Estimating Credit Union Exposure to Climate-Related Physical Risks

Estimating Credit Union Exposure to Climate-Related Physical Risks

Executive Summary

Weather and climate-related disasters pose significant risks to providers of financial services, including credit unions. In an effort to improve our understanding of these risks and the extent to which credit unions are exposed, staff in the Office of the Chief Economist (OCE) paired credit union call report data with FEMA’s National Risk Index (NRI), which measures a community’s vulnerability to the negative effects of 18 natural hazards.1 OCE’s analysis suggests that roughly one-quarter of federally insured credit unions, accounting for one-third of system-wide assets, are located in communities classified as having relatively high or very high risk of experiencing negative effects due to natural hazards. Minority depository institutions (MDIs) face a substantially higher risk than the credit union system in the aggregate. Just over half of MDIs are at relatively high or very high risk of experiencing negative effects from natural hazards. These findings are consistent with the literature on vulnerable communities that shows natural disasters have disproportionate negative effects on their health, well-being, economic security, and mobility. OCE’s estimates indicate that credit unions most at risk of negative outcomes due to natural hazards tend to be located in coastal areas, particularly in California, Texas, and Florida.

Introduction

Each year, natural disasters like hurricanes, wildfires, droughts, and floods impose a substantial financial toll on households and businesses alike. In 2021, the United States experienced 20 separate billion-dollar weather and climate disaster events, which caused an estimated $153 billion in damage. The frequency and cost of these events is on the rise. As the red line on the chart below shows, 2021 was the third most costly year on record for these types of events and it was the seventh consecutive year in which 10 or more billion-dollar weather and climate disaster events have occurred in the United States.2 In 2022, there were an estimated 15 billion-dollar disaster events making it the eighth straight year with 10 or more billion-dollar disaster events.

Natural disasters and climate change more broadly present several complex conceptual and practical challenges for individuals and for individual credit unions. A natural disaster can cause assets to reprice, change the cost or availability of credit, and affect the timing or reliability of cash flows. For example, a natural disaster could drive down the value of affected real estate, impacting real estate loans and mortgage-backed securities, as well as the financial health of the households and/or businesses that occupy the affected property. This chain of events would, ultimately, influence the performance of the institution that holds the loan and bears credit risk.

The risks posed to credit unions are not hypothetical. Natural disasters have caused considerable losses for credit unions and resulted in liquidations. For example, Hurricane Katrina led to the closure of two credit unions located in areas that took a direct hit from the storm in 2005: Orleans Public Schools Federal Credit Union and Chalmette Refinery Federal Credit Union. In both cases, employees and members were displaced across the country. Communications challenges following the hurricane coupled with displacement of members and employees led Chalmette to close its doors in early 2006. Without employees and members, Orleans Public Schools Federal Credit Union could not survive and was liquidated in 2007.

The figure shows the number and inflation-adjusted cost of disaster events in the united states annually between 1980 and 2022.

Analysis and Results

OCE matched 2021Q4 credit union call report data with FEMA’s National Risk Index (NRI) to estimate the credit union system’s potential exposure to climate-related physical risks. The NRI assigns one of five risk ratings to each community, ranging from very low risk to very high risk on a five-level ordinal scale. Our analysis makes use of this ordinal scale. For simplicity, a credit union’s headquarters, branches and ATM locations are identified at the county level by its physical address reported on the call report. The FEMA risk indicators are matched to NCUA’s 2021Q4 call report data at the county level using FIPS codes.

Credit Union System Exposure to Negative Effects of Natural Hazards

In OCE’s preliminary analysis, 25 percent of credits unions, accounting for 34 percent of system-wide assets, are headquartered in areas at relatively high or very high risk of experiencing negative effects due to natural hazards.3 These credit unions account for 32 percent of all credit union loans outstanding and 34 percent of credit union deposits4. (Note that there are other ways to gauge credit union system exposure to risk, including using measures of supervisory risk. OCE’s analysis focuses on a few select measures, including assets, loans and deposits.)

The figure is a bar chart showing the percentage of credit unions, credit union members, credit union assets, credit union loans, and credit union deposits in various National Risk Index Categories. The categories are very low, relatively low, relatively moderate, relatively high, and very high. The proportion of credit unions in the very low, relatively low, relatively moderate, relatively high, and very high categories are 13%, 30%, 32%, 17%, and 8% respectively. The proportion of credit union members in the very low, relatively low, relatively moderate, relatively high, and very high categories are 20%, 24%, 25%, 24%, and 7% respectively. The proportion of credit union assets in the very low, relatively low, relatively moderate, relatively high, and very high categories are 18%, 24%, 25%, 25%, and 9% respectively. The proportion of credit union loans in the very low, relatively low, relatively moderate, relatively high, and very high categories are 19%, 24%, 25%, 24%, and 8% respectively. The proportion of credit union deposits in the very low, relatively low, relatively moderate, relatively high, and very high categories are 17%, 24%, 25%, 25%, and 9% respectively.

We also looked at the exposure of MDIs and low-income designated credit unions to natural disasters. Our preliminary estimates suggest that MDIs are especially vulnerable. We found that just over half of MDIs are at relatively high or very high risk of experiencing negative effects stemming from climate, weather-related, and other natural disasters. Nearly two-thirds of MDI assets are concentrated in these high-risk communities. These findings are consistent with the literature on vulnerable communities, that shows natural disasters have disproportionate negative effects on their health, well-being, economic security, and mobility.5

The bar chart shows the percentage of all credit unions, low-income designated credit unions, and minority depository institutions distributed across communities by level or risk, as measured by FEMA’s National Risk Index. The categories are very low, relatively low, relatively moderate, relatively high, and very high. The proportion of all credit unions in the very low, relatively low, relatively moderate, relatively high, and very high risk categories are 13%, 30%, 32%, 17%, and 8%, respectively. The proportion of low-income designated credit unions in the very low, relatively low, relatively moderate, relatively high, and very high risk categories are 12%, 30%, 32%, 19%, and 8%, respectively. The proportion of minority depository institutions in the very low, relatively low, relatively moderate, relatively high, and very high risk categories are 4%, 14%, 31%, 29%, and 23%, respectively.

Geographic Concentration of Credit Unions at High Risk

As shown in Figures 4 and 5, credit unions that are at relatively high and very high risk of experiencing negative consequences of weather and climate-related disasters tend to be located along the coasts, particularly in California, Texas, and Florida. These three states together account for 11 percent of credit unions located in communities that are at an elevated risk and 22 percent of credit union assets.6

Figure 4: Total Assets in Billions ($) by FEMA’s Risk Categories

Note: The size of the bubble corresponds to the magnitude of assets, with larger bubbles indicating higher assets.
Source: FEMA (2021) National Risk Index, NCUA’s Call Report Data 2021Q4 and OCE Calculations

This map shows the location of credit unions in the United States. Credit unions are reflected by a bubble. The size of the bubble corresponds to the magnitude of its assets, with larger bubbles indicating higher assets. The color of each bubble coresponds to FEMA’s National Risk Index risk categories. Colors range from green, which indicates very low risk to red, which indicates very high risk.

This map shows the exposure of credit union assets to natural hazard risk by state. The share of credit union assets in relatively high-risk and very high-risk communities are grouped in one of three categories: 0% to 1%, 1% to 3%, and 3% to 15%. Washington, Arizona and Oregon are in 1% to 3% category. California, Texas and Florida are in the 3% to 15% category. All other states are in the 0% to 1% category.

In California, nearly half (48 percent) of credit union headquarters and branches are located in very high-risk communities and another 44 percent are located in relatively high-risk communities. Together, these credit unions account for over 90 percent of members, assets, loans, and deposits of California-based credit unions. All MDIs and low-income designated credit unions in California are at an elevated risk as well.

This bar chart shows the percentage of all credit unions, credit union members, credit union assets, credit union loans, and credit union deposits distributed across communities in California by level or risk, as measured by FEMA’s National Risk Index. The categories are very low, relatively low, relatively moderate, relatively high, and very high. The proportion of credit unions in the relatively low, relatively moderate, relatively high, and very high categories are 1%, 7%, 44%, and 48%, respectively. The proportion of credit union members in the relatively moderate, relatively high, and very high categories are 4%, 59%, and 37%, respectively. The proportion of credit union assets in the relatively moderate, relatively high, and very high categories are 4%, 60%, and 37%, respectively. The proportion of credit union loans in the relatively moderate, relatively high, and very high categories are 4%, 60%, and 36% respectively. The proportion of credit union deposits in the relatively moderate, relatively high, and very high categories are 4%, 59%, and 37%, respectively.

As for Texas, 30 percent of credit unions are in relatively high-risk communities and 24 percent are located in communities at very high risk. These credit unions account for over 80 percent of members, assets, loans, and deposits of credit unions with headquarters or branches in the state. As is the case elsewhere, MDIs are even more vulnerable, with 73 percent located in relatively high-risk or very high-risk communities in Texas. Just over half (51 percent) of low-income designated credit unions have headquarters or branches in these high-risk communities.

Lastly, of the credit unions with headquarters or branches in Florida, 16 percent are in very high-risk communities and 39 percent are in relatively high-risk communities. Together, these Florida-based credit unions account for more than 60 percent of members, assets, loans, and deposits of credit unions with facilities in the state. MDIs and low-income designated credit unions located in Florida face greater risk, with 78 percent of MDIs and 60 percent of low-income designated credit unions in relatively high– or very high-risk communities.

This bar chart shows the percentage of all credit unions, credit union members, credit union assets, credit union loans, and credit union deposits distributed across communities in Texas by level or risk, as measured by FEMA’s National Risk Index. The categories are very low, relatively low, relatively moderate, relatively high, and very high.The proportion of credit unions in the very low, relatively low, relatively moderate, relatively high, and very high categories are 2%, 15%, 29%, 30% and 24%, respectively. The proportion of credit union members in the relatively low, relatively moderate, relatively high, and very high categories are 2%, 16%, 64%, and 18%, respectively. The proportion of credit union assets in the relatively low, relatively moderate, relatively high, and very high categories are 2%, 14%, 65% and 19%, respectively. The proportion of credit union loans in the relatively low, relatively moderate, relatively high, and very high categories are 2%, 13%, 68% and 17%, respectively. The proportion of credit union deposits in the relatively low, relatively moderate, relatively high, and very high categories are 2%, 14%, 64% and 19%, respectively.

This bar chart shows the percentage of all credit unions, credit union members, credit union assets, credit union loans, and credit union deposits distributed across communities in Florida by level or risk, as measured by FEMA’s National Risk Index. The categories are very low, relatively low, relatively moderate, relatively high, and very high. The proportion of credit unions in the very low, relatively low, relatively moderate, relatively high, and very high categories are 3%, 5%, 37%, 39%, and 16%, respectively. The proportion of credit union members in the relatively low, relatively moderate, relatively high, and very high categories are 1%, 34%, 58%, and 7%, respectively. The proportion of credit union assets in the relatively low, relatively moderate, relatively high, and very high categories are 1%, 35%, 57%, and 7%, respectively. The proportion of credit union loans in the relatively low, relatively moderate, relatively high, and very high categories are 1%, 35%, 59%, and 6%, respectively. The proportion of credit union deposits in the relatively low, relatively moderate, relatively high, and very high categories are 1%, 35%, 57%, and 7%, respectively.

Credit Union Exposure to the Costliest Natural Hazards

We also looked at credit union exposure to specific natural disaster risks, with a focus on the 10 most costly natural disasters based on FEMA’s estimates of expected annual losses.7 Together these top 10 disasters account for 97 percent of total expected annual losses from the natural disasters included in FEMA’s National Risk Index. Pairing the NRI data with NCUA’s call report data, we found that roughly half of credit union assets are in areas at relatively high or very high risk of experiencing tornado, the fourth costliest disaster type. A similarly large share of credit unions is exposed to strong winds, the ninth costliest disaster type.

This bar chart shows Credit Union Assets in High-Risk Areas by Natural Disaster Type. The reported numbers are in Billions of Dollars. The categories are relatively high and very high. Credit union assets exposed to earthquake in relatively high and very high risk areas are $180 billion and $245 billion, respectively. Credit Union assets exposed to ravine flood in relatively high and very high risk areas are $598 billion and $133 billion, respectively. Credit Union assets exposed to hurricane in relatively high and very high risk areas are $181 billion and $53 billion, respectively. Credit Union assets exposed to tornado in relatively high and very high risk areas are $626 billion and $383 billion respectively. Credit Union assets exposed to wildfire in relatively high and very high risk areas are $304 billion and $99 billion respectively. Credit Union assets exposed to hail in relatively high and very high risk areas are $260 billion and $124 billion respectively. Credit Union assets exposed to heat wave in relatively high and very high risk areas are $323 billion and $57 billion respectively. Credit Union assets exposed to coastal flood in relatively high and very high risk areas are $72 billion and $40 billion respectively. Credit Union assets exposed to strong winds in relatively high and very high risk areas are $466 billion and $475 billion respectively. Credit Union assets exposed to drought in relatively high and very high risk areas are $182 billion and $61 billion respectively.

Table 1 disaggregates the information provided in Figure 9 and reports percentages of credit unions, members, total assets, MDIs, and low-income designated credit unions in the relatively high-risk and very high-risk communities that are exposed to each of the top 10 natural disasters.

Rank FEMA cost estimates ($billions) Natural Disaster Number of Credit Unions* Members* Assets* Minority Depository Institutions (MDIs)* Low Income Credit Unions (LICUs)*
1 $176 Earthquake 9 17 20 14 8
2 $64 Ravine Flood 32 34 35 45 30
3 $51 Hurricane 12 16 16 25 16
4 $45 Tornado 50 48 48 59 46
5 $14 Wildfire 9 17 20 11 9
6 $10 Hail 15 18 18 14 13
7 $9 Heat Wave 25 23 23 41 23
8 $7 Coastal Flooding 13 11 11 14 15
9 $7 Strong Winds 41 47 45 50 36
10 $6 Drought 11 13 14 3 10

* Share Exposed by Type of Disaster (%)

Note that exposure is flagged when a credit union’s mailing address is in an area with elevated exposure to the identified disaster.
Source: FEMA (2021) National Risk Index, NCUA 5300 Call Report (2021Q4), OCE calculations.

Concluding Thoughts and Future Research

Our analysis provides a preliminary look at where credit unions are located in relation to the geographies most at risk of suffering from natural disasters, with a particular focus on credit union exposure to risk in three states. We also examined the exposure of the credit union system to specific natural disaster risks, with a focus on the 10 most costly natural disasters based on FEMA’s estimates of expected annual losses. OCE’s estimates are intended to provide context to discussions about the credit union system’s exposure to climate-related financial risk and should be interpreted with care.

Our analysis did not make use of member- or loan-level data that could provide more precise estimates of credit union exposure to natural hazards. Unfortunately, the micro-level data necessary to facilitate such work are not readily available. Moreover, our analysis does not consider the likelihood of specific disasters affecting credit unions or estimate what credit union losses might be given the occurrence of these disasters. For the latter, further investigation into credit unions’ historical experience could provide helpful guidance. However, several major challenges will remain in forecasting losses. For instance, what should be assumed about government intervention in the wake of a natural disaster? In the past, the effects of many natural disasters have been mitigated by local, state, and federal aid. It is difficult to anticipate the extent to which government intervention will ameliorate the negative effects of natural hazards in the future and whether the effect of mitigation efforts will differ from those in the past.

Future extension of this analysis would benefit from the use of scenarios to evaluate potential credit union losses from natural hazards. We understand that the Federal Reserve Board is exploring scenario analysis. OCE will be monitoring this initiative with great interest and conducting further analysis of credit union-specific data as additional information related to climate and weather events becomes available.

link