About | HeinOnline Law Journal Library | HeinOnline Law Journal Library | HeinOnline

LOG IN
LOG IN


21 Law, Prob. & Risk 1 (2022)

handle is hein.journals/lawprisk21 and id is 1 raw text is: Law, Probability and Risk (2022) 21, 1-20 https://doi.org/10.1093/lpr/mgac010 Advance Access publication on November 12, 2022 Sampling risk evaluations in tax audits: Some modelling issues JOSTEIN LILLESTOLt Department of Business and Management Science, Norwegian School of Economics (NHH), Bergen, Norway The context of this article is the use of sample data to support claims of tax evasion at eateries, where the possibilities are overreporting of take-away sales and underreporting of cash payments. Ratios of sales amounts of alternative types are computed from the sample and used as estimates of the true yearly ratios. Decisions are made by comparison with the reported ratios in the taxpayer's yearly income statement, allowing for sampling risk. To this end, a 'risk distribution' is established and its quantiles used as decision limits. There are different ways of doing the calculation and to establish the accompanying risk distribution, among them models based on Gamma-assumptions, as detailed in Lillestsl (2019, Sample Statistics as Convincing Evidence: A Tax Fraud Case. Law, Probability and Risk, 10, 149-176). They may lead to different results, more or less favourable to the taxpayer. The chosen method must therefore be fair and defensible. In this connection, the question of conditioning turns out to be relevant. The objective of this article is to explore these issues and provide some recommendations on the choice of method. Keywords: audit sampling; tax evasion; ratio estimates; gamma distribution; resampling 1. Introduction Audit sampling has a long history, e.g. see the work of Neter (1952) and the bibliography of applica- tions by Trueblood and Monteverde (1954). Later on, audit sampling became established as a field with handbooks and texts (Arkin, 1963, 1982). These focused on applying general sampling theory to auditing. Then methods were developed for specific issues arising in auditing (Leslie et al., 1979; Arens and Loebecke, 1981) which controlled sampling error. However, legitimate uses of sampling evidence in courts were often questioned, see Sprowls (1957). An early account on sampling in tax audits is given by Heintz and Wendt (1976). The state of this field into the new century may be found in Yancey (2002) and in the bibliography Yancey (2010). By the late 1980s, the statistical community had established statistics and the law as a new field of study, see DeGroot et al. (1986), Fienberg (1989), Gastwirth (1988, 1992, 2000) and Finkelstein and Levin (1990). Here the chal- lenges are discussed in general, exemplified by cases from many fields, among them in litigation and in discrimination cases. Statistical sampling in tax audits with a legal perspective was examined by Bright et al. (1988), and reprinted in Kadane (2008). More recently the professional associations in statistics and in auditing have jointly developed guides for the use of statistics in the legal setting, see Aitken et al. (2010) and COIC & RSS (2017). tCorresponding author: jostein.lillestol@nhh.no © The Author(s) (2022). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com