JATA - Fall 1990

Volume 12, No. 1

Estimates of the Price Elasticity of Charitable Giving: A Reappraisal Using 1985 Itemizer and Nonitemizer Charitable Deduction Data

James R. Robinson


Traditionally studies of the demand for charitable contributions have utilized data restricted to taxpayers who itemize deductions (itemizers) and have employed models which assume constant price and income elasticities across income levels. During the period from 1982 through 1986 nonitemizers were allowed to deduct a portion of charitable contributions. In this study price and income elasticities were estimated by income level using 1985 data for itemizers and nonitemizers in four states with different state-tax characteristics. The estimated demand equations differed significantly for itemizers and nonitemizers using ordinary least squares regression. The estimated price elasticity was -1.43 for itemizers and -8.50 for nonitemizers, and these estimates did not differ greatly over four income levels. Since only 37.7 percent of nonitemizers claimed charitable contributions in this sample compared to 94.1 percent of itemizers, a maximum likelihood estimation technique was employed to correct for bias from taxpayer reporting behavior. After correcting for reporting behavior, nonitemizer charitable behavior was indiscernible from itemizer behavior, and the estimated price elasticity was -0.67. Price elasticity increased with income level and exceeded one (in absolute value) only for the highest income level (over $100,000). The results call into question the usual assumptions of constant price elasticity across income levels, cast new doubts upon the standard specification of the equation used to estimate the demand for charitable giving, and challenge the standard wisdom that the charitable contribution deduction is an efficient subsidy.

Testing for Framing Effects in Taxpayer Compliance Decisions

A. Schepanski and D. Kelsey


This paper reports upon experimental tests of framing effects in taxpayer compliance decisions. Subjects were asked to indicate whether they would report the correct amount of tax due or take advantage of an evasion opportunity where possible outcomes and related likelihoods were described. Subjects participated in one of three framing conditions -- a loss condition, a refund condition and a final asset condition. In the loss condition, outcomes were expressed as deviations from the status quo. Outcomes in the refund condition were expressed as reductions from an unexpected increase in wealth of $700, whereas in the final asset condition outcomes were expressed in terms of final wealth states that would occur under each state-choice pair. Even though the objective information was equivalent in all conditions, significant framing effects were observed. Subjects responses in both the refund and the final asset conditions were more risk-averse than those in the loss condition. These results suggest that the manner in which outcomes are described to taxpayers can exert considerable influence on their risk attitudes. This in turn suggests that it may be possible to utilize the framing phenomenon to improve taxpayer compliance by changing the relative attractiveness of the compliant alternative. Alternative strategies that might be explored in the future research are discussed.

A Note on Professionals' Judgments of Tax Authority

Craig E. Bain and Bob G. Kilpatrick


Tax professionals collect, weigh, and combine information from various sources to formulate an evaluation of the level of authority for a given tax position. Because no explicit, objective guidelines exist for performing such an evaluation, there are no clear-cut "correct" answers available with which to compare individual professional judgments of most tax issues. Until there are such guidelines, the ambiguity of tax authority judgments must be resolved through consensus, i.e., the degree to which tax professionals concur in their evaluations of tax authority. This research explores tax professionals' evaluations of the level of authority for generic tax situations involving "gray" areas, i.e. areas in which there exists support for and against a tax position. Judgments of 56 tax personnel from a Big Eight CPA firm concerning the level of authority for 130 context-free tax scenarios were analyzed. Results include (1) a rank-order analysis of the weightings of seven sources of authority allowed under Treasury Regulations, (2) judgment consensus analysis of the tax professionals' evaluations of the level of authority in a controlled experiment, and (3) consistency analysis to determine the stability of the tax professionals' judgments over time.

The Income Tax Withholding Phenomenon: Evidence from TCMP Data

Otto H. Chang and Joseph J. Schultz, Jr.


Using data provided by the Internal Revenue Service from the Taxpayer Compliance Measurement Program, this study attempts to determine whether individual taxpayers who owe additional tax when they file their returns are generally less compliant than those who are due refunds. The findings suggest that differences in compliance rates exist and seem to be persistent regardless of the size of the balance or refund due, the filing status, the level of adjusted gross income, or the primary sources of income.

Measuring Tax Faculty Research Publication Records

Robert M. Kozub, Debra L. Sanders and William A. Raabe


This article examines the publication records of tax accounting faculties for the years 1981 through 1988 to analyze the publication environment that faces the business-school tax researcher. This study reviews: (1) the tax research output of tax faculty members by school of residence and school of degree; (2) the effects of co-authorship on the results; and (3) the implications of the results.

A Test of the Accuracy of State Tax Valuation Models for Depreciable Assets

Paul Wertheim, Allen Ford and Maurice Joy


For property tax purposes, state taxation authorities often estimate current market values for depreciable assets with "trending models." Trending models are formulas used to estimate an asset's current market value by adjusting its depreciated historical cost for inflation effects. This study examines the accuracy of state tax valuation models in estimating market values using a sample of depreciable assets. Results show that state models currently used vary widely in estimation accuracy. However, when the variables are chosen properly, trending models can give reasonably accurate estimates of current market values of depreciable assets.