Indicators of tax morale: an exploratory study

Abstract

Taxpayer compliance research has tended to focus on why people evade their taxes rather than on why the vast majority of people do willingly comply with their tax obligations. Whilst tax administrations globally seek to improve the efficiency of their revenue collections, there is growing recognition of the need to have a deeper understanding of why taxpayers comply voluntarily. A person’s internal motivations to comply are commonly characterised as his/her ‘tax morale’, the ‘key’ to the puzzle of understanding taxpayer compliance behavior. Thus tax morale is often seen as a ‘black box’ or the error term for the difference between predicted and observed behavior, a puzzle within a puzzle. This paper initially explores the origins and likely determinants of tax morale, followed by a discussion of measurement difficulties. The key feature of this paper isour attempt to use real taxpayers whose actual compliance behavior is known (rather than self-reported) to identify factors that could be indicative of taxpayer morale from data reported in individual tax returns. Using data from the Internal Revenue Service’s National Research Program from the audit of 1,101 cases with only sole proprietor income, we tested six indicators that theoretically have some correlation with tax morale.Our main findings are threefold. Firstly, IRS random audit studies suggest a possible tax morale component to taxpayer compliance based on the distribution of reporting compliance rates. Secondly, it is extremely difficult to separate this tax morale component from other factors that are at least equally significant in deterring underreporting. Thirdly, although much of the focus in the more recent tax morale literature has focused on religiosity as a causal factor, a more secular explanation may be simply one’s personal integrity (or moral rules and norms) irrespective of religious beliefs if any.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 89 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.