Central Banks and Climate Policy
“One thing that is said about climate change and central banking is that climate change could be very bad for financial stability, and I agree. Wars are very bad for financial stability, too… Should central banks ration the provision of credit to suppliers of arms manufacturers?” Sir Paul Tucker, Oral Evidence to the Select Committee on Economic Affairs of the U.K. House of Lords, 2 February 2021.
Avoiding a climate catastrophe requires an urgent global effort on the part of households, firms and governments to reduce our reliance on fossil fuels. Like many economists, we support a carbon tax. We also favor generous fiscal support for R&D to substitute for fossil fuels and remove carbon from the atmosphere.
What role should central banks play in this global effort? That is the prime focus of this post. We argue that central banks must preserve the independence needed for effective monetary policy. That implies only a modest role in addressing climate change.
Central banks are involved in both financial regulation and monetary policy. In each case, there are some things that central bankers can and should do to help counter the threat posed by climate change. As financial regulators, they should implement an improved disclosure regime and develop tools to ensure the financial system is resilient to climate risks.
In conducting monetary policy, central bankers should follow a simple, powerful principle: do not influence relative prices. To be sure, it is and should be standard practice to use interest rates to influence relative prices between consumption today and tomorrow. However, central banks ought not influence relative prices among contemporaneous activities. We will see that achieving this form of relative price neutrality may require central bankers to shift the composition of their assets and to alter the treatment of collateral in their lending operations.
Most people view central bankers as the guardians of price stability. They manage their balance sheets, lend to intermediaries, and control interest rates to keep inflation low and promote full employment. To help achieve these goals, monetary policymakers are granted considerable independence. As independent authorities, central bankers can and should avoid making decisions that are inherently distributional: they should not pick winners and losers. This is a form of tax policy, so it is the proper function of elected officials and fiscal authorities, not central banks.
In practice, monetary policymakers focus on the path of the aggregate price level. What this means is that individual price changes are of concern only to the extent that they lead to (or signal) broad-based inflation. Insofar as possible, monetary policy strives to be neutral with respect to the distribution of price changes in the economy. Admittedly, some sectors are more interest rate sensitive than others, so the goal of cross-sectional neutrality can never be more than aspirational.
In addressing climate concerns, the same neutrality principle ought to apply. Consider, for example, central banks’ balance sheet composition and lending policies. In some cases, officials already have a neutrality concept in mind. Those that purchase private securities (which does not include the U.S. Federal Reserve) may attempt to replicate the market portfolio. That is, they hold securities in the proportion in which they are issued. For lending, a broad range of banking system assets is accepted as collateral, with haircuts that depend on an asset’s perceived credit and market risk.
While straightforward, neither of these approaches to neutrality is, in fact, neutral with respect to the economy’s carbon emissions or climate risk. Put differently, central banks are (inadvertently) influencing relative prices. The reason is that bond issuers tend to be carbon intensive relative to the economy as whole, so a portfolio that is bond-market neutral, or a collateral policy that makes it attractive to deliver bonds, unintentionally subsidizes firms that contribute more to climate risk.
The following chart, based on information collected by Papoutsi, Piazzesi, and Schneider (2021), makes this point. The plot ranks seven sectors of the euro area economy by their shares of direct emissions (black bars). Admittedly, it is difficult to measure emission intensity, which also may change over time (see Bank of England’s recent analysis here), so central banks should be humble about their ability to make policies that depend greatly on such measurements (see King and Katz).
Euro Area: Sectoral shares of emissions, bond market, and capital income, end-2017
Source: Interpolated from Papoutsi, Piazzesi, and Schneider (2021) Figures 2 and 4. “Dirty manufacturing” includes oil and coke, chemicals, basic metals, and nonmetallic manufacturing. “Other manufacturing” is food, beverages, tobacco, textiles, leather, wood, paper, pharmaceuticals, electronics, electrical equipment, machinery, furniture, construction, and other manufacturing. Emissions are directly associated with the sector. See Papoutsi, Piazzesi, and Schneider for details.
The other bars show the shares of bond market capitalization (in red) and the shares of capital market income (in blue). Capital income, computed as valued added less wage payments, is a proxy for the overall capitalization of the sector. Looking at the chart, we see that bond-market shares more closely match emissions than do shares of capital income. (The correlation of the bond market and emissions shares is 0.3, while that between the capital income and emissions shares is minus 0.2.)
To see the implications of this, we can compute the emissions associated with two portfolios, one using bond-market shares as weights and a second using capital-income shares. The following chart, which normalizes total emissions for the bond-market weighted portfolio to 1, shows the results. Importantly, the emissions from utilities have a much lower capital income weight (4%) than bond weight (18%), so this sharply reduces their capital-income weighted emissions. At the other end of the spectrum, services—which are not very carbon intensive—have a much larger capital income share than their bond market share. Overall, shifting to a capital-income share weighted portfolio reduces direct emissions intensity by 27%.
Emissions of bond-market vs capital-income share weighted portfolios (Bond-market portfolio normalized to 1.0)
Source: Constructed from information in the previous chart.
To be clear, holding assets based on capital income shares avoids influencing relative prices, consistent with the neutrality principle that we propose. It avoids the problem that Paul Tucker highlights in the opening quote, where the central bank would have to make tax-like judgments regarding the social desirability of specific economic functions. Engaging in such quasi-fiscal operations that are rightly the purview of elected officials would put a central bank’s independence at risk.
Turning to lending policies, every central bank has a haircut schedule. For example, in its refinancing operations, the Eurosystem accepts more than 25,000 different debt instruments at haircuts between ranging from 0.4% and 44.6%. The haircuts are designed to ensure that the authorities can recover on a defaulted loan by selling the collateral. This means larger haircuts for riskier collateral. In practice, issuers with relatively high ratings enjoy smaller haircuts. Again, because firms in carbon-intensive industries typically issue the most bonds, accepting their collateral in line with bond weights means that they will be subsidized. Instead, to achieve relative price neutrality, the collateral system should reflect the structure of the entire economy, not just that portion relying more heavily on debt financing.
What about the role of financial regulators and supervisors—including those that are central banks? Here central bankers frequently are responsible for guarding the safety and soundness of the financial system and for protecting consumers and investors.
Perhaps the most important means for financial regulators and supervisors to address climate change is through enhanced disclosure. For example, climate change has the potential to create systemic risk. To assess and anticipate such risks, regulators can require intermediaries to disclose their exposure to carbon-producing sectors and to natural disasters associated with climate change. Not surprisingly, the Financial Stability Board established in 2015 a task force on climate-related disclosure that is working on a set of international standards (see here).
Regulators also should independently assess and disclose the climate resilience of financial intermediaries, especially bank and non-bank financial intermediaries. Again, this is a global problem that requires very detailed information and technical knowledge. From the perspective of both authorities and market participants, it is useful to apply uniform scenario analysis across all institutions, without regard to their home jurisdictions.
Here, we point to two strands of work. First, academic researchers are producing climate risk assessments at the level of individual institutions. Jung, Engle, and Berner (2021) is a recent example. For a set of 15 U.S. and U.K. banks, they construct a measure of the capital shortfall in a climate stress scenario defined as a 50% decline in the value of each institution’s climate-related stranded asset portfolio. Second, the Network for the Greening of the Financial System—a group of 95 financial supervisory organizations around the world—is constructing common scenarios for use in climate resilience assessments.
Importantly, if regulators aim to follow the relative-price neutrality principle, there are things that they ought not do. In theory, for example, regulators could restrict the assets that intermediaries may acquire or adjust the risk-weights on those assets they hold. In either case, however, authorities would have to dictate what is green and what is not. To be effective, they also would have to alter these portfolio guides whenever conditions (including firm-level exposures) change. In practice, drawing and maintaining lines of this sort is quite fraught and something we normally don’t assign to unelected bureaucrats.
In addition to monetary policy and financial regulation, central banks also provide a set of public goods in the form of data and research. On these, there is scope for a greater contribution. For example, as ECB Executive Board Member Isabel Schnabel argues, climate change can affect a central bank’s ability to achieve its price stability objective. With that in mind, central bank researchers can develop and disseminate new indicators for assessing the impact of both climate change and carbon emissions mitigation policies on the economy and financial system.
Finally, we have a few general comments about financial stability. First, in the presence of effective disclosure, we assume that investors would have considerable incentive to assess resilience, to price climate risks, and to penalize those institutions and sectors that are the most exposed. However, experience teaches us that an underpricing of long-term risks is not uncommon. In general, we expect that even professional asset managers are focusing relatively little attention on the distribution of possible surprises 10 year hence, much less those that could emerge in 25 or 100 years.
What about the financial stability risks arising directly from climate change? In our experience, financial crises arise from sudden shifts in the prices of risky assets. The burst of the dot-com boom is a classic example of an asset price bubble that did not trigger a crisis. In contrast, in the runup to the 2007-09 crisis, there was a large buildup of mortgage-related debt based on unrealistic expectations for the path of residential housing prices. The system collapsed when housing prices turned down.
What does this have to do with climate risk? In theory, sudden asset price shifts could follow both physical climate shocks (arising from more frequent natural disasters and extreme weather disturbances) and transitional shocks (resulting from large-scale changes in technologies or regulations that reallocate resources across sectors). In practice, we suspect that the greater threat to financial stability arises from the possibility of a sudden political decision, such as an unanticipated (however welcome) imposition of a carbon tax that could cause a swift repricing of a broad array of assets (see, for example, Cochrane).
To sum up, everyone cares deeply about climate change. The primary responsibility for enacting policies to mitigate climate risk rests with fiscal authorities, who can introduce a carbon tax and subsidize R&D to lower carbon use. Financial regulators have a role in promoting disclosure and the use of common scenarios for assessing climate resilience and encouraging more accurate market pricing of climate risk. Finally, central banks need to take the principle of relative price neutrality to heart and avoid inadvertent subsidies to carbon-intensive industries.
Acknowledgements: We thank Rob Engle, Mervyn King and Dick Berner for discussions and exchanges that helped us to understand various aspects of this subject. This post is based partly on comments one of us (Steve) presented at the E-Axes Forum Launch Event on 31 August 2021.