Policy Shocks and Stock Market Returns Evidence from Chinese Solar Panels Meredith Crowley & Huasheng Song University of Cambridge & Zhejiang University September 2015 MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 1 / 32
Introduction: Chinese Solar Panels and Policy Shocks MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 2 / 32
Introduction: Growth of Yingli Solar’s Output (Megawatts) Grey = Wafers, Orange = PV Cells, Yellow = PV Modules MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 3 / 32
Chinese Solar Panels and Policy Shocks: Why Care? In 2011, China’s share of the EU market for solar panel modules hit 80%. In 2012, China exported e 21 billion in solar panel products to the EU. Chinese solar panels comprised about 7% of total Chinese exports to the EU. In July 2012, a German firm filed an antidumping petition claiming that Chinese firms were pricing their products unfairly and should be subject to antidumping tariffs. As the EU’s antidumping case proceeded over 2012-2013, Chinese solar panel producers were hit with a series trade policy and domestic industrial policy shocks. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 4 / 32
What can we learn from the Chinese solar panel case? The basic facts: 1. The Chinese solar panel industry is large and diverse, comprised of private firms and State Owned Enterprises. 2. Access to financing apparently varies across firms, with Chinese solar panel firms listed in the Hong Kong, Shanghai-Shenzhen, and New York stock markets. 3. The EU antidumping process is characterized by scheduled announcements of tariff increases and/or quota restrictions for “investigated” products. 4. During the EU’s antidumping investigation, the Chinese government announced two policies regarding the development of the Chinese solar panel industry. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 5 / 32
What can we learn from the Chinese solar panel case? The important questions: 1. Do firms that produce the same product experience the same change in value in response to a demand shock? 2. What accounts for the heterogeneity of abnormal returns across firms experiencing the same event? 3. What can we deduce about the effectiveness of stock markets in guiding resource allocation in China? MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 6 / 32
What do we do in this paper? We estimate the abnormal returns of Chinese firms that are publicly listed in three different stock markets: Shanghai-Shenzhen, New York, and Hong Kong. We find that the abnormal returns vary by labor productivity, export share, the market in which a firm lists, a firm’s size and corporate structure, and a firm’s position on the value chain of production. The punchline: The EU’s import restrictions on Chinese firms had a negative impact on the profitability of private sector firms, especially those which listed in New York, but had no effect on China’s publicly listed State Owned Enterprises. The Chinese government policies benefited firms listed in New York, but had almost no impact on publicly listed State Owned Enterprises. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 7 / 32
Background: European trade policy announcements The events of the EU’s antidumping case: 2012-2013 Table 1: Events in the Solar Panel Market, 2012-2013 Event Date Description Petition 24 Jul. 2012 EU PV firms filed petition for AD protection against Chinese imports Preliminary Ruling 4 Jun. 2013 Provisional AD duty announced Development Guideline 15 Jul. 2013 Guideline announced by the State Council of China Amendment 2 Aug. 2013 Provisional AD duty amended to voluntary quota Subsidy Scheme 30 Aug. 2013 National Development and Reform Commission announced the solar panel subsidy scheme Final ruling 2 Dec. 2013 Application of voluntary quota & import tariffs MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 8 / 32
Background: The EU market for solar panels Table 2: Evolution of the EU solar panel market: 2009-2012 Indicator 2009 2010 2011 IP Module 100 251 462 408 Import volume index Cell 100 303 554 582 Wafer 100 551 926 748 Module 63% 71% 80% 80% Market share Cell 8% 16% 22% 25% Wafer 6% 22% 32% 33% Module 100 79 64 36 Price index Cell 100 73 70 58 Wafer 100 73 73 60 Source: Commission Regulation (EU) No 513/2013 of 4 June 2013. This document describes the analysis performed by the EC in its preliminary investigation into the allegation of dumping by Chinese firms. Tables 1-a, 2-a, 3-a, 4-a, 5-a, and 7-a of the Commission’s report display data in physical units of megawatts and e per kilowatt as well as indices based in 2009. These underlying data were collected by Europressedienst, an independent consultancy employed by the European Commis- sion. The authors reorganized the data reported in Commission Regulation (EU) No 513/2013 to make this table. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 9 / 32
Data We construct a sample of 37 publicly-listed Chinese producers of photo voltaic (PV) products. 18 firms are listed in the Shanghai-Shenzhen stock market. 11 are listed in New York. 8 are listed in Hongkong. These sample are among the largest PV producers in China. Stock price information comes from Wind, WRDS, and CRSP. Information on assets, employment, revenues, age, leverage, R&D, and products were collected from the annual reports of each firm. Information on the EU trade policy investigation were collected from the Official Journal of the European Union . Information on Chinese industrial policy announcements were collected from Chinese government agencies. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 10 / 32
Data: Summary statistics of sample firms Table 3: Summary Statistics Export R&D Market Statistics Assets ∗ Emp. Revenue ∗ Age Leverage Share Intensity CN mean 25 8170 9.8 14.2 0.589 .309 .0334 sd 46 9747 15 6.31 0.172 .304 .0186 HK mean 10 3018 3.7 9.25 0.504 . .025 sd 18 5046 6.2 5.86 0.225 . .0277 US mean 16 9680 7.9 8.5 0.770 .765 .0196 sd 10 5903 4.9 3.25 0.139 .165 .0133 Total mean 19 7475 7.9 11.4 0.622 .472 .0275 sd 34 8200 11 6.05 0.201 .342 .0199 ∗ in billions of Chinese renminbi Notable points: US-listed firms are younger than China-listed firms. US-listed firms are larger by employment than China-listed firms. Hong Kong-listed firms are smallest by employment. China-listed firms are largest by revenues and assets. US-listed firms have a higher export share. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 11 / 32
Figure 2: The Value-Chain of China-listed Firms Key: Red blocks indicate the main sales activity; blue blocks indicate production line activity. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 12 / 32
Figure 3: The Value-Chain of US-listed Firms Key: Red blocks indicate the main sales activity; blue blocks indicate production line activity. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 13 / 32
Figure 4: The Value-Chain of Hong Kong-listed Firms Key: Red blocks indicate the main sales activity; blue blocks indicate production line activity. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 14 / 32
Model: Estimating Abnormal Returns (MVRM) t + 4 ∑ (1) R it = α i + β i R mt + θ is D s + ǫ it s = t − 2 where R it = the return on firm i ’s security α i = intercept β i = systematic risk of firm i ’s security R mt = market return D s = dummy variable equal to one on dates s around the event date θ is = excess return for stock i on date s ǫ it = regression residual for security i in t MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 15 / 32
Model: Estimating Abnormal Returns, market model (2) R it = α i + β i R mt + ǫ it where R it = the return on security i on day t α i = the intercept β i = systematic risk of security of period t R mt = market return ǫ it = regression residual for security i in t MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 16 / 32
Model: Estimating Abnormal Returns From regression (2) we obtain the expected or predicted return, E ( R it ) . Then the abnormal return, AR it , is calculated as the difference between the observed return and the predicted return: (3) AR it = R it − E ( R it ) The cumulative abnormal return (CAR) for firm i during the event window ( − k , + l ): + l ∑ (4) CAR i = AR it t = − k We construct the CAR for each firm in our sample for each of the events in the EU’s antidumping investigation and for the Chinese government’s policy announcements. MC & HS (Cambridge and Zhejiang) Policy Shocks and Stock Market Returns September 2015 17 / 32
Recommend
More recommend
