to an unfair distribution of the tax burden or when a policy of promoting a fair distribution of income leads to efficiency losses (low growth rates etc.). Academic discussion also revolves around whether equity and efficiency are always inversely related or whether they can both rise or improve at once. Studies from IMF, OECD and World Bank suggest that, if we disregard Pareto analysis, then economic performance and income equality can indeed rise in concert.

    The main objective of the present study is to obviate the need to resolve the amorphous equity-efficiency dilemma. Instead, we shall endeavor to assign this "privilege" to a well-known entity that fits perfectly into the real world's social-political democratic norms: The electorate. Each household-taxpayer-voter is assumed to reveal their preferences about their most desirable socio-economic priorities in every election period. This is done by voting the political party (or coalition of parties) on the basis of a long array of issues, the most important of which are assumed to relate to the allocation of resources (i.e. tax revenue) to the equity objective (income redistribution) and/or to the efficiency objective (growth). Any political party that strives to win the next election or to stay in power must insistently frame the issues regarding the design of the tax structure narrowly, in terms of the volition and consent of the majority of voters.

    To formalize a tax system that would potentially achieve the objectives of equity and/or efficiency, fiscal authorities have to define the proper mix of direct/indirect tax rates. Such a definition implies that policy makers must determine a range of low or high (in)direct tax rates to capture their effect on the objectives of equity/efficiency. This is so because the incidence of each of those two classes of taxes differs greatly between equity and efficiency: High marginal (and average) income tax rates are argued to discourage work effort, savings-investment and growth prospects (see, for example, Lambert & Yitzhaki, 1995; Cremer et al, 1990; Bordignon, 1993; Cuccia & Carnes, 2001; Saez, 2001; Musgrave, 1990,1993; Aronson et al, 1999; Infanti, 2008; Berliant & Strauss, 1993; Elkins, 2006; Fuest et al, 2008; King & Sheffrin, 2002; Gouveia & Strauss, 1994; Kaplow, 1995; Hyun & Lim, 2005; Avi-Yonaht, 2002; Sorensen, 1994; Auerbach & Hassett, 2002; Nicita, 2004; Listokin, 2012), whereas high indirect tax rates are claimed to put barriers up to prevent a fair income distribution (see, for example, Virmani, 1989; Gooroochurn & Milner, 2005; Anderson et al, 2001; Saez, 2004; Emran & Stiglitz, 2005; Truyts, 2012; Naito, 1999; Mayeres & Proost, 2001; Keen & Lockwood, 2010; Auerbach & Hines, 2002; Blackorby & Brett, 2004; Genser et al, 1996; Bucovetsky & Smart, 2006; Decoster et al, 2010; Dixon and Rimmer, 1999; Kaplanoglou &Newbery, 2003, 2004; Karakosta et al, 2014; Makdissi& Mussard, 2008; Yitzhaki & Mayshar, 1995; Peitz & Reisinger, 2014; Boadway & Pestieau, 2002; Smart, 2002; Wang & Zhao, 2009).

    The effects of (in)direct taxes on incentives and income distribution have been studied in the above literature on optimum taxation but no agreement seems to have been reached by scholars on a number of key issues. As a result, the equity/efficiency problems raised are not of the kind which can be resolved unambiguously. The contribution that formal theory can make is primarily to create a framework, within which the policy issues can be discussed in a logically consistent manner. Consider, for example, the following cases:

    (1) The progressive personal income tax is considered to be the most drastic tax instrument to both guarantee a fair distribution of income on the one hand and harm productivity incentives on the other. However, flat income taxes are perceived by a number of researchers (see, for instance, Jacobs et al, 2010) to be efficient in creating better incentives than progressive tax structures. The introduction of the flat tax by some Central and Eastern European countries has come along with a rise in tax revenue, a boost in economic growth and an improvement in labor market performance.

    (2) The early literature focused on the efficiency role of commodity taxes in the context of the Ramsey rule dealing with a uniform tax on commodities: Goods that are more complementary with leisure should bear higher commodity tax rates. The Ramsey rule is usually applied to excise taxation (e.g. specific taxes on premises used in spending free-time as a pleasure fair consisting of amusements).

    The present study broadens the definition of indirect-tax inefficiency by employing an average consumption tax rate to provide an intuitive insight into the process of identifying the inefficiency properties of this tax. A plausible assumption is that the majority of population consumes a portion of their disposable income and that consumption in absolute terms rises with income (agents' monotonicity) but falls as a proportion of income when income rises (diminishing marginal propensity to consume). Thus, an indirect tax is regressive in terms of a household's aggregate earnings. Apparently, two sources of distortions are generated by the introduction of indirect taxation in the discussion on the equity-efficiency dichotomy:

    * Given that the indirect tax to income ratio falls as income increases, the content of the equity principle is reversed: Indirect taxes encourage income redistribution in favor of the higher income levels.

    * The alleviation of the relative indirect tax burden on upper income groups, which are characterized by a high propensity to save, encourages savings and investment projects which lead to higher rates of economic growth.

    Due to the fact that the equity/efficiency characteristics of taxation play the key role in describing the profile of fiscal policy, the novel approach adopted in the present study to evaluate the optimal values of direct/indirect tax rates may be interpreted as constituting the second major objective that is set for developing our modelling structure in Section 4. Average direct/indirect tax rates have been chosen in order to simplify the analysis, without inflicting any damage to the validity of our results.

    The present paper demonstrates that an optimal level of tax evasion can be derived from maximizing a utility function with respect to (in)direct tax rates. The manipulation of the first-order conditions by using a novel mathematical approach leads to an infinite number of optimal direct/indirect tax rates. The selection of the optimal mix of these tax rates depends on households' preferences over equity/efficiency, as they are formulated by voters' volition in election periods. A simulation procedure helps understanding how the optimal tax-evasion level is chosen and how the optimal allocation of tax revenue to the fiscal objectives of equity/efficiency can be achieved, in the context of a panel data set including a large number of countries.

    The rest of the paper is organized as follows. Section 2 describes the institutional framework underlying the model. A simple theoretical description of how the main objectives of the study can be achieved is contained in Section 3, whereas a detailed econometric and mathematical presentation of the modelling structure is given in Section 4 and Appendix 3. Section 5 concludes the discussion laying out directions for future work.


    As it has been discussed above, it is the total direct tax burden as a proportion of income (average direct tax burden) and total indirect tax burden as a proportion of total consumption expenditure (average indirect tax burden) that are taken into account by the household-voter (together with other issues) and tilt the balance in favor or against each particular political party in the voting process. In our estimates of the data, we aggregate corporate profits taxes (which are computed as a proportion of profits), social security contributions (computed as a proportion of wage income), property taxes (computed as a proportion of property values) and progressive personal income taxes. The sum of all these tax revenues expressed as a percentage of GDP is taken to represent the average direct tax burden.

    On the other hand, it could be argued that excise taxes on luxuries are indirect taxes that seem to be targeted at some groups of taxpayers to further equity goals, while different degrees of progressivity in income taxes have equity implications, too. In contrast, the proportional social security taxes in some countries are levied on income up to a predetermined level, so that they tend to become regressive.

    There are two primary reasons for employing average tax rates in our study, instead of marginal tax rates:

    i) Marginal (in)direct tax rates are invariably used to measure the extent to which a macro(micro)economic variable, such as GDP or consumption, changes following a one-percentage unit increase (decrease) in a given tax rate. In contrast, the average tax rate represents the total tax burden as a proportion of total income or consumption. It is natural to expect that a tax evasion decision that is to be made by any taxpayer is dependent on the total amount of tax that should be paid to the government relative to his income than a small fraction of the tax corresponding to the marginal tax rate.

    ii) For the purpose of our research, it is of no interest to the taxpayers that some specific forms of indirect taxes (such as excises on luxuries) meet equity criteria. It is the decreasing ratio of consumption expenditure (and hence of total...

To continue reading