Fiscal illusion and fiscal obfuscation: tax perception in Sweden.

• Author: Sanandaji, Tino
• Position: Essay

Fiscal illusion refers to "the notion that systematic misperception of key fiscal parameters may significantly distort fiscal choices by the electorate" (Oates 1988, 65). The premise is that the tax system's design can lead to underestimation of the costs of public expenditure, with the public not being fully informed of taxation's total costs. Fiscal illusion is an example of a collective-action problem in public policy, where the benefits of individual voters' gathering and processing information are shared by many, but the costs are placed solely on the individual (Caplan 2001; Congleton 2001). In this article, we investigate the extent and nature of fiscal illusion in Sweden, drawing evidence from a nationwide survey of approximately one thousand randomly selected Swedish adults.

Sweden constitutes an unusually suitable testing ground of fiscal illusion. First, it has the highest tax rate as share of national income in the world, raising the question of how public support for high tax rates is maintained. Second, in Sweden most tax revenue is collected through indirect taxes rather than direct taxes, which increases the likelihood of fiscal illusion. Finally, the Swedish tax system is both flat and simple. There are few deductions, and taxes are collected on an individual rather than a household level.

Fiscal illusion consequently can be separated more easily from other systematic misperceptions of taxes and income.

Previous Research on Fiscal Illusion

John Stuart Mill ([1848] 1994) suggested that the burden of indirect taxes would be systematically underestimated. Italian economist Amilcare Puviani contributed to the field with more substantial work on fiscal illusion in 1903 (Baker 1983). Puviani argued that the ruling authorities attempt to create a "fiscal illusion"--an underestimation of the real tax burden among the taxed subjects--by means of various fiscal instruments. The notion of fiscal illusion was thereafter left largely unexplored until James Buchanan (1960) restored attention to this hypothesis. Developing Puviani's original intuition, Buchanan distinguished three main strategies that authorities attempting to take advantage of fiscal illusion would ideally employ in order to hide the collection of revenue. One is to use state-owned property to produce income, thus preventing the individualization of net opportunity costs. The second is to use indirect rather than direct taxation, which makes appreciation of the private part of consumption expenditures more difficult for the taxpayer-consumer. Finally, the authorities can raise revenue by means of inflation.

In contrast, the expenditure side of fiscal illusion has been used to reach the conclusion that the public sector is in fact too small. In particular, Anthony Downs (1961) considers the potential results of the complexity inherent in the tax system, where information costs cause rational individual voters to be ignorant of specific aspects of public spending. Downs argues that remote government benefits will tend to be less apparent than indirect taxation. As a result, a vote-maximizing government will keep public expenditures at a "suboptimally low" level as assessed against public preferences under perfect information. Similar arguments have been put forth by John Kenneth Galbraith (1958). Johan Fall and Anders Morin (2001), in contrast, find that the Swedish public overestimated the share of public spending that went to more popular core services, such as health care and schooling, compared to public programs with less measured popularity. The results suggest limited knowledge not only about the level of taxation, but also about the distribution of public spending, with people overestimating the share spent on the activities that the voters value the most.

Empirical work on determining the extent of tax illusion has attempted to link the size of the public sector with measures of fiscal complexity, based on the so-called revenue-complexity hypothesis, which maintains that voters underestimate taxes in fragmented tax systems. The first test of whether fiscal complexity influences the size of the public sector was undertaken by Richard Wagner (1976), who applied the Herfindahl index to empirical investigations of fiscal illusion. According to this index, perfect concentration (where all tax revenue comes from one source) corresponds to a value of unity, with higher dispersion of tax sources resulting in lower values. Hence, a fiscal system is conceived of as being more complex if its revenues derive from a greater number of tax sources. Wagner then regressed total public expenditures for fifty large U.S. cities on the Herfindahl index, controlling for a set of socioeconomic variables. He found that more dispersed tax sources were indeed associated with higher spending--evidence in favor of the fiscal-complexity hypothesis. Following this study, a number of authors (Clotfelter 1976; Munley and Greene 1978; Pommerehne and Schneider 1978; Baker 1983) have attempted to replicate Wagner's findings, using more sophisticated techniques and other datasets, with varying degrees of success. (1)

Wallace Oates (1988) and Brian Dollery and Andrew Worthington (1996) survey the empirical results on fiscal illusion, finding mixed results. In a carefully designed recent experiment, Raj Chetty, Adam Looney, and Kory Kroft (2009) demonstrate that tax salience has economically significant behavioral implications, which indicates that tax visibility matters both for consumer choice and for public policy. (Other studies on fiscal perception include...