The microeconomic effects of temperature changes

Highlights

• The microeconomic effects of temperature changes are non-linear.

• Adaptation actions can lead to higher costs, resulting in capital misallocation.

• High temperatures reduce labor productivity, resulting financial performance loss.

• Firms in developed regions is more sensitive to high temperatures.

Abstract

This paper uses financial data from listed Chinese manufacturing companies, combined with meteorological data, to estimate the microeconomic effects of temperature changes. By constructing seasonal average temperatures, we find that average summer temperatures significantly reduce the financial performance, while average winter, spring, and autumn temperatures do not have a significant effect on the corporate financial performance. We further find that the effect of temperature changes on corporate financial performance is nonlinear through temperature bins variables. Both extreme low temperatures (<-12 °C) and high temperatures (>27 °C) have a negative impact on corporate financial performance. The mechanism tests find that high temperatures not only reduce the labor productivity of firms, but also increase the corporate adaptation actions, leading to increased adaptation costs, crowding out productive funds and causing a capital mismatch. Further heterogeneity analysis confirms that the magnitude of the negative effect of temperature changes is limited by the enterprise nature and the region in which the firm is located. The financial performance of state-owned enterprises, firms in the south and developed regions are more vulnerable to high temperatures. This study provides a policy basis for mitigating the negative effects of temperature changes and provides new empirical evidence for climate governance.

Introduction

According to the Fifth Assessment Report of the IPCC, the global average temperature has increased by nearly 1 °C since the 21st century, which has seriously affected the normal operation of productive activities such as the work of workers and the operation of machines (Tawatsupa et al., 2010; Zivin and Neidell, 2014; Cai et al., 2018; Zhang et al., 2018; Chen and Yang, 2019). Therefore, scholars have begun to pay extensive attention to the causal relationship between temperature changes and economic output (Hsiang, 2010; Dell et al., 2012; Zhang et al., 2018; Chen and Yang, 2019). Most research has found that higher temperatures have a significant negative (Hsiang, 2010; Dell et al., 2012), or inverted U-shaped (Burke et al., 2015), effect on the total economic output of a country or region. The above findings are also confirmed in the agricultural sector (Schlenker and Roberts, 2009; Deschênes and Greenstone, 2012; Marshall and Kyle, 2016). For instance, Chen et al. (2016) and Cui (2020) conducted studies on the relationship between temperature changes and agricultural yields in China and the United States, respectively, based on setting temperature bins. They found that yields of maize and soybean began to decrease when temperatures rose to 28 °C, and that extremely high temperatures above 34 °C accelerated the rate of yield decline.

Considering the important share of manufacturing industry in the national economy of many countries (Cai et al., 2018; Zhang et al., 2018), some scholars have started to explore the impact of temperature changes on the output of manufacturing enterprises at the micro-firm level in recent years, which has revealed the microeconomic effects of temperature changes to a certain extent. To be specific, Cachon et al. (2012) constructed a temperature bin using the number of days per week above 32 °C, and used data from 64 automotive companies in the United States over the period 1994–2005 to quantify that extremely hot temperature (>32 °C) for more than six days a week would reduce weekly vehicle production by approximately 8%. In fact, in addition to the extremely hot temperature that can affect the output of manufacturing companies, the impact of low temperature cannot be ignored. Zhang et al. (2018) used the data from Chinese manufacturing scale firms from 1998 to 2007 and found an inverted U-shaped relationship between temperature and corporate total factor productivity in a quantitative study by setting a temperature bin between −12 °C and 32 °C of 5.5 °C wide. Based on the above methods, Chen and Yang (2019) further controlled sunshine duration to mitigate endogeneity problems caused by the omission of other climatic factors in the model, and again confirmed a non-linear relationship between temperature changes and corporate output.

Our paper has different findings on the microeconomic effects of temperature changes. Specifically, firstly, much of the existing literature has focused on the results of the impact of temperature changes on the output of manufacturing companies, although some scholars have found significant heterogeneity of the impact of high temperatures on the output of capital-intensive and labor-intensive companies, revealing to some extent that the impact of high temperature on the output of companies is constrained by labor (Cai et al., 2018). The existing research failed to reveal how capital allocation changes in the face of temperature variations, and further did not quantitatively confirm the above theoretical mechanisms. Therefore, in this paper, we take a look at the factors of production, such as labor and capital, to explore the mechanisms inherent in the microeconomic effects of temperature changes.

Secondly, companies usually take adaptation actions such as welfare subsidies and installation of cooling equipment when faced with extremely hot temperatures (Cai et al., 2018). In fact, considering that state-owned enterprises are stronger in implementing adaptation actions compared to non-state-owned enterprises, existing studies found that temperature changes have a much smaller impact on the output of state-owned enterprises through the heterogeneity analysis, which only revealed the impact of adaptation actions on the estimated results indirectly (Cai et al., 2018). If the adaptation actions are not controlled in the econometric model, it may underestimate the negative impact of the microeconomics of temperature changes. Our paper then uses the cost of electricity and water and welfare as proxy variables for the measure of adaptation actions, and constructs a regression to test whether the implementation of adaptation actions by firms has an impact on the benchmark results.

Finally, some research has attempted to mitigate the endogeneity problem in econometric models by controlling other climatic factors like sunshine duration, as well as year and firm fixed effects. Research still neglected city-level factors that affect both temperature and microeconomic effects of temperature changes. For instance, the size of population, level of economic development, and intensity of environmental regulations would also cause a bias in the estimation results. More crucially, taking into account that variations in a company's output affect its CO₂ emissions and thus affect local temperature changes (Ang, 2007), there may also be an inverse causal relationship on the microeconomics effects of temperature changes. Therefore, the endogeneity problem in the existing literature remains to be overcome. Based on this, our paper further mitigates the endogeneity problem in the econometric model by controlling the relevant variables at the city level. We also use the forest growth rate measured by the existing temperature-forest growth model as an instrumental variable for temperature with the Two-stage least square method.

Our results can inform the development of climate governance policies. In fact, we find that temperature changes have a negative effect on corporate financial performance. Although the existing literature suggests that implementing adaptation actions are effective in enhancing employee workforce productivity and reducing the negative effects of temperature changes, our study suggests that too many adaptation actions increase corporate adaptation costs and cause capital misallocation. State-owned enterprises, in particular, often have to pay more for adaptation costs and are more severely affected by temperature changes. Therefore, local governments should formulate corresponding adaptation policies and provide subsidies for enterprises to reduce the cost of adaptation and maintain the healthy development of local enterprises. However, since regional adaptation policies are often constrained by the financial situation of local governments, to mitigate the long-term impact of temperature changes on the corporate financial performance, the central government should adhere to the principle of green development, promote industrial restructuring, and develop a clean energy structure so as to reduce greenhouse gas emissions.

The rest of the paper is organized as follows. Section 2 presents the data and method. Section 3 describes the results and the discussion of the results. Section 4 presents the mechanism test and the heterogeneity analysis while Section 5 gives a conclusion and offers policy implications.