Climate change has emerged as a major challenge for central banks. On the one hand, there is a debate about appropriate actions of central banks to limit climate change. In this regard, the ECB has taken a leading role, recognising that the “urgency of this topic… requires all policymakers to explore their roles in tackling this challenge” (Schnabel 2021) and noting that while “we are not in the driving seat”, this “does not mean that we can simply ignore climate change, or that we do not play a role in combating it” (Lagarde 2021). On the other hand, it is widely accepted that climate change matters for monetary policy, at least to the extent that it gives rise to financial stability risks (Brunnermeier and Landau 2020).
Measuring expectations about the near-term economic impact of climate change
Against the background of these far-reaching and fundamental considerations, we show in a new paper that climate change represents a much more direct and immediate challenge for monetary policy (Dietrich et al. 2021). Our point of departure is the fact that irrespective of when and how climate change actually plays out, what matters for monetary policy is how people expect it to play out. After all, expectations feed back into economically relevant decisions today and these decisions matter for monetary policy.
We solicited these expectations through a representative survey of US consumers. Specifically, our data come from a larger, nationally representative daily survey of consumers sponsored by the Federal Reserve Bank of Cleveland. We asked more than 10,000 respondents how they expect climate change to impact the economy in the near term, that is, over the next 12 months. Specifically, we asked them about (1) the impact of climate change on GDP growth, (2) likely damages caused by climate change to economic activity and (3) the probability of larger, climate-change related disasters (causing damages of as much as 5% of GDP). We also obtained information on respondents’ socioeconomic background and checked their ability to compute probabilities.
Three main results on the effect of climate emerge, summarised in Table 1. The top panel of the table pertains to all respondents while the bottom panel summarises statistics for those respondents with high numerical ability. There are no systematic differences across the two panels. First, the expected impact of climate change on GDP growth is approximately zero across respondents. Second, the average for climate change-related damages over the next 12 months is 1.5% of GDP (we asked respondents to assign probabilities to five brackets ranging from zero damages to damages of 5%). Our third question zooms in on large disasters and asks respondents to state the probability of such a large disaster. Here we find quite high values: the median response is 12% across all respondents, and 15% for respondents with high numerical ability.
Notes: statistics are weighted using survey weights as well as Huber-robust weights. High numerical ability respondents answer a question on probabilities with an error margin of at most 2 percentage points.
There are various possibilities for why the perceived probability of disaster is so high, and in fact much higher than what historical data suggest. For instance, respondents may think we have been lucky in the past, just like in the case of ‘peso problems’ - in the relatively short sample under consideration, adverse events may have simply materialised less often than what the objective probability would imply. Alternatively, natural disasters due to climate change may be much more frequent in the future because we may have reached so-called ‘tipping points’. Yet another possibility is that we are picking up a salient ‘Greta effect’ – people overestimate the risk of natural disasters because of a media focus on climate change, consistent with research that has documented how media focus can be an independent source of business cycle fluctuations (Chahrour et al. 2020).
In support of this last possibility, we find in the survey that respondents that are not exposed to media at all report a significantly lower estimate of the probability of natural disasters. Moreover, we formally complement our survey analysis with several information treatments: a ‘newspaper treatment’, which shows respondents sections of a USA Today newspaper article on the 2020 wildfire and hurricane season; a ‘Lagarde treatment’, which is a recent statement by ECB President Lagarde on the importance of climate change for the ECB's monetary policy; and two treatments that provide respondents with information about the frequency and extent of large disasters in the past.
These treatments have a sizeable impact on the answers of the survey respondents. In Figure 1, we show the distribution of responses to the third question (on the probability of rare disaster within the next 12 months). The red solid line is the distribution across all respondents (median: 12%). The newspaper and the Lagarde treatment, shown by the blue-dotted and the green-dashed lines in the top panel, respectively, stand out in the way they shift the mass of the distribution to right – people who receive information about these disasters think they are more likely to happen.
In addition, we correlate the responses with other covariates and detect very plausible patterns. For instance, personal experiences of disasters raise the perceived probability of disasters. Likewise, we find that Republicans assign smaller probabilities to large climate change-related disasters; the opposite holds for Democrats and independent voters (see the bottom panel of Figure 1). The tenor of our findings here is that expectations of climate change-related disasters in the near future are very high and vary in a meaningful way with a number of socioeconomic characteristics. More importantly still, we find that reported expectations are also quite sensitive to new information.
a) Disaster probability distribution: Info treatments
b) Disaster probability distribution: Political affiliation
Implications for monetary policy
Expectations about climate change-related disasters are ‘bad news’ about the future and induce a contraction of current expenditures according to theory – an instance of ‘Keynesian supply shocks’ (Guerrieri et al. 2020). We establish this formally in a New Keynesian model. The model is due to Fernández‐Villaverde and Levintal (2018) and features rare disasters, but also nests the textbook version of the New Keynesian model for which we establish a number of closed form results. In particular, we show that the natural rate of interest declines in response to expected disasters, both with the probability and the size of natural disasters. Monetary policy plays a key role in shaping the adjustment of the economy to these disaster expectations. In the basic model there are no supply-side effects of disaster expectations as such, and monetary policy may fully stabilise the economy at potential if it adjusts the policy rate in sync with the natural rate of interest.
We calibrate the full model to capture key results of our survey and solve it numerically. Table 2 summarises the effect of disaster expectations on key variables under the baseline scenario according to which a disaster is expected to occur with a 12% probability. It reduces productivity and destroys a fraction of the capital stock such that output declines by 5% on impact. What matters in our analysis, however, are expectations of the disaster, not the disaster itself.
Notes: The table gives simulation results for different disaster calibrations. Numbers represent deviations from the no disaster steady state.
These expectations reduce the natural rate by about 65 basis points. This is a sizeable effect and may put central banks in a difficult position, notably if policy rates are low to begin with – if monetary policy is unable or unwilling to lower policy rates, a large recession might ensue. For our baseline scenario we assume a conventional Taylor-type interest rate rule which ensures that the demand contraction remains fairly contained – the output gap and inflation decline by about 0.2 and 0.3 percentage points, respectively. Yet disaster expectations also lower potential output because the expected damages to the capital stock in the event of a disaster reduce investment significantly.
And finally… a paradox of communication
In sum, climate change represents a major challenge for central banks and central banks have started to confront this challenge in various ways – not least by communicating frequently about the issue. This is certainly laudable. But there is a risk that the immediate implications of climate change – namely, those that operate through the expectations channel – are going unnoticed because of an undue focus on how to battle climate change, a task central banks are perhaps not particularly well-equipped to deal with. What’s more, there may even be a paradox of communication inherent to central bank activity. To the extent that central bankers engage in the debate, they may themselves contribute to the media focus and salience of climate change. This contribution, in turn, may exacerbate adverse expectations about future climate-change related disasters. In this way, by trying to tackle a major global challenge upfront, central banks actually make their tasks harder today – because interest rates are low and further reductions in the natural rate are hard to accommodate.
Authors' note: The views stated in this column are those of the authors and are not necessarily those of the Federal Reserve Bank of Cleveland or the Board of Governors of the Federal Reserve System.
Brunnermeier, M and J Landau (2020), “Central banks and climate change”, VoxEU.org, 15 January.
Chahrour, R, K Nimark and S Pitschner (2020), “Sectoral Media Focus and Aggregate Fluctuations”, mimeo.
Dietrich, A, G Müller and R Schoenle (2021), “The Expectations Channel of Climate Change: Implications for Monetary Policy”, CEPR Discussion Paper 15866.
Guerrieri, V, G Lorenzoni, L Straub, and I Werning (2020), “Macroeconomic implications of COVID-19: can negative supply shocks cause demand shortages?”, NBER Working Paper 26918.
Lagarde, C (2021), “Climate change and central banking”, Keynote speech at the ILF conference on Green Banking and Green Central Banking, 25 January.
Fernández‐Villaverde, J and O Levintal (2018), “Solution methods for models with rare disasters”, Quantitative Economics 9: 903-944.
Schnabel, I (2021), “From green neglect to green dominance?” Intervention by Isabel Schnabel, Member of the Executive Board of the ECB, at the “Greening Monetary Policy – Central Banking and Climate Change” online seminar, organised as part of the “Cleveland Fed Conversations on Central Banking”, 3 March.