內生的 稅務調查確率을 가정한 納稅協力模型

• 영문명: A Model of Tax Compliance with Endogenous Probability of Audit
• 발행기관: 한국재정학회(구 한국재정·공공경제학회)
• 저자명: 孫光洛(Gwang L. Son)
• 간행물 정보: 『재정논집』재정논집 제7권, 252~278쪽, 전체 27쪽
• 주제분류: 경제경영 > 경제학
• 파일형태: PDF
• 발행일자: 1993.03.31

영문 초록

This paper develops and estimates a model of tax compliance in which the Internal Revenue Service uses its resources to maximize tax revenue. The model differs from previous models in two important respects. (ⅰ) It reflects the natural temporal sequence of decision making; i.e., a taxpayer first files a tax return and then the IRS decides whether or not to audit. (ⅱ) The model does not restrict the basic variables to only two possible levels as several previous models have done. The theoretical analysis establishes that the probability of audit depends on many variables such as the taxpayer’s reported income, the IRS’s auditing costs, the tax rate, and the penalty rate. The IRS’s budget is related to the probability of audit only when the IRS faces binding budget constraint. An increase in a taxpayer’s risk aversion results in higher compliance and less auditing. The model shows that minimizing the variance of the IRS’s estimate of a taxpayer’s income does not always improve compliance. There is a unique variance for this estimation error which maximizes a taxpayer’s compliance, and this optimal variance for the estimation error depends on the taxpayer’s degree of risk aversion. Several authors have questioned whether it is useful for the IRS to keep secret the formula it uses to estimate taxpayer’s incomes. We show that if the estimation errors are sufficiently small, the IRS would benefit from revealing the estimation formula and committing to audit any taxpayer with reported income less than the estimate.

목차

Ⅰ. 머리말

Ⅱ. 模型

Ⅲ. 比較靜學

Ⅳ. 所得推定 정확성, 納稅者의 위험기피도, 그리고 納稅協力水準

Ⅴ. 所得推定公式의 비밀유지에 관하여

Ⅳ. 맺음말

參考文獻