RESUMO
Using a permanent income hypothesis approach and an income-giving status interaction effect, a double hurdle model provides evidence of significant differences from the impact of household income and various household characteristics on both a household's likelihood of giving and its level of giving to religion, charity, education, others outside the household, and politics. An analysis of resulting income elasticity estimates revealed that households consider religious giving a necessity good at all levels of income, while other categories of giving are generally found to be luxury goods. Further, those who gave to religion were found to give more to education and charity then those not giving to religion, and higher education households were more likely to give to religion than households with less education. This analysis suggests that there may be more to religious giving behavior than has been assumed in prior studies and underscores the need for further research into the motivation for religious giving. Specifically, these findings point to an enduring, internal motivation for giving rather than an external, "What do I get for what I give," motive.
