======================= HES POSTING ==================== One recent authoritative work is H. Margolis, SELFISHNESS, ALTRUISM, AND RATIONALITY, Cambridge 1982. Another is an edited volume by J. Mansbridge BEYOND SELF-INTEREST. Without explaining what I mean by the term, the latter includes some neo-classical but mostly non-neo-classical points of view. For somewhat older points of view, I wonder if Burton Weisbrod's work on the non-profit economy would be worth looking into? There are at least two strands of altruism in the literature that I don't think have been mentioned. One is the inter-generation macro models of Barro and others. Each generation has utility for itself and (discounted) for its children. Solving recursively yields a first-period patriarch whose utility is consistent with the entire dynasty. Another strand is in the public finance literature about public goods and donations to charity, with links to "mechanism design" in game theory. J. Andreoni offers a competing hypothesis for altruism: giving itself enters the utility function. People receive a "warm glow," he calls it, just from giving, but don't necessarily care about the outcome. This is in reaction to the standard view which is taken to be that people have utility for the public good and supply it accordingly (of course, the result is underprovision, but not necessarily zero provision). Likewise, they have altruism for others and so give to charities like any other "commodity." But this view would imply 100% crowding out: If the government taxes me a dollar and gives it to the public good or charity, I can still obtain my old utility-maximizing bundle by decreasing my donation by exactly one dollar. (Subject to a boundary condition: the government can't tax me by more than I was giving in the first place.) Experiments and empirical studies have generally rejected 100% crowding out, prompting Andreoni's suggestion. Spencer Banzhaf <[log in to unmask]> Economics, Duke University ============ FOOTER TO HES POSTING ============ For information, send the message "info HES" to [log in to unmask]