INSIGHTS FROM:Gregory B. Fairchild; Megan Juelfs
WRITTEN BY: Salem Zelalem
Minority-owned banking has been integral to the United States for more than 150 years, both as a staple of communities and as a driver of racial equity. Still, Minority Depository Institutions (MDIs) have received a lot of criticism; the general perception is that they are risky, low-performing and ill-advised banking alternatives. MDIs suffer from the same racially driven perceptional biases that convince investors to label minority businesses as “risky,” despite data indicating otherwise. The argument against MDIs goes: As recipients of specific government support, investment in MDIs is potentially costly to all citizens, including those who are not clients. These speculations swirl at all times, but especially in times of financial crisis, they rise to the surface with more intensity. Some critics are already questioning if smaller, community-oriented banks can survive the economic fallout from the coronavirus.
However, recent research by Professor Gregory Fairchild and his co-authors belies the public perception, finding financial evidence that supplements societal drivers to support MDIs.
Criticism of MDIs has propagated for decades. One early detractor was Andrew Brimmer, the first black governor of the Federal Reserve Board. In 1971, he published an article in Journal of Finance stating that these minority banking institutions were operating at a systematic disadvantage to their larger, non-minority counterparts. Due to segregation and resulting wealth disparities, minority banks were often located in areas of poverty and faced greater risk. In addition, he noted higher operating costs, lower margins and high failure rates among local small businesses. Historically, segregation had kept African Americans from studying foundational bank management skills like finance, accounting and economics, resulting in inefficient management decisions.
In the very same 1971 issue of the Journal of Finance, Edward Irons, chair of the Howard University School of Business, wrote an article in which he noted the same foundational issues within minority banks. However, unlike Brimmer, Irons did not believe lowered performance followed naturally from minority banks’ intrinsic characteristics, pointing instead to external factors. If minority-owned banks were granted the same initial resources, he argued, they would not be declining at these high rates.
While Brimmer’s position has been the norm in scholarly conversation, Greg Fairchild, Young Kim, Megan Juelfs and Aron Betru find that MDIs are not systematically less efficient than their non-MDI counterparts, despite higher rates of closure (the bulk of closures are voluntary mergers). Their analysis of relative productive efficiency examines how well an institution transfers a set of resources into outputs, like small-business loans or return on assets. This analysis allows investigators to determine how well MDIs as a group perform in maximizing outputs if operating within an equal set of inputs.
Fairchild and his co-authors used a data envelopment analysis to study MDI efficiency relative to comparable banks, finding that MDI and non-MDI institutions perform with equivalent efficiency in terms of return on assets (ROA) and small business loans in the vast majority of quarters examined. Even during the Great Recession, MDIs performed, in most quarters, with equivalent efficiency. Furthermore, when broken down by MDI type, they found that black MDIs tended to perform at the highest relative efficiency, greater than that of non-MDI institutions in 29 out of 41 quarters, beginning in the second quarter of 2004. Hispanic MDIs tend to operate equivalently to non-MDI institutions, excepting the two quarters of data during and immediately after Hurricane Maria — since many Hispanic MDIs operate in Puerto Rico, this is a likely explanation for momentary deviation. Native American MDIs were also found to be equivalently efficient to non-MDIs, while Asian MDIs appear to perform less efficiently than non-MDI institutions.
The overall success of MDI management may be surprising, as the data also show that black MDIs tend to operate in relatively lower income neighborhoods, have lower average rates of ROA, and a lower rate of return on small business loans.
However, the difference in performance is fully due to a lack of initial assets. The research indicates that, if granted the same resources from the outset, Hispanic and Native American MDIs operate efficiently enough to perform as well as or better than non-minority banks. Furthermore, the rates of efficiency show that black MDIs would actually be able to produce better results than non-MDIs. The differences in endowment currently obscure black MDIs’ high efficiency — exactly as Irons intuited almost 50 years ago.
IMPACT OF RESEARCH
MDIs have faced a substantial amount of skepticism, in part because they receive tailored government support. Past detractors have claimed this money is inefficiently allocated, due to a perception of the insurmountable challenges MDIs face, both internal and external. However, this new study demonstrates that the problems are perception and are not endogenous to the banks themselves. Even in hard economic times, MDIs perform at the same level as non-MDI institutions. And in many instances, MDIs do more with the resources they have, and with more investment, MDIs have the capacity to overcome decades of discrimination, efficiently serving the financial needs of members of their communities.
Greg Fairchild co-authored “Good Money After Bad? The Comparative Efficiency of Minority Depository Institutions (MDIs),” forthcoming at Journal of Developmental Entrepreneurship, with Young Kim of the Darden School of Business, Megan Juelfs of Darden’s Institute for Business in Society and Aron Betru of the Milken Institute.
This article was developed with the support of Darden’s Institute for Business in Society, at which Gregory Fairchild is an academic director, Megan Juelfs is associate director of research initiatives and Salem Zelalem is a research assistant.