Tax Preferences, Fiscal Transparency, and the Meaning of Welfare: An Experimental Study
| Author | Liam Stanley,Todd K Hartman |
| Published date | 01 November 2018 |
| Date | 01 November 2018 |
| DOI | http://doi.org/10.1177/0032321717731661 |
| Subject Matter | Articles |
https://doi.org/10.1177/0032321717731661
Political Studies
2018, Vol. 66(4) 830 –850
© The Author(s) 2017
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0032321717731661
journals.sagepub.com/home/psx
Tax Preferences, Fiscal
Transparency, and the
Meaning of Welfare:
An Experimental Study
Liam Stanley1 and Todd K Hartman2
Abstract
What is the effect of providing personally tailored budgetary information on public attitudes to
tax and spending? We address this question with a survey experiment based on the annual tax
summaries introduced by the UK tax authorities in 2014. By subtly manipulating the categories
of state spending – in particular, the controversial category of ‘welfare’ – to invoke a sense of
unfairness, we show how budget information in general and the United Kingdom’s annual tax
summaries in particular impact support for state spending. Though the stated aim of providing
personalised tax receipts to income taxpayers is to enhance fiscal transparency, doing so may also
damage support for state spending if the information provides a sense that existing redistribution
is unfair. The article contributes to political science debates about public attitudes to tax and
spending, the character and trade-offs of fiscal transparency, and the framing effects of welfare.
Keywords
tax, survey experiment, UK, welfare, budgets
Accepted: 18 August 2017
In November 2014, the United Kingdom’s HM Revenue and Customs (HMRC) began
posting an annual tax summary to every income taxpayer providing them with an item-
ised breakdown and visualisation of how the government has spent their tax over the past
year. For example, a taxpayer earning £30,000 will be informed that they have contrib-
uted £1663 to welfare and £1280 to health at the top end, and £78 to overseas aid and £51
to the European Union (EU) budget at the bottom end. Costing an estimated £5.3m
(Sweet, 2015), then Chancellor George Osborne justified the scheme as a ‘revolution in
1Department of Politics, The University of Sheffield, Sheffield, UK
2Sheffield Methods Institute, The University of Sheffield, Sheffield, UK
Corresponding author:
Liam Stanley, Department of Politics, The University of Sheffield, Sheffield S10 2TN, UK.
Email: l.m.stanley@sheffield.ac.uk
731661PSX0010.1177/0032321717731661Political StudiesStanley and Hartman
research-article2017
Article
Stanley and Hartman 831
transparency’ that ‘will show how hardworking taxpayers have to pay for what govern-
ments spend’ (Gov.uk, 2014).
The media coverage that followed the launch highlighted serious concerns about
how unorthodox and unconventional accounting and categorisation practices ensured
that ‘welfare’ was the largest spending category (BBC, 2014). Dame Anne Begg, for-
mer Chair of the House of Commons Work and Pensions Committee, claimed that the
tax summaries were part of an attempt to ‘make the phrase welfare almost appear as a
dirty word’ by giving ‘people the impression [that] the bulk of welfare goes to work-
ing-age unemployed people when in reality that is a very small proportion’ (Guardian,
2014). Meanwhile, Frances O’Grady, General Secretary of the Trade Union Congress,
went as far to call the documents ‘party political propaganda masquerading as neutral
information’ (BBC, 2014). The nonpartisan Institute for Fiscal Studies (IFS) stated
that ‘this is not spending that would normally be classed as “welfare”’ (Hood and
Johnson, 2014).
Underpinning this policy debate is a broader question of relevance to political science:
What is the effect of providing personally tailored budgetary information on public atti-
tudes to tax and spending? In addressing this policy-relevant and theoretical question, we
draw on a novel experiment embedded in a national survey based on the UK annual tax
summaries. Rather than manipulating information regarding expenditure flow – for
example, by seeing whether participants react differently to hypothetical information
about either a 40% or 20% welfare spend – we instead manipulated the way in which real
spending data are categorised and calculated. To do so, we used alternative annual tax
summaries created by civil society actors in which welfare was disaggregated into smaller
categories. By randomly assigning survey participants to receive one of these three tax
summaries (and a fourth no-information control condition) prior to a survey that meas-
ures attitudes towards tax, spending and welfare, we are able to test whether the catego-
ries used on the HMRC annual tax summaries affect public support for current state
spending arrangements.
Our results demonstrate that participants exposed to the HMRC treatment are less
likely to agree with how tax money is spent and less likely to suggest that the current
allocation of public funds is a good use of taxpayers’ money. Despite the subtle character
of the experimental manipulation, our results therefore demonstrate that the way in which
budgetary information – and, in particular, welfare – is categorised and presented can
influence support for public spending. Ultimately, while providing budgetary information
in the name of greater transparency may be considered part of good governance, budget-
ary information can be utilised for political ends given that national accounting categories
are contestable and indeed contested.
Our findings contribute to three interrelated debates in political science. First, by test-
ing theories about state–taxpayer relations, we contribute to a long-standing literature on
taxation and democracy (Levi, 1989; Martin et al., 2009; Moore, 2004; Ross, 2004;
Steinmo, 1993), as well as contemporary political science research that draws on meth-
odological innovations and experimental methods in studying attitudes towards tax and
state spending (Brockmann et al., 2015; Paler, 2013; Zhang et al., 2016). Second, our
work contributes to the literature on policy-relevant discussions about fiscal transparency
by highlighting the potential trade-offs and political dynamics of providing ostensibly
objective budgetary information (Alt et al., 2014; Alt and Lassen, 2006; Benito and
Bastida, 2009; Heald, 2003, 2012; International Monetary Fund (IMF), 2007; Organisation
for Economic Co-operation and Development (OECD), 2002). Finally, by placing the
To continue reading
Request your trial