A Perspective on the Experimental Analysis of Taxpayer Reporting.
分析美国个人纳税人申报决策的各种方法，指出基于预期效用理论的逃税模型存在根本问题，因为审计率低且惩罚力度不足，无法完全解释纳税遵从行为。
Abstract The article analyzes various approaches on individual taxpayer's reporting decision in the United States. Tax evasion appears to be a large and growing problem in the country. Despite obvious difficulties in measurement, the U.S. Internal Revenue Service estimates that the tax gap, or the amount of underreported federal income taxes, was $83-94 billion in 1987, and had grown at an annual rate of over ten percent since 1973. The analysis of the individual reporting decision has taken a variety of approaches. The underlying premise of nearly all of these approaches, at least those in economics, has been the same: individuals pay taxes because they fear detection and punishment. This economics-of-crime approach is based on traditional expected utility theory, which views a rational individual as weighing the expected utility of benefits from successful underreporting against the uncertain prospect of detection and punishment. There are, however, several fundamental problems with the existing applications of expected utility theory of taxpayer reporting. Although it is clear that detection and punishment cannot explain all compliance behavior. The frequency of audit in the country has fallen to less than one percent, and the additional penalties constitute only a fraction of the unpaid tax liability.