Does an Altruistic Model of Charity Predict Crowding Out?

From, “Does Welfare Spending Crowd Out Charitable Activity? Evidence from Historical England under the Poor Laws”

The theoretical foundation of crowding out is based on the traditional public good model of
charitable giving. Agents derive utility from a public good, in this case welfare provision or
the well-being of others, and regard their own and other agents' contributions to the public
good as perfect substitutes. This means the agent is purely altruistic, in that he is only
concerned with the total amount of welfare provided, such that the model predicts perfect
(i.e. dollar-for-dollar) crowding out between government provision of welfare and private
charity (see for example Warr 1982 or Bergstrom, Roberts and Varian 1986). However, since
the prediction of perfect crowding out is not empirically supported and the predicted level
of giving is unrealistically low, the model has been extended in several directions. One of
these extensions is the impure altruist model developed by Andreoni (1989 and 1990).2 Here,
agents are said to be impurely altruistic as they derive utility from their own contribution to
charity as well as the total level of welfare. One explanation could be that agents not only
care about the well-being of others but also wish to donate to charities `to do the right thing'
or `to do good'. This leads to a situation where crowding out is less than perfect, i.e. less
than one-for-one. Another explanation for less than perfect crowding out is, for example, a
signaling e ect of wealth from charitable giving, as in Glazer and Konrad (1996). However,
the predicted relation is still negative.