Abstract ABSTRACT: The purpose of this study is to test the feasibility of using a model derived from the judgments of a group of tax experts in controlled experimental conditions to predict actual compliance behavior. A judgment model, using amount of income, source of income, penalty for cheating, and rate structure as independent variables, was derived from CPA tax professionals. The model then was used successfully to predict actual taxpayer compliance using IRS Taxpayer Compliance Measurement Program data. The model also was used to infer the relative importance of the determinants of taxpayer compliance used in the study. Source of income was found to be about three times more important than the next most heavily weighted variable. These results demonstrate the potential for employing a derived judgment model as an efficient means of predicting the effect of proposed changes in tax policy on taxpayer compliance.

[The purpose of this study is to test the feasibility of using a model derived from the judgments of a group of tax experts in controlled experimental conditions to predict actual compliance behavior. A judgment model, using amount of income, source of income, penalty for cheating, and rate structure as independent variables, was derived from CPA tax professionals. The model then was used successfully to predict actual taxpayer compliance using IRS Taxpayer Compliance Measurement Program data. The model also was used to infer the relative importance of the determinants of taxpayer compliance used in the study. Source of income was found to be about three times more important than the next most heavily weighted variable. These results demonstrate the potential for employing a derived judgment model as an efficient means of predicting the effect of proposed changes in tax policy on taxpayer compliance.]