Tax Complexity,Tax Salience and Tax Politics

What is Tax Salience? An Introduction to the Literature

The starting point for this article is Gamage and Shanske’s suggestion that the problematics of tax complexity should be considered within the context of the interdisciplinary tax salience literature, in part because this literature ‘measures how taxpayer [behaviour] departs from key assumptions of neoclassical economic theory’ (2011: 19, 33). In this sense, they challenge an existing orthodoxy. They advance the basic precept that the complexity of tax legislation is inextricably linked with taxpayers’ perceptions of it, and, thus, it is from the taxpayers’ perspective that calls to simplify or reform legislation must be considered. The issue is thus entirely one of subjective perception, of which there are two categories, namely, what tax legislation actually requires and what taxpayers think that the law requires (Gamage and Shanske, 2011: 19, 33). In traditional, political discourse, the gap between the two should be as small as possible (Gamage and Shanske, 2011).

Gamage and Shanske suggest, however, that there are some problems with the traditional approach to tax complexity (2011: 33). First, focusing on narrowing the gap, and on making tax legislation as clear and as simple as possible, necessarily places simplicity at the top of a list of priorities. This prioritization may occur without sufficient consideration, for example, of the role of taxpayer errors (Gamage and Shanske, 2011: 33). Perhaps legislation is perceived to be complex by taxpayers because they have misunderstood it or because they have applied it incorrectly to their circumstances. The challenge is that it is not necessarily clear if every error should be dismissed as the product of poor drafting. Second, the priority given to tax salience as a goal – or, in other words, the assumption that it is important and valuable that taxpayers should be able to understand the law – may be exaggerated when compared with other goals of tax legislative drafting (Gamage and Shanske, 2011). Gamage and Shanske do not suggest that salience does not belong on a list of priorities for tax legislation drafting but, rather, the argument is that salience does not necessarily belong at the top of that list.

There are identifiable consequences for both high and low salience systems. In a tax system with complete salience, taxpayers are aware of ‘actual’ taxes when they make economic decisions, or political choices (see, inter alia, Auerbach, et al., 2010; Congdon, et al., 2009; Finkelstein, 2009: 972). In a system with lower salience, taxpayers do not realize the full costs of taxes but they do possess a ‘perception’ of them (Auerbach). Further, one of the purported consequences of decreasing tax salience is growth in the size of government, a concern that can be found in the writings of Mill (Finkelstein, 2009: 970, citing Gamage and Shanske, 2011–2012: 19; Mill, 1848 [2004]). Indeed, Buchanan, in particular, argued that ‘tax misperception’ is ‘exploited’ by governments in order to justify expansion of the state (Finkelstein, 2009). Within these arguments, governments traditionally are perceived as beneficiaries of low tax salience or, simply, of tax complexity. Discernible within these assumptions is a lingering suspicion that, perhaps, governments deliberately conspire to render tax legislation as complex as possible, such that taxpayers will not be able to tell when they are overpaying.

As mentioned infra, there is not a great deal of empirical research considering whether taxpayers struggle to comply with, or to protect themselves adequately from, complex legislation. Perhaps the reason for this is that the negative consequences of tax complexity are perceived to be self-evident. Yet one interesting, recent study did attempt to put these theories to the test (and, as a consequence, has been widely cited) (Finkelstein, 2009, cited by, inter alia, Auerbach et al., 2010; Chetty et al., 2009). Finkelstein examined whether any link could be discerned between the introduction of electronic toll passes in the United States and the rate of the toll (Finkelstein, 2009). Electronic toll passes generally work by affixing a plastic box to the windscreen on a toll payer’s car, so that the driver may drive quickly through toll stops, without the need for personal interaction with a toll attendant. Given the seamless experience of payment enabled by the electronic toll pass, Finkelstein queried, are drivers less aware of the amount of the toll that they are paying? Additionally, is this lack of awareness exploited by governments, who then proceed to raise the rates? Finkelstein’s study concluded that there is a ‘robust’ link between levels of electronic toll pass usage, and higher rates of tolls, so this is a clear possibility (although she emphasizes that the ‘normative implications of these findings are ambiguous’) (Finkelstein, 2009: 1009).

The reason for this ambiguity, Finkelstein suggests, is that evidence has not been collected on ‘what is done’ with the extra revenue that electronic toll passes enable to be collected ( Finkelstein, 2009). She also emphasizes that a taxpayer’s response to lower salience for a comparatively small toll payment may significantly differ from the interest taken in higher value taxes (Finkelstein, 2009). ¹ The hypothesis, however, is that tax salience can have particular resonance in consideration of topics such as the potential introduction of value added taxes (as has been considered in the United States), the reduction of budget deficits, the desirability of tax expenditures and electronic tax return filing, among others (Gamage and Shanske, 2011: 19).

Although this article has discussed salience as if the term were synonymous with complexity, broadly understood, it is worth emphasizing that tax salience is a term with several connotations of meaning. As Gamage and Shanske explain, there are two categories of tax salience within the literature, namely, ‘economic salience’ (as coined by Schenk), which indicates the impact of the presentation of tax upon economic decisions and ‘political salience’, which refers to the impact of tax presentation on the political process and election outcomes (Gamage and Shanske, 2011: 20, citing Schenk, 2011: 272–273). Gamage and Shanske emphasize that these two categories must be distinguished because increased political salience can result in reduced economic salience (and, indeed, the converse) (Gamage and Shanske, 2011: 20). Additionally, whilst they do not suggest that either economic or political salience should be ‘ignor[ed]’, they also caution as to the importance of remembering that claims about salience in ‘real-world policy’ are not supported by empirical evidence (Gamage and Shanske, 2011).

Salience, as an interdisciplinary term, exists, as a concept, at the borderline of behavioural economics and political discourse. What is its potential value in legal analyses? One potential benefit of introducing terms from behavioural economics into analyses of tax law and policy is the possibility of a clearer, public understanding of the impact of complex tax legislation. Indeed, in an earlier piece, Schenk emphasized the importance of clarifying terms such as ‘transparency’, ‘complexity’ and ‘salience’ and warned that they are often confused, especially in the context of the public discourse surrounding ‘hidden taxes’ (Schenk, 2011: 255). Especially when salience is presented as a remedy for purported hidden taxes, Schenk emphasizes, it is important to clarify what, exactly, transparency, complexity and salience mean (Schenk, 2011: 255). Transparency, traditionally lauded as a very positive goal, is valued both as a virtue, and for its own sake, and as a tool with instrumentalist potential for improving the political process (Schenk, 2011: 257–259). The means by which transparency might be achieved, however, are less clear (Schenk, 2011: 259).

Schenk writes that one of the precepts behind the public discussion of complexity is ‘[t]he general idea … that the government should operate in the open and not in secret and should strive to make information available to the public’ (Schenk, 2011: 259). This idea, or value, is evident in some aspects of the political process (e.g. when legislatures publicly debate proposed legislation) but not in all (i.e. when taxpayers are unable to contribute to the hearings or to the process) (Schenk, 2011). It is against this background that the impact of tax complexity should be assessed.

Schenk defines the term complexity itself as, unsurprisingly, a complicated term, but one which essentially indicates a taxpayer’s capacity to understand, and to comply, with legislation (Schenk, 2011: 261). The complications emanate from the different categories of complexity, including compliance, ‘transactional’ (or behavioural responses to law, undertaken in the hope of paying less tax) and ‘rule complexity’ (simply, difficulty understanding the law) (Schenk, 2011: 261). These categories of reactions both ‘overlap’ and obscure each other (Schenk, 2011). Schenk, however, utilizes the term to refer to rule complexity, for the purposes of her analyses of tax salience.

As, in the context of tax, salience analyses involve discussions about rules, it follows that specificity of expression is important in this area, not least because different types of complexity have different types of impact. Additionally, some forms of impact are more important than others. Economic, or market, salience is not considered to be a significant problem because taxpayers have the capacity to ‘learn from experience’ (Gamage and Shanske, 2011: 21). In fact, attempting to curb low market salience could cause more problems than are solved. An increase in political salience, conversely, could be construed as essentially providing taxpayers with ‘false or arbitrary information about tax costs’ (Gamage and Shanske, 2011: 21). The reason for this is that ‘[n]either the fields of philosophy nor of voter psychology are sufficiently developed to guide us as to what information voters ought to focus on when assessing real-world fiscal policies’ (Gamage and Shanske, 2011: 21). Given this, ‘[d]emocratic values thus provide no indication as to whether political salience should be made higher or lower’ (Gamage and Shanske, 2011: 21).

Gamage and Shanske’s ambivalence over the value of political salience is echoed by Weisbach who has suggested that the desirability of increasing knowledge of tax law depends upon three questions (Weisbach, 2013). These include ‘beliefs’ about tax legislation ‘in the absence of knowledge’, ‘the quality of the tax law’ (i.e. how well is it designed?) and ‘the type of tax’ (Weisbach, 2013: 188). On the third point, Weisbach explains that it is most likely desirable that taxpayers should understand the intention behind and design of the so-called Pigouvian taxes or taxes that are designed to correct negative externalities (Weisbach, 2013). As Pigouvian taxes are designed to correct inefficiencies in the market, they, simply, are meant to impact upon taxpayer behaviour. ² They differ in this way from taxes that are designed only to raise revenue, and for which, Weisbach argues, there is a less compelling argument in favour of increased taxpayer understanding (Weisbach, 2013).

The clear implication from Weisbach’s arguments is that, sometimes, the capacity of a taxpayer to understand certain types of taxes is not a very important point to consider in assessment of its overall value. Indeed, Weisbach emphasizes that it is crucial that the specific context of each tax is clarified, before assumptions are made about the desirability of increasing taxpayer understanding (Weisbach, 2013: 189). For example, thresholds and marginal rates for tax credits in the United States are designed in a way that makes them difficult to understand (Weisbach, 2013). Should they be simplified and rendered easier to understand? Perhaps, it is worth recalling that some forms of general anti-avoidance legislation appear to be predicated to some extent on the belief that it is sometimes undesirable for taxpayers to understand legislation well or, at the least, well enough so as to be able to spot the loopholes (Weisbach, 2013).

Weisbach concludes that ‘ … unfortunately, … whether knowledge of the tax law improves the efficiency of the system is contingent on a number of hard to observe factors’ and thus, ‘clean’ answers to the questions surrounding the desirability of increasing taxpayer understanding may not be possible (Weisbach, 2013: 210). Further, given that some tax law is designed with the objective of influencing taxpayer behaviour as little as possible, it then may follow that it would be useful to aim to ensure that tax legislation is difficult to understand (Weisbach, 2013: 187). It should be emphasized though that this would be quite a specific type of tax and this suggestion should not be applied to all forms of tax legislation.

What is Tax Complexity? Consideration of the Target of the Tax Salience Literature

Despite Weisbach’s warning against clean answers to the question of whether it is desirable for taxpayers to understand legislation, and despite the very specific nature of the types of taxes to which the desirability of comprehensibility may/not apply, the radical suggestion that tax complexity in some instances, indeed, may be positive does provide a useful transition to the following questions. When is tax complexity a good thing? Is high tax salience, generally, of dubious, or at least relative value? One way of addressing these questions is through separate contexts. First, the value of tax salience can perhaps be understood in the area of consumer protection. Studies have shown that if, as is common in the United States, taxes are not included in the display price of a product, there is a potential that consumers will underestimate the ultimate cost of a product and, thus, overconsume (Nussim, 2010: 222). Second, the issue of taxpayers claiming back tax presents an example of an issue for which it is appropriate to ask, for whom does ‘value’ have relevance? Deductions for pensions, and charitable giving, are claimed back in the United Kingdom at notably low rates, presumably because taxpayers do not understand that they are entitled to make the claim (Greenwood, 2010; Hills, 2006). Yet when taxpayers are presented with the option of automatic assessment – of not receiving a letter through the post from HM Revenue and Customs (HMRC) – on the whole, the discomfort with which many taxpayers view HMRC interaction may lead them to assent readily, without awareness that this may lead to paying more tax than is required. ³ (See generally, albeit in an Australian context, Braithwaite, 2013.)

In this sense, perhaps, low tax salience has value but perhaps for the tax collector, and not for taxpayers. Yet tax complexity may arise not simply because it benefits the tax authority but also as a consequence of pursuing worthwhile and necessary objectives in the interest of the taxpayer. For example, complex rules may be necessary to measure income accurately or to ensure that benefits are received by the taxpayers who need them (U.S. Government Accountability Office (GAO), 28 June 2011). Additionally, complex rules may be an effective response to non-compliance, which has the potential to be viewed as positive action taken on behalf of taxpayers who do comply ( GAO, 28 June 2011).

A third possible source of complexity (in addition to the pursuit of goals that benefit either the tax authority or the taxpayer) is that, perhaps, taxpayers ask too much of tax legislation or place unreasonable demands upon the capacity of written legislation to communicate, effectively, what at first glance might appear to be simple concepts – for example, the tax terms of asset or gain. It may be the case that the rules themselves have been designed with unrealistic expectations or without the benefit of a clearly articulated policy. Hart, in his account of the ‘open texture’ of language, famously referred to the relative ignorance of fact and the relative indeterminacy of the aim of the legislator requiring reassessment at the point of application (Hart, 1961: 7; Shaw, 2013). This applies clearly to much of the tax lexicon.

On the point of policy, Sawyer and others have suggested that clear policy, in fact, is the sine qua non of clear tax legislation. Indeed this argument is that for tax legislation to be considered successful, or ‘effective’, the policy should be clear from the statute itself (Sawyer, 2013: 321, citing Oratore, 2010). Similarly, if the policy underpinning the legislation is unclear, then it is difficult, Sawyer argues, for the statute to be effective (p. 322).

He explains that the tax legislative process in New Zealand is based upon these principles, which are highlighted as priorities within a system of extensive collaboration between tax academics, the government and the tax authority but comparatively limited input from politicians. Sawyer suggests that the heavy involvement of tax academics in the design and review of tax legislation is both an advantage, in terms of the lack of politicization, and a disadvantage, in terms of a struggle for legitimacy (Sawyer, 2013: 329). Turning his attention from New Zealand to the United Kingdom, Sawyer proposes that the need to ‘fast-track’ legislation through the Parliament has meant that efforts towards clarity, sometimes, are simply lost (pp. 330–331). Reminding us that the UK’s tax legislation is the longest in the world (p. 334), Sawyer writes that he welcomed the establishment of the OTS as a step in the direction of addressing the simplification of tax policy, in its purest sense, and not just a nod to improving the form of tax legislation (p. 342).

Salience, Complexity, and the Mechanics of (Tax) Law Production

In this sense, Sawyer is evocative of much of the literature surrounding the benefits of ‘principles-based drafting’ (Cooper, 2010; Freedman, 2010). Principles-based drafting has been at the heart of discussions in the United Kingdom about the reform of complex legislation, at least since Avery Jones’ famed call to place it at the heart of reform efforts (Avery Jones, 1996) and Freedman’s clarifying return to the question in 2010 (Freedman, 2010). Principles-based drafting is a subject of frequent discussion but less frequent efforts towards definition – and, indeed, Cooper explicitly declines to define principlesbased drafting, although he describes it as similar to ‘an operative rule’ (Cooper, 2010: 341). That is to say, ‘[i]t is not a statement of the object or purpose that is to be accomplished by another rule, nor is it an aid to interpreting the operative provision’ (Cooper, 2010: 341). Whilst the principles-based drafting literature has a comparatively long history in the United Kingdom (Avery Jones, 1996), Freedman has cautioned that, in addition to acknowledging its promise for shorter legislation and fewer anti-avoidance provisions, it is not a ‘ … solution to all ills’ (Freedman, 2010: 717, 735). One reason for caution is the variety of statutes themselves, which range in importance (Katzmann, 2012: 645). A further reason to view proposed solutions to all ills with suspicion is the variety of cultural, and indeed, structural differences between different forms of legislation. Braithwaite also has advised caution and suggests that general principles should be balanced with specific rules (Picciotto, 2007: 15, citing Braithwaite, 2002, 2003, 2005).

The aim of this section is not to revisit the principles-based drafting debate, which has been developed amply by leading authorities in the United Kingdom, Australia and beyond but to step behind the question of ‘how are draughtspersons starting the process of design – is it with a clear principle in place?’ ⁴ to a more basic issue surrounding the nature of the draughtspersons themselves. Is the capacity of a legislature to articulate policy within legislation in any way influenced by superficial factors, such as size? Put differently – and remaining mindful of Freedman’s warning that principles-based drafting is not an all-purpose panacea – are some types of legislatures more capable of producing principles-based legislation (in all of its complexity of meaning) than others? If so, do these legislatures prioritize high levels of taxpayer comprehension, or salience, in the legislative process? Ultimately, is there a connection between tax salience and principles-based drafting?

Principles-based drafting is very much about the mechanics of law production. The consideration of principles-based drafting for the United Kingdom largely has proceeded over the years with comparative glances elsewhere. This is not unusual, and, indeed, it has been common, perhaps at least since (in the United States) Surrey, to find that comparative analyses of tax systems are underpinned by reference to mechanical difference. Although these differences are highlighted as the point of the comparative exercise, the pragmatic value or normative underpinning of the exercise itself is less clear. An assumption of commonality among differing common law tax collection systems is discernible. In Surrey’s case, he suggested that coherent tax law may be easier to produce in the United Kingdom, than in the United States, because of the presence of the parliamentary whip system in the United Kingdom (1956: 1154–1157). Similarly, as mentioned, infra, Sawyer suggests that a significant, comparative (with New Zealand) feature of UK tax law making is the ‘fast tracking’ procedure (2013: 230–231).

Given the general agreement about the value of principles-based drafting, does it come more easily to some systems than to others? Is one form of legislative ‘mechanics’ preferable to another? Significantly, does the value placed on salience vary between these systems? There are some interesting points of reference in the literature exploring this question. Waldron’s exploration of whether the size of a legislature, normally disproportionate to the size of the highest court, ‘tells us anything interesting about law’ is instructive in this context (2000: 510). The size of legislatures, generally, is a particular concern of the tax salience hypothesis, which rests on the assumption that insufficiently salient tax legislation causes excessively large governments (if not, specifically, legislatures) (Nussim, 2010: 259). Although the lack of empirical evidence for this assumption has been noted in the literature, its basis for tax policy appears to persist (Nussim, 2010: 259). A large government (as an administrative structure) is suggested as an undesirable outcome because of the perceived costs and other dangers. For example, the quality of the legislation produced by large legislatures may suffer, simply because there is likely to be more of it or of accompanying items such as explanatory notes. Explanatory notes are a particular concern of Waldron in this context who suggests that they carry the danger of ‘muddy[ing] the waters’, rendering it more difficult to ascertain exactly what was enacted (2000: 533–534).

Additionally, the larger the legislature, the longer the debates over legislation have the potential to last, especially if every member of the legislature aims to make a contribution. Waldron emphasizes that the end product of such debates – the records – needs to be treated with caution. He explains that although thorough and even prolonged debate of plans for legislation does have potential value, it is important to ensure that not every sentence of the debate is treated as if it were legislation (pp. 533–534). This, Waldron explains, is the significance of voting to sift out what should count as law and what should not (Waldron, 2000: 533–534). Voting is important because it has the capacity to counter the possibly negative effects of both large legislatures and prolonged debates. It is also inherently valuable, ‘… secur[ing] for us a legislative process that is both deliberative and decisive’ (Waldron, 2000: 534). It is ‘our notion of formal validity’ (Waldron, 2000: 534).

Voting may be an important part of both legislative quality and legitimacy but it cannot control for every potential deficiency of process. ⁵ For example, it is not possible to write a rule that will cover every eventual contingency, for the simple reason that it is impossible to predict the future (Picciotto, 2007: 14; Stumpff, 2013: 681). If a legislator is overly ambitious and tries to behave as if this is not true, then a number of costs may be imposed (Picciotto). The problem, however, is that these costs may be unrecognized (Stumpff, 2013: 676). These dangers must be balanced against the standard supposition that uncertainty is costly, and, thus, it is important to attempt to create a rule that will address most possible situations (Stumpff, 2013). Even if an exhaustive set of rules is not possible, the reasoning follows, transaction costs should be lowered in proportion to the number of rules that are introduced and possibilities that are covered (Stumpff, 2013).

One response to this dichotomy might be to suggest that costs seem not to be the problem, so much as different types of costs and the preferences of the person viewing the problem. That is to say, that some taxpayers may object to the costs associated with uncertainty and others may object to the costs that result from rules underpinned by unrealistic expectations as to the extent to which it is possible to predict future contingencies. In this context, it may be worth recalling the warnings within the literature that tax law has less promise in its interaction with economics, even as compared with other areas of law (Raskolnikov, 2013: 524–525). Generally, however, it seems sensible to follow the approach of Gamage and Shanske, Schenk and other writers within the salience literature and to treat uncertainty, realistic approaches to future contingencies and even salience as values to be included on a sort of ‘wish list’ of useful aspects in legislation, to be prioritized in different orders, for different taxes. Thus, for example, Schenk’s distinction, infra, is between the value of salience for revenue-raising taxes (low) and for taxes designed to change behaviour (high).

A number of potential points of interaction between the salience literature and Waldron’s work with the size of legislatures are evident at this point. First, it is worth heeding the admonitions of Schenk, Gamage and Shanske and others that salience is a layered term that should be employed with specific reference to political, economic and legal contexts. Second, the conclusions of some portions of the tax salience literature that legislative clarity should be included only within a list of priorities for legislative drafting, and not necessarily at the top of that list, is evocative of existing analyses of principles-based drafting in the United Kingdom, New Zealand and elsewhere. Neither salience nor principles-based drafting have the potential, on their own, to ‘fix’ the problem of tax complexity.

The potential contribution of Waldron’s work with the size of legislatures to the tax salience and tax complexity literature likely will hinge on the impact of the voting process on the production of principles-based legislation. Is voting an important part of this process or in some ways an afterthought to the more important efforts of the legislative draughtsperson? Does voting protect, promote or hinder principles in legislation? The focus in the previous section, on one very particular aspect of Waldron’s wider investigation into whether the size of legislatures provides any insight into the law they produce, turns on voting as a key part of securing legitimacy of process and forcing decisiveness. Through voting, Waldron suggests, one is able to determine what ‘counts’. It is important however to consider these insights in light of what has been described as ‘the voting paradox’ (Katz, 2011: 4–6). It is probably best summarized, briefly, by the following, favoured joke of rational choice voting theorists, ‘… a man … goes into a restaurant, sees chicken and steak on the menu, and chooses the chicken. The waitress tells him that there is a fish special that day, whereupon he chooses the steak instead’ (Katz, 2011: 8). It is all a matter of ranking, runs the implication and the results of such a system are quite likely to be ‘irrational’ (Katz, 2011).

The implication of the fish/steak/chicken joke for tax complexity is that whilst voting may indeed be an important part of conferring legitimacy onto the products of legislatures, simply because the results are legitimate, it need not follow that they are very well written or, indeed, rational. The larger issue of the tax salience debate, however, deals not with process, or form, but goes to the heart of the rule itself – put simply, is it ever acceptable, or indeed, moral, for a legislature to produce a rule that its populace cannot understand, with deliberation and by design? Evocative as this is of the Hart/Fuller debate, the intention of legislatures, at the outset, must be to attempt to design rules that citizens will understand, regardless of the value of lower tax salience (Hart, 1958; Fuller, 1958).

Salience, Confidence and Fairness

There is a danger in assuming that there is a generalized agreement at the core of rules, which delegates all presumed complexity to the peripheral or grey areas. Complexity in tax legislation can produce an undesirable environment of uncertainty, which can increase the amount of ‘cat and mouse’ gameplaying (Picciotto, 2007: 15). In this article’s earlier examination of the connection between tax salience and tax compliance, it was noted that there has been limited empirical evidence into the connection between the comprehensibility of fiscal legislation to taxpayers and levels of taxpayer compliance. The phrasing of this question, however, was very narrow. It does not consider, for example, whether tax complexity may produce other effects and whether these indirect impacts may lead to reduced compliance.

Among a selection of proposals to improve taxpayer compliance, Picciotto has advanced the reduction of complexity (Picciotto, 2007). He argues that if taxpayers do not have confidence in the tax system as a consequence of the environment of uncertainty, then opportunities, and the motivation, for tax avoidance may increase (Picciotto, 2007: 23). Additionally, the role of formalism in the creation of tax complexity is significant. A traditionalist, formalist approach to rule making may produce overly detailed and complex legislation, which may lead to uncertainty and tax avoidance – and, then, in order to combat tax avoidance, detailed rules are deemed to be ever more important (Picciotto, 2007). This is a vicious circle.

There may be relatively little empirical evidence that levels of tax salience correspond with levels of tax compliance but, as Picciotto explains, there is ‘ample’ evidence to support the suggestion that if taxpayers do not view a tax system as fair, then tax avoidance is more likely (Picciotto, 2007: 24, citing Braithwaite, 2003). This may be due to the fact that compliance is linked to whether or not a taxpayer has a sense of identification with the state (Picciotto, 2007). Identification, in particular, is undermined during periods of ‘rapid social change’, such as the forces of globalization (Picciotto, 2007), or the aftermath of the financial crash. The creation of offshore financial centres, or tax havens, also risks that taxpayers may view systems as enormously complex (involving competing legal systems) and unfair (Picciotto, 2005). Then, if tax complexity undermines perceptions of fairness, and identification with the state, it may be linked in these ways to an increase in levels of tax avoidance.

It is important at this point to distinguish between two, different approaches within Picciotto’s analyses to reducing tax complexity, namely, focusing on improving tax clarity or focusing on tax simplification (Picciotto, 2005). Simplifying legislation actually may increase the likelihood of uncertainty, he argues, and thus may contribute to the tax avoidance cycle (Picciotto, 2005). Both tax simplification and the primacy of clarity of policy articulation have been prioritized within the United Kingdom by the relatively recent establishment of two, separate governmental offices, namely, the OTS and the Office for Budget Responsibility (OBR). The OTS, which commenced operation in 2010 in the United Kingdom, has the advancement of simplified tax legislation as a clear mandate and aims to provide advice to the government on this issue. ⁶ Towards this end, in October 2013, the OTS launched a ‘tax complexity project’, which has the objective of considering tax complexity in the context of the entire UK tax system, in a holistic sense. (The Office of Tax Simplification, n.d.) In its introduction to the project, the OTS identified several problematic factors that are likely to give rise to systemic complexity, namely, length, language, style of drafting and the variety of difference taxes (OTD, n.d., citing Broke, 1999). The OTS identified two further pressures, namely, politics and the introduction of new ‘policy initiatives’ (OTD, n.d.). This section will conclude by considering these points and their impact upon salience, confidence and fairness.

The OTS guidance document notes that whilst the ‘oft-quoted’ fact that the United Kingdom has the longest tax legislation in the world has been considered in detail by simplification efforts, it is not necessarily the only factor of importance (OTD, n.d.). The OTS do not dismiss length of legislation as unimportant, however, and suggest that length actually may present a significant ‘psychological’ barrier for taxpayers (OTD, n.d.). This is evocative of Picciotto’s suggestion that the perceived fairness of a tax system may be connected to compliance (infra). Although Picciotto does not refer to psychological barriers specifically, it would appear that both Picciotto and the OTS agree that a taxpayer’s relationship with the tax system is important. ⁷

The proposal within the tax salience literature that the comprehensibility of legislation needs to be placed lower on a list of desirable factors should be considered against this background. Put differently, if the persistence of tax complexity presents psychological barriers, or in some way contributes to mistrust of the state, then its place on this last of factors should be reconsidered. This proposal, however, raises the question of for whom tax legislation is intended. Picciotto suggests that the audience for the Tax Law Rewrite Project, for example, ultimately, was the profession (2007: 24–25).

In what sense could a reform of tax legislation target only the profession and not taxpayers? Consider Pechman’s argument, as far back as 1975, that tax legislation generally reflects more about the political compromises than any decisions that may have been taken on the grounds of what may be best for an equitable tax system. (1975: 57). The prioritization of the pragmatic is neither novel nor new. Perhaps these pragmatic choices ultimately only impact what could be described as tangential or side issues, and not necessarily inherent ideas (e.g. the difference between income and capital), which can be challenging to articulate. Is it possible ever really to simplify an idea that, given its basis, is necessarily complex? A further issue to consider is the engagement of the Parliament with this process. If one considers the engagement with voting theory, earlier in this article, then the core idea behind legislation may be underpinned with as much logic as choosing (in the famed illustration) chicken over steak. Focusing on the simplification of language used in legislation, in these circumstances, can appear to involve little more than paying attention only to the periphery of the task.

The OBR, which, like the OTA, also was formed in 2010, has the role of providing independent oversight for governmental fiscal policy, not necessarily to provide permanent, final advice on questions such as principles-based drafting, or the ideal length of tax legislation (OBR, n.d.). So whereas the purpose of the OTS could be described as providing clarity of legislation, the purpose of the OBR (at least tangentially) could be described as providing clarity of policy. These two government offices would appear to have been influenced in their inception by the long discussion of principles-based legislation in the United Kingdom and, even if not necessarily stated explicitly, by the value in political discourse of calls for high salience tax systems. The OBR, like the OTS, has considered its remit from a long-term perspective. For example, in a July 2011 report, the OBR acknowledged that governmental tax policy is ‘rarely’ long term and indicated that it would be unrealistic to expect it to be (OBR, July 2011, para. 23). Too many variables – armed conflicts, economic depression, even human life expectancy – would be very difficult to predict over, say, a 50-year period. Yet tax policy is necessarily concerned with the priorities, and debates, of today.

Organizations such as the OTS and the OBR are very much a product of the modern times. The British Broadcasting Corporation (BBC) compared the establishment of the OBR in the early days of the coalition government with Gordon Brown’s establishment of the Financial Services Authority and transfer of independence from the Bank of England in 1998 (BBC, 2012). The political independence of the OBR, the BBC suggested, was likely to make its job ‘easier’ (BBC, 2012). The reporter, Stephanie Flanders, commented, however, that ‘… the world has not had an enormous amount of experience with these bodies’ ( 2012). Both the OBR, in its reference to ‘political factors’, and the OTS appear very aware that tax laws and budgetary processes are as much about distribution of power as they are about distribution of funding. That government advisory offices may be party neutral, but tax law seldom is, perhaps most evident against the backdrop not of the United Kingdom, but of the European Union (EU). As Kauppi explained, the contents of tax policy, and the budgetary priorities from which it flows, are intimately connected to power, ‘[s]ince all nations would welcome more spending in their nation and all would resist a cut in spending, the EU budget allocation across nations is one manifestation of power that is both observable and quantifiable’ (Kauppi and Widgrén, 2004: 224). The countries with the most investment are the presumed ‘winners’ in any EU Budget. This is not to impute dissent where it may not necessarily have existed – perhaps all other member states may be agreed that a single member state is a funding priority. A reduction in funding for a specific country may be a product of consensus and the country in question may even be content to accede to more deserving causes.

Additionally, although tax and budgetary policy, if clarified as part of a process towards a high salience system, may be exposed as hinging on expressions and acknowledgements of power and political victories, power may not be the only point. A budget may have a variety of functions, beyond simply serving as a bill or an outline of intentions. The major ‘functions’ of a budget may include a response to ‘private wants’, an adjustment of the ‘distribution of income’ or assisting ‘stabilization’ (Musgrave, 1956: 333, emphases removed). These factors will be culturally and economically specific. Towards these ends, and as the literature that introduced this article originates in the United States, it is worth considering it against the background of the specific, cultural and historical budgeting and tax legislative history there. ⁸ In particular, it may be worth recalling the ‘great unraveling’ of the federal budget process in the United States (Rubin, 2007). Among other difficulties, health care and eldercare have suffered from underfunding in the United States since 1998 due to the costliness of wars and natural disasters. The situation in 2007 had become critical, especially as the baby boomer generation then was set to retire, and, apparently, it appeared politically necessary to take decisive action (Rubin, 2007). By 2009, the situation had ‘become far worse than anyone anticipated’, and, given the combination of the financial market crisis and the Iraq War, the US federal budget had reached ‘chronic deficit’ (Posner, 2009: 233). A taxpayer’s ability to understand the law might appear to be less important against this background. Put simply, in the United States, salience was not prioritized in the 21st century. The US literature with which this article has engaged sought to find a (relativized) place for salience.

By way of contrast, turning attention back to the United Kingdom, the OTS and the OBR were established at the end of period described above. Indeed, both offices were established in the midst of the post-2008 financial crisis, and it may be that economic crises can produce a strained form of thoughtfulness. This is true possibly not only from the perspective of governments but from the perspective of legal scholarship as well, and the literature discussed within this article may be an illustration of this. From the embers of prosperity that, all too recently, seemed sound, healthy and well protected by legislation and regulatory processes forged out of previous economic trouble, interdisciplinarity seems to take an easier hold on public discourse – enter, perhaps, the salience debate and the insights of behavioural economics more generally. Interdisciplinarity has become part of the process of determining how to produce suggestions that will allow prosperity to return. Thus, behavioural economists, and economic sociologists, have been asked to provide an academic input on processes that seem to have derailed completely. Their input is generally treated respectfully, if not necessarily followed completely, especially if their recommendations appear too expensive. Ultimately they have been asked to help determine what went wrong. Interdisciplinarity in this context may lend itself to determinations as to how we may build the economy, and legislation, that is really wanted, as opposed to the one that developed by default.

Conclusion

It remains important to acknowledge the political element of tax legislation, generally. This has been evident in this article’s review of the tax salience literature and consideration of the contributions it may make to existing, theoretical analyses of tax complexity. Yet this article’s concluding emphasis echoes its earlier evocation of the Hart/Fuller debate. As Picciotto also emphasized, the decision as to whether or not a particular rule applies to a given situation necessarily ‘entails a value judgment’ (2007: 16). It follows that the decision as to whether or not a rule is clear can, similarly, involve such assumptions. Importantly, for the purposes of this article, the decision as to the emphasis to be placed upon a taxpayer’s ability to understand legislation will involve value-laden choices.

As behavioural responses to what might be described as ‘low salience taxes’ have increasingly been studied by tax legal scholars interested in the design of fiscal legislation (Schenk, 2011: 255), the concept of tax salience has played a sometimes obstructive role in US political debate, such that proposals that might have improved tax efficiency have been rejected because of concerns over ‘fiscal illusion’ (Gamage and Shanske, 2011: 20, citing Finkelstein, 2009: 970). Of course, that tax complexity is unpopular would appear to make perfect sense, as it is difficult to imagine that taxpayers will be capable of complying with rules that they cannot understand. When a mysterious tax assessment is collected, taxpayers are unlikely to agree that the assessment is fair if they cannot grasp the means by which it was reached. Indeed, it is taken as understood that the ability of taxpayers to comprehend tax legislation is hindered by its complexity. Yet, although ‘[t]he behavioural economics revolution has finally reached the study of taxation’ (Gamage and Shanske, 2011: 19), it need not follow that this form of unpopularity is less important, touching as it does fundamental questions of tax compliance and collection and, perhaps, trust in the state.