Doctoral Dissertation Research: The Private Provision of Public Goods, Classifying American Philanthropy

There are 1.4 million tax-exempt charitable organizations in the United States. These organizations can take multiple legal structures but primarily consist of public charities and private foundations. Many of these organizations fund academic research. Despite their number, relatively little is known about how these organizations fund research and if this research contributes to the public good. The lack of knowledge is caused in part from a lack of classification. Nonprofit, foundation, and association are terms used interchangeably by charitable organizations across legal structure and funding strategy. Further, there is a tendency by third-party agencies to group charitable organizations by their focus area, such as education, health, or community development, and evaluate general attributes like their level of administrative expenses, transparency, and financial sustainability. Thus, organizations cannot easily be compared across fields nor can their approach to philanthropy and impact be evaluated. This project creates a classification scheme to categorize charitable organizations based on their legal structure and philanthropic strategy. This will allow charitable organizations to be analyzed with their peers based on how they are organized and how they give.

This project develops a theoretical model to advance our understanding of charitable organizations using a two-way system of classification based on their legal structure and their philanthropic strategy. The model divides charitable organization first by revenue source -- endowment or donation -- and then by strategy --mission or broadly focused -- creating four categories to capture their operations. The model is then vetted with data from the IRS Form 990 for a sample of approximately 500 charitable organizations that is geographically representative and structurally diverse. The first part of the analysis assigns each organization in the sample to a category from the classification scheme. The second part of the analysis examines how the categories vary by organization's characteristics and giving behavior. Using a multinomial logit model, the four categories are regressed on a series of organizational characteristics and giving behavior of the charitable organizations to analyze what attributes are linked to a classification category. Philanthropists from a set of these organizations will be interviewed to assess if this classification scheme fits within their vision and understanding of their mission and purpose. The interview data will be used to vet the classification model to assess if organizations have been properly categorized by their own interpretations and to understand how that affects the decisions they make. This project advances our knowledge of philanthropic organizations and begins to test how these strategies impact the public good.