by Andreas Charisiadis, Research Assistant, Cambridge Centre for Finance and Cambridge Endowment for Research in Finance
Mitigating climate change has become one of the defining challenges of our time. Indeed, there is global concern about the potentially disastrous long-term consequences of unmitigated climate change. The risks arising from climate change have far-reaching implications for the real economy and financial markets in particular. Two types of climate risk are particularly prevalent: direct environmental risk and policy (or regulatory) risk. The former relates to the occurrence of extreme climatic events, such as storms, floods, droughts, or wildfires, whereas the latter refers to the uncertain impact and timing of regulations aimed at mitigating climate change, such as the introduction of carbon taxes or emission caps. Several recent contributions explore how such risks impact capital markets, and specifically how these risks – and the way they are perceived by investors – can affect trading behaviour.
Climate risks and investor behaviour
Alok et al. (2020) explore whether professional money managers misestimate the risk of climatic disasters. The authors hypothesise that misestimation of such risks may be driven by salience bias, ie the behavioural tendency to overweight more readily available information. Tversky and Kahneman (1973) document that this type of bias can induce individuals to overestimate the risk of salient events depending on the ease with which instances thereof can be recalled. On that basis, Alok et al. (2020) hypothesise that fund managers may overestimate the risk of extreme climatic events if they have experienced such events in the (recent) past themselves and may therefore substantially reduce their holdings of assets exposed to such risks. The distance of funds from climatic disaster zones – defined as areas directly hit by an extreme weather event – serves as an exogenous source of variation in the salience of climatic disasters for money managers. While all funds tend to reduce their exposure to equities of firms located in a disaster area, this decrease in portfolio weights is significantly more pronounced for funds located closer to the affected area. Alok et al. (2020) attribute this asymmetric portfolio reallocation to overestimation of the disaster risk by fund managers located in the vicinity of the disaster zone. The trading behaviour ensuing from such salience bias can be shown to hurt financial returns: a zero-cost portfolio consisting of long positions in disaster zone stocks which closely located fund managers underweighted the most, and short positions in stocks which were underweighted the least, yields significant risk-adjusted excess returns for a holding period of two years following the disaster.
Choi et al. (2018) study how investors update their beliefs about climate risk. They find that investors revise their expectations about climate change when experiencing an episode of unusually high temperatures. During abnormally warm months there is a significant increase in attention to climate change as measured by the volume of Google searches for ‘global warming’. On that basis, Choi et al. (2018) explore whether and how the ensuing updated investor beliefs about climate risk affect prices and trading behaviour in financial markets. They document that retail investors exhibit a tendency to divest from carbon-intensive equities when experiencing abnormally warm episodes, resulting in a relative underperformance of such stocks vis-à-vis their low-carbon counterparts. The authors also find that institutional investors appear less prone to be swayed by such transitory weather events and therefore – in contrast to retail investors – do not systematically react to abnormal temperatures.
Hedging climate risks
An important strand of the literature explores the methods which investors can use to insure themselves against climate risks. For instance, Engle et al. (2020) study a dynamic approach for forming portfolios of publicly traded assets in order to hedge climatic risks. This method allows investors to (partially) insulate themselves from risks which would otherwise be difficult to insure, as the inherently long-run and systemic nature of climate risk impedes the implementation of standard insurance contracts. Rather than relying on securities which generate positive returns in the event that climate risks materialise, Engle et al. (2020) construct portfolios whose short-term returns hedge innovations in news about climate change. Using textual analysis of newspapers, the authors extract climate change related news, which carry information about the perceived and actual level of climate risk. This allows constructing a climate news index which tracks the intensity of climate change reporting over time. On that basis, equity portfolios which hedge innovations in the time series of climate news can be constructed. These portfolios overweight stocks which rise in value upon the arrival of (negative) climate change news and contain short positions in stocks whose prices decline during such events. The resulting hedge portfolios require continuous rebalancing based on the latest information about the relationship between equity returns and climate news, similar to the dynamic hedging approach of Black and Scholes (1973) and Merton (1973). Continued updating will eventually yield a portfolio which provides compensation for losses incurred as a result of the materialisation of climate risks over the long run. Interestingly, the resulting hedge portfolios do not necessarily align with the common prior that optimally hedging climate risks primarily relies on placing industry bets, ie holding long positions in ‘clean’ industries, such as renewable energy, and short positions in ‘dirty’ industries, such as fossil fuels. In a series of out-of-sample performance tests, Engle et al. (2020) document that their methodology yields hedge portfolios which outperform alternative methods of constructing climate risk hedges (for instance, using industry tilts via positions in energy ETFs).
Andersson et al. (2016) present a dynamic investment strategy with which long-term passive investors can hedge climate policy risks while avoiding substantial sacrifices in returns. They describe a procedure for ‘decarbonising’ standard equity indices in order to construct a hedge against the risk of the introduction (or the tightening) of carbon reduction policies. The formation of their ‘green’ index follows a two-step procedure: First, the k most carbon-intensive stocks are excluded from the chosen benchmark (say, the S&P 500). These firms are the ones most vulnerable to climate change mitigation policies. The remaining stocks in the index are then optimally re-weighted in order to minimise the tracking error with respect to the benchmark. As a result, the only significant difference in aggregate risk exposure between the benchmark and the decarbonised index is with respect to carbon risk. The authors show that their approach allows the tracking error to be almost eliminated, while simultaneously achieving a substantial reduction in the exposure to carbon risk. In fact, as long as carbon reduction policies are pending, the decarbonised index achieves returns comparable to the reference index. However, once regulatory risks materialise (ie more stringent carbon reduction policies are introduced, or are expected to be introduced) the ‘green’ index is bound to outperform its benchmark.
Alok, S., Kumar, N., and Wermers, R. (2020) “Do fund managers misestimate climatic disaster risk?” Review of Financial Studies, 33(3): 1146-1183.
Andersson, M., Bolton, P., and Samama, F. (2016) “Hedging climate risk.” Financial Analysts Journal, 72(3): 13-32.
Black, F. and Scholes, M. (1973) “The pricing of options and corporate liabilities.” Journal of Political Economy, 81: 637-654.
Choi, D., Gao, Z., and Jiang, W. (2020) “Attention to global warming.” Review of Financial Studies, 33(3): 1112-1145.
Engle, R.F., Giglio, S., Kelly, B., Lee, H., and Stroebel, J. (2020) “Hedging climate change news.” Review of Financial Studies, 33(3): 1184-1216.
Merton, R.C. (1973) “Theory of rational option pricing.” Bell Journal of Economics and Management Science, 4: 141-183.
Tversky, A. and Kahneman, D. (1973) “Availability: A heuristic for judging frequency and probability.” Cognitive Psychology, 5(2): 207-232.