From heatwaves and hurricanes to wildfires and high tide flooding, what role is there for US financial regulators in managing the physical risks of climate change? On Dec. 4, 2020, the Climate Impact Lab, the Federal Reserve Bank of San Francisco, the Energy Policy Institute at the University of Chicago, the Goldman School of Public Policy at the University of California, Berkeley, the Kenneth C. Griffin Applied Economic Incubator, and Rhodium Group convened a virtual conference for policymakers, regulators, and leading researchers on the topic of physical climate risk to the US financial system. This event featured keynote speakers from the Federal Reserve, the Commodity Futures Trading Commission, and Capitol Hill, and panel discussions on the macroeconomic and regional risks posed by climate change.

Slides from the event are available at this link. 

Identifying climate risk research gaps

Held in the wake of the 2020 US election, this one-day virtual conference brought academics studying the economic risks of climate change together with regulators who are thinking about these topics, to discuss the gaps in our understanding. Attendees heard a Federal Reserve perspective on climate risk, as it relates to monetary policy and the stability of the financial system, as well as a readout on the Commodity Futures Trading Commission’s (CTFC) groundbreaking report on managing climate risk. Observations from these regulators carry extra weight, given their direct authority to leverage research in addressing the problem.

Conference co-host Glenn Rudebusch, of the Federal Reserve Bank of San Francisco, identified the greatest sources of uncertainty related to forecasting climate damages, including how future emissions will evolve, potential tipping points in the climate system, and the financial repercussions of a given climate outcome. Rudebusch pointed to the feedback loop from policy, economic, and financial developments back to the path of carbon emissions as a critical gap in current understanding. He shared his views on how further research can help financial risk managers understand how climate risk is linked to balance sheet risk. Rudebusch offered real estate as a “concrete” example, with property assets and property markets increasingly threatened by storm surge, hurricanes, wildfires, and increased temperatures. He suggested that property damage could lead to credit or collateral losses, as properties that support loans are destroyed or becomes less productive.

Rudebusch proposed specific guidance to researchers on how they can support microprudential and macroprudential oversight. At the micro level, financial institutions need the right governance, models, and data sources to assess credit, market, and operational risks. On the macro level,

Rudebusch suggested researchers should focus on improving the ability to conduct climate stress tests. Relative to existing stress tests, climate change’s physical effects take place over a much longer time horizon and involve much more complex interactions, he said. Rudebusch noted two recent reports from the Federal Reserve researchers—one on supervision and regulation and another on financial stability—that provide a two-prong official acknowledgment of climate change’s importance.

Keynote speaker Robert Litterman, chairman of the task force convened by the CTFC to study climate risk, contributed more advice to researchers on supporting climate-conscious financial regulation. He encouraged federal agencies to take a role in leading, coordinating, and funding research on climate risk, as well as understanding the implications of that research. There is also a tremendous amount of expertise in the private sector, he added, as evidenced by the composition of the CTFC task force. Research is at an early stage, Litterman acknowledged, with a great need for both “bottom up” data that captures local, asset-level exposure. However, regulators also need data on the systemic potential of risk to the financial system. Tailoring research to many different levels of aggregation may be one of the biggest challenges, Litterman said.

Scoping the magnitude and geography of climate risk across the US

To understand how much to invest in mitigating the problem, regulators need to gauge the scale of climate change’s impact on society and understand which locations or sectors of the economy may be most exposed, said Climate Impact Lab Co-Director Solomon Hsiang of the University of California at Berkeley in his remarks—which stressed the key role of data science. Hsiang emphasized how the technology of governance has improved over time, empowered by more accurate information. Detailed data and quantified estimates of climate change impacts will be vital to developing an effective regulatory response to climate change, he said.

The conference showcased how far researchers, including members of the multi-institutional Climate Impact Lab, have come in quantifying climate risk—both from a macroeconomic perspective and through a more localized hazard-specific lens. In the first academic panel of the day, researchers explored the overall impact of climate change on the US economy: its effect on growth, employment, productivity, and the magnitude of long-term risk. Pete Klenow of Stanford University presented a theoretical approach to assessing where growth effects might come into play. Stanford’s Marshall Burke and University of Illinois’ Tatyana Deryugina highlighted their empirical research: Burke on how growth rates respond to temperatures, and Deryugina on the mechanisms that drive climate damages. The discussion, moderated by University of Chicago’s Amir Jina, hit on climate stress testing, the uncertainty around tipping points that could lead to horrible outcomes, and how to square empirical results with theoretical mechanisms.

Adaptation and the costs associated with attempts to minimize climate damage took center stage. Fully adapting to the impacts of climate change may be possible, members of the panel suggested, but at what cost? The conversation also focused on transfers of wealth to impacted areas and how social safety nets like unemployment insurance, food stamps, and other income transfer assistance may offset some of the losses driven by more extreme events. Deryugina’s research on the fiscal cost of hurricanes prompted a discussion about how to get a clearer picture of climate change’s effect on social insurance programs. According to research from Jina and Hsiang, a significant amount of the cost from these storms actually manifests years down the road, as resources that would otherwise go towards productive output must be diverted to rebuilding and recovery.

Following the macroeconomic risk discussion, a second panel zoomed in on regional risks to the US—from flooding and coastal storms to extreme heat and more costly wildfires. Wolfram Schenkler of Columbia University spoke to the effect of climate change on agricultural yields and the implications for rural communities. In the medium-term, Schenkler said crop insurance would protect US farmers against the yield losses driven by climate-related weather shocks, and farmers would potentially see increased prices for their crop as extreme heat drives decreased yields. Further out into the future, climate models predict significant increases in heat, nearing “Dust Bowl” conditions. Rural communities are predicted to see population declines, Schenkler said, as cities grow. The biggest impact of climate change’s effect on domestic agricultural yields may be felt outside the US, with food prices rising in developing countries.

Looking to the West, University of California at San Diego’s Judd Boomhower discussed mounting wildfire damages as climate change worsens the size, frequency, and severity of these events. Boomhower said structure loss is the most immediately obvious effect but not the only category of economic impacts. Many fires are extinguished before they do any property damage, meaning federal agencies and states are spending more on wildland firefighting and pumping resources into fuels management. By thinning vegetation, officials can reduce the intensity of eventual fires. Boomhower noted California’s recent public safety power shutoffs have significant economic impacts on people who lose service. He pointed to researchers’ growing understanding of air pollution costs from wildfire smoke as a critical area for future study.

Flooding from coastal storms, sea-level rise, and changing precipitation trends drove much of the regional risk discussion. Miyuki Hino of the University of North Carolina at Chapel Hill shared her insights on how households adapt to the effects of sea-level rise inundation, showing that without public policy to support equitable outcomes, those with the least resources would be most vulnerable to the encroaching ocean. Though news headlines reinforce the costly effects of hurricanes on property, Hino said those damage estimates often overlook long-term reductions in income and economic growth, mental and physical health impacts, and the mounting price tag associated with non-storm flooding in impacted coastal communities.

While researchers have a better sense of where the floodplain is, much less is known about who is exposed, University of Iowa’s Eric Tate showed in his presentation. Tate’s research links storm impacts to the lives of people who live in its path: especially residents of mobile homes, low-income and minority neighborhoods, and renters. Existing social vulnerability indicators, such as the formula used by the Department of Housing and Urban Development’s Community Development Block Grant Programs, are generalized, not flood-focused or mitigation-focused. Tate suggested future research should work to develop customized measures of vulnerability.

Recommending policies to better manage climate risk

The conference aimed to explore specific regulatory and policy responses to climate risk, in addition to laying out a research agenda. Participants asserted that doing nothing to avert climate change would be costly and exacerbate inequality. Many pointed to putting a price on carbon as the most efficient mechanism to mitigate climate risk, but a range of targeted solutions emerged over the course of the event. The third panel of the day delved into managing risk in corporate lending, housing, and insurance markets.

Harnessing the power of markets to try to build resilience to climate impacts will take multiple policies across different scales and regulatory bodies, said Carolyn Kousky, executive director of the Wharton Risk Management and Decision Processes Center at the University of Pennsylvania. Kousky spoke to shortcomings related to pricing climate risk, including in US insurance markets. Because insurance contracts are typically annual, Kousky said they are not a good tool for reflecting the future risk of flooding or extreme events. A better example of pricing in long-term risk comes from local governments that have adopted policies to require a climate analysis before issuing development permits. Kousky pointed to California’s recent wildfire season as an example of the strain climate change is already causing for disaster insurance markets. She stressed that government intervention, which can help advance social policy goals like making policies affordable, can also undermine insurance markets. When we fail to provide sufficient information to market participants then future losses do not influence today’s decisions, Kousky said.

The Fed’s Ricardo Correa showed the increasing severity of certain natural disasters is beginning to be priced into corporate lending markets in a presentation of his research on the topic. Consider two firms receiving loans from the same bank: a hurricane hits one, but the other is not impacted. In the year after the storm, Correa’s research finds the impacted firm pays a higher interest rate on its borrowing. Correa said the same effect is evident when climate impacts are more salient in the news. The interest rate spread for at-risk firms almost doubled in years after major reports from the U.N. Intergovernmental Panel on Climate Change. Further work is needed to assess whether these price changes are temporary or permanent, he said. Correa said his research also indicates the need for better data to assess banks’ and firms’ exposure to climate-related risks.

How do Americans’ beliefs about climate risk shape coastal housing markets? University of California at Santa Barbara’s Lint Barrage overviewed her research on this topic. Ideally, if everyone was perfectly informed about the threat posed by sea-level rise, safer homes would be sold at a premium, and more exposed homes would get discounted, providing proper incentives to construct houses in safer areas. However, field surveys show that those who live in high-risk flood zones are less worried about flooding than their neighbors. These misperceptions contribute to an overvaluation of risky assets, potentially creating a coastal property bubble that could be burst by climate change. Barrage suggested climate skepticism may be delaying adaptation in the housing market. She called accurate and forward-looking risk information “critical” to the housing market’s stability, pointing to Federal Emergency Management Agency (FEMA) maps as vital but out-of-date and the First Street Foundation’s floodfactor.com as a project leading the way. Barrage also suggested flood risk disclosure laws needed to be improved to ensure a channel for better information to reach home buyers.

Rep. Patrick McHenry (R-N.C.), in his remarks to the conference, reinforced the need to improve flood maps that are the basis for the National Flood Insurance Program (NFIP). Bipartisan legislation that McHenry co-sponsored calls for increased funding for expanded and improved flood mapping and mitigation programs. The NFIP should account for more adverse weather events related to climate change and use that in their assessment of year-over-year risk to the taxpayer, McHenry suggested. Additionally, he wants greater visibility on flood risk to the portfolios of federally-backed mortgage companies Fannie Mae and Freddie Mac. As the Republican leader of the House Financial Services Committee, which provides oversight of the Federal Reserve, the Treasury Department, and other federal financial regulators, McHenry said he thinks about climate risk as it impacts the bottom line to government expenditures and applauded the Fed for early steps toward understanding the danger.

Sen. Brian Schatz (D-Hawaii) commended the Fed and the CTFC for incredible work on assessing the threat posed by climate change, while knocking other agencies for “thinking too narrowly” about impacts to financial stability. Schatz, a member of the Senate Banking, Housing, and Urban Affairs Committee and chair of the Senate Democrats’ Special Committee on the Climate Crisis, last year unveiled legislation that would require the Fed to subject banks to a stress test for climate. Even without new action by Congress, Schatz said federal financial regulators have an obligation under current law to make sure risk is quantified—including the climate-related calamities like wildfires and hurricanes. Because climate risk has been a subject of political polarization and because it is difficult to quantify, it has been “booked at zero,” Schatz said.

The conference ended with a peek at cutting-edge research by the Climate Impact Lab that aims to provide the sort of local, granular data that policymakers and regulators can use to assess climate damages. Lab co-founder Michael Greenstone, who directs the Energy Policy Institute at the University of Chicago, presented findings from a working paper assessing temperature’s toll on public health. It shows that with continued high emissions of greenhouse gases, the death rate from climate-induced changes in temperature would surpass the current death rate for all infectious diseases combined—outside of COVID-19. These results mean that the economic costs of climate-induced health risks are at least an order of magnitude larger than has previously been understood, Greenstone said. Data science is enabling the “bottom up” view of climate impacts that Greenstone views as critical to local planning and much more nuanced, large-scale policy responses.

A recording of the full conference is available to watch here: