FROM THE LAB TO THE FIELD: A REVIEW
OF TAX EXPERIMENTS
Institute of Development Studies
Tax experiments have been gaining momentum in recent years, although this literature
dates back several decades. With new developments in methods and data availability, tax experiments
have gradually moved away from lab settings and towards the ﬁeld. This movement from the lab to
the ﬁeld has happened against the background of the ‘credibility revolution’ in applied economics,
which has seen more rigorous methods applied to policy relevant questions, and of the availability
to researchers of administrative data from tax returns. These developments have allowed signiﬁcant
advances in the experimental literature on tax compliance. This paper reviews this literature, giving
particular attention to ﬁeld experiments using administrative data, but putting them in the broader
context of the compliance literature. A particular effort is made to take a global perspective, in a
literature that is only recently seeing the emergence of evidence from Africa, Latin America and Asia.
Administrative data; Randomisation; Tax compliance; Tax experiments
Tax non-compliance is one of the greatest obstacles that revenue administrations face. However, increasing
compliance is not only important to collect more government revenue. The way that objective is
achieved has implications also on the practice of tax administration. Collecting taxes through aggressive
enforcement and coercion is more costly than encouraging taxpayers to collaborate with tax collectors.
Furthermore, tax compliance is deeply intertwined with the political environment that shapes the social
contract between citizens and the state. Low corruption and government accountability have been deﬁned
as pre-conditions for countries’ successful tax performance (Bird et al., 2008).
Clearly, in any country, taxpaying is the result of both quasi-voluntary compliance and enforcement.
It would be simplistic to think that taxpayers comply with tax laws just because they think it is a good
thing to do. An element of coercion is present in most cases. For example, withholding procedures
make it very difﬁcult for employees to evade taxes on their wages. Therefore, most of them do not
have the choice of whether to evade or not (Kleven et al., 2011). At the same time, it is also clear that
the observed level of tax compliance is too high to be only the result of coercion (Alm et al., 1992b;
Andreoni et al., 1998). Most countries simply do not have the capacity to control and audit a large
number of taxpayers. Therefore, the right question to ask may be ‘Why do people pay tax?’ rather than
‘Why do they evade?’ (Alm et al., 1992b). The answer is likely to be a mix of deterrence, related to
the probability of audit and sanctions, and a range of factors often referred to as tax morale. Although
there is no uniform deﬁnition of tax morale, a recent review (Luttmer and Singhal, 2014) unpacked
Corresponding author contact email: email@example.com Tel: +44(0)1273 915765
Journal of Economic Surveys (2018) Vol. 32, No. 2, pp. 273–301
2017 John Wiley & Sons Ltd.