ASSET DIVERSIFICATION VERSUS CLIMATE ACTION

Published date01 August 2024
AuthorChristoph Hambel,Holger Kraft,Frederick van der Ploeg
Date01 August 2024
DOIhttp://doi.org/10.1111/iere.12691
INTERNATIONALECONOMIC REVIEW
Vol. 65, No. 3, August 2024 DOI: 10.1111/iere.12691
ASSET DIVERSIFICATION VERSUS CLIMATE ACTION∗
By Christoph Hambel, Holger Kraft, and Frederick van der Ploeg
Tilburg University, The Netherlands; Goethe University, Germany; University of Oxford, U.K.
Asset pricing and climate policy are analyzed in a global economy where consumption goods are pro-
duced by both a green and a carbon-intensive sector. Given that the economy is initially heavily dependent
on carbon-intensive capital, the desire to diversify assets complements the attempt to mitigate economic dam-
ages from climate change. In the longer run, however, a trade-off between diversif‌ication and climate action
emerges. Wederive the optimal carbon price and the equilibrium risk-free rate, and risk premia. Climate disas-
ters signif‌icantly decrease the risk-free rate but increase risk premia on f‌inancial assets, especially if no climate
policy is implemented.
1. introduction
Climate change impacts all areas of human life and impacts economic activity.1To avoid
carbon-dioxide emissions and climate change, emissions-free technologies and renewable en-
ergies are developed. Depending on the perceived severity of the consequences of climate
change, there are different opinions about how urgent it is to transition to a less carbon-
intensive or carbon-free economy. We are interested in the interplay between f‌inancial consid-
erations and policies to mitigate climate change and answer three key policy questions. First,
does the f‌inancial need to diversify assets hamper or help the f‌ight against climate change and
how does it affect optimal carbon pricing? Second, how does climate change and the desire to
combat it affect the risk-free interest rate and precautionary saving? We highlight that there
is a subtle dynamic interdependence between the f‌inancial goal to diversify assets in portfo-
lios and the environmental goal to reduce carbon emissions. Third, we allow for a falling cost
of renewables driven by learning by doing (LBD) during the green transition, and examine
whether the key insights still hold.
∗Manuscript received September 2022; revised November 2023.
We thank Daniel Andrei, Patrick Bolton, Julia Braun, Jesus Fernandez-Villaverde (the editor), Ruediger Kiesel,
Thomas Lontzek, Colin Mayer, Stavros Panageas, Armon Rezai, Suphi Sen, Eduardo Schwartz, Gerhard Stahl,
Christian Traeger, Frank Venmans, and the participants of the Finance seminar at McGill, seminars at Humboldt
University, the University of Lausanne, EARES, University of Brussels, Paris School of Economics, University of
Western Australia, the Banca d’Espana, the Bundesbank and de Nederlandsche Bank, the European Finance As-
sociation (EFA) Meeting 2020, the European Economics Association (EEA) Meeting 2020, the SURED Meeting
2020, the EAERE Meeting 2020, the EBI Global Annual Conference 2020, the German Finance Association Meeting
2021, and the Verein fuer Socialpolitik (VfS) Meeting 2021, the DEARE Meeting 2023, the DVfVM meeting 2023,
as well as three anonymous referees for very helpful comments and suggestions. All remaining errors are our own.
Christoph Hambel and Holger Kraft gratefully acknowledge f‌inancial support by the Deutsche Forschungsgemein-
schaft (DFG). Please address correspondence to: Christoph Hambel, Tilburg University, Department of Economet-
rics and Operations Research, P.O.box 90153, 5000 LE Tilburg, TheNetherlands. Phone: +31 (0) 13 466 8908. E-mail:
c.hambel@tilburguniversity.edu.
*Vander Ploeg is also aff‌iliated with the University of Amsterdam, CEPR, and CESifo.
1See, for example, Scheffers et al. (2019).
1323
© 2024 The Authors. International Economic Review published by Wiley Periodicals LLC on behalf of the Economics
Department of the University of Pennsylvania and the Osaka University Institute of Social and Economic Research
Association.
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs Li-
cense, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-
commercial and no modif‌ications or adaptations are made.
1324 hambel, kraft, and van der ploeg
Our economic framework is a stochastic macroeconomic growth model with two capital
stocks and two energy sources. The green sector takes carbon-free or green energy as input.
The brown sector is carbon-intensive and requires fossil fuel whose combustion leads to car-
bon emissions. Although our contribution is more methodological than empirical, the brown
sector includes coal-f‌ired power stations, steel, cement, aluminum, fossil-based transport (in-
cluding sea- and airports), and the oil, gas, and coal sectors. The green sector includes the sec-
tors that do not depend excessively on oil, gas, or coal but depend on solar or wind energy.
There are two types of capital stocks, brown and green. Investments and capital realloca-
tion from the brown to the green capital stock are both subject to intertemporal and intersec-
toral adjustment costs. Capital stock accumulation is exposed to diffusive shocks as well as the
risk of macroeconomic disasters. We have a two-sector endogenous growth (AK) model and
abstract from directed technical change toward green technologies in the spirit of Bovenberg
and Smulders (1996), Acemoglu et al. (2012), and Casey (2024), but we do allow for LBD in
only the production of the green good which thus captures some features of directed techni-
cal change. Emissions are proportional to fossil fuel use and temperature is driven by cumula-
tive emissions.2We allow for two effects of climate change on economic activity: the standard
one is that higher temperature leads to a higher share of damages in predamage output as in
the seminal DICE-2016R2 model (e.g., Nordhaus 2017); the novel one is that higher tempera-
tures increase the Poisson risk of climate-related disasters (cf. Bansal et al., 2019; Karydas and
Xepapadeas, 2022).
We f‌irst analyze the interplay between the intensity of climate action proxied by the carbon
price and the economic motive to diversify. Given that the economy is initially dominated by
the brown capital stock, there are two complementary goals. The f‌irst is to mitigate climate
change by decarbonizing the economy. The second is to diversify the economy. Both goals re-
quire decision makers to initially reduce the brown capital stock. The speed of transition to-
ward a low-carbon economy is thus initially amplif‌ied by the diversif‌ication motive. Over time,
however, the two goals start to conf‌lict. From a diversif‌ication perspective, the process should
be stopped if there is a balance between green and brown capital. But from an environmen-
tal perspective the brown capital stock should eventually be run down completely. Our vari-
ous calibrations show, however, that this does not occur unless damages from global warming
are extremely severe relative to those in the DICE model of Nordhaus (2017). Effectively, cli-
mate policy drives the brown capital stock below the fully diversif‌ied level; diversif‌ication con-
siderations may prevent policymakers from driving it to zero.
Second, we investigate the interplay between climate change and asset prices. We focus on
the risk-free rate and the precautionary savings motive during the transition from a carbon-
intensive toward a low- or zero-emissions economy. To separate economic from climate ef-
fects, our model includes the risk of macroeconomic disaster shocks as in Barro (2006, 2009)
and Pindyck and Wang (2013). Hence, our model can generate the high equity premium and
low risk-free rate observed in historical data even if climate change has no signif‌icant impact
on the economy. Taking the effects of climate change into account, we f‌ind that the risk-free
interest rate decreases in response to rising temperatures ref‌lecting the additional precaution-
ary savings needed to cope with global warming.3
Third, we allow for falling costs of renewable energy as a result of LBD (also known as
Swanson’s or Wright’s law). This requires both a carbon tax to internalize the global warm-
ing externalities and a renewable energy subsidy to internalize the LBD externalities. The
green transition now occurs more quickly, but the diversif‌ication objectives frustrating the lat-
ter phases of the green transition still show up.
2See Matthews et al. (2009), Allen et al. (2009), IPCC (2014), van der Ploeg (2018), and Dietz and Venmans
(2019), among others, for further references.
3Risk premia are only signif‌icantly affected if we allow for potential climate disasters for which the probability of
them occurring increases with temperature. Then, the risk premia increase with higher temperature. Without such
temperature-dependent disasters, this effect is modest.
asset diversification versus climate action 1325
Our integrated assessment model (IAM) enriches an asset pricing framework with rare
macroeconomic disasters as in Barro (2006, 2009) or Wachter (2013) by adding the climate
module of an IAM. Our representative agent has recursive utility as in Bansal and Yaron
(2004) and Pindyck and Wang (2013). If the sizes of the two sectors would be exogenous and
the effect of climate change is disregarded, the two-tree model analyzed in Cochrane et al.
(2007) arises as a special case. In our IAM, we have a stochastic two-sector growth model with
temperature driven by cumulative emissions.
Our two-sector IAM thus differs from one-sector IAMs which do not focus on asset pric-
ing. For example, the DICE model has one sector and is widely used to study optimal car-
bon abatement and carbon pricing. It combines a Ramsey-type model for capital allocation
with deterministic dynamics of the economy, emissions, carbon dioxide, and global tempera-
ture (e.g., Nordhaus 1992, 2017). Crost and Traeger (2014), Jensen and Traeger (2014), Ack-
erman et al. (2013), Bretschger and Vinogradova (2019), van den Bremer and van der Ploeg
(2021), Hambel et al. (2021a), and Lemoine (2021) use recursive utility to analyze stochastic
models of the economy and the climate. Cai and Lontzek (2019) extend the DICE model with
stochastic growth and the risk of tipping points. Golosov et al. (2014) obtain closed-form so-
lutions in a one-sector framework with log utility, Cobb–Douglas production, full depreciation
in one discrete time period, and damages that are an exponential function of the atmospheric
carbon stock. Traeger (2023) generalizes this setting to recursive preferences and provides a
description of the carbon cycle and the climate system. He also allows for epistemological un-
certainty and anticipated learning.
Few papers combine asset pricing with an IAM.4Barnett et al. (2020) analyze a stochas-
tic one-sector macroeconomic dynamic stochastic general equilibrium (DSGE) model with
stochastic economic growth rates and endogenous investments in fossil fuel reserves. They
also address preference-based concerns about ambiguity and model misspecif‌ication. Barnett
(2023a) uses an extended Fama–French three-factor model to estimate negative effects of
temperature on excess returns for brown portfolios and positive effects for green portfolios.
He also derives optimal climate policy from a two-sector DSGE model. Whereas this study
in contrast to our analysis allows for model uncertainty, we allow for macroeconomic and cli-
matic disasters.
Bansal et al. (2017, 2019) quantify the impact of local temperature on asset prices. They
study a global long-run risk model that simultaneously matches the observed temperature and
consumption growth dynamics. Furthermore, their model can generate a low risk-free interest
rate and a high equity premium (cf. Donadelli et al., 2017). Karydas and Xepapadeas (2022)
study an economy with two assets, but focus on a Lucas-tree endowment economy where the
agent cannot actively control the transition to a low-carbon economy. Finally, Daniel et al.
(2019) show in an asset pricing setting that recursive preferences with general resolution of
uncertainty about climate change can lead to a declining carbon price.5
Section 2 introduces our IAM. Section 3 characterizes the social optimum and discusses
how the social optimum can be decentralized in the market economy. Section 4 discusses the
diversif‌ication motive. Section 5 discusses our calibration strategy. Section 6 presents our re-
sults on the relation between the diversif‌ication motive and climate action. Section 7 discusses
how climate change affects the equilibrium risk-free rate, the precautionary savings motive,
and risk premia. Section 8 concludes. The appendices give proofs, calibration details, and re-
sults on risk premia.6
4van den Bremer and van der Ploeg (2021) study a one-sector production economy with endogenous climate
change and a wide range of economic and climatic uncertainties that generates low risk-adjusted interest rates and
a high risk premium. Like Lemoine (2021), they f‌ind that the effect of covariance between climate shocks and econ-
omy shocks depends on whether the coeff‌icient of relative risk aversion exceeds one or not. Dietz et al. (2018) def‌ine
the “climate beta” to be the elasticity of damages with respect to economy activity and show that the optimal carbon
price increases in this beta.
5The feature of a declining carbon price is derived from a binomial tree with a f‌ixed horizon.
6For additional material such as the solution approach and a battery of robustness checks, see http://christoph-
hambel.de/wp-content/uploads/2024/01/HKvdP-IER- 2024-Online-Appendix.pdf

Get this document and AI-powered insights with a free trial of vLex and Vincent AI

Get Started for Free

Unlock full access with a free 7-day trial

Transform your legal research with vLex

  • Complete access to the largest collection of common law case law on one platform

  • Generate AI case summaries that instantly highlight key legal issues

  • Advanced search capabilities with precise filtering and sorting options

  • Comprehensive legal content with documents across 100+ jurisdictions

  • Trusted by 2 million professionals including top global firms

  • Access AI-Powered Research with Vincent AI: Natural language queries with verified citations

vLex

Unlock full access with a free 7-day trial

Transform your legal research with vLex

  • Complete access to the largest collection of common law case law on one platform

  • Generate AI case summaries that instantly highlight key legal issues

  • Advanced search capabilities with precise filtering and sorting options

  • Comprehensive legal content with documents across 100+ jurisdictions

  • Trusted by 2 million professionals including top global firms

  • Access AI-Powered Research with Vincent AI: Natural language queries with verified citations

vLex

Unlock full access with a free 7-day trial

Transform your legal research with vLex

  • Complete access to the largest collection of common law case law on one platform

  • Generate AI case summaries that instantly highlight key legal issues

  • Advanced search capabilities with precise filtering and sorting options

  • Comprehensive legal content with documents across 100+ jurisdictions

  • Trusted by 2 million professionals including top global firms

  • Access AI-Powered Research with Vincent AI: Natural language queries with verified citations

vLex

Unlock full access with a free 7-day trial

Transform your legal research with vLex

  • Complete access to the largest collection of common law case law on one platform

  • Generate AI case summaries that instantly highlight key legal issues

  • Advanced search capabilities with precise filtering and sorting options

  • Comprehensive legal content with documents across 100+ jurisdictions

  • Trusted by 2 million professionals including top global firms

  • Access AI-Powered Research with Vincent AI: Natural language queries with verified citations

vLex

Unlock full access with a free 7-day trial

Transform your legal research with vLex

  • Complete access to the largest collection of common law case law on one platform

  • Generate AI case summaries that instantly highlight key legal issues

  • Advanced search capabilities with precise filtering and sorting options

  • Comprehensive legal content with documents across 100+ jurisdictions

  • Trusted by 2 million professionals including top global firms

  • Access AI-Powered Research with Vincent AI: Natural language queries with verified citations

vLex

Unlock full access with a free 7-day trial

Transform your legal research with vLex

  • Complete access to the largest collection of common law case law on one platform

  • Generate AI case summaries that instantly highlight key legal issues

  • Advanced search capabilities with precise filtering and sorting options

  • Comprehensive legal content with documents across 100+ jurisdictions

  • Trusted by 2 million professionals including top global firms

  • Access AI-Powered Research with Vincent AI: Natural language queries with verified citations

vLex