Community Development Financial Institution (CDFI) program evaluation: a luxury but not a necessity?

ABSTRACT

Community development financial institutions (CDFIs) are an integral component of US community economic development policy. These organizations differ in structure and client base, but they all seek to promote sustainable and equitable growth at the local level. By injecting capital into under-served markets, CDFIs facilitate development that may not otherwise occur.  Attempts to measure and evaluate CDFI performance have often yielded disparate results. A systematic review of the literature reveals that capacity constraints have resulted in CDFI evaluations with wide variance in scope and rigor.  Making comparisons across these institutions is difficult because there are few standardized performance metrics that could inform evaluations. Major CDFI funders in the philanthropic and public sectors often have competing demands and do not consistently use evaluations to guide their decisions. To incrementally advance CDFI research and promote a more accurate understanding of their impact, we recommend a shift towards utilization-focused evaluations (UFEs).

KEYWORDS:

• Community development financial institutions
• cdfi
• community development financial institutions fund
• cdfi fund
• community economic development
• performance measurement
• impact measurement
• program evaluation
• outputs
• outcomes
• summative impact
• small business lending
• small business administration guarantees
• riegle community development and regulatory improvement act
• community reinvestment act
• cra
• utilization-focused evaluations
• united states federal policy

Acknowledgments

Numerous researchers, development practitioners, and leaders in the CDFI community provided feedback and suggestions for this article. The authors are deeply grateful for the input of Rodrick Banks, Paige Chapel, Michael Eggleston, Eric Hangen, Helen Leung, Emily Paranjape, Natalie Prochaska, Khaliid Scott, and Michael Swack. We also extend our thanks to Rohan Kumar and Kathryn Obenshain, whose research assistance on this article was invaluable.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Our unit of analysis is US CDFIs, but similar types of organizations exist throughout the world. The use of financing for social impact is becoming an increasingly important component of global development policy (Finnegan et al., 2021; Murtagh & Goggin, 2015).

2. 12 CFR § 1805.201(3) stipulates CDFIs provide 60% of their financial services to a targeted population(s) and/or (2) investment area(s). Targeted populations include lower income individuals, racial and ethnic minorities, or other marginalized constituencies. An investment area is a community in economic distress as measured by a formula of macroeconomic indicators.

3. The act of performance measurement is sometimes treated as being separate from program evaluation. Theodos and Seidman (2017) conceptualize evaluations as being more episodic in nature.

4. The financial strength and performance rating (FSP) is based on an assessment of a CDFI’s capitalization, asset quality, management, earnings, and liquidity (CAMEL) (Aeris, 2018, p. 4).

5. This number does not include all CDFIs which Aeris has ever rated, only those rated recently. Additionally, some CDFIs voluntarily report financial data to Aeris but do not pay to be rated.

6. Not all institutions/individuals allow disclosure of their subscription, the total number may be higher.

7. Though many CDFI funders, and CDFIs themselves, appear to perceive the rating as way to evaluate social impact (Greene et al., 2020, p. 11).

8. This includes regulations which incentivize providing capital to CDFIs (The Community Reinvestment Act, Pub. L, 1977), the establishment of the CDFI Fund (Riegle Community Development and Regulatory Improvement Act, 1994), and the launch of the SBA’s major lending programs that are commonly used by CDFIs (Community Advantage Pilot Program, 2011; Small Business Act, 1953; Small Business Investment Act, Pub. L, 1958).

9. The ILR was discontinued in 2017 and was replaced with a reporting format called the Annual Certification and Data Collection Report (ACR). CDFIs submit the ACR within 90 days of the the organization's fiscal year end.  (Community Development Financial Institutions Fund, 2020).

10. Another perspective is that taking on more risk means the Fund’s grants are having their intended effect, because it suggests the CDFI is injecting capital into high need communities.

11. A limitation of our analysis is that CDFIs are not the only users of the SBA’s lending programs. They are also used by other finance institutions. Only some of the results are attributable to CDFI lending.

12. Our focus is on small business support, but there is also little clarity on how the law has shaped consumer lending. The CRA has incentivized regulated banks to remain open in low-income areas (Ding & Reid, 2020). However, it has made little headway in improving economic conditions within distressed communities that were subject to historical redlining (Park & Quercia, 2020).

13. Defined by the authors as the average proportion of inflation-adjusted (U.S. Bureau of Labor Statistics, 2017) operating revenues according to the CDFI Fund’s (Community Development Financial Institutions Fund, 2019b) ILR data for 2003 to 2017. Public sector sources are defined as funds from federal, state, local, and other government entities.

14. Measured by the authors using the previously noted methodology. Philanthropic sources are defined as receipts from foundations, religious institutions, and other nonpublic charities.

15. Regression discontinuity allows for the evaluation of CDFI interventions without random assignment, preventing the dilemma of assigning potential beneficiaries to a non-treatment group solely to evaluate a program’s effectiveness (Imbens & Lemieux, 2008).

16. Unique articles are the total count of articles within the category. Each article can contain multiple primary findings if it includes multiple types of evaluation questions.