Widely used nonprofit efficiency tool doesn't work

At issue is something called the overhead ratio, which is the amount of money a nonprofit organization spends on overhead -- such as infrastructure, executive compensation and day-to-day management -- relative to overall spending. Overhead spending does not include funding spent on implementing programs, such as program staff salaries.

The overhead ratio is important because it is used by many nonprofit rating sites, donors and nonprofit scholars to assess a nonprofit's efficiency at using its resources to accomplish its organizational goals.

"But the overhead ratio doesn't actually measure efficiency, for two reasons," says Jason Coupet, an assistant professor of public administration at NC State and lead author of a paper on the study. "First, the overhead ratio doesn't account for what organizations are actually doing with their resources. And second, the ratio doesn't account for what organizations are accomplishing with their non-overhead spending.

"Some nonprofit researchers have raised concerns about the accuracy of overhead ratios as a means of assessing efficiency, but our work is the first to approach it using efficiency theory -- and we were able to demonstrate the problem using real-world data," Coupet says.

For this study, the researchers collected data on the overhead, programmatic and fundraising expenditures of 666 Habitat for Humanity affiliates around the country, as well as how many houses they were able to offer low-income home buyers.

To better understand the value of overhead ratio as a means of measuring efficiency, the researchers turned to two well-established tools for assessing efficiency: Data Envelopment Analysis (DEA) and Stochastic Frontier Analysis (SFA). The tools rely on different techniques, but have been proven repeatedly to provide valid and reliable assessments of efficiency across multiple types of organizations.

The researchers used the Habitat for Humanity data to calculate the efficiency of each affiliate using DEA and SFA, as well as calculating each affiliate's overhead ratio.

The researchers found that DEA and SFA efficiency scores were highly correlated. In other words, if one tool found that an affiliate was efficient, the other tool usually did too.

However, the overhead ratio was negatively correlated with both tools.

"In short, this demonstrates that not only is the overhead ratio bad at assessing efficiency, but also that using it to assess efficiency may actively mislead donors," Coupet says.

"We're hoping to work with the nonprofit sector to adopt more accurate efficiency assessment tools," Coupet says. "That may be tricky, since many nonprofits have organizational goals that are difficult to quantify, but we're optimistic that better tools can offer richer insights for many organizations. Ultimately, we want to help nonprofit organizations get the most out of their resources, and to identify those nonprofits that are actually operating efficiently.

"If nothing else, we hope people stop using overhead ratios as proxies for efficiency."