Guohao Tang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Guohao Tang has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Fuwei Jiang
Center for Macroeconomic Research, School of Economics and Wang Yanan Institute for Studies in Economics, Xiamen University, Xiamen, China
4 shared publications - Guofu Zhou
Washington University in St. Louis
3 shared publications Hanrui Xu
3 shared publicationsCunfei Liao
Hunan University - College of Finance and Statistics
2 shared publicationsXinlin Qi
Global Markets Department Industrial and Commercial Bank of China (ICBC) Beijing China
2 shared publicationsYu Jiang
2 shared publications
Latest Publications
- Open Access
- 17 November 2026
- Published by Elsevier in SSRN Electronic Journal
Beyond Policy Alignment: Residual Heterogeneity and the Cross-Section of Stock Returns in China
Almost every Chinese listed firm claims to serve the national strategy and to align its strategy with China's policy priorities. Separating the parts of a firm's strategy that echo policy language from the parts that go beyond it helps reveal whether the strategy serves those priorities and gives the firm an edge over its peers. Using annual reports of Chinese A-share firms from 2001 to 2024, we use a large language model (LLM) to score them against a common industrial policy benchmark and construct a residual heterogeneity index (RHI). Firms with high RHI earn significantly higher stock returns over the following year, and the premium is not explained by size, book-to-market, momentum, or profitability. The alphas of the high-minus-low portfolio remain significant for holding periods of up to 48 months. Similarity to policy texts does not predict returns. High-RHI firms subsequently invest more and employ a larger share of R&D personnel, and they achieve higher profitability and operating efficiency. Under the same policy direction, future returns are associated with firm-specific strategic content in disclosures rather than with similarity to policy language. Our findings show how firms create differentiated value under industrial policy and how investors distinguish substantive responses to industrial policy from rhetorical ones.
- 17 November 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessBeyond Policy Alignment: Residual Heterogeneity and the Cross-Section of Stock Returns in China
Almost every Chinese listed firm claims to serve the national strategy and to align its strategy with China's policy priorities. Separating the parts of a firm's strategy that echo policy language from the parts that go beyond it helps reveal whether the strategy serves those priorities and gives the firm an edge over its peers. Using annual reports of Chinese A-share firms from 2001 to 2024, we use a large language model (LLM) to score them against a common industrial policy benchmark and construct a residual heterogeneity index (RHI). Firms with high RHI earn significantly higher stock returns over the following year, and the premium is not explained by size, book-to-market, momentum, or profitability. The alphas of the high-minus-low portfolio remain significant for holding periods of up to 48 months. Similarity to policy texts does not predict returns. High-RHI firms subsequently invest more and employ a larger share of R&D personnel, and they achieve higher profitability and operating efficiency. Under the same policy direction, future returns are associated with firm-specific strategic content in disclosures rather than with similarity to policy language. Our findings show how firms create differentiated value under industrial policy and how investors distinguish substantive responses to industrial policy from rhetorical ones.
- Open Access
- 27 October 2026
- Published by Elsevier in SSRN Electronic Journal
The State-Dependent Pricing of Panic Risk: Evidence from China's Stock Market
This paper proposes a real-time panic index for the Chinese A-share market and uses it to price panic risk in the cross-section of stock returns. The index combines market-wide stress across prices, investor holdings, and information flows with a prospect-theory-based measure of portfolio drawdown pain. We find that exposure to panic innovations is negatively priced. The long-short spread is significant after excluding the smallest 30% of stocks and becomes substantially larger during high-panic periods. The premium accrues for at least four trading weeks and survives competing volatility, tail-risk and uncertainty betas and realistic trading costs. Following panic shocks, margin balances contract more where financing intensity is higher, consistent with a leverage channel. These findings highlight the importance of real-time information and competing short-horizon return characteristics for identifying state-dependent panic risk premia.
- 27 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe State-Dependent Pricing of Panic Risk: Evidence from China's Stock Market
This paper proposes a real-time panic index for the Chinese A-share market and uses it to price panic risk in the cross-section of stock returns. The index combines market-wide stress across prices, investor holdings, and information flows with a prospect-theory-based measure of portfolio drawdown pain. We find that exposure to panic innovations is negatively priced. The long-short spread is significant after excluding the smallest 30% of stocks and becomes substantially larger during high-panic periods. The premium accrues for at least four trading weeks and survives competing volatility, tail-risk and uncertainty betas and realistic trading costs. Following panic shocks, margin balances contract more where financing intensity is higher, consistent with a leverage channel. These findings highlight the importance of real-time information and competing short-horizon return characteristics for identifying state-dependent panic risk premia.
- Open Access
- 29 September 2026
- Published by Elsevier in SSRN Electronic Journal
The Mirror of Erised: Misperception of Foreign Economic Policy Uncertainty and Stock Market Volatility
Does Chinese media misperceive U.S. economic policy conditions? We propose a U.S. economic policy uncertainty index perceived by Chinese media. This index exhibits notable divergence from the U.S. economic policy uncertainty index reported by U.S. media. Article-level evidence from BERTopic attributes this discrepancy primarily to Chinese media misperception of the sources and magnitude of U.S. economic policy uncertainty. Empirical results show that the U.S. EPU perceived by Chinese media exert a significant positive impact on Chinese stock volatility, whereas the U.S. EPU reported by U.S. media does not. This positive effect remains significant after controlling China-specific policy and is more pronounced among large, young, high transparent, non-manufacturing, and low trade dependency firms. Leveraging the large language model (LLM) to extract textual information and capture emotional changes, we identify two underlying mechanisms explaining this effect: concerns among corporate management and investors’ overconfidence in private information valuation induced by Chinese media misperception about the U.S.
- 29 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe Mirror of Erised: Misperception of Foreign Economic Policy Uncertainty and Stock Market Volatility
Does Chinese media misperceive U.S. economic policy conditions? We propose a U.S. economic policy uncertainty index perceived by Chinese media. This index exhibits notable divergence from the U.S. economic policy uncertainty index reported by U.S. media. Article-level evidence from BERTopic attributes this discrepancy primarily to Chinese media misperception of the sources and magnitude of U.S. economic policy uncertainty. Empirical results show that the U.S. EPU perceived by Chinese media exert a significant positive impact on Chinese stock volatility, whereas the U.S. EPU reported by U.S. media does not. This positive effect remains significant after controlling China-specific policy and is more pronounced among large, young, high transparent, non-manufacturing, and low trade dependency firms. Leveraging the large language model (LLM) to extract textual information and capture emotional changes, we identify two underlying mechanisms explaining this effect: concerns among corporate management and investors’ overconfidence in private information valuation induced by Chinese media misperception about the U.S.
- Open Access
- 26 September 2026
- Published by Elsevier in SSRN Electronic Journal
Margin Trading, Short Selling, and Return Predictability: Evidence from the Chinese Stock Market
Using Chinese A-share stocks eligible for margin trading and short selling from 2010 to 2025, this study constructs measures of margin-financing and short-selling activity along three dimensions: changes in outstanding balances, relative trading intensity, and sentiment momentum. Principal component analysis is then used to construct a composite factor capturing the relative strength of margin financing versus short selling. The factor significantly predicts lower future stock returns. A long-short portfolio that is long low-factor stocks and short high-factor stocks earns significant positive returns at both daily and weekly frequencies, and the predictive relation remains robust after controlling for conventional risk factors and firm characteristics. Predictive power declines as the holding horizon lengthens. Return predictability is stronger when investor sentiment is more intense or optimistic and among stocks with stronger lottery-like characteristics, consistent with short-term mispricing and subsequent price correction. The main findings remain robust after accounting for transaction costs, factor orthogonalization, and separate tests in bull and bear markets.
- 26 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMargin Trading, Short Selling, and Return Predictability: Evidence from the Chinese Stock Market
Using Chinese A-share stocks eligible for margin trading and short selling from 2010 to 2025, this study constructs measures of margin-financing and short-selling activity along three dimensions: changes in outstanding balances, relative trading intensity, and sentiment momentum. Principal component analysis is then used to construct a composite factor capturing the relative strength of margin financing versus short selling. The factor significantly predicts lower future stock returns. A long-short portfolio that is long low-factor stocks and short high-factor stocks earns significant positive returns at both daily and weekly frequencies, and the predictive relation remains robust after controlling for conventional risk factors and firm characteristics. Predictive power declines as the holding horizon lengthens. Return predictability is stronger when investor sentiment is more intense or optimistic and among stocks with stronger lottery-like characteristics, consistent with short-term mispricing and subsequent price correction. The main findings remain robust after accounting for transaction costs, factor orthogonalization, and separate tests in bull and bear markets.
- Open Access
- 14 September 2026
- Published by Elsevier in SSRN Electronic Journal
Do Policymakers Listen to Markets? Evidence from China's Industrial Policy Dismantling *
Do policymakers listen to stock markets when deciding which industries lose key policy support? We examine this question in the context of China's industrial policy dismantling across the 12th, 13th, and 14th Five-Year Plans (2011–2021). Dismantling refers to removing or downgrading an industry from key supported status. We identify dismantling by translating expert reading and keyword-matching procedures into structured large language model (LLM) prompts that rely on retrieval-augmented generation (RAG). Using probit models, we link pre-decision stock returns to the probability of dismantling. Industries with lower past cumulative returns (PastCR) during the information-gathering stage of policy drafting are significantly more likely to lose key support. This effect is stronger where public information is scarce, as in industries with limited media coverage and highly similar annual reports. These results are consistent with the view that stock prices convey complementary information rather than merely reflect common fundamentals. A Karlson–Holm–Breen decomposition attributes approximately one-tenth of the total effect to a reassessment of industry innovation capacity. Our results extend the price-informativeness literature to the policy domain. The proposed LLM-based approach provides a reproducible pipeline for extracting industry-level policy variables.
- 14 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessDo Policymakers Listen to Markets? Evidence from China's Industrial Policy Dismantling *
Do policymakers listen to stock markets when deciding which industries lose key policy support? We examine this question in the context of China's industrial policy dismantling across the 12th, 13th, and 14th Five-Year Plans (2011–2021). Dismantling refers to removing or downgrading an industry from key supported status. We identify dismantling by translating expert reading and keyword-matching procedures into structured large language model (LLM) prompts that rely on retrieval-augmented generation (RAG). Using probit models, we link pre-decision stock returns to the probability of dismantling. Industries with lower past cumulative returns (PastCR) during the information-gathering stage of policy drafting are significantly more likely to lose key support. This effect is stronger where public information is scarce, as in industries with limited media coverage and highly similar annual reports. These results are consistent with the view that stock prices convey complementary information rather than merely reflect common fundamentals. A Karlson–Holm–Breen decomposition attributes approximately one-tenth of the total effect to a reassessment of industry innovation capacity. Our results extend the price-informativeness literature to the policy domain. The proposed LLM-based approach provides a reproducible pipeline for extracting industry-level policy variables.
- Open Access
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
The Mirror of Erised: Local Narratives of Foreign Economic Policy Uncertainty and Stock Market Volatility
We propose a U.S. economic policy uncertainty index based on Chinese media. This index exhibits notable divergence from the conventional U.S. economic policy uncertainty index. Further investigation attributes this discrepancy primarily to inaccurate coverage on U.S. economic policy conditions by private Chinese media. Empirical results show that the Chinese media-based U.S. EPU exert a significant positive impact on Chinese stock volatility, whereas the conventional U.S. EPU does not. This positive effect is stronger among large, young, high transparent, non-manufacturing, and low trade dependency firms. Leveraging the large language model (LLM) to extract textual information and capture emotional changes, we identify two underlying mechanisms explaining this effect: concerns among corporate management and investors’ overconfidence in private information valuation induced by Chinese media narratives about the U.S.
- 29 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe Mirror of Erised: Local Narratives of Foreign Economic Policy Uncertainty and Stock Market Volatility
We propose a U.S. economic policy uncertainty index based on Chinese media. This index exhibits notable divergence from the conventional U.S. economic policy uncertainty index. Further investigation attributes this discrepancy primarily to inaccurate coverage on U.S. economic policy conditions by private Chinese media. Empirical results show that the Chinese media-based U.S. EPU exert a significant positive impact on Chinese stock volatility, whereas the conventional U.S. EPU does not. This positive effect is stronger among large, young, high transparent, non-manufacturing, and low trade dependency firms. Leveraging the large language model (LLM) to extract textual information and capture emotional changes, we identify two underlying mechanisms explaining this effect: concerns among corporate management and investors’ overconfidence in private information valuation induced by Chinese media narratives about the U.S.
- Open Access
- 20 July 2026
- Published by Elsevier in SSRN Electronic Journal
Brown Firms Matter: An Asset Pricing Perspective
This paper examines whether brown discrimination is priced in the cross-section of Chinese stock returns. We define brown discrimination as mutual funds' tendency to tilt their holdings toward green firms relative to brown firms. Using Chinese mutual fund holdings and firm-level green invention patent data, we construct a market-level brown discrimination index and estimate firm-level sensitivity to this index through rolling regressions. We find that firms with higher brown discrimination sensitivity earn significantly lower subsequent returns. This return penalty remains robust to standard risk adjustment, controls for conventional firm characteristics, and alternative index constructions. We also find an information-based interpretation: the premium is attenuated when firms provide observable greentransition signals and when investor attention is higher, but it is stronger among firms with higher absolute carbon emissions. Overall, the findings suggest that the differential capital allocation between green and brown firms especially in high carbon emission markets.
- 20 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessBrown Firms Matter: An Asset Pricing Perspective
This paper examines whether brown discrimination is priced in the cross-section of Chinese stock returns. We define brown discrimination as mutual funds' tendency to tilt their holdings toward green firms relative to brown firms. Using Chinese mutual fund holdings and firm-level green invention patent data, we construct a market-level brown discrimination index and estimate firm-level sensitivity to this index through rolling regressions. We find that firms with higher brown discrimination sensitivity earn significantly lower subsequent returns. This return penalty remains robust to standard risk adjustment, controls for conventional firm characteristics, and alternative index constructions. We also find an information-based interpretation: the premium is attenuated when firms provide observable greentransition signals and when investor attention is higher, but it is stronger among firms with higher absolute carbon emissions. Overall, the findings suggest that the differential capital allocation between green and brown firms especially in high carbon emission markets.
- Open Access
- 14 June 2026
- Published by Elsevier in SSRN Electronic Journal
Macroeconomic Expectations and Expected Returns: Evidence from China
This paper examines whether subjective macroeconomic expectations predict equity risk premia in China. Using professional macro forecasts from the Wind database over 2011-2023, we construct a composite macroeconomic expectation index via PLS. We find that the index significantly and positively predicts future equity premia both in-sample and out-of-sample, implying a countercyclical equity premium. The predictive power is robust to a wide range of macro-financial controls and delivers economically meaningful gains in portfolio allocation. Further analysis shows that controlling for forecast errors does not attenuate predictability, and the subjective index largely subsumes the information in realized macro variables, suggesting that it captures forward-looking macroeconomic conditions rather than belief bias. Cross-sectional and long-horizon results reinforce this interpretation.
- 14 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMacroeconomic Expectations and Expected Returns: Evidence from China
This paper examines whether subjective macroeconomic expectations predict equity risk premia in China. Using professional macro forecasts from the Wind database over 2011-2023, we construct a composite macroeconomic expectation index via PLS. We find that the index significantly and positively predicts future equity premia both in-sample and out-of-sample, implying a countercyclical equity premium. The predictive power is robust to a wide range of macro-financial controls and delivers economically meaningful gains in portfolio allocation. Further analysis shows that controlling for forecast errors does not attenuate predictability, and the subjective index largely subsumes the information in realized macro variables, suggesting that it captures forward-looking macroeconomic conditions rather than belief bias. Cross-sectional and long-horizon results reinforce this interpretation.
- Open Access
- 7 May 2026
- Published by Elsevier in SSRN Electronic Journal
Macroeconomic Expectations and Expected Returns: Evidence from China
This paper examines whether subjective macroeconomic expectations predict equity risk premia in China. Using professional macro forecasts from the Wind database over 2011-2023, we construct a composite macroeconomic expectation index via PLS. We find that the index significantly and positively predicts future equity premia both in-sample and out-ofsample, implying a countercyclical equity premium. The predictive power is robust to a wide range of macro-financial controls and delivers economically meaningful gains in portfolio allocation. Further analysis shows that controlling for forecast errors does not attenuate predictability, and the subjective index largely subsumes the information in realized macro variables, suggesting that it captures forward-looking macroeconomic conditions rather than belief bias. Cross-sectional and long-horizon results reinforce this interpretation.
- 7 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMacroeconomic Expectations and Expected Returns: Evidence from China
This paper examines whether subjective macroeconomic expectations predict equity risk premia in China. Using professional macro forecasts from the Wind database over 2011-2023, we construct a composite macroeconomic expectation index via PLS. We find that the index significantly and positively predicts future equity premia both in-sample and out-ofsample, implying a countercyclical equity premium. The predictive power is robust to a wide range of macro-financial controls and delivers economically meaningful gains in portfolio allocation. Further analysis shows that controlling for forecast errors does not attenuate predictability, and the subjective index largely subsumes the information in realized macro variables, suggesting that it captures forward-looking macroeconomic conditions rather than belief bias. Cross-sectional and long-horizon results reinforce this interpretation.
- 8 November 2025
- Published by Elsevier in Journal of International Money and Finance
- Vol. 160
Taming the global factor zoo
- 8 November 2025
- Published by Elsevier in Journal of International Money and Finance
- Vol. 160
Taming the global factor zoo