Overview
With over a million and a half nonprofits in the U.S. alone, there is a seemingly infinite variety of nonprofit services. But needs seem to always outpace supply. To provide these services, nonprofits need resources. Donations may be the first resource that comes to mind. But nonprofits can also leverage volunteers to put their plans into action. In some cases, it may even be possible to enlist the help of beneficiaries themselves.
But what do people think about using beneficiaries as volunteer labor. On the one hand, it could reduce nonprofit costs while dissuading beneficiaries from simply taking advantage of free services. On the other hand, it could deter beneficiaries who are unable to offer their labor or are hesitant to do so.
Perhaps an even more important question is what effect would such an exchange-based model have on donations? Might donors view such a model as frugal, innovative, or cost-effective, and thus be more likely to give? Research on effective altruism suggests donors may be more emotional than rational (Berman, et al., 2018). At worst, donors might view such exchange-based services as mildly exploitative, and in turn refuse to donate. Political beliefs may also play a role. Those that lean economically conservative might be in favor of a work-for-services nonprofit model, while liberals might not.
But what do people think about using beneficiaries as volunteer labor. On the one hand, it could reduce nonprofit costs while dissuading beneficiaries from simply taking advantage of free services. On the other hand, it could deter beneficiaries who are unable to offer their labor or are hesitant to do so.
Perhaps an even more important question is what effect would such an exchange-based model have on donations? Might donors view such a model as frugal, innovative, or cost-effective, and thus be more likely to give? Research on effective altruism suggests donors may be more emotional than rational (Berman, et al., 2018). At worst, donors might view such exchange-based services as mildly exploitative, and in turn refuse to donate. Political beliefs may also play a role. Those that lean economically conservative might be in favor of a work-for-services nonprofit model, while liberals might not.
The Experiment
We conducted a within-subjects experiment with 205 people on the online research platform Prolific to test whether people are more or less likely to donate to a nonprofit that provides its services in exchange for beneficiary labor. Participants were instructed to read a few brief descriptions of nonprofits, including one that offered its services for free and one that offered its services in exchange for beneficiaries’ work. Then, participants answered a survey question asking how likely they’d be to donate to each nonprofit.
We used two hypothetical nonprofits for this experiment, randomizing which nonprofit offered free services while the other offered services for labor. The two nonprofits included Hunger Haven, “An organization that distributes food to people in need,” and HomeNow, “An organization that provides temporary housing for homeless people.” Each brief description was appended with one of the following phrases, either “for free” or “in exchange for their help [stocking the shelves / cleaning the property].” We randomized which condition and nonprofit each participant saw, as well as the order in which the two nonprofits were presented, for counterbalancing.
Below is one combination of descriptions that participants could have seen. Before each description, participants were told to "Please read the nonprofit description below."
We used two hypothetical nonprofits for this experiment, randomizing which nonprofit offered free services while the other offered services for labor. The two nonprofits included Hunger Haven, “An organization that distributes food to people in need,” and HomeNow, “An organization that provides temporary housing for homeless people.” Each brief description was appended with one of the following phrases, either “for free” or “in exchange for their help [stocking the shelves / cleaning the property].” We randomized which condition and nonprofit each participant saw, as well as the order in which the two nonprofits were presented, for counterbalancing.
Below is one combination of descriptions that participants could have seen. Before each description, participants were told to "Please read the nonprofit description below."
Hunger Haven
An organization that distributes food to people in need, free of charge.
An organization that distributes food to people in need, free of charge.
HomeNow
An organization that provides temporary housing for homeless people, in exchange for their help cleaning the property.
An organization that provides temporary housing for homeless people, in exchange for their help cleaning the property.
After reading each nonprofit description, participants were asked “How likely would you be to donate to this nonprofit? (1 = Very Unlikely, 7 = Very Likely)” on a 1-7 scale (Lee, et al., 2014). This was our outcome measure of interest, which we've used for prior nonprofit studies.
We also measured participants political beliefs with a brief demographic survey at the end of the study. Participants were asked “Generally speaking, do you usually think of yourself as a Republican, a Democrat, an Independent, or something else?”
We also measured participants political beliefs with a brief demographic survey at the end of the study. Participants were asked “Generally speaking, do you usually think of yourself as a Republican, a Democrat, an Independent, or something else?”
Results
A paired samples t-test revealed a statistically significant difference in likeliness to donate (p < 0.001), such that donors were less likely to donate to a nonprofit using an exchange-based service model (avg. = 4.99) than nonprofit with a free services model (avg. = 5.41). But this difference was fairly small, only 9%, or 0.30 standard deviations. The figure below graphically illustrates this difference.
This effect is largely explained by political beliefs. Using a mixed effects linear regression model, we found that Democrats were significantly less likely to donate to the exchange-based nonprofit than the free nonprofit (diff. = -0.60), whereas Republicans and Independents were indifferent (diff. = -0.13), (p = 0.021). Given that our sample was comprised of 57% Democrats, 30% Independents, and 13% Republicans (10% other or no preference), the main effect is largely attributable to this interaction.
More details regarding our methodology and statistical analysis can be found here.
Conclusion
Nonprofits serving those in need are always in need of resources. But the results here suggest that adding an exchange-based element into your service model might not be the way to do it. The likely trade-off in donations may negate the benefit of such an approach. Then again, if such an exchange-based service model was able to save costs of more than 10% of donation revenue, or if your donor base does not lean liberal, such an approach might pay off. Although a single study can’t definitively answer such an important question, it certainly provides some interesting data to consider.
References
Berman, J. Z., Barasch, A., Levine, E. E., & Small, D. A. (2018). Impediments to Effective Altruism: The Role of Subjective Preferences in Charitable Giving. Psychological Science, 29(5), 834–844. https://doi.org/10.1177/0956797617747648
Lee, S., Winterich, K. P., Ross, W. T. (2014). I'm Moral, but I Won't Help You: The Distinct Roles of Empathy and Justice in Donations. Journal of Consumer Research. 41(3), 678–696, https://doi.org/10.1086/677226
Lee, S., Winterich, K. P., Ross, W. T. (2014). I'm Moral, but I Won't Help You: The Distinct Roles of Empathy and Justice in Donations. Journal of Consumer Research. 41(3), 678–696, https://doi.org/10.1086/677226
Methods Note
We used a paired sample t-tests to test for significant differences in donation likeliness between our hypothetical high and low effective altruism nonprofits. For significant differences, the difference between the two groups' averages would be large and its corresponding “p-value” would be small. If the p-value is less than 0.05, we consider the difference statistically significant, meaning we'd likely find a similar effect if we ran the study again with this population of people. To test for interaction effects, we used mixed effects linear regression models. In addition to politics, we also tested whether the results differed based on the nonprofits themselves, Hunger Haven or HomeNow. They did not. We found no significant interaction effect between nonprofit and whether its services were free or exchange-based (p = 0.714).