How earthquakes affect firms risk response: Evidence from Chinese public firms

Sudden natural disasters introduce operational hazards into firms. Earthquakes are frequent natual disasters, strong and quick with clear event time. Rather than meteorological disasters like hurricanes and floods, earthquakes are more unexpected and would not last for a long period. Furthermore, destructive great earthquakes usual cause numerous assets lossof frms, bring threats to people and trigger panic in the whole economy.

Although the eathquakes are suddend events, the risk propagation on the economy is long-lasting through multiple channels. Earthquakes’s impact on firms’ stock price and total assests are not limited to a short-term, neither in the direct covered areas([valizadeh2017ripple]). From the ex-post perspective, firms with factories factories located in areas affected by natural disasters are much less profitable([hsu2018natural]). In another way, earthquake shock could also spur the rise of innovation ability ([rao2021]). To sum up, earthquakes do have comprehensive effects on firm operation and economy development.

First, I want to illustrate that a single strong earthquake would result in operation distress on the firms. The Strong Lushan Earthquake in 2013 is chosen because it is almost the most severe earthquakes in the past two decades ranked only second to the Great Wenchuan Earthquake in 2008. The magitude of Lushan Earthquake is 7.0. The direct economic loss consulted in Sichuan Province exceeds 130 billian dollars. Wenchuan Earthquake and Lushan Earthquake both happened in Sichuan province, because Sichuan province and Chongqing are located on the eauthquake-prone areas, i.e. Longmenshan Fault. Thus, the treat group is listed firms in Sichuan and Chongqing because the two city is close geographically and similarly in economic structure. Considering the similarity in industry composition of public firms, Henan Province and Hebei Province are combined to serve as the control group. The control group firms are distributed mostly in manufacturing and production industry. Besides, Henan and Hebei are also located at Taihang earthquake zone, where happend the Strong Tangshan Earthquake (magnitude 7.8) in 1976. Thus, the model is constructed as below:

$Special\ items$ usually is a large, one-time expense or income that are not expected by firms. Since no such same term in Chinese balance sheet, I used some general terms which could be high correlated to the sudden fluctuations caused by earthquakes. Specifically, $special_{\_}items$ refers to cash flow of each public firm in each year. Sales is annual total sales. The controls $X_{\_}{st}$ include $CFA,GROW$ $,LRL,ROA,ROE,$ $eq\_times$ in each year for each firm. $CFA$ is the ratio of net cash flow and total assets. $GROW$ is the ratio of retaqined earnings and net assets. $ROE$ is return on equity. $ROA$ is return on assets, reflecting the profitability. $LRL$ is the ratio of net profits and sales, i.e. profit margin of primary business. $eq_{\_}times$ is the moderate and strong earthquake times (magitude is higehr than 4.5) of the city area where the firm’s office is located at, merged from the earthquake database established before by administrative code. $\epsilon$ is a stochastic error. $\theta_{\_}i$ is firm-level fixed effect, $\delta_{\_}t$ is year fixed effect. $post_{\_}{i,t}$ takes the value of 1 after the earthquakes happen, i.e. including 2013 and after. Parameter $\tau$ captures the average impact of unexpected natural diasater. All statistics are annual.

Event study([dou2022competition]) on effect of strong earthquake happened at different time. Given the preliminary results, the effect of single earthquake on the firms is still ambiguous. Actually, I have been trying this part in both annual and semester level, altough the results is not satisfied. Risk propagation through ([hyun2019spillovers]),([chen2021regulating]). In addition to the traditional variables, the paper attempts to add variables such as node degree and network topology centrality.