After-life consumption and charitable giving

American Journal of Economics and Sociology, The, July, 2004 by Warren B. Hrung

I

Introduction

U.S. TAXPAYERS REPORTED ALMOST $100 billion of contributions to charity in 1997 (Internal Revenue Service 1999: 64). Most studies of charitable giving focus on total contributions, while studies on the economics of religion naturally focus only on religious contributions. This study investigates religious giving and nonreligious giving separately. A positive relationship between age and religious giving is found to exist, but no such relationship is found for nonreligious giving. The results are consistent with theoretical predictions that religious giving is fundamentally different from nonreligious giving. Religious giving is assumed to be directly related to after-life consumption, while nonreligious giving is not related to after-life consumption. Previous studies focusing only on total charitable contributions should therefore examine whether or not their results hold for both religious and nonreligious giving.

Section II presents a literature review of studies on charitable giving and the economics of religion. A theoretical model of charitable giving with after-life consumption is presented in Section III. The empirical results are found in Section IV, and Section V concludes.

II

Literature Review

STUDIES ON CHARITABLE GIVING typically focus on total contributions and the price elasticity of contributions (for example, see Feldstein and Clotfelter 1976; Steuerle 1987; Randolph 1995; Auten and Joulfaian 1996; Duquette 1999; Joulfaian 2000). This focus is due to the tax deduction for charitable contributions. A high price elasticity suggests a role for tax policy in terms of influencing contributions in an "efficient" manner (Barrett 1991). In most of these studies, there is a positive relationship between age and total contributions to charity. (See Clotfelter 1985 and Steinberg 1990 for a review of this literature.) Other studies on charitable contributions focus on topics such as the substitutability between time and money devoted to charity (Duncan 1999) and the interdependence of gifts to charity among similar individuals (Andreoni 1988; Andreoni and Scholz 1998).

Studies on the economics of religion focus on religious contributions of time and money. Economic analysis of religious behavior began with Azzi and Ehrenberg (1975) (AE, hereafter). Their model assumes that households maximize an intertemporal utility function that depends on secular consumption and expected after-life consumption. The time and money devoted to religious activity determine the level of after-life consumption. The assumption that the marginal product of religious activity does not decrease with age leads to the prediction that religious activity will increase with age. Their empirical analysis as well as later work (Iannaccone 1998) finds support for this prediction. Since AE, a large literature has emerged examining a wide range of religious behavior. Iannaccone (1998) provides an excellent review of this literature.

Very few studies examine religious giving separately from total charitable contributions. This is largely due to the fact that many data sets do not provide a breakdown of total contributions into their religious and nonreligious components. The use of tax data is limited because regular tax data only provide information on cash/noncash charitable contributions. The only tax data with information on religious and nonreligious giving is estate tax data. Morgan, Dye, and Hybels (1977), Joulfaian (1991), and Clain and Zech (1999) are some of the few studies that do examine religious giving separately from total charitable contributions. Morgan, Dye, and Hybels (1977) show that religious giving declines as a percentage of total giving as income increases. Contributions to higher education is the main category that increases as a percentage of total giving as income increases. They also find that the relationship between age and any type of charitable giving disappears after income is taken into account. Joulfaian (1991) examines estate tax data and finds a positive relationship between age and all types of charitable bequests. However, his use of estate tax data results in a sample that is primarily over age 60 at time of death. Clain and Zech (1999) examine whether or not nonreligious charitable activity crowds out religious charitable activity. They find that nonreligious charitable activity is a complement to religious charitable activity. With respect to age, however, they find no relationship between age and contributions of money to religious groups. There is thus still no consensus in the literature as to the relationship between age and religious/nonreligious giving.

III

Theoretical Framework

THIS SECTION WILL PRESENT a simplified version of the AE model. However, I will allow for nonreligious charitable giving in addition to religious charitable giving. A household faces the following intertemporal maximization problem:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

A household can spend its exogenous income (Y) on contributions to a nonreligious charity (C) or contributions to a religious organization (R). Utility is a function of nonreligious charitable contributions and expected after-life consumption (A). Religious contributions influence utility only indirectly through after-life consumption, which is a positive function of religious contributions. This model simplifies the AE model by considering only contributions of money and not contributions of time by assuming exogenous income. Contributions to nonreligious groups are included to examine any differences between the two types of giving that arise from this model.