Stochastic framework for carbon price risk estimation of real estate: a Markov switching GARCH simulation approach

DOIhttps://doi.org/10.1108/JPIF-12-2021-0104
Published date14 February 2022
Date14 February 2022
Pages381-397
Subject MatterProperty management & built environment,Real estate & property,Property valuation & finance
AuthorCay Oertel,Ekaterina Kovaleva,Werner Gleißner,Sven Bienert
Stochastic framework for carbon
price risk estimation of real estate:
a Markov switching GARCH
simulation approach
Cay Oertel and Ekaterina Kovaleva
IntReal, Hamburg, Germany
Werner Gleißner
FutureValue Group AG, Leinfelden-Echterdingen, Germany, and
Sven Bienert
International Real Estate Business School, University Regensburg,
Regensburg, Germany
Abstract
Purpose The risk management of transitory risk for real assets has gained large interest especially in the
past 10 years among researchers as well as market participants. In addition, the recent regulatory tightening in
the EU urges financial market participants to disclose sustainability-related financial risk, without providing
any methodological guidance. The purpose of the study is the identification and explanation of the
methodological limitations in the field of transitory risk modeling and the logic step to advance toward a
stochastic approach.
Design/methodology/approach The study reviews the literature on deterministic risk modeling of
transitory risk exposure for real estate highlighting the heavy methodological limitations. Based on this, the
necessity to model transitory risk stochastically is described. In order to illustrate the stochastic risk modeling
of transitory risk, the empirical study uses a Markov Switching Generalized Autoregressive Conditional
Heteroskedasticity model to quantify the carbon price risk exposure of real assets.
Findings The authors find academic as well as regulatory urgency to model sustainability risk
stochastically from a conceptual point of view. The own empirical results show the superior goodness of fit of
the multiregime Markov Switching Generalized Autoregressive Conditional Heteroskedasticity in comparison
to their single regime peer. Lastly, carbon price risk simulations show the increasing exposure across time.
Practical implications The practical implication is the motivation of the stochastic modeling of
sustainability-related risk factors for real assets to improve the quality of applied risk management for
institutional investment managers.
Originality/value The present study extends the existing literature on sustainability risk for real estate
essentially by connecting the transitory risk management of real estate and stochastic risk modeling.
Keywords Sustainability, Monte Carlo simulation, Markov chain, Real estate risk management, Carbon
pricing, Generalized autoregressive conditional heteroskedasticity
Paper type Research paper
Introduction and recent regulatory tightening in the EU
The issue of ecological sustainability and the transition of real assets toward decarbonization
have recently experienced a heavy increase in the public as well as academic interest. Since
real estate accounts for approximately 29% of the overall greenhouse gas offset in the EU, the
building stock is highly relevant for the abovementioned transition (Hirsch et al., 2020). In line
with this development, institutional real estate investors are now facing the risk exposure for
property investments arising from transitory risk. For the transitory risk exposure, the
Carbon Real Estate Risk Monitor (CRREM) has been the central analytical, yet deterministic,
framework to identify the risk of stranding and ensuing carbon costs of real assets that are
not aligned with Paris Climate Accord targets (Hirsch et al., 2020). The Paris Climate Accord
Stochastic
framework for
carbon price
risk
381
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/1463-578X.htm
Received 1 December 2021
Revised 26 January 2022
Accepted 27 January 2022
Journal of Property Investment &
Finance
Vol. 40 No. 4, 2022
pp. 381-397
© Emerald Publishing Limited
1463-578X
DOI 10.1108/JPIF-12-2021-0104
formulates decarbonization targets, which are ambitious especially for the existing building
stock, because corresponding greenhouse gas targets aim at reductions of 40% by 2030
compared to the 1990s. Thus, carbon price risk exposure is highly relevant to real assets [1].
In addition to the Paris Climate Accord, the European Commission published the EU
Sustainable Finance Action Plan as central regulatoryinstrument to build a comprehensive
legislative framework explicitly for financial markets in March 2018 [2]. Essentially, the
specified plan aims at the redirection of capital flows toward ecologically sustainable
investment products, the management of financial risk arising from climate change as well
as increasing the transparency of climate change related factors, which affect the value of
investment positions (European Commission, 2018). Based on these fundamentals of this
Action Plan, the EU concretized necessary actions of financi al market partici pants
(FMP) and advisers, by clarifying sustainability-related disclosure and fiduciary duties.
The EU communicated these clarifications in the Sustainable Finance Disclosure
Regulation (SFDR).
In the context of the SFDR, FMPs are among others defined as firms offering commercial
portfolio management of a financial product. Typical examples for affected FMPs in the real
estate industry, which manage large capital stocks, are alternative investment fund firms. These
investment firms are accordingly obliged to disclose information about the actual implementation
of their sustainability risk management, especially in the investment process (Art. 3 (1) SFDR).
Interestingly, the SFDR in fact provides an own definition of the term sustainability risk,by
describing it as environmental, society or governance event or condition if it occurs, could cause an
actual or a potential material negative impact on the value of the investment(Art. 2 (22) SFDR).
Based on this definition, FMPs are thus facing the actual challenge to translate the Plans
fiduciary duties into applied methodology. For alternative investments such as real estate, risk
modeling is generally challenging and, both, highly data-sensitive and -dependent (Oertel, 2018).
The data scarcity and methodological uncertainty surrounding new risk factors in the field of
sustainability increase the challenge to implement risk management systems even more.
Nonetheless, concerning the SFDRs definition of sustainability risk including the terms ifand
couldas well as potential negative impactis in line with a quantitativeapproach to model the
risk by including uncertainty (ifand could)using probability distributions and monetary risk
exposure (potential negative impact). Thus, the stochastic modeling of the risk factor appears to
be the only methodological approach to satisfy the condition of economically valuable and legally
meaningful applied risk management. Accordingly, the central research question for the present
paper can be summarized as such: How can carbon risk exposure be modeled stochastically for
real estate investment positions?
Essentially, the article aims at contributing to the existing body of literature in two ways:
Firstly, we outline the general idea of a stochastic approach to sustainability risk of real estate
investment positions. Secondly, we illustrate the superiority of multiregime models to
account for structural breaks in the volatility process of the accompanying sustainability-
related risk exposure. Therefore, we chose the carbon price risk as prominent example to
illustrate the quantitative risk modeling of sustainability-related risk factors for real estate. In
order to provide insights, the structure of the article can be summarized as such: The next
chapter reviews the literature on carbon price risk modeling of the chosen EU ETS price data
in general. The following section describes the current state of the literature on quantitative
risk management for real estate. The ensuing section describes the methodology to model the
carbon price risk, including the chosen empirical model to parameterize and simulate from.
Then, the article presents the data and the univariate analysis. We then report the results of
the empirical calibration of the model and the simulation results for the carbon risk exposure.
The final section concludes, names limitations of the study and outlines potential further
research.
JPIF
40,4
382

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT