Physical Climate Risk and Foreign Direct Investment: Is China Different?

This study deploys newly available data to examine the exposure of multinational companies’ overseas investments to physical climate risk. Globally, we find that foreign investment in the agriculture and mining sectors is most associated with physical risk. We also examine China, as it is fast becoming one of the largest centers of both inward and outward foreign investment across the globe. We find that foreign facilities located in China are associated with higher hurricanes and typhoon risk than their domestic counterparts in China. For Chinese firms operating abroad, we find that China’s overseas facilities are associated with higher water stress, floods, and hurricanes & typhoon risks across host countries, compared with non-Chinese companies. Within host countries, however, climate risks of Chinese facilities are comparable to that of non-Chinese facilities.

Faced with these threats, governments, companies, investors, and communities are becoming more active in addressing climate risks in recent years 20 . For instance, France started to require public companies to disclose climate risk information since 2016 21 . The Task Force on Climate-related Financial Disclosures (TCFD) was established in 2015 to improve and increase reporting of climate-related financial information 22 . The Network for Green the Financial System was established in 2017 to share best climate risk management practices among central banks and supervisors on a voluntary basis 23 . The 2020 version of the Equator Principles has incorporated climate risk assessment into its guidelines and called for climate-resilient infrastructure 24 . The Green Investment Principles (GIP) for the Belt and Road also encourage projects to disclose climate risks under the TCFD guidelines 25 . Also, institutional investors are urging companies to disclose and reduce climate risks [26][27][28][29] .
Despite the importance and increasing awareness of climate risk, little is known about firms' physical climate risk and how firms incorporate the risk into their business strategies. Significant work has been conducted to date focusing on transitional climate risk -that is, risk arising from policy transitions to a low-carbon economy, such as regulatory risk, technology risk, market risk, and reputational risk [30][31][32][33][34][35][36] . It is surprising that few studies focus on firms' physical climate risk [37][38] . This paper represents an initial foray into the neglected research area and examines the physical climate risk of multinational companies through the lens of overseas investment. Facility investment is an important business strategy, and at the same time one method firms take in response to the actual or expected physical climate and its effects 39 . We use newly available physical climate risk data of public companies made available from Four Twenty Seven (a Moody's affiliate). We assess the climate exposure of firms' overseas facilities across the industries, host countries, and headquarter countries. Also, we pay attention to Chinese overseas investments. This is because 1) China is now a major investor in least-developed countries and in developing Asia 40,41 ; and 2) Chinese overseas investments are perceived as qualitatively unique 42 : they may not be solely profit-seeking 43 , and are frequently criticized for bringing negative social and environmental impacts while helping local economic development 44 . Figure S.1 in the Supplement Information summarizes the framework of the study. We use the term country and jurisdiction interchangeably in this paper for simplicity.

RESULTS
In this study, we start by documenting a series of stylized facts pertaining to the physical climate risks of multinational companies' overseas facilities across the globe. First, we find that physical climate risks of firm's overseas facilities vary by industry, with agriculture and mining industries having the highest aggregate physical climate risk.

Global Landscape of Physical Climate Risks of Public Companies' Overseas Facilities
Physical climate risk by industry. Figure 1 summarizes the climate risk scores of firms' overseas facilities by industry according to the SIC groups. On average, agriculture and mining industries have the highest aggregate climate risk, while public administration sector has the lowest climate risk. Specifically, agriculture, forest, and fishing industry have the highest heat stress risk; manufacturing industry has the highest water stress; mining industry has the highest floods risk; and whole trade industry has the highest sea level rise and cyclone risks. These make sense as resource-intensive sectors such as agriculture, mining, and manufacturing are more directly affected by chorionic risks 41 such as heat and water stresses, while trade and transportation sectors are more directly affected by sea level rise and cyclone risks as their assets are usually close to seaports.      Analysis is based on physical climate risk scores and facility statistics of 2,233 public companies from Four Twenty Seven. FDI outflow stocks are based on the World Bank data.     rise risks. This is probably because climate risks of facilities in China are high (except for heat stress), as shown in Figure 2.b. Also, with the exception of water stress, the results are consistent with the global average as presented in Table S.3 of Supplementary Information. China. There are a couple of reasons that may explain why Chinese overseas facilities are located in jurisdictions with higher climate risks. One possible reason is that facilities in China have high climate risks in general (see Figure 2.b), and thus companies are willing to take higher than average climate risks when investing overseas. As suggested in Table   3, although the average climate exposures of Chinese overseas facilities are high, they are lower than that of facilities located in China, except for heat stress and sea level rise.
Another possible factor contributing to the difference is that some Chinese companies may not be solely profit-seeking 45 , and they may be willing to invest in some countries for political or strategic reasons, regardless of their climate risks. It is also possible that Chinese companies have to invest in some undesirable locations with higher physical climate risk exposure because the desirable locations have been taken by earlier investors or China is investing in strategic assets for non-commercial reasons 47,48 .
This study has several contributions. To the best of our knowledge, this is the first To assess the difference of physical climate risks of overseas facilities owned or operated by Chinese companies within countries, I estimate Model 2 for different physical climate risk drivers: where i indexes firm, j indexes industry, c indexes host country, h indexes headquarters country. Return on Assets (ROA) is the ratio of operating income before depreciation to the book value of total assets. Leverage is the ratio of debt (long-term debt plus short-term debt) to the book value of total assets. Cash holding is the ratio of cash and short-term investments to the book value of total assets. We also include GDP per capita of headquarters countries as a control variable as headquarters countries' economic development level may have impact on firms; overseas location choice.
We also test the difference of physical climate risks of facilities located in China owned or operated by non-Chinese companies using the following specification.
ClimateRisk ij = ! + 3 × + " #$ + #!$ (3) where i indexes firm, j indexes industry, h indexes headquarters country. α j are industry fixed effects. The coefficient of interest is β3, which measures the relationship between non-Chinese ownership and physical climate risks of facilities located in China.
In addition, we explore whether the physical climate risks of overseas facilities owned and operated by Chinese companies are different from their assets located in China using the following regression.
ClimateRisk ij = ! + 4 × + " # + #! (4) where i indexes firm, j indexes industry. α j are industry fixed effects. The coefficient of interest is β4, which measures the relationship between overseas facilities and physical climate risks of facilities owned or operated by Chinese companies.