Donor-advised funds (DAFs) are becoming increasingly popular. DAFs receive a growing share of all charitable donations every year and control a sizable proportion of grants made to other nonprofits. The growth of DAFs has generated some controversy over their function as intermediary philanthropic vehicles. Much of what is written is fueled by anecdotal evidence or aggregated national statistics. Using a panel data set of 996 DAF organizations from 2007 to 2016, this article provides an empirical analysis of DAF trends. We conduct longitudinal analyses of key DAF metrics, such as grants and payout rates. We find that a few large organizations heavily skew the aggregated data for a rather heterogeneous group of nonprofits. Next we use macroeconomic indicators to analyze changes in DAF metrics during economic recessions. We find payout rates increased during times of recession, as did a new variable we call the flow rate.
When nonprofit organizations deliver services on behalf of the government, the government agency has the opportunity to select the optimal number of providers to maximize performance. Should more providers deliver services across smaller areas to increase local tailoring or should contracts be consolidated so fewer providers deliver services across larger areas to take advantage of economies of scale? This paper examines a series of contract consolidations aimed at improving the performance and reducing the costs of the Combined Federal Campaign (CFC), the Office of Personnel Management’s workplace giving program for federal employees, which was administered by contracts with nonprofit intermediaries. Using a difference-in-differences analysis based on waves of contract consolidations over time, I find that larger service areas typically had lower giving and costs on a per employee basis. The consolidation process itself tended to decrease average giving further but had no additional effect on costs. Combined, these effects yield no change in costs per dollar raised for larger or consolidated service areas; the benefits of contract consolidation were more modest than CFC administrators had hoped.
The Effect of Simultaneous Solicitations on Charitable Giving: Experimental Evidence on Donor Decision Making (Revised July 2019), Submitted [Survey Instrument]
Simultaneous solicitations are an important part of the modern, competitive philanthropic landscape. Typical simultaneous solicitations are time-limited giving opportunities involving a list of approved charities. Examples include Giving Tuesday, community giving campaigns, and workplace giving campaigns. To grapple with this phenomenon, it is important to understand how donor decision making regarding lists of charities differs from their decision making regarding a series of single-organization solicitations. This paper presents evidence from an online survey experiment comparing simultaneous (list) vs. sequential (one-by-one) solicitation decision making. While the donation amount did not differ between groups, simultaneous respondents donated to fewer organizations. Simultaneous respondents were also more likely to compare the organizations to each other, which disadvantaged low-familiarity organizations. Simultaneous respondents reported higher donor satisfaction and lower decision difficulty, which may explain the rising popularity of these giving schemes. Finally, simultaneous respondents were better able to remember the amount they donated, which indicates that sequential giving may benefit from a lower incidence of mental budgeting behaviors. Taken as a whole, these results suggest that simultaneous solicitations induce simpler decision making rules, which may lead to a shift in the distribution of gifts among organizations as such solicitations become more popular.
The number of nonprofits in the United States is growing dramatically. Between 2003 and 2013, the United States experienced a 19.5% increase in 501(c)3 public charities. If aggregate giving does not increase proportionately, this trend could force nonprofits to operate with smaller average donations. Using aggregate and employee-level data from the Combined Federal Campaign (CFC), the nation’s largest workplace giving campaign, this paper asks if increasing the number of nonprofit organizations affects giving. I find that the addition of new nonprofits did not substantially increase the charitable resource pool. It did not increase the proportion of individuals who gave. Furthermore, although donors gave slightly larger gifts after new organizations entered, the amount was not large enough to prevent the average nonprofit from losing revenue. Unless it is accompanied by substantial efficiency gains, fragmentation will increase total sector spending on overhead and other fixed costs, leaving less for mission-related activities.
A Donor-based Concept of Nonprofit Competition (Formerly titled “The Structure of Nonprofit Competition and Its Relationship to Overhead”)
This paper proposes a new definition of nonprofit markets based on individual-level donor behavior and donor-nonprofit network ties. Empirically defining markets in the nonprofit sector has been difficult, because the well-established empirical industrial organization method of market definition uses price data, and it is therefore not applicable to nonprofit donations. Instead, nonprofit scholars rely on market definitions based on an organization’s subsector and geographic location. However, these definitions fail to capture important facts about donor behavior. This paper defines nonprofits as competitors if they have overlapping donors. The definition is validated with data from the Combined Federal Campaign (CFC). The validation exercise shows that the new market definition predicts donor substitution among organizations 58% more accurately than the standard nonprofit market definition based on an organization’s subsector and geographic location. The CFC data and this donor-based market definition are also used to examine an important nonprofit policy issue—the relationship between market concentration and nonprofit spending on overhead.
This work explores the effects of time-limited, multi-organization giving days, such as the popular and successful Giving Tuesday, on nonprofit organizations. Some charitable organizations are concerned that this day may displace, rather than increase, overall year-end giving. New solicitations may generate donations by lifting overall giving or by shifting it across organizations or time. Understanding the degree of lift versus shift in charitable giving is one of the most important, yet unresolved, questions in the charitable giving literature (Scharf et al. 2017). This paper presents evidence from a multi-wave online survey experiment to simulate typical end-of-year giving conditions. Experimental subjects in the treatment group received three solicitations, including a multi-organization solicitation in the style of Giving Tuesday, while subjects in the control group received two solicitations. The Giving Tuesday treatment causes an increase in total donations, and does not cause a significant shift away from giving in the later waves. The benefits of the Giving Tuesday treatment are concentrated among a few “high-familiarity” organizations.
Which Nonprofit Industries are the Most “Local”? (Extended abstract and conference slides available, please email for more information)
Researchers studying nonprofit competition usually define markets based on geopolitical boundaries and, as a result, restrict their analyses to “local” nonprofit industries. This research note examines whether researchers are, in fact, using the most local industries for their analyses. It finds that some industries which are commonly assumed to be local actually have a substantial number of regional and national organizations. This paper uses data from the Combined Federal Campaign to evaluate nonprofit industries’ localness of operations and donations and rank the most local nonprofit industries. Incorrectly identifying local nonprofit industries may affect our understanding of nonprofit markets and competitive dynamics. Insufficiently-local industries should be excluded from future analyses relying on local markets.