Recent developments in financial markets have shifted researchers' focus from the financial development of the real sector to the development of financial investments. When firms in the real sector engage in too much financial investment, i.e., when they participate in financial market activities and neglect the development and innovation of the main business, the result may be phenomena such as “financialization”, which involves a switch from a “real” economy to a “virtual” economy. Previous studies have examined the negative impact of the financialization of the real economy on corporate management, innovation, and so on, but only a few have focused on a more crucial problem: why do firms favor financial investments？ Studies using appropriate theoretical settings and modeling are urgently needed. One important research question is whether there is any heterogeneity in the financial investment mechanisms of different types of enterprises, and what are the differences？ A scientific answer to this question will undoubtedly provide policy makers with a more targeted and differentiated approach to regulation.
This study uses the portfolio choice model featuring fixed and financial assets, relaxing the model hypothesis and constructing a cross-period portfolio choice model to analyze the heterogeneity of the mechanism that drives the financialization of different types of enterprises. We set up the empirical model based on theoretical expressions, and apply GMM estimation using the panel data on China's A-share listed non-financial firms from the 2006-2016 period.
Our analysis focuses on the heterogeneity of the mechanism driving the financial investments of different types of enterprises. We separate financial assets into cash and non-cash financial assets. We divide the main sample into two subsets based on the type of ownership and industry: state-owned enterprises vs. non-state-owned enterprises and manufacturing vs. non-manufacturing groups.
The results show that there are obvious differences in the mechanisms driving financial investment. The non-state-owned enterprises face stronger external financing constraints and higher operational risks; therefore, the motivation to hold financial assets is mainly to avoid risk. State-owned enterprises are mainly restricted by internal cash flow and leverage ratio, and risk aversion is not a significant driving factor. The income from the main business of manufacturing firms depends on fixed assets, which encourages firms to hold financial assets to avoid the fixed investment risk, whereas the financial investment of non-manufacturing firms is mainly driven by the profit motive.
We also find that there are apparent differences in the mechanisms driving choices in cash and non-cash financial assets. The cash investment of state-owned enterprises is mainly driven by the profit motive, whereas non-state-owned enterprises do not show this characteristic. The non-cash financial assets investments of non-state-owned, manufacturing and non-manufacturing firms are driven by uncertainty in economic policy, whereas state-owned enterprises are unaffected by this uncertainty. The results are robust to the removal of long-term equity investments in financial assets.
For policy makers interested in promoting “financial sectors returning to source” and “financial sectors serving the real economy”, this study's findings have specific policy implications. Strengthening the regulation of investment risk, clearing financing channels, and reducing the uncertainty of real investment should be key targets for policy makers. More importantly, macro policies that are generalized to all enterprises fail to recognize the heterogeneity of enterprises and the different preferences in financial investment. To guide the real economy to participate in investment activities and to help realize the optimal allocation of resources across departments and industries, policies are needed to address the unfair financing environment for non-state-owned enterprises, the “soft budget constraints” of state-owned enterprises, and the high risk of investment in manufacturing industries.

Over the past few decades, there have been frequent global and regional economic crises. As fluctuations in economic cycles accelerate, it is becoming difficult for central banks' current behavior rules to effectively achieve counter-cyclical goals. This raises the need for more monetary policy innovations. In the past, Taylor's monetary policy rule has been successfully applied in Western countries, but it has weak applicability in China. For some periods, it cannot smooth economic fluctuations while also playing a pro-cyclical role. In addition, although discretionary monetary policy itself plays a strong role in economic stability, it may cause the problem of dynamic inconsistency. It cannot smooth economic fluctuations, but it may also play a pro-cyclical role. It can be seen that the design of a monetary policy with both regular and discretionary components has important reference significance for central banks' counter-cyclical regulations.
In this paper, we carry out a pertinent design and construct a regime-switching Taylor rule. However, unlike previous research, this paper internalizes Markov regime transformation probability and makes it time-varying. In addition, considering that the main discretionary monetary policy tools are money supply, open market operations, and deposit reserve ratios, we make the Markov regime transformation probability time-varying depending on the monetary base, credit of the banking system, exchange rate, and deposit reserve ratio. In this way, we construct a Markov regime-switching Taylor rule with time-varying transition probabilities. In this monetary policy, time-varying Markov regime transition probability is the key link. First, it links the Taylor rule to determine the regime in which it can play the role of an automatic stabilizer. Second, it links the discretionary monetary policy tools so that monetary policy no longer uses interest rates as a direct operational tool but as an intermediate target. In the meantime, the method of responding to inflation gaps and output gaps for interest rates under different regional systems indirectly changes with the money base, credit level of the banking system, exchange rate, and deposit reserve ratio. In addition, these direct operating tools can also achieve a counter-cyclical function as a discretionary monetary policy.
The study finds that there is an asymmetric effect for the automatic stabilizer function of the rule component in the monetary policy. In regime one, there is no automatic stabilizer function in the rule component; in regime two, there is a favorable automatic stabilizer function in the rule component. In addition, it is found that low average interest rates are the reason why the monetary policy in regime one does not have automatic stabilizer function; the zero-interest-rate problem limits the central bank's ability to regulate the macro economy through short-term nominal interest rates. In regime one, the volatility of short-term nominal interest rates is small, which also illustrates the central bank's lower intervention in interest rates. The situation is the opposite in regime two, which explains why the monetary policy in regime two has a strong automatic stabilizer function.
Based on the analysis results, the monetary policy designed in this paper gives rise to the following operation modes. During recessions, the central bank should use the money base and window guidance as the direct operation tools, and take nominal short-term interest rates as the intermediate target. On one hand, when the economy is in a downturn, increasing the money base growth rate and loosening the window guidance has an anti-cyclical function, which can warm up the economy; on the other hand, it can guide the economic system to switch to the regime where the rule component has a favorable automatic stabilizer function, thus producing a positive and dynamic regulating effect between short-term nominal interest rates and inflation and output. During boom periods, the central bank should use exchange rates and deposit reserve ratios as the direct operation tools, and take nominal short-term interest rates as the intermediate target. Raising the exchange rate and deposit reserve ratio has an anti-cyclical function, which can cool down the economy. It can also guide the economic system to switch to the regime where the rule component has a favorable automatic stabilizer function, thus producing a positive and dynamic regulating effect between short-term nominal interest rates and inflation and output.

The 2008 global financial crisis is attributed to excessive risk-taking by commercial banks and the failure of financial regulation. In December 2010, the Basel committee officially released the Basel III standards, which revised multiple defects identified in the Basel II standards. Based on the Basel Accords, the China Banking Regulatory Commission (CBRC) introduced the New Capital Accord and planned to implement a series of reforms labeled Advanced Methods of Capital Management (AMCM). The AMCM consists of an internal ratings-based (IRB) approach to credit risk measurement, an internal model approach to market risk measurement, and an advanced measurement approach to operating risk measurement. In April 2014, the CBRC approved six banks to act as pilot adopters of AMCM by replacing the previous credit risk standard, under which risk weights were determined uniformly by the regulatory agency, with the IRB approach. This raises several issues in need of further study. Will AMCM implementation influence the effect of capital regulations, change commercial banks' risk preferences, or cause structural adjustments to credit allocation？
The partial implementation of AMCM enables us to use the difference-in-differences (DD) and triple differences (DDD) methods for empirical analysis. Using semi-annual and annual reports of listed commercial banks from 2010 to 2016, we collect data on 16 banks (6 banks as the treatment group and 10 banks as the control group). We use risk-weighted assets and the proportion of individual industry loans to total loans as the explanatory variables and other characteristics as the control variables. We further rely on the industry-specific non-performing loan ratio published by the CBRC as the measure of industry-level credit risk and then match this measure to our banking data to settle on 13 industries.
Empirical results using the DD method show that the pilot banks significantly reduce risk-weighted assets after AMCM implementation, which suggests that this policy successfully reduces banks' risk appetite. More importantly, this change in risk appetite may influence how credit is allocated to industries with different credit risk levels. Using the DDD method, we find that AMCM implementation prevents pilot banks from making loans to high-risk industries and has a nonlinear effect on loans to less risky industries; i.e., loans increase not for the least risky industries, but rather for ones with slightly higher risk. This illustrates banks' tradeoff between risks and returns. Due to the pro-cyclical nature of commercial bank credit, the influence of AMCM on credit allocation may have implications for economic activities. By dividing industries into virtual and real sectors, we find that pilot banks reduce loans to the real estate (virtual), manufacturing (real), and construction (real) industries, and they significantly increase loans to the financial industry (virtual). Therefore, pro-cyclical effects and banks' loan adjustment behavior may hinder the reallocation of credit from virtual to real sectors, which is not conducive to structural economic change.
Based on the above empirical results, this paper has several policy implications. For one thing, regulators should pay attention to the impacts of AMCM and encourage commercial banks to improve their customer data collection and risk analysis abilities to ensure the effectiveness of IRB systems. For another, regulators should implement macroeconomic regulations, such as counter-cyclical capital buffers, to mitigate the pro-cyclical effects arising from the IRB approach.
This paper contributes to the literature in several ways. First, there is presently disagreement over the effect of IRB approaches, and our paper enriches these discussions. Second, unlike previous domestic studies that explore bank credit behavior from the perspective of capital requirements, this paper analyzes the impact of changes in capital measurement methods on bank credit behavior. Finally, this paper offers insights on whether IRB methods can aid in the credit allocation flight from virtual to real sectors, sheds further light of the issue of financial resource flow from virtual to real sectors, and has value as a reference for regulatory authorities in setting policy.

Economic policy is important for the transformation and structural change of China's economy. However, the complexity of formulating specific economic policies and unpredictability during their implementation leads to economic policy uncertainty. Studies have shown that such uncertainty significantly affects economic development and corporate behavior. However, few studies have focused on the mechanisms by which economic policy uncertainty affects macroeconomic operation and corporate behaviors. China's financial system is characterized as bank-dominated, and bank credit is an important source of financing for microeconomic entities, which makes bank credit an important channel through which economic policy uncertainty can affect the real economy.
This paper examines the effect of economic policy uncertainty on the cost of bank loans to firms, focusing on two main issues. First, we examine how banks adjust banks loan pricing in response to economic policy uncertainty. We not only pay attention to the average effect of economic policy uncertainty on bank loan costs to firms but also to the heterogeneous effects on different types of bank and firm. Second, we examine whether economic policy uncertainty affects firm loan costs through a deleterious effect on firm credit defaults. These two issues provide a better understanding not only of how economic policy uncertainty affects firm behavior and the macro-economy, but also of the importance of consistency and transparency of economic policy in maintaining financial stability.
Using the indicator constructed by Baker et al. (2016) and representative bank loan data from 2010 to 2015, this paper empirically explores the impact of economic policy uncertainty on firm bank lending costs. Several findings are worth mentioning. First, we show that economic policy uncertainty has a significantly positive effect on firm bank loan costs, especially for loans issued by smaller banks. We explain this as follows. A bank's incentive to self-insure is stronger during periods of higher uncertainty, as greater uncertainty makes it difficult to form accurate expectations of future liquidity demand and the frequency and intensity of the implementation of future macroeconomic policy. Banks therefore tend to identify borrowers by increasing loan costs. Second, our heterogeneity analysis indicates that economic policy uncertainty has a greater effect on the bank lending costs of micro-enterprises and private enterprises. Third, we show that the default risk of bank loans decreases when economic policy uncertainty increases. This indicates that the increase in bank loan costs when economic policy uncertainty increases is not due to an increase in default risk. It is more likely that banks strengthen their self-insurance incentive and transfer their policy uncertainty risk to firms when policy uncertainty increases.
Based on our empirical analysis, we put forward the following policy suggestions. First, amplitude and frequency should be taken into consideration when applying economic policies to regulate the economy, especially the potentially detrimental effect of policy adjustment on bank credit allocations. Second, the management and guidance of banking sector expectations needs strengthening to improve bank anticipation of government policy and thus encourage banks to better serve the real economy. Third, the government must consider the potential heterogeneous effects of economic policy adjustment on different types of firms and banks, to improve the pertinence and effectiveness of economic policies.
Our study contributes to the literature in the following ways. First, it enriches the literature on the effect of economic policy uncertainty and provides a better understanding of how it affects macroeconomic fluctuations and firm behavior from a micro-perspective. Second, we find that increased economic policy uncertainty intensifies the distortion of credit allocation in China's banking sector, indicating that the amplitude and frequency of economic policies are important in improving the stability and efficiency of its banking system. Future research should comprehensively explore the influence of economic policy uncertainty on bank credit supply and micro credit demand.

Since the 2007-2009 global financial crisis and the ensuing worldwide economic downturn, China has implemented a series of policy maneuvers including monetary stimulus to revitalize its financial markets and the economy. Our study examines the performance of China's stock market in the short windows around the central bank PBOC's related announcements. We delve into two important questions. Is the Chinese stock market responsive to announcements of possible changes in monetary policy？ Do equity prices reflect any uncertainty variation associated with potential policy shifts？ The results of this study not only provide additional rationale for PBOC policies that substitute away from quantity-based instruments but also point to a better scheme for providing forward guidance that will stabilize the financial markets without weakening the transmission of China's monetary policy.
Previous research has focused on the ex-post impacts of unexpected monetary policy changes on the real economy and financial markets. Particularly for China, few studies have examined the ex-ante reactions of stock markets to potential changes to monetary policy before the PBOC has announced an updated policy stance. Our study specifically evaluates the performance of China's stock market a few days prior to such announcements, when investors' uncertainty about policy changes is high.
For our sample period from January 2011 to June 2017, we identify 78 PBOC's monthly announcement events during which monetary aggregates data are released. For a window of three days before and after a monetary announcement, we use a dummy regression framework and examine the excess returns of China's stock market constructed from Wind A Share Index, a comprehensive index that covers all of the A shares traded on the Shanghai and Shenzhen exchanges. We are able to document a statistically and economically significant pre-announcement equity premium for China. That is, China's stock market cumulatively builds up gains over a three-day period that reaches its peak on the day of announcement; subsequently, the market flattens out. By looking into the cumulative returns aggregated from the high frequency returns of five-minutes trading blocks, we detect a pre-announcement drift of returns, which is consistent with our regression results.
We then provide evidence that squares well with the implications derived from uncertainty-reduction based theory, which associates equity premium with policy uncertainty. First, our regression analysis shows that equity premium is largely associated with more delayed announcements when greater uncertainty has been accumulated among investors. Second, stock return volatility declines from its peak three days before an announcement. Our results suggest that ex-ante reduction of accumulated uncertainty about policy changes leads to the accumulation of excess returns prior to an announcement. For robustness, we further run estimations of alternative specifications and rule out the channel of “news-effect” as the main explanation. That is, as stock prices jump only upon policy stimulus, equity premium can be realized because investors are informed of good data ahead of time or they simply expect the stimulus ex-ante. The results based on the regression of interaction terms show that the magnitude of M2 growth changes, regardless of whether it is indicative of policy tightening or laxness, does not affect our baseline-estimated size of the pre-announcement premium.
The main contributions of our study are as follows. (1) We are the first to document the pre-announcement reactions of China's stock markets to incoming PBOC's announcements of monetary aggregates data. (2) Our study finds that the size of the pre-announcement premium depends on the relative timeliness of the arrival of public announcements. (3) Exploiting the variation in timeliness across announcement events, we provide empirical evidence of a more general theory that helps rationalize the pre-announcement premium found both in China and the U.S.
Our findings have important policy implications. (1) PBOC should quickly adopt price-based policy instruments. We show that announcements of monetary aggregates data, the key quantity-based instruments, destabilize the financial markets by triggering cyclical fluctuations between months. In contrast, real-time information about interest rates should be readily accessible to the financial market, which would help to prevent investors from building up too much policy uncertainty, which may lead to financial turmoil. (2) PBOC could design a better scheme for its forward guidance and communications with the market. For example, it can largely mitigate the extra market volatilities driven by unexpected delays of data release by pre-fixing and pre-informing the market of announcement dates.

As one of the three pillars of the modern financial industry, insurance plays an important role in economic compensation, capital financing, and social management, and it is known as the “patron saint” of socioeconomic development. Although China's commercial insurance market has developed rapidly over the past 40 years into the world's second largest insurance market, the insurance industry is still in the primary stage of development from the perspective of insurance depth and density. The literature generally points to limited participation among residents as an important reason that China's commercial insurance industry is not strong despite its large size. The key to solving the problem of limited participation is to identify factors that affect residents' purchasing of commercial insurance. At present, there are relatively few studies on the influence of social interaction on family commercial insurance purchases, and no consistent conclusions have been reached so far.
To address this research gap, the present paper studies the influence of social interaction on family commercial insurance purchases and its channels using data from the China Household Finance Survey (CHFS) 2013. First, by selecting eight variables related to social interaction and using the iterative principal factor method, this paper constructs a comprehensive social interaction index. Second, Probit and Tobit models are used to conduct empirical analysis, and two channels are verified through the variables of financial knowledge and degree of trust in commercial insurance. Third, potential endogeneity problems related to social interaction are overcome by constructing instrumental variables, and heterogeneity analysis and robustness tests are conducted. Finally, based on tracking data from the 2013-2015 CHFS, logical problems related to the time inconsistency between social interaction and household commercial insurance purchases are overcome. The results show that social interaction promotes household purchases in the commercial insurance market, and the effect is still significant after solving the endogeneity problem. In addition, this paper confirms two channels of social interaction that affect commercial insurance purchases. According to heterogeneity analysis, social interaction plays a more important role in promoting insurance purchases among families in the eastern and central regions, those with medium levels of education, those with high income levels, and those in low-purchases-rate communities.
The main innovations of this paper are fourfold. First, the use of CHFS data to examine the relationship between social interaction and household commercial insurance purchases allows our findings to be strongly representative, as the data span 29 provinces and more than 28,000 households. In this way, we avoid the drawbacks of previous studies that fail to provide a comprehensive picture of residents' insurance purchases in China. Second, by using the iterative principal factor method to construct a comprehensive social interaction index, we avoid the concern that a single variable cannot accurately describe social interaction. Third, by verifying two channels of social interaction's impact on household commercial insurance purchases, our paper allows for a more thorough understanding of the effect of social interaction. Finally, potential bias in regression results caused by endogeneity problems and time inconsistency is solved by constructing instrumental variables.
This paper is the first to explore the development of China's commercial insurance market from the perspective of social characteristics. Efforts to promote insurance market development should go beyond economic factors and also recognize the importance of residents' social characteristics. Therefore, relevant agencies should pay attention to the role of social interaction in the development of insurance. It is also necessary to further standardize China's insurance market and strengthen the trust of commercial insurance conditions to improve the depth and breadth of insurance purchases and promote the industry's development.

China's economic development has always been plagued by the difficulty and high cost of corporate financing. To improve the financing environment for firms, the government of China has promoted market-oriented financial reforms, adjusted the structure of direct financing and indirect financing, and encouraged the development of small and medium-sized banks. However, issues related to corporate financing remain unresolved. According to a survey conducted by the All-China Federation of Industry and Commerce in 2017, the cost of financing is still the most important cost factor affecting the development of firms. To address this issue, the importance of supply-side structural reform and measures to reduce firm costs were emphasized at the 19^th National Congress of the Communist Party of China. In addition, a 2019 government work report included making corporate financing easier and less costly as a key task. Thus, the question of how to reduce the cost of corporate financing has drawn the attention of policymakers, entrepreneurs, and researchers. Particularly in light of China's constantly changing legal system, one important problem is how the cost of corporate debt responds to the reform of the security interests system, which is exogenously induced by the enactment of the Property Law.
As China has transitioned from a planned economy to a market economy, new laws have been continuously introduced to improve the legal environment of the market economy. Compared with previous laws, the Property Law enacted in 2007 introduced a broader and more efficient system of security interests and reduced the transaction costs of secured lending transactions. In particular, the menu of assets legally accepted as collateral was enlarged to include accounts receivable, inventories, and other liquid assets. The reform of the security interests system induced by the Property Law has reduced the risk of secured lending transactions and helped enhance the bargaining power of borrowing firms; this is expected to affect the cost of corporate debt. Thus, this provides an ideal setting for us to study how legal reforms affect the cost of corporate debt financing.
Using the Property Law's enactment as a natural experiment and a difference-in-differences method, we construct a sample of manufacturing firms based on the Annual Survey of Industrial Firms in China during 2003-2008 and study the effect of the law on the cost of corporate debt. We find that the reform leads to a significant reduction in the cost of debt, and this effect is more pronounced for firms with a lower proportion of fixed assets. Furthermore, consistent with the economic intuition that the Property Law reduces the cost of debt financing by expanding the range of collateralized assets, triple-difference tests indicate that the reduction effect shows rich diversity. First, the reform more strongly reduces the cost of corporate debt in provinces with less developed legal or financial environments. Second, the reform more strongly reduces the cost of debt for firms subject to severe financial constraints.
Researchers have not reached a consensus on the determinants of the cost of corporate debt. In particular, although many empirical studies using cross-sectional data find that legal systems are an important determinant, few studies explore whether and how legal reforms in an emerging and transitional economy like China's affect the cost of corporate debt. This paper not only identifies the causal relationship between creditor protection-strengthening legal reforms and a declining cost of corporate debt but also clearly reveals that reform of the security interests system reduces the cost of corporate debt by strengthening creditor protections and enhancing the bargaining power of borrowing firms. This paper provides fresh empirical evidence from China's emerging and transitional economy for the law and finance literatures, and it improves our understanding of how creditor protections influence the cost of debt.
In addition, this paper has important policy implications. To alleviate the difficulty and high expense of corporate financing, we must continuously promote the benign change of the basic financial system through market-oriented legal reforms and consolidate the basic institution of the capital market.

Innovation plays a critical role in establishing competitive advantage and promoting economic growth, and the factors that affect innovation have received significant research attention. According to the literature, a sophisticated incentive system can reliably promote corporate innovation. However, the system should provide effective incentives for both managers and employees because studies have shown that stock options for employees, labor protection measures, and labor unions have important effects on corporate innovation.
In recent years, the income structure of employees has diversified with the development of the economy in China. However, salaries are still the dominant form of income. Moreover, there may be significant differences between employees' nominal pre-tax salaries and their actual after-tax salaries due to the high personal income tax rates. In this case, payroll tax avoidance activities may affect the effectiveness of compensation incentives, and thus affect corporate innovation.
Two competing hypotheses are theoretically related to the consequences of payroll tax avoidance activities. First, unlike standard routine work, innovation projects require employees to commit substantial effort in continually learning to adapt to the new environment created by innovation. Thus, it is costly for employees to carry out innovative projects. If the level of compensation does not cover the high private cost of being involved in corporate innovation, the employees will lack the motivation to participate. In particular, the high personal income tax rates in China significantly reduce employees' actual after-tax salaries, and therefore weaken the effectiveness of compensation in promoting corporate innovation. Therefore, because payroll tax avoidance effectively lowers the tax rate of employees, it is likely to improve the effectiveness of compensation in promoting corporate innovation.
Second, the risk preferences of deciders are stable at any particular time. Because the variance in after-tax salaries increases the perceived risk of employees to balance their risk exposure, the employees have an incentive to reduce their involvement in high risk innovation. Moreover, employees' payroll tax avoidance activities may increase the volatility of their after-tax salaries over time. For instance, employees' nominal pre-tax salaries will be much higher when company earnings are higher. However, the progressive tax rates for wage and salary earners reduce the difference between the peaks and troughs of the compensation. In other words, payroll tax avoidance can also hinder corporate innovation.
In this study, we investigate whether and how payroll tax avoidance affects corporate innovation using a sample of Chinese A-share listed firms from 2007 to 2015. Following the literature, we measure corporate innovation using the invention patent applications in a given year. Our findings are as follows. (1) The extent of payroll tax avoidance is significantly and positively correlated with corporate innovation. (2) The positive correlation is more pronounced when there is high taxable pressure on employees, and the reform of the individual income tax system in 2011 inhibited the positive correlation. These results suggest that payroll tax avoidance promotes innovation by reducing the adverse effects of payroll tax and strengthening the compensation incentive effect. (3) For state-owned companies and companies located in high tax enforcement regions, payroll tax avoidance plays a greater role in promoting innovation. (4) This positive correlation is significantly attenuated for firms' with higher per capita salary fluctuations. (5) The positive correlation is more significant in companies that provide significant benefits for innovators.
We contribute to the literature in three ways. First, different with the literature focusing on the stock options of employees, labor protection measures, and labor unions, we provide new evidence on the relationship between the motivation and innovation of employees by studying their payroll tax avoidance activities. Second, the effect of corporate tax avoidance on innovation has been discussed in the literature. Thus, we add to the literature on tax avoidance and corporate innovation by examining the impact of payroll tax avoidance on corporate innovation. Third, this study extends the literature on the economic consequences of payroll tax avoidance. Although studies have investigated the influence of payroll tax avoidance on firms' accounting performance, to the best of our knowledge, we provide the first evidence tying payroll tax avoidance to a vital dimension of corporate performance, namely, corporate innovation.

Innovation plays an essential role in economic growth and corporate development. However, the literature focuses mainly on corporate innovation from the perspective of innovation quantity and efficiency, while rarely investigating the different types of innovation and their determinants, such as incremental versus radical innovation (Dewar and Dutton, 1986; Cardinal, 2001). According to the People's Daily, the number of patent applications in China has multiplied by 10 times over the past 10 years, and China has received the highest number of patent applications worldwide in the past 5 years. Nevertheless, the overall quality of innovation in China remains low. Thus, investigating the heterogeneity of innovation using radical and incremental innovation contributes not only to the innovation literature, but also to government policy design for a transition economy and sustainable development.
This paper investigates the effects of customer concentration on radical and incremental innovation by firms. Prior studies show that major customers can contribute to corporate innovation by providing information on consumer demand and new product development (Fang, 2008; Coviello and Joseph, 2012). However, few differentiate between radical versus incremental innovation. Theoretically, collaboration between major customers and firms can increase radical innovation, as information from major customers can mitigate the technological and financial risks. Christensen (2014) argues that closely collaborating with and learning from major customers can benefit firm research and thus provide short-term profits. However, in the long term, radical innovation is vital to firm development, yet the lack of radical innovation increases a firm's fiasco risk. Therefore, the effect of customer concentration on radical and incremental innovation by firms is uncertain and requires empirical evidence.
Our initial sample consists of all firms listed on China's Shenzhen and Shanghai Stock Exchanges from 2009 to 2016. We manually collect customer data from voluntary disclosures in the annual reports of A-share listed firms. All financial data are from China Stock Market & Accounting Research (CSMAR), ownership data are from the China Center for Economic Research (CCER), and patent data are from the China Research Data Services Platform (CNRDS). We exclude firms in the financial industry, firm-year observations with incomplete or incorrect financial and ownership information, and firm-year observations with zero invention and utility patents. Finally, we winsorize the continuous variables by 1% at the top and bottom of the sample and end up with 9,046 firm-year observations. Our results show that firm customer concentration is significantly positively associated with radical innovation. However, this relation only appears among state-owned enterprises (SOEs) and firms with high geographic proximity to their major customers. In addition, we find that among firms with such a relation, higher customer concentration increases firm invention patents rather than utility patents.
This paper contributes to the literature in two ways. First, its contributes to the corporate innovation literature by exploring the heterogeneity of innovation types, namely radical and incremental innovation, whereas prior studies focus on innovation quantity and efficiency. In addition, we provide new measures of radical innovation against the Chinese institutional background. Second, we contribute to the literature on how customer concentration affects corporate investment. Prior studies have documented evidence on conservatism, cost of capital, performance, auditor choice, and auditing fees. Whereas Chu et al. (2018) investigate how geographic location between firms and customers affects innovation, we investigate the effect of customer concentration on a firm's radical innovation.
This paper has two limitations that could be addressed in future studies. First, although we present new measures for radical innovation, our measures for both radical and incremental innovation could be more accurate using broader data on patents, such as citations for patents. Second, future studies could explore more aspects of the customer perspective in addition to customer concentration, such as customer operational and financial status.

In recent years, the relationship between stock market uncertainty and stock pricing has attracted widespread attention. One question is whether uncertainty is always “bad.” In fact, “good” uncertainty does exist. A typical example is when a company is ready to launch a new product; the market is optimistic about the new product, but uncertain how much profit it will ultimately bring. Thus, uncertainty can be decomposed into “good” uncertainty and “bad” uncertainty. Investors prefer stocks with “good” uncertainty exposure and dislike stocks with “bad” uncertainty exposure. To obtain stocks with high “good” uncertainty, investors must pay higher prices and accept lower expected returns. In comparison, stocks with high “bad” uncertainty have lower stock prices and high expected returns.
Following Barndorff-Nielsen et al. (2010), we use the realized positive semi-variance and realized negative semi-variance to represent “good” and “bad” uncertainty, respectively. “Good” uncertainty has a positive impact on stock prices. However, uncertainty also results in large fluctuations in stock prices. As positive semi-variance measures the price volatility as it relates to positive returns, it can effectively reflect the “good” uncertainty. A similar reasoning applies to the use of negative semi-variance to measure “bad” uncertainty. We further subtract the realized negative semi-variance from the positive semi-variance and then standardize it to get our key indicator – relative signed variation (RSV).
This study uses a sample of China's A-shares from the 2007 to 2017 period. High-frequency stock data are drawn from CSMAR, and non-high-frequency stock data are from RESSET. In addition, China's Economic Prospective Index is obtained from CEIC. The main conclusions of this study are as follows. First, consistent with our theoretical analysis, regardless of the sorting method we apply (single sort method and double sorts method), there is a significantly negative relationship between RSV and stock portfolio returns. When the sample is divided into five portfolios based on the RSV in ascending order, we find that the corresponding returns monotonically decrease from Portfolio 1 to Portfolio 5. This pattern remains even after we apply controlling variables such as realized volatility and realized skewness using a double sorting method, indicating that RSV is an important stock pricing factor independent of these variables. In comparison, the higher the RSV level, the higher the absolute value of the high-minus-low portfolio returns based on realized volatility. In addition, realized skewness loses its stock pricing capability after controlling for RSV. Second, we use a firm-level cross-sectional regression method to further verify the negative relationship between RSV and stock returns. After controlling for common pricing factors such as realized volatility, realized skewness, market beta, firm size, book-to-market ratio, momentum, and illiquidity, the significantly negative relationship between RSV and stock returns remains. Third, the relationship between “good” and “bad” uncertainty and risk premium is state-dependent. Based on the China's Economic Prospective Index, we separate the full sample into periods of high economic prosperity and low economic prosperity. The RSV corresponds to lower stock risk premiums during periods of high economic prosperity, but higher stock risk premiums in periods of low economic prosperity, which suggests that the stock risk premium is more sensitive to “good” and “bad” uncertainty risk exposures during period of low economic prosperity. Finally, using the “good” uncertainty and the “bad” uncertainty, we construct a low-minus-high portfolio and find that our constructed portfolio outperforms the market excess return portfolio, SMB portfolio, and HML portfolio as measure by the mean return and Sharpe ratio.
This study makes three academic contributions. First, the study decomposes the uncertainty at the micro level into “good” uncertainty and “bad” uncertainty, and finds that RSV is an important pricing factor in China's stock market. To the best of our knowledge, this is the first study to discuss the impact of “good” and “bad” uncertainty on Chinese stock market pricing. Second, a large number of studies have proven that skewness is a significant stock pricing factor. Although in economic logic there are some similarities between realized skewness and RSV, this study demonstrates that RSV has a stronger stock pricing power than realized skewness. Third, this study finds that the impact of “good” and “bad” uncertainty on stock prices is state-dependent, and RSV leads to a higher stock risk premium when the economy is less prosperous.

At the 19^th National Congress of the Communist Party of China, the authorities emphasized the importance of financial sector institutional reform in China, particularly increasing the proportion of direct financing and promoting the healthy development of a multilevel capital market. As a vital part of the multilevel capital market, China's A-share market plays an important role in optimizing information transfer and allocating capital resources. However, a large body of literature argues that herding behaviors may negatively affect the information transparency and pricing efficiency of the stock market. Such behaviors can even cause financial turmoil in severe cases. Short-term speculative herding behaviors also lead to a high turnover rate of capital flow and impede the formation of long-term capital, which is harmful to economic growth. Thus, understanding herding behaviors in China's A-share stock market is critical to improve pricing efficiency, increase the proportion of direct financing, and support the healthy development of the real economy.
Many theoretical studies find that herding behaviors are short-lived and fragile. However, research on herding behaviors in China is mainly based on the quarterly holdings data of institutional investors. First, short-lived herding behavior and its pricing effect are highly likely to be missed when using quarterly data. Second, correlated trades at the quarterly level tend to reflect changes in the fundamental values of stocks rather than herding behaviors. In a word, the limitations of using quarterly holdings data may result in significant deviations in measuring herding behaviors.
For the above reasons, this paper improves the LSV method developed by Lakonishok, Schleifer, and Vishny (1992) and creatively uses daily trading data to obtain a more precise measure of short-term herding behaviors in China's A-share stock market. Based on this measure, we further investigate stock-specific characteristics that affect herding behaviors and the pricing effects of herding behaviors.
We have four major findings. (1) The degree of herding monotonically increases with trading frequency. It is 3.92%, 2.48%, and 1.64% over the daily, weekly and monthly horizons, respectively. This result is consistent with the theoretical prediction that herding behaviors are short-lived and fragile. (2) Asymmetric information, the proportion of institutional investors, and stock size significantly affect the degree of herding behaviors. Herding behaviors are more severe in stocks with higher levels of asymmetric information or higher proportions of institutional investors, and there is a U-shaped relationship between the degree of herding and firm size. (3) There is a price reversal after the herding: a positive (negative) abnormal return is gained after the sell-side (buy-side) herding behaviors, and the price reversal is more significant following a higher degree of herding. (4) The price reversal effect after the herding behaviors is asymmetric: it is more pronounced for large and liquid stocks after buy-side herding than after sell-side herding.
This paper makes three major contributions: First, we are the first to use daily trading data to obtain a more precise measure of short-term herding behaviors in China's A-share stock market, which overcomes the limitations of using quarterly data in previous studies. Second, based on this more accurate measure of herding behaviors, we deeply examine the determinants of herding behaviors and the effects of herding on future prices. Our research serves as an important supplement to the literature on herding behaviors in China's A-share market. Third, contrary to findings based on quarterly holdings data of institutional investors, we find that short-term herding is subsequently followed by price reversals, which supports the argument that herding behaviors negatively affect the price discovery function of the stock market. These findings have great value in deepening our understanding of investor behavior and improving the price discovery function of China's A-share stock market.