A CRA To Meet the Challenge of Climate Change

A federal law passed more than 40 years ago to address the discriminatory practice of redlining in low-income neighborhoods and communities of color can and should be modernized to address other systemic racism-fueled inequities. These inequities, including the disproportionate exposure to environmental hazards and climate-related challenges, have been exacerbated by the coronavirus pandemic.

The Community Reinvestment Act (CRA) was enacted in 1977 to combat redlining—the practice of systematically denying mortgages and other financial services to communities based on their racial makeup—and other forms of racial discrimination in lending. The CRA should be updated to spur lending, investment, and other services that address climate resilience in low-income communities of color, which are particularly vulnerable to extreme weather and climate-related events.

The coronavirus pandemic has disproportionately hit low-income Black and brown communities in great measure because they are underresourced.¹ African Americans, Latinos, and Native Americans get sick and die from the virus at rates higher than their white counterparts and higher than their shares of the population.² The unequal burden that COVID-19 is placing on communities of color is not coincidental. Because of decades of housing discrimination and residential segregation, people of color disproportionally reside in low-income neighborhoods.³

Moreover, due to decades of systemic environmental racism, these communities are far more vulnerable to the effects of climate change and are disproportionately affected by pollution and environmental degradation and disasters.⁴ Consequently, these communities are particularly vulnerable to the coronavirus, as their excessive exposure to environmental hazards, such as air pollution, contributes to respiratory and heart diseases that make individuals more likely to suffer the worst consequences of the virus.⁵

In the wake of COVID-19’s disparate impact on communities of color, it is time to create a new social compact with the financial system—banks and other institutions—to address the disproportionate way that climate change and environmental racism affect those communities’ health and economic futures. Tackling environmental racism and climate resilience in communities of color is important not just for those directly targeted by racial discrimination but also for society at large, as research shows that racism harms the whole economy in different ways.⁶ If modernized, the CRA can represent a useful tool to effectively address climate resilience and environmental racism in low- and moderate-income (LMI) communities of color.

Fortunately, a lot can be done at the agency level, without the need for congressional action. The Center for American Progress proposes a practical analytical framework that would enable CRA examinations to boost investment in low-income communities of color. In particular, federal examinations conducted to evaluate compliance with the CRA should encourage banks to provide loans, investments, and other services that address climate resilience in a way that focuses more strongly on geographic racial and ethnic disparities.

This report provides the following recommendations:

Explicitly target low-income communities of color to boost climate resilience where it is critically needed. Financial regulators should take environmental factors into account in their CRA examination criteria. The analysis presented in this report provides a concrete way forward on how to fine-tune geographic targets with the inclusion of race and environmental justice criteria in CRA examinations. This should incentivize banks to provide loans and boost investment in communities of color that face critical challenges related to climate change and environmental racism.

Create a climate resilience and environmental justice finance mandate for the CRA. Financial regulators should ensure that the CRA focuses on quality investment in the types of projects that have the strongest potential to advance community resilience in the most climate-vulnerable communities. Projects that count toward an institution’s environmental investment score could take a number of forms, among them the development and construction of energy efficient and climate resilient affordable housing; the installation of energy efficiency improvements in homes and buildings; the creation and expansion of green jobs with family-sustaining wages and in small businesses; the deployment of community solar projects; and the creation of green infrastructure, including parks and green spaces, to address flood and heat risks, to name a few.⁷

Expand CRA coverage. Extending CRA coverage to institutions that are not currently covered by the law is crucial to updating the CRA. In particular, regulators should extend the CRA to include credit unions, which are already required to some extent to serve their member-owners. Banks and nonbanks should also be incentivized to invest outside their traditional assessment areas. This requirement will expand access to credit in banking deserts, including central cities and rural areas.

Strengthen CRA enforcement. Providing more stringent, quantitative benchmarks for institutions has the benefit of establishing clear and uniform standards and mitigating both grade inflation and regulatory uncertainty. In addition, bank performance should be measured by outcomes in assessment areas; by tracking social and economic outcomes; and by assessing the responsiveness of the investments to community environmental needs by, for example, monitoring carbon emission levels.

Increase public accountability. The role of community benefits agreements should be formalized within the CRA to ensure that communities of color have a role in identifying investment needs and to increase accountability for financial firms. Also, CRA regulators should expand and improve the data they report publicly in uniform formats to allow for independent analysis.

Environmental racism consists of “the institutional rules, regulations, policies or government and/or corporate decisions that deliberately target certain communities for locally undesirable land uses and lax enforcement of zoning and environmental laws, resulting in communities being disproportionately exposed to toxic and hazardous waste based upon race.”⁸ For decades, legal forms of discrimination, racially biased housing policies, and racist lending practices have played a critical role in segregating people of color, particularly African Americans, into neighborhoods that face chronic disinvestment and higher levels of lead exposure, poorer air quality, and exposure to toxic chemicals due to their close proximity to landfills, hazardous waste sites, and other industrial facilities.⁹ Race represents the main determinant of the placement of toxic facilities in the United States.¹⁰ Black and brown communities represent the majority of nearly 2 million Americans who reside within a mile of sites that are vulnerable to flooding.¹¹ African Americans and Latinos are more exposed to fine particulate matter (PM2.5) air pollution, the largest environmental health risk factor in the United States. Such exposure is disproportionately caused by consumption of goods and services—such as automobiles and electricity—mainly by non-Hispanic whites but disproportionately inhaled by people of color.¹² The disproportionate presence of people with disabilities among communities of color makes their vulnerability to climate change and natural disasters even more critical.¹³

One of the most well-documented examples of environmental racism is a region along the Mississippi River in Louisiana known as “Cancer Alley.” This area, predominantly inhabited by African Americans, is home to more than 30 chemical plants that emit carcinogenic pollutants into the air. These plants often promise to bring jobs to areas but end up hiring mostly outside the communities that they are negatively affecting. Meanwhile, residents of these communities see alarming rates of cancer and miscarriages among friends and family.¹⁴ Flint, Michigan, represents another notable example of environmental injustice, among numerous others.¹⁵ The city’s failure to adequately treat its municipal water system after changing water sources resulted in the exposure of thousands of residents, the majority of whom where people of color, to elevated lead levels. The water crisis in Flint has had a severe impact on children exposed to lead, with 80 percent having diagnosable learning disabilities.¹⁶

Environmental racism is unquestionably related to climate change because it determines who is most likely to suffer most from the consequences of activities that produce global warming. Low- and moderate-income communities of color find themselves on the front lines of climate change, as their often outdated housing and infrastructure—including a lack of adequate insulation and air conditioning—is more vulnerable to the adverse effects of extreme weather and climate change.¹⁷ These communities are often located in hazardous areas, such as floodplains and fire zones.¹⁸ As the global sea level rises, African American coastal communities in the South are at great risk of displacement.¹⁹

Communities of color have the fewest resources with which to prepare for extreme climate events. The impact of Hurricane Katrina on African American New Orleans residents was greatly exacerbated by residential segregation and the barriers that residents faced in preparedness and evacuation, including limited access to personal vehicles and greater dependence on public transportation.²⁰ In fact, in 2011 and 2012, the majority of communities most harmed by costly natural disasters were LMI families.²¹ Moreover, communities of color do not always receive as many federal relief dollars as do wealthier, predominantly non-Hispanic white ones, which exacerbates income inequality and the racial wealth gap.²²

Numerous studies have documented the disproportionate exposure of people of color to land uses and activities that exacerbate climate change. Historical redlining, the siting of affordable housing, and past uneven disinvestment have greatly shaped the character of urban development and the uneven distribution of ecological benefits, including access to amenities such as greenspace.²³ Federal programs, such as those that provided incentives for major highway construction across low-income neighborhoods of color, have also increased the likelihood of these communities being exposed not only to higher levels of pollution but also to higher levels of heat.²⁴ Extreme heat is considered one of the most serious threats to human health in urban areas across the United States.²⁵ Heat accounted for more deaths than flooding and hurricanes combined from 1990 to 2019.²⁶ Because of climate change, extreme heat events are becoming more common and more intense, and studies of extreme heat point to racial disparities in heat-related mortality.²⁷ Land cover characteristics in racially segregated areas contribute to heat-related health disparities.²⁸ Some studies also connect land use planning and zoning to the urban heat island effect because of the influence that they have on the location, density, mix of buildings and structures, and construction materials of the built environment in cities.²⁹ Further, areas that have experienced systematic disinvestment driven by racial bias through practices such as redlining are more vulnerable to heat because their built environments often feature heat-retaining materials and limited greenspace.³⁰

Addressing climate change and systemic environmental racism is an urgent matter, requiring policymakers to use every tool at their disposal to promote equitable community development and climate resilience—the ability of a community to respond and adapt to climate change.³¹ For decades, activists and community groups—seeing the effects of climate change and pollution firsthand—have fought for environmental justice. However, more needs to be done at a larger scale. One arena in which government can tackle climate change and environmental racism is in financial protection and community investment. Low- and moderate-income communities and communities of color need more resources to build healthier, more resilient economies, housing, and infrastructure. Specifically, investment in these communities should simultaneously advance climate resilience and reverse the effects of environmental racism.

One of the most dependable tools to boost community investment is the 1977 Community Reinvestment Act—a landmark federal law that requires depository institutions to meet the credit needs of the local communities in which they are chartered. The law affirms that banks must serve their entire community and directs the relevant enforcement bodies to “assess the institution’s record of meeting the credit needs of its entire community, including low- and moderate-income neighborhoods” and “take such record into account in its evaluation of an application for a deposit facility by such institution.”³² The Office of the Comptroller of the Currency, the Federal Deposit Insurance Corporation, and the Federal Reserve Board jointly enforce the CRA by conducting regular evaluations of depository institutions. The CRA statute instructs the relevant regulatory agency to take the CRA record of a bank into account when it seeks to establish a new branch or relocate an existing branch. The most persuasive incentive to earn a high CRA rating is that it is considered when banks seek to merge or make an acquisition.³³ CRA examinations evaluate depository institutions’ performance in assessment areas—that is, the geographic areas in which banks have their main office, their branches, and their deposit-taking automated teller machines. A bank’s assessment area also includes the surrounding geographic area where the bank has originated or purchased a large portion of its loans.³⁴ Regulators assign banks one of four grades: outstanding, satisfactory, needs to improve, and substantial noncompliance. Banks are evaluated on different rubrics depending on their size.³⁵ Large banks are scored on three criteria to determine whether they are adequately fulfilling their obligations to their assessment area: lending, investment, and service.³⁶

Since its passage, the CRA has been instrumental in investing in LMI communities. While it’s difficult to measure the impact of the CRA on lending, many studies suggest that the CRA does increase credit access in LMI communities.³⁷ Studies highlight how essential the CRA is to expanding access to credit in underserved communities, including by ensuring the continued presence of brick-and-mortar bank branches in these communities.³⁸ According to the National Community Reinvestment Coalition (NCRC), banks have made almost $2 trillion in small-business and home loans in LMI neighborhoods since 1996.³⁹ In 2018 alone, CRA-covered institutions invested $103 billion in community development loans and $255 billion in CRA-qualifying small-business loans.⁴⁰ A study conducted by the Federal Reserve Bank of Philadelphia showed that when a metropolitan area or county is removed from a CRA-covered area, lending from that bank drops by almost one-fifth in some cases.⁴¹ Another study showed that the census tracts below the moderate-income line covered by the CRA presented more lending activity than similar census tracts just above the line.⁴² Although the CRA generally has a positive effect on credit availability for small firms located in minority neighborhoods compared with small businesses located in other areas, inequalities persist depending on the race and ethnicity of small firms’ owners.⁴³

Climate change and environmental racism pose existential threats to the very communities the CRA was meant to protect, underscoring the need for agencies to revisit the CRA’s evaluation criteria and use the law to promote climate resilience in communities of color. Paradoxically, the law does not mention race, and the racial composition of communities is not taken into consideration in CRA examinations, despite the fact that reversing redlining in communities of color was one of the major motivations for the passage of the law.⁴⁴ Redlining refers to the practice of systematically denying mortgages and other financial services by the federal government, local governments, and the private sector—including banks and other financial firms—to geographic areas demarcated based on their racial and ethnic makeup. The practice, which started in the 1930s, resulted in the consistent deflection of investment money away from central cities where people of color were concentrated. The Home Owners’ Loan Corporation (HOLC), a government-sponsored corporation established in 1933 as part of the New Deal, institutionalized redlining to evaluate the quality of neighborhoods. Neighborhoods with large populations of African Americans and other people of color typically received the lowest ratings and were deemed too risky to secure government-backed mortgages. Subsequently, the Federal Housing Administration and the Veterans Administration loan programs adopted the HOLC rating system when determining where to approve mortgages.⁴⁵ Research has shown that the 1930s-era HOLC ratings have continued to have lasting and significant effects on urban neighborhoods’ levels of disinvestment, access to credit, and racial segregation patterns.⁴⁶ According to NCRC research, the neighborhoods classified as risky by the HOLC have remained predominantly minority, lower income and more vulnerable to COVID-19.⁴⁷

The criteria currently adopted for CRA examinations are largely based on the income levels of communities that depository institutions serve.⁴⁸ Lending, services, and investments only count toward a positive CRA assessment if they are made to low- and moderate-income people or places; distressed or underserved middle-income nonmetropolitan census tracts; or designated disaster areas.⁴⁹ Income levels are defined in the following ways:

• Low income: 0 percent to 49 percent of the median area income, defined as the median income of the metropolitan area or nonmetropolitan part of the state
• Moderate income: 50 percent to 79 percent of the median area income
• Middle income: 80 percent to 119 percent of the median area income
• High income: 120 percent or more of the median area income⁵⁰

Distressed or underserved middle-income nonmetropolitan census tracts are defined based on the following characteristics:

• Rates of poverty, unemployment, and population loss
• Population size, density, and dispersion—that is, distance from a population center.⁵¹

The NCRC has proposed to add serving underserved census tracts as a criterion on component tests in CRA examinations.⁵² Underserved census tracts are identified based on levels of retail lending per housing unit and small business.⁵³ NCRC research notes that a significant portion of underserved census tracts are characterized by a large presence of “minority” populations.⁵⁴

The inclusion of race on CRA exams is important from a climate change and environmental justice perspective, as low-income and people of color are less likely than the rest of the U.S. population to be prepared for extreme weather and climate-related events.⁵⁵ No indicators related to environmental justice and climate change are currently included in the criteria adopted for CRA examinations. The criteria used for CRA examinations must be reconsidered to encourage banks to provide loans, investments, and other services that address climate resilience with a stronger focus on geographic racial and ethnic disparities.

The following analysis illustrates an example of how race and environmental and climate factors could be incorporated into a set of revised criteria for Community Reinvestment Act examinations. The framework of the analysis is largely based on University of Virginia professor Bev Wilson’s study of urban heat management and the legacy of redlining.⁵⁶ Using residential security maps created in the 1930s by the Home Owners’ Loan Corporation and overlaying them on estimates of land surface temperature (LST) derived from Landsat 8 satellite imagery, Wilson’s study focuses on the cities of Baltimore, Maryland; Dallas, Texas; and Kansas City, Missouri, to illustrate the relationship between historic redlining and uneven exposure to heat in these cities.

The purpose of CAP’s analysis is to identify readily available environmental indicators that are highly correlated with heat exposure—a climate change factor—and with current disinvestment patterns. Such indicators can then be combined with income levels and racial characteristics at the census tract level in order to fine-tune geographic targets for CRA examination purposes. CAP’s analysis focuses on the Baltimore metropolitan statistical area (MSA) and, like Wilson’s study, performs a thermal analysis with satellite imagery collected by the U.S. Geological Survey from the Landsat sensor. The imagery is processed to derive estimates of LST and vegetative cover throughout the Baltimore metropolitan area during a summer 2020 day with limited cloud coverage. Figure 1 illustrates average LSTs across the metropolitan area’s census tracts.

Figure 1 shows that the central part of the metropolitan area contains census tracts featuring the highest temperatures. These are located predominantly in the city of Baltimore, contiguous census tracts, and those southwest of the city.

To compare the distribution of LSTs with current disinvestment, CAP’s analysis employs 2018 CRA and 2018 Home Mortgage Disclosure Act (HMDA) data on originated small-business, small-farm, single-family, and multifamily loans at the census tract level. (see Figure 2)⁵⁷

A visual inspection of Figures 1 and 2 suggests that census tracts with the smallest shares of loans tend to also feature the greatest exposure to land surface heat.

Adopting LST as a climate-related indicator for CRA purposes would be impractical, given the amount of processing time and resources that would be required to estimate exposure to heat throughout the entire U.S. territory. It is possible, however, to identify readily available environmental indicators that are highly correlated with LST. The U.S. Environmental Protection Agency (EPA), for instance, provides a set of environmental indicators at the census block group level⁵⁸ that can be compared with the LST distribution across a geographic area. These indicators can be accessed through the Environmental Justice Screening and Mapping Tool (EJScreen) that was established to combine consistent environmental and demographic data to address environmental justice issues.⁵⁹ The environmental indicators provided by EJScreen at the block group level include the following:

• National-scale Air Toxics Assessment (NATA) air toxics cancer risk, measuring lifetime cancer risk from inhalation of air toxics
• NATA respiratory hazard index, measuring the ratio of exposure concentration to health-based reference concentration
• NATA diesel particulate matter level in air
• Particulate matter levels in air
• Ozone summer seasonal average of daily maximum eight-hour concentration in air, in parts per billion
• Traffic proximity and volume
• Potential lead paint exposure, measured by the percentage of housing units built before 1960
• Proximity to Risk Management Plan sites—that is, facilities presenting the risk of chemical accidents
• Proximity to hazardous waste facilities
• Proximity to National Priorities List sites, also known as Superfund sites
• Wastewater discharge indicator based on stream proximity and toxic concentration

EJScreen provides the absolute values of each indicator as well as percentiles,⁶⁰ which are useful to assess how local residents compare with everyone else in the United States, thus providing context for the numbers. The analysis presented in this report performs calculations based on percentiles.

The first table in the downloadable spreadsheet illustrates the correlations among the estimated land surface temperature by block group and the environmental indicators listed above in order to identify the indicators that present the strongest relationship with LST and can thus be utilized as proxies of climate change vulnerability.

As the table shows, most environmental indicators present a statistically significant correlation with each other and with land surface temperature. The correlation with LST is particularly strong for five environmental indicators, highlighted in bold font in the first column, which present very high correlation coefficients (above 0.70).⁶¹

Figures 3 through 7 below illustrate the geographic distribution of the five indicators by census tract. As the maps suggest, there is in general a clear overlap in the distribution of the five indicators, especially in the central part of the metropolitan area. Most importantly, when these distributions are visually compared with the distribution of LST and lending throughout the Baltimore MSA, one can see that the census tracts presenting the worst exposure to poor air quality and the highest proximity to hazardous facilities also tend to present high land surface temperatures and low levels of lending.

The second table in the downloadable spreadsheet indicates that LST, exposure to poor-quality air, and proximity to hazardous sites present statistically significant correlations with the percentage of minority population, the income levels, and the lending volume in census tracts. Specifically, the higher the values of LST and environmental indicators, the larger the percentage of minority population, the lower the income level, and the lower the amount of all lending, particularly small-business and single-family home mortgage lending.

While CRA exams do not include any criteria based on race, the underserved areas designated to guide the government-sponsored enterprises (GSEs) in meeting their location-based affordable housing goals are delineated based on both income and minority status. Specifically, underserved areas consist of the following: 1) census tracts or block numbering areas in which the median income does not exceed 80 percent of the area median income (AMI); and 2) census tracts with a minority population of at least 30 percent and a median income of less than 100 percent of the AMI.⁶²

Adopting these underserved areas for CRA exam purposes would help promote lending in both low-income and minority areas. It would also have the following advantages:

• Underserved areas are revised each year, and revised lists of census tracts are readily available through the Federal Housing Finance Agency (FHFA) portal.
• The standardization of criteria for CRA exams and for scoring mortgage purchases has the potential to streamline efforts to promote investment and lending in low-income and minority communities by both individual lenders and the GSEs.

Figure 8 illustrates the distribution of underserved census tracts in the Baltimore MSA.

Examining environmental and climate characteristics of underserved areas can further help identify the census tracts that are most vulnerable from a climate change perspective. (see Table 1) As Table 1 shows, underserved areas received just 26 percent of all home mortgage, small-business, and small-farm loans in 2018. In these census tracts, the average percentage of minority population was 68 percent, and the AMI was about half of the median income in all other census tracts. Most importantly, underserved areas present much larger average values of environmental and climate indicators than all other census tracts in the metropolitan area.

Table 1

Given the strong correlation between LST and the other two environmental indicators, targeting underserved census tracts that are highly exposed to poor-quality air and in close proximity to environmentally hazardous facilities would be a practical way to highlight areas that are also vulnerable from a climate change perspective.

For example, Figure 9 shows the distribution of hypothetical target areas identified through the combination of income, minority status, and environmental indicators. In particular, the targeted census tracts present simultaneously high values of all five environmental criteria.⁶³ Note that the difference in the average land surface temperature between the target areas and all other census tracts is nearly 7 degrees Fahrenheit.

Several issues would need to be addressed when adopting this approach:

• The U.S. Environmental Protection Agency provides a number of environmental indicators and environmental justice indexes. The indicators are generally correlated with each other and with LST. To fine-tune proposed geographic targets, it would be reasonable to explore other environmental indicators besides those chosen for this analysis. It is also important to take into consideration the measurements in which such indicators are provided. This analysis has chosen percentiles. The data are provided by block group, so it is necessary to aggregate data by census tract in order to combine environmental indicators with underserved census tract data coming from the FHFA.
• This analysis focused on one metropolitan area. It is important to replicate the analysis for other metropolitan areas to gauge whether this approach yields consistent results. Moreover, this methodology should be tested in rural areas and other communities with more dispersed populations. The analysis employed satellite imagery for just one day. A more robust analysis would need to employ thermal imagery for multiple days.
• While this approach focuses on metropolitan areas, environmental indicators should also be included in the delineation of distressed or underserved middle-income nonmetropolitan census tracts.

Expanding the CRA to directly target communities of color would address the governmental and market failures associated with systemic environmental racism that continue to affect low- and moderate-income communities and communities of color and their climate resilience capacity.⁶⁴

While geographic targeting is very important in order to boost investment in communities of color that face critical challenges related to climate change and environmental racism, it isn’t the only key. Just as crucial is incentivizing the types of investment activities that would help reverse decades of environmental racism and promote economic stability, hazard mitigation, and climate resilience.

There is a growing acknowledgment of the role that financial institutions have played in financing greenhouse gas-emitting industries—including oil and petrochemical companies, coal mining, natural gas extraction, and power plants that burn fossil fuels, among others—which contribute to climate change.⁶⁵ Some banks, however, are actually beginning to take action themselves to green their portfolios by moving their investments away from fossil fuels. The Partnership for Carbon Accounting Financials is an organization of more than 66 financial institutions with combined assets of $5.3 trillion. This group is dedicated to increasing the disclosure of carbon investments at its institutions.⁶⁶

While banks are beginning to evaluate the environmental impacts of their practices, the demand for green investment vastly outstrips what is currently supplied by the financial industry. Massive public and private investments in clean and pollution-free technologies and climate-ready and sustainable housing and infrastructure must be made soon to adapt to and mitigate the impacts of climate change.⁶⁷ It is estimated that meeting these challenges would require between $166 billion and $322 billion in additional lending globally each year.⁶⁸

Banks are currently judged on their community development investments in their areas. For investments to count under the Community Reinvestment Act, they must create or retain jobs, promote workforce education or training, or promote the development of affordable housing and transportation.⁶⁹ Many environmentally friendly investments already count as community development under the current regulations. Financial institutions, however, are not regulatorily assessed on investments that are specifically made in healthy, pollution-free, and climate-resilient housing and infrastructure projects. In addition to traditional metrics, regulators could evaluate how sustainable the investments of financial institutions are and take that into account as part of their grade.

Projects counting toward an institution’s environmental investment score could take a number of forms. For instance, the financing of net-zero buildings—structures with energy consumption that is equal to their energy input—could contribute to cutting down on carbon emissions, water consumption, and the amount of solid waste that is transported to landfills.⁷⁰ Financial institutions’ investments could target the improvement of infrastructure in coastal cities in order to increase weather resiliency and cut down on insurance costs. Moreover, financing the development of green spaces in urban communities could contribute to reducing the impacts of urban heat islands, which are serious concerns in low-income communities of color.⁷¹

CRA coverage and enforcement need improvement

The CRA’s potential for enhancing lending and investment in low- and moderate-income communities and communities of color has been weakened by important changes in the financial landscape over the past few decades.

CRA coverage

Since the passage of the CRA, both the banking industry and the credit needs of the United States have shifted radically.⁷² Repeated bank mergers have consolidated much of the market, leading to the decline of many independent community banks. Mortgage lending has shifted from banks to affiliates or independent mortgage companies not covered by the CRA.⁷³ Fintechs—technology firms operating as an alternative financial channel to traditional banks and providing services predominantly online—have been playing an increasing role in shaping financial and banking landscapes.⁷⁴ Fintech lenders are not subject to the same rigorous oversight as depository institutions. Online banking and financial technology companies have loosened the relationship between geography and credit access.

In recent years, some banking activities and their inherent risks have migrated from banks to nonbanks. While banks have increased their share of outstanding loans since the financial crisis, a significant portion of residential mortgage lending and leveraged lending has migrated out of banks. The shift in mortgage lending from CRA-covered financial institutions to mortgage lending affiliates has greatly undercut the strength of the CRA. By 1997, more than 40 percent of mortgage loans made by CRA-covered entities were made by affiliates and more than half of all mortgage loans were disbursed by mortgage companies.⁷⁵ These affiliates may choose to have their activities evaluated under the CRA, but it is not required. Independent mortgage companies are exempt from the CRA altogether. CRA exemption leaves borrowers of color exposed to discriminatory practices such as steering and redlining as well as predatory lending.⁷⁶

Also, a growing share of mortgage lending occurs outside depository institutions’ assessment areas. This is problematic because the CRA is applicable only to mortgage lending and other services provided within a CRA-regulated institution’s assessment area. Pointing to technological change and the growing shift to online banking, policymakers and advocates contend that the geographic definitions of assessment areas are too narrow.⁷⁷ Moreover, some experts say that the original geographic definition excludes many areas in need of investments, usually referred to as “CRA deserts.”⁷⁸ The narrow definition of service areas may discourage investment in rural areas, which by current metrics often lie outside the defined service areas of depository institutions. Fine-tuning the geographic guidelines would incentivize banks to turn their investments to these areas, rather than to “CRA hotspots” in already saturated markets.⁷⁹ The steady closure of bank branches in LMI communities since 1989 has reduced the effectiveness of the CRA in underserved communities.⁸⁰ Because banks are evaluated by their practices near branch locations, many LMI communities are not covered by the CRA.

CRA enforcement and public accountability

The effectiveness of any law hinges on the strength of its enforcement. Unfortunately, the strength of the CRA is eroded by grade inflation. In 2020, only two banks were given a grade of “substantial noncompliance.”⁸¹ In part because of the vague nature of CRA examinations, there is little clarity about how much CRA grades reflect the actual performance of banks, although many fair lending advocates blame lax enforcement more than the text of the regulations itself.⁸² At any rate, research shows evidence of significant grade inflation.⁸³ However, the persistent credit gaps in communities of color and LMI communities suggest that there is some room for improvement, despite the fact that only a handful of institutions fail to receive a satisfactory evaluation. In 48 U.S. cities, Black consumers are significantly more likely to be denied a loan than their white counterparts. In 25 cities, the same is true for Latino consumers.⁸⁴

Community development investment works best when it is done with input from the front-line stakeholders directly affected. Studies have shown that community engagement through grassroots organizing and community benefits agreements (CBAs) results in higher CRA performance.⁸⁵ Some communities have employed local grassroots coalitions to engage with lenders and negotiate community development strategies, yielding better outcomes for organized communities. CRA stakeholders often make CBAs with banks that plan to merge, requiring financial institutions to spell out how they plan to maintain their commitment to their communities after the merger is consummated.⁸⁶ These forms of bargaining parallel the wage board model to some extent, representing negotiations between private and public entities with the interest of the public in mind.

Finally, regulators must publish their results in uniform formats to enable researchers and stakeholders to conduct meaningful analysis. CRA evaluations, however, are not published in a standard format. In some reviews, a number may be reported in a table, while in another it may be in a graph or even presented as an in-text statistic.⁸⁷ Moreover, reported CRA data lack any spatial dimension and are largely provided at the aggregated national level, making it difficult to fully evaluate the extent to which an institution is truly serving all of its geographies.⁸⁸

It is imperative that the Community Reinvestment Act be modernized with an enhanced focus on climate resilience and environmental justice. To achieve the needed updates, policymakers and financial regulators should consider the following recommendations.

Explicitly target low-income communities of color to boost climate resilience where it is critically needed

Climate-conscious financing and investment should flow to communities in which the need is most urgent. The criteria used for CRA examinations should be reconsidered to strongly direct banks to invest and provide other services that address environmental racism and climate resilience in targeted low-income communities of color. The analysis presented in this report provides an example of how to fine-tune geographic targets with the inclusion of race and environmental justice criteria.

CRA exams should be largely based on financial institutions’ activities in areas that coincide with the underserved areas based on which government-sponsored enterprises’ mortgage purchases are scored. The adoption of underserved areas for CRA purposes would address the need for CRA exams to include racial minority status as a key criterion for evaluations.

Targeting the census tracts that are currently considered underserved for GSEs’ evaluation purposes and simultaneously feature poor air quality and high exposure to environmentally hazardous facilities would be a practical way to highlight areas that are also vulnerable from a climate change perspective. In addition, streamlining geographic targets across the regulatory system has the potential to boost lending and investment in underserved areas that are environmentally vulnerable.

Create a climate resilience and environmental justice finance mandate for the CRA

Besides specifically targeting low-income communities of color that are vulnerable to climate change and environmental racism, the CRA should focus on quality investment in the types of projects that have the strongest potential to advance community resilience in the most climate-vulnerable communities. Projects counting toward an institution’s environmental investment score could take a number of forms, among them the development and construction of energy efficient and climate resilient affordable housing; energy efficiency improvements for homes and buildings; the creation and expansion of green jobs with family-sustaining wages and in small businesses; community solar projects, green infrastructure, such as parks and green spaces, to address flood and heat risks; public transit; electric buses; bike shares; job training programs; climate resilient and energy efficient community centers, hospitals, and day care centers; and toxic waste and industrial site cleanup, among others.⁸⁹ All banks should have to meet a threshold of climate financing activities based on asset size and other metrics, including financed emissions for large financial institutions that are currently contributing to high levels of emissions. If banks fail to meet the financing threshold, penalties would include merger restrictions, possible asset caps, and restrictions on capital distributions, among others.

Possible mechanisms to facilitate equitable and climate-conscious investment are climate-focused New Markets Tax Credits, climate-focused investments supported by community development financial institutions, and the purchase of high-standard green municipal bonds. Municipal bonds would have to be certified as serving low- and moderate-income communities and communities of color, meet green guidelines modeled after either the Green Bond Principles established by the International Capital Market Association or the Climate Bonds Standard established by the Climate Bonds Initiative⁹⁰ and meet the Principles of Environmental Justice.⁹¹ Financial institutions may also receive green credit for loans to small businesses and farms that have an environmental justice or green mission or that aim to make investments that will reduce their pollution and carbon footprint. Examples include lending to a farm that needs capital to be certified organic to provide access to fresh organic food in underserved communities or to transition toward a sustainable farming or carbon-farming model, or providing a business loan to a construction company looking to purchase the equipment necessary to build more energy efficient homes.

Expand CRA coverage

Extending CRA coverage to institutions that are not currently covered by the law is crucial to updating the CRA. Banks and nonbanks should also be incentivized to invest outside their traditional assessment areas.

Nonbanks lend far more to LMI people than banks covered by the CRA. The Community Reinvestment Modernization Act of 2009 would have, among other things, expanded CRA coverage to independent mortgage companies and credit unions.⁹² If it had been passed, it would have expanded assessment areas to include areas where a bank does a significant amount of business, regardless of whether those areas contain physical branches. This provision is essential to extending more equitable credit to communities that may be underbanked or unbanked, which are likely in most need of investment. It also provides a possible model for assessing online banking services that do not have any physical branches. Credit unions, given their cooperative structure that requires them to serve their members, would receive a streamlined evaluation.

Fintech institutions may represent a challenge in terms of how they are evaluated since their service areas do not conform to traditional assessment areas, which are largely delineated based on the presence of brick-and-mortar bank branches. The National Community Reinvestment Coalition has indicated that current CRA regulations for fintechs result in narrow assessment areas that do not reflect the geographic areas where fintechs conduct their business.⁹³ According to the NCRC, fintechs’ assessment areas should be delineated based on where they perform most of their lending or take the majority of their deposits. Consistent with the NCRC’s recommendations, assessment areas should also include rural areas, where internet availability is limited. Further, fintech lending and investment should be evaluated based on how much of these activities takes place in the proposed targeted geographic areas within their assessment areas.

Strengthen enforcement

Regulators must address grade inflation in order for any CRA reforms to have a meaningful impact. There are several avenues by which regulators may accomplish this. One method is to provide more stringent, quantitative benchmarks for institutions.⁹⁴ This approach would have the benefit of establishing clear and uniform standards and mitigating both grade inflation and regulatory uncertainty. On the other hand, the analysis involved is complex, and quantitative metrics alone may inadvertently incentivize larger development projects. They also do not fully capture the spatial dimension of investment.

Another approach would be to measure banks’ performance by outcomes in their assessment areas by tracking social and economic outcomes, such as how many jobs or affordable housing units are created, or by assessing the responsiveness of the investments to community environmental needs—for example, by monitoring carbon emission levels. If investments are harmful for the environment, they should not count toward CRA credit. At the same time, if green community investments do not benefit LMI communities and communities of color—for instance, if they displace them—they should not count toward CRA credit. In contrast, green community investments that target and primarily benefit LMI communities and communities of color should be weighed more favorably and receive extra CRA credit compared with activities that do not simultaneously address green finance and target LMI communities and communities of color.

Increase public accountability

The role of community benefits agreements should be formalized within the CRA. Community benefit plans must be required for all mergers that indicate via performance measures how mergers will increase lending, investment, and services in underserved communities. Banks that receive grades below “satisfactory” should also be required to draft community benefit plans or CBAs with communities in which they have failed to meet their obligations in order to establish a plan for improving their performance.

In order to inform community engagement and third-party evaluation and research, CRA regulators must expand and improve the data that they report publicly and do so in uniform formats. In addition, providing census tract information and adding a common institutional identifier code between Home Mortgage Disclosure Act data and Federal Financial Institutions Examination Council CRA reports would allow researchers to merge those datasets, opening the door to a whole new world of analysis.

Like nearly every policy designed for the 21st century, confronting climate change and environmental racism is essential for achieving racial justice and equity. If climate resilience planning and policy do not pay attention to issues of racial equity and racial justice, they risk benefiting only a segment of society while perpetuating racial segregation and health inequality, in addition to widening the wealth gap. Recognizing these challenges, particularly in the wake of COVID-19, with its disparate impacts on communities of color, it is more than clear that the time is now to create a new social compact with the financial system to address the ways climate change and environmental racism disproportionately affect communities’ health and economic futures. If modernized, the Community Reinvestment Act can be a useful tool to effectively address climate resilience and environmental racism in low- and moderate-income communities of color.

Michela Zonta is a senior policy analyst for Housing and Consumer Finance Policy at the Center for American Progress.

Caius Z. Willingham is a research associate for Economic Policy at the Center.

The authors would like to thank Rebecca Cokley, Andy Green, Linda Jun, Cathleen Kelly, Mara Rudman, Josh Silver, and Andres Vinelli for their review; Malkie Wall and Galen Hendricks for their fact-checking assistance; and CAP’s Editorial and Art teams for their guidance.

Willingham previously published under the name Zoe Willingham.

The original impetus for the Community Reinvestment Act (Title VIII of the Housing and Community Development Act of 1977) was to combat redlining and other forms of racial discrimination in lending. It was critical that government policy find a way to address redlining, in part because New Deal policy helped contribute to it. Though the Home Owners’ Loan Corporation helped millions of families obtain safe and sound 30-year fixed-rate mortgages, its rating system unjustly labeled African American neighborhoods as more risky investments. As a result of these biased risk assessments, and the refusal of the Federal Housing Administration to insure loans in these areas, banks were reluctant to lend in these neighborhoods, starving them of capital and shutting out borrowers of color.⁹⁵

Because of the structural barriers Black Americans and other people of color encountered in the housing market as a result of public and private discrimination, they were forced to live in overcrowded, expensive rental housing with substandard living conditions and few local job opportunities. A commission convened by President Lyndon B. Johnson to study the root of the racial unrest following the assassination of Martin Luther King Jr. pointed to the lack of investment in Black neighborhoods as a key source of discontent. As stated in the 1968 Kerner report: “The private sector must be brought into the production and financing of low and moderate rental housing to supply the capabilities and capital necessary to meet the housing needs of the nation.”⁹⁶ However, even after the Fair Housing Act prohibited discrimination in home financing and real estate, communities of color continued to be highly segregated.⁹⁷

The Fair Housing Act was just the start of a series of laws passed in the coming decade to address credit access for underserved communities. Six years later, Congress passed the Equal Credit Opportunity Act, which banned discrimination in issuing credit—beyond the residential lending covered by the Fair Housing Act. At the urging of civil rights advocates, Congress also passed the Home Mortgage Disclosure Act in 1975, which required lenders to report the race and gender of homebuyers and the location of purchased properties—data that are essential to the enforcement of fair lending laws and the evaluation of lending practices. However, while each of these laws prohibited discrimination against certain groups seeking credit, none of them provided an affirmative incentive for banks to expand their service to low- and moderate-income communities.⁹⁸

The early revelations of the data collected through the HMDA showed clearly that the Fair Housing Act and Equal Credit Opportunity Act were not enough to address barriers to credit. Banks were reluctant to lend in redlined neighborhoods because they were not familiar with the housing stock and creditworthiness of their residents. It was a vicious cycle. Decades of discrimination had increased the cost of lending in these neighborhoods.⁹⁹ The CRA was needed to induce banks to serve disadvantaged neighborhoods and neighborhoods that faced discrimination. Sen. William Proxmire (D-WI), the Community Reinvestment Act’s champion in Congress, held that the public charters of banks endowed them with the responsibility to serve the public interest and by extension, these communities.¹⁰⁰ Moreover, he argued that the CRA was necessary because federal investment alone was not enough to address the economic woes of impoverished communities and neighborhoods without help from depository institutions.

Banking regulators and banks themselves pushed back against Sen. Proxmire’s proposal as duplicative of existing banking regulations and restrictive of regulators’ discretion to uphold their other duties. Detractors also argued that the regulation would, by requiring an increase in supply to underserved communities, inadvertently reduce the flow to other areas.¹⁰¹ Ultimately, they argued that the market knows how best to distribute credit.

Despite these criticisms, the case for the proposed CRA was strengthened by the success of similar policies enacted by the states of Massachusetts and Connecticut. After Massachusetts required that banks serve all areas in which they do business before they are permitted to open new branches, banks began to take steps to expand services at all branches. The Connecticut Banking Commission, finding significant disparities in bank operating hours between neighborhoods, placed a moratorium on branch applications, to which banks immediately responded by announcing extended hours of operation at many branches.¹⁰² These successful policies, combined with the indisputable evidence from the first set of HMDA data, helped drive the CRA through Congress and eventually to President Jimmy Carter’s desk in 1977. The CRA requires depository institutions to be evaluated regularly based on their contributions to the credit needs of the entire communities they serve. Their records are taken into consideration when assessing their applications for deposit facilities, including mergers and acquisitions.

The CRA remained largely unchanged until the Clinton administration. The 1994 Riegle-Neal Interstate Banking and Branching Efficiency Act allowing bank holding companies to purchase out-of-state branches retained the responsibility of banks to provide credit to communities served by all of their branches, ensuring continued compliance with CRA requirements.¹⁰³ In 1993, President Bill Clinton directed the regulatory agencies to formulate new CRA rules to improve enforcement and provide clarity on the requirements imposed on banks. The statute itself provided no guidelines for how regulators were supposed to evaluate banks, resulting in highly opaque ratings. After numerous public hearings and much controversy from all sides of the issue, regulators finalized the rules in 1995, establishing guidelines for how to score banks’ performance.¹⁰⁴ These 25-year-old regulations shaped the CRA into the law it is today.

The Community Reinvestment Modernization Act of 2009 represents the most significant and comprehensive attempt to strengthen the law in recent years.¹⁰⁵ The legislation would have extended CRA coverage to nonbank entities such as securities and insurance firms; required that predatory or deceptive practices resulted in grade penalties; and repealed certain regulations issued by the Office of the Comptroller of the Currency (OCC), the Federal Reserve, and the Federal Deposit Insurance Corporation (FDIC). Another proposal, the American Community Investment Reform Act of 2010, would have, among other things, authorized the Consumer Financial Protection Bureau (CFPB) and the U.S. Securities and Exchange Commission (SEC) to conduct evaluations of applicable institutions, clarify qualifying community development and related activities, include affiliate performance as part of an institution’s score, and limit “outstanding” scores to special circumstances.¹⁰⁶

More recently, in 2019, Sen. Elizabeth Warren (D-MA) introduced S. 787, which would amend the CRA to combat ratings inflation by increasing the number of ratings from four to five, adding a rating called “sufficient,” and require an improvement plan for any institution that receives a rating of “sufficient” or lower as well as a public hearing on banks failing to meet expectations.¹⁰⁷ S. 3213, introduced in January 2020 by Sen. Warren and Rep. Maxine Waters (D-CA), would amend the Bank Holding Company Act of 1956 so that mergers and acquisitions would be denied to banks that received a rating below “outstanding” in two of their last three exams and would require banks deemed to be in “substantial noncompliance” to draft a community benefits plan.¹⁰⁸ These proposals would both significantly strengthen the CRA by bolstering the incentives for banks to strive for outstanding grades.

On the other hand, a number of lawmakers have attempted to weaken the CRA throughout the years. Take, for example, efforts to raise the threshold for what is considered a large bank under the CRA, reducing the number of banks that qualify for the more comprehensive performance evaluation.¹⁰⁹ In 2013, Libertarian Rep. Justin Amash (MI) boldly introduced H.R. 3550, which would have completely repealed the CRA had it passed.¹¹⁰

The CRA came under significant attack recently. Joseph Otting, the former comptroller of currency under the Trump administration, made reviewing the CRA his top priority at the OCC. Before he stepped down from his position in 2020, the OCC finalized rules that significantly expanded what activities qualify for CRA credit.¹¹¹ These rules reduce the amount of information that must be publicly reported. The dominant measure in the new OCC exams is a ratio of the dollar amount of CRA activities divided by deposits that would provide an incentive for banks to seek out large deals and neglect smaller-dollar mortgage and small-business lending needed in lower-income communities. The rule also promulgates unjustified expansion of activities that count, including many activities that do not directly benefit low- and moderate-income people and communities.¹¹² The most problematic change is revising the presumptive CRA rating thresholds to dramatically increase CRA ratings inflation.¹¹³

The OCC proposed the rule without conducting even cursory data analysis—forging ahead of the Federal Reserve as it formulated its rule and making CRA-enforcing agencies even more out of step with each other. The rule weakens CRA grading to a significant degree; a bank can score a passing grade from the OCC even if it fails in a significant number of assessment areas.¹¹⁴ Dismayed by the deep cuts to consumer protection, lawmakers in both chambers of Congress introduced a resolution to disapprove of the rule but failed to pass it.¹¹⁵ The FDIC, which approved the initial proposed rules, did not sign onto the finalized OCC rule.¹¹⁶