Nonprofit organizations depend on charitable contributions to provide more than 20 percent of all cash revenue. In 2004, these revenues amounted to $250 billion or a little more than 2 percent of GDP. Philanthropic support does not come cheaply, however. In 2001, for example, 200 major charities spent almost $2.5 billion on their efforts to raise funds. To transform contributions "in kind" into cash, charities of all sizes often use auctions. Despite this, neither the theory nor the practice of efficient fund-raising--and, in particular, charity auctions--has received sufficient attention from economists.
This project draws on both theoretical and experimental methods to provide an examination of three sets of charity auction mechanisms. The first set has already received some attention from both theorists and empiricists: lotteries, first price sealed bid winner pay and first price sealed bid all-pay auctions. The second comprises those that have been discussed in the theoretical literature but have seldom been tested in the lab or field: for example, second price sealed bid winner pay, k-th price sealed bid all-pay, English, Dutch, and silent auctions. The third and final set offers a new format that is motivated by the investigators' experience conducting charity auctions in the field.
This project has two complementary phases. In phase one (years one and two of the project) all three sets of mechanisms will be tested in a controlled laboratory setting; in phase two (years two and three of the project) the three most effective mechanisms will be reevaluated where it matters most, in the field. The first phase is expected to produce the first comprehensive assessment of charity auction mechanisms, and that the results will advance both the theoretical and empirical literatures. The second phase is expected to provide valuable practical advice for real philanthropic organizations at a time when "every dollar counts." Furthermore, the first protocol to allow for and analyze mechanism-specific differences in participation is a key outcome of this project. In particular, endogenous participation is central to the theory of efficient fund-raising, yet it has received virtually no attention in the literature.
The project will thus inform the following questions: Which fund-raising mechanism raises the most money? Which mechanism is most efficient? Are there mechanism-specific participation effects? How does the interaction between format and participation affect revenues? Does the effect of participation last or are participants "treated" by continued exposure? Does bidding behavior conform to theoretical predictions?
The project also integrates research and education in several ways. For example, to prepare for the field experiments the investigators will team teach a research methods course for advanced undergraduates with sections on auction theory, experimental design and statistical analysis. Middlebury College students will be hired to help design and conduct the experiments and some will write senior theses based on the experiments. Research assistants will also help develop classroom experiments to be used as an illustration of auction theory, and the importance of participation and selection, more generally. Lastly, research assistants will help disseminate the results through publications and presentations (most notably, to the organizations that participated in the auctions).
To summarize, this project will not just extend the theoretical and empirical literatures on charity auctions and endogenous participation, but will also provide valuable advice to nonprofit organizations on mechanism design and revenue generation.
