26 Pages Posted: 15 Feb 2013
Date Written: January 23, 2013
A substantial amount of evidence – anecdotal and empirical – suggests that modern tax systems, especially those in developed countries, are highly complex. For example, in Australia a Federal Court judge remarked that tax “legislation in general is simply far too complex” (Perram 2010: 184). In the same year a United States report noted that “taxpayers and businesses spend 7.6 billion hours and incur significant out of pocket expenses each year complying with federal income tax filing requirements” as a direct result of tax complexity (President’s Economic Recovery Advisory Board (PERAB) 2010: 3). And in July 2010 the United Kingdom Chancellor of the Exchequer George Osborne established the Office of Tax Simplification (OTS) charged with the responsibility to “advise the Chancellor on delivering a simpler tax system, providing independent advice on options for addressing existing complexity in the tax system” (HM Treasury 2010: 1).
These three examples from three different countries illustrate the widespread concern – which is not a new phenomenon – about tax system complexity. It is therefore not surprising that the quest for means to reduce tax system complexity – or to move towards its antonym of greater simplicity – has been a fixation of governments and others for many years. Surrey and Brannon (1968: 915) have noted that “simplification is the most widely quoted but least widely observed of the goals of tax policy”. It has been used (and abused) as a primary justification for tax reform over the last century, and typically it is seen as “a good thing” – “to say that one is in favour of tax simplification is tantamount to stating that one is in favour of good as opposed to evil” (Cohen et al. 1975: 7).
Conceptually an important step towards tax simplification is measuring and monitoring the level of tax complexity. As a contribution to the literature on tax simplification, the principal aim of this paper is to propose the construction of an index of tax system complexity. Such an index can be interpreted as a summary indicator of the overall complexity of a tax system at a particular point in time so that a series of such an index can be used to monitor the changing level of tax system complexity over time. Since the required data for empirical application are not yet readily available, the focus of the paper is on the methodology of the proposed construction. Although the discussion primarily draws from Australian examples, the proposed methodology is of sufficient generality to be applied to any other country.
The present paper has its genesis in a (successful) Australian Research Council (ARC) Linkage grant application in 2010 by the authors and colleagues from other Australian universities. The paper is, in addition, motivated by two specific considerations. First, as elaborated in the next section, tax complexity truly is a multidimensional concept that means different things to different people. As a result, tax complexity can be measured in a variety of ways and those alternative indicators may not necessarily be consistent. This gives rise to the need for the construction of a summary index that encompasses most (if not all) relevant indicators of tax complexity.
The second motivation for the paper is that there exists no universally accepted single overall measure of tax system complexity. This is in sharp contrast with efficiency and equity where theoretical summary measures are well established and applied empirically by tax researchers. For example, Diamond and McFadden (1974) developed an index of the efficiency costs of different types of taxation; and Musgrave and Thin (1948) proposed a tax progressivity index (based on the well known Gini index of income inequality) to measure certain equity aspect of the tax system.
The lack of any single measure of tax system complexity may in part reflect the relative neglect of the concepts of tax complexity and tax simplicity in the public finance and related literature. Ironically, tax simplicity was explicitly recognised at the birth of modern economic science almost two and half centuries ago (Smith 1776: Book Five, Chapter II). Since then, and until relatively recent years, it has been somewhat neglected. It was largely ignored by early 19th century classical economists such as John Stuart Mill (see Sandford et al. 1989: 25–26) and the first empirical study of tax compliance costs only appeared just over 75 years ago (Haig 1935). More recently, as pointed out by Slemrod and Yitzhaki (1996: 172), by the beginning of the 1960s, administration, compliance and enforcement issues had almost disappeared from the literature, excluding that on developing countries. This disregard has, to a lesser extent, persisted until today, especially in the discipline of public finance, and is evident by a quick examination of leading textbooks and academic journals in this field.
The reasons for this neglect are manifold and they have been identified elsewhere in the literature (see, for example, Tran-Nam et al. 2000: 230).
A similar situation exists at the tax policy making level. In the various tax reforms in the English-speaking world, simplicity has typically ranked well below efficiency/neutrality and equity as one of the goals of tax reform. This view has influenced the wave of substantial tax reforms around the world in the past 35 years. While these reforms have exhibited many efficiency-enhancing features (such as lower income tax rates, broader tax bases and more uniform and neutral tax structures), more often than not only lip service has been paid to the need for tax simplification.
Notwithstanding the relative neglect of issues relating to complexity and simplification in the public finance literature and in the tax policy making process, there has nonetheless been a steady growth of international interest in tax operating costs, particularly tax compliance costs, both by academic researchers and by governments (for a summary refer to Evans 2008), due to a variety of factors as discussed by Sandford (1995: 5-7).
However there is still a paucity of empirical evidence, at both the national and the international comparative level, of tax complexity, although the literature relating to inter-temporal and international comparisons of tax system complexity is not entirely empty. For example, in collaboration with the World Bank and International Finance Corporation (IFC), PwC has been, for the past eight years, publishing an overall “Paying Taxes” ranking of most countries on an annual basis (see, for example, PwC 2013). It will be suggested in Section 4 that the methodology and application of the PwC “Paying Taxes” indicators and ranking are not sufficiently rigorous. In addition, the OTS (2012) has recently been active in this area and released a working paper on tax complexity. The OTS’s proposed methodology is much closer in spirit to that expounded in this paper although there are also some major differences, highlighted later in this paper.
The organisation of the remainder of the paper is as follows. Section 2 provides a succinct overview of tax complexity. This section discusses the meaning, causes, impact and measurement of tax complexity, drawing heavily on previous work of the authors (see, for example, Evans and Tran-Nam 2010). Section 3 presents a brief overview of index numbers. This include the meaning, use and construction of index numbers. Section 4 then discusses in detail the proposed methodology for constructing a tax system complexity index. This section sets out criteria for constructing such an index, and focuses on measurement issues arising from the empirical application of the proposed approach. Summary remarks are then given in the concluding section.
Keywords: tax system complexity, tax simplification, basic number index theory
JEL Classification: H29, K34
Suggested Citation: Suggested Citation
Evans, Chris and Tran-Nam, Binh, Towards the Development of a Tax System Complexity Index (January 23, 2013). UNSW Australian School of Business Research Paper No. 2013 TABL 1001. Available at SSRN: https://ssrn.com/abstract=2216322