The impact of regional economic incentives on underwriters ’ market share in China

Purpose – To examine whether and how the different levels of regional economic incentives would have an effect on underwriters ’ market share in general. Design/methodology/approach – Drawing on Chinese IPO firms during the period 2006-2016, this study examines the impact of different levels of regional economic incentives on underwriters ’ market share. Findings – Theauthorsfindthatregionaleconomicincentiveshaveapositiveimpactonunderwriters ’ market shareandthatlocaleconomicincentiveshaveasignificantlystrongerimpactthancentraleconomicincentives.Furthermore,theauthorsfindthatIPOfirmswithunderwritersdrivenbyregionaleconomicincentivesexperienceworsepost-IPOperformancethanfirmswithunderwritersdrivenbycentraleconomicincentives,whichdonotexperienceasignificantdeclineinpost-IPOperformance. Originality/value – Taken together, the authors ’ findings are consistent with the notion that performance assessment motivates officials at various levels of government to bring companies in their jurisdiction to the IPOmarketprematurely.Inaddition,theresultsindicatethatcentraleconomicincentivesplayasignificantroleindrivingChina ’ s macroeconomic development and market-oriented system reforms. As such, they are one of the major driving forces behind China ’ s market-oriented system reforms. driven by local economic incentives facing fierce competition tend to lower their client selection criteria, adopt more lenient practice standards and take ineligible firms public prematurely.Byproviding an in-depth study of how different levels of economic incentives impact underwriters ’ incentives, our paper informs progressive policy-making in China ’ s capital market. Our paper also highlights problems (i.e. a lack of integrity, awareness of the rule of law and due diligence on the part of some underwriters) in the current sponsorship system. Our findings are pertinent to the policy debates on how to further reform China ’ s financial market system to optimize resource allocation and how to effectively monitor underwriters so that they can function effectively as gatekeepers in China ’ s capital market. Our findings may serve as a point of reference for policy-making aimed at strengthening financial market regulation and penalizing regulatory breaches by underwriters.


Introduction
We use the sponsorship system in China's IPO market and study the role of underwriters, who are key players in this heavily government-regulated process. Given its fundamental role as a capital-raising mechanism in China, the central government has implemented the IPO sponsorship system since 2003 to promote the orderly development of the capital market. Under this sponsorship system, the sponsor acts as both the coordinator and leader of the overall IPO process. The sponsor is also responsible for verifying the truthfulness of IPO application documents and information disclosures and for critically commenting on the professional opinions provided by other intermediaries. The sponsor takes overall charge of preparing the IPO application documents and assumes joint and several liabilities for the authenticity, accuracy and completeness of the documents. By doing so, the sponsor becomes the ultimate coordinator of all of the intermediaries involved in the IPO process, acting as the most important intermediary in the firm's IPO process and, consequently, is the key gatekeeper in China's capital market. As such, the sponsor plays an important role in ensuring the smooth operation of China's IPO market. However, in consideration of China's status as a transitional economy and the fact that different levels of governments and officials have discretion and control over huge amounts of public resources (Tian & Zhang, 2013), the efficiency of the IPO process depends heavily on effective communication within the government.
Although a few studies examine financial intermediaries in China (e.g. Chen, Guan, Zhang, & Zhao, 2017), there is limited research into the varying impacts of regional economic incentives on financial intermediaries and the heterogeneity of the mechanisms that underlie and drive different levels of economic incentives. Motivated by this void in the literature, we focus on sponsors as a financial intermediary and investigate the impact of regional economic incentives on the market share of financial intermediaries in the Chinese market. Our empirical findings shed light on the operational efficiency of China's capital market and resource allocation efficiency and provide guidance on how to support the progressive market-oriented reform of China's capital market.
This paper's sample comprises all IPO firms in China during 2006-2016. We use the work experience of the general manager or board chairman of the underwriter as a proxy for regional economic incentives, divided into central economic incentives and local economic incentives. To proxy for market share, we use the share of a given underwriter's IPO underwriting revenues in the total IPO underwriting revenues of all underwriters in a specific year. We then conduct an in-depth investigation of the impacts of regional economic incentives on underwriters' market share and the heterogeneity in such impacts across different levels of economic incentives [1].
Our results reveal that regional economic incentives have a significant positive impact on underwriters' market share and that local economic incentives have a stronger positive impact on underwriters' market share than central economic incentives. We further examine the performance change of the IPO firms in the 6, 12 and 18 months after the IPO is listed, separately for underwriters driven by central economic incentives and those driven by local economic incentives. We fail to find a significant change in the post-IPO performance of firms with underwriters driven by central economic incentives. However, for firms with underwriters driven by local economic incentives, the post-IPO performance experiences a significant decline that worsens over time. In addition, when comparing the post-IPO performance of firms across underwriters associated with different levels of economic incentives, we find that the firms taken public by underwriters driven by local economic incentives experience worse performance post-IPO. This finding is consistent with the tournament theory of political promotions for local government officials in China, where the performance of officials is assessed by the number of IPOs (an important indicator) within their jurisdiction. Faced with pressures for political promotions according to the rank order in a tournament, officials at different levels of government are motivated to speed up IPO activities within their jurisdiction even when some of the firms are not mature enough for an IPO. Consequently, these firms suffer deterioration in performance post-IPO. In contrast, the central government's economic incentives are more focused on economic fundamentals and macroeconomic development nationwide. Furthermore, central government administrators have generally developed a certain reputation within the system and they tend to be reluctant to risk damaging their reputation and career by attempting to push unqualified firms through an IPO. In addition, relative to local government officials, central level administrators are subject to more stringent supervision by their superiors (Chen, Kim, Li, & Liang, 2018) and are thus more prudent when handling potential IPO firms.
We make several contributions to the literature. First, moving away from the narrow approaches of prior studies, our paper is among the first to use the political identity of top executives (mainly the board chairman, the general manager, or both) to proxy for regional economic incentives to investigate how such incentives impact capital market development in China. Moreover, by further dividing regional economic incentives into central and local economic incentives, we explore the difference in their impacts and the reasons for these differences. Our findings are consistent with the tournament theory of political promotion for Chinese local government officials. To our knowledge, this is the first paper to apply the concept of regional economic incentives to study China's capital market. In addition to providing a new and deep perspective into the efficient operation of China's capital market, our findings also provide empirical support for research on the role of regional economic incentives in the capital market. As such, our research has both theoretical and empirical implications.
Second, by studying the impact of regional economic incentives on underwriters' market share, we demonstrate that regional economic incentives can interfere with fair competition in the primary market, in turn negatively affecting resource allocation. Notably, we find that when faced with the tournament-style performance assessment, regions with stronger economic incentives tend to interfere more with economic activities within their jurisdiction, which runs against the central government's market reform directives. Therefore, our conclusions not only contribute to the literature on financial intermediaries but also provide new insights into the nature of the progressive market-oriented system reform in China's capital market and demonstrate empirical support for economic reform and innovations in China's transition economy.
Third, we conduct an in-depth analysis of the different impacts of the 1994 tax reforms on central and local governments. We show that under the widespread pressure of fiscal deficits and tournament-style performance assessment, local governments and their officials have strong incentives to intervene in economic activities within their jurisdiction. Therefore, our study not only adds to the research on China's Tax-Sharing System, but also informs the ongoing tax reforms in China that aim to combine the administration of state and local tax collection by provincial and sub-provincial governments.
Fourth, our study has policy implications for strengthening financial regulation in China. On June 6, 2018, the China Securities Regulatory Commission (CSRC) issued Amendments to Administrative Measures for IPO and Public-Listing, which stipulated that "Where the Issuer reports a loss in the year of the IPO, the CSRC shall suspend the qualification of the underwriter for a period of 3 months, from the date the loss is confirmed and shall revoke the qualification of the Sponsor Representative." Our findings indicate that different levels of economic incentives have different impacts on financial intermediaries and lead to different degrees of interference in economic activities. As such, our study is particularly relevant to the ongoing policy debate on reform measures in China's transition economy.

Theoretical framework and hypothesis development
Since China implemented its "opening up" policy, economic construction has taken center stage in the basic principles of the Chinese Communist Party (CCP). When assessing the performance of local government officials, the central government has endeavored to avoid erroneous assessments due to the use of ambiguous criteria, better motivate officials and promote regional economic development using a set of well-defined criteria. To achieve this, the central government turned to easily quantifiable and observable indicators, which has gradually led to the use of a tournament-style assessment approach to the political promotion of local officials (Zhou, 2007). As an important indicator of capital market development, the number of IPO firms within a jurisdiction has become a point of reference for local government officials' performance (Piotroski & Zhang, 2014). Consequently, facing the double pressure to manage a fiscal deficit and to obtain a promotion, local governments and their officials are incentivized to intervene in the economic activities of firms or financial intermediaries in their region. For instance, they may bring an IPO firm within the jurisdiction to market prematurely.
In contrast to developed countries with mature capital markets, China is still transitioning from a planned economy to a market-oriented economy (Deng & Zeng, 2009;Zhou & Qiu, 2013). Governments and officials at different levels have control and discretion over huge amounts of public resources (Tian & Zhang, 2013) and tend to interfere with the market via various administrative and regulatory systems (Piotroski & Zhang, 2014). Additionally, China is a relationship-based society with a long tradition of official rank culture. In a social environment where the government and officials control many crucial economic resources, the government's "invisible hand" plays a critical role in the numerous areas that involve resource allocation (Cai, Tian, & Guo, 2017). As a result, regional economic incentives exert a pivotal impact on enterprises' economic activities.
In China's current transition economy, there is huge uncertainty surrounding the success of a firm's IPO, from the preliminary qualifications review and offer price to its floatation in the subsequent secondary market. All of these factors are inextricably intertwined with effective communications with the government (Zheng, Xu, Bai, & Qin, 2017). The sponsorship system of share issuance currently in use in China is essentially an administrative procedure. With the government exerting tight control over key resources and the CSRC's Issuance Examination Committee in charge of approving an application, a potential IPO firm hoping to increase its chance of approval has strong incentives to leverage all possible channels and establish connections with the government to obtain critical information that could help with the IPO (CSRC, 2008;Lu, Wan, & Yang, 2015). Chen et al. (2017) find that underwriters whose top executives have worked in government departments and agencies are able to maintain good communications with the government, which then tends to be more friendly to the clients of such underwriters.
The above reasoning leads to H1: H1. Regional economic incentives are positively associated with underwriters' market share.
Since embarking on its "opening up" policy in 1978, China has gradually phased in a fiscal split system reform. However, the central government's fiscal income as a proportion of the nation's total fiscal income actually decreased after the reform (Zhang, 2007). Therefore, the central government set out to reclaim important economic powers in some key areas, aiming to re-adjust the balance of fiscal relations between the central and local governments. This led to the Tax-Sharing System reform in 1994 (Zhang, 2006). One key aspect of this reform is the division of taxes into central taxation, local taxation and tax sharing between the central and the local governments. Since implementing the Tax-Sharing System reform, the central government's share of fiscal income as a percentage of total fiscal income has steadily increased, but local governments are still responsible for local fiscal expenditures. This has led to huge imbalances in local government deficits and placed local governments under substantial fiscal pressure (Li & Chen, 2013). According to Dong (2007), although the tax reform measure has relieved the central government of its responsibility for expenditures, the concentration of fiscal power has increased, placing local governments under huge fiscal pressure. Zhang and Gong (2005) also point out that the implementation of the Tax-Sharing System reform indeed improved the central government's fiscal capability and that since 1995; the central government's fiscal income share in the country's total fiscal income has continued to rise. As a result, local governments' fiscal dilemmas have motivated them to interfere with economic activities within their jurisdictions; correspondingly, regional economic incentives are positively associated with interference in capital market activities. For example, governments' regional economic incentives mean that they have greater motivation to push companies within their jurisdictions to seek an IPO. However, in reality, most companies' operations are subject to the administrative influence of local governments, and, to a large extent, their survival and development depend on local government support and cooperation. According to L€ u and Nie (2014), although the Tax-Sharing System has promoted economic growth, it has also led to problems such as local protectionism. Conversely, Allen, Qian and Qian (2005) suggest that China's non-official systems based on reputation, relationship and trust may supplement the formal systems and support enterprise growth. Central administrators, who have built a good reputation within the system, could suffer losses to their reputation and future career prospects if they push unqualified IPOs. Furthermore, relative to local government officials, central administrators are more likely to be subject to stringent monitoring by their superiors (Chen et al., 2018). As a result, central administrators tend to be more prudent when pushing through an IPO.
Additionally, although the Tax-Sharing reform gave the central government greater control over resources, as China's economic system reform continues to progress and market mechanisms have continued to improve, the central government has sought to reduce government interference with enterprise operations and to withdraw from microeconomic activities. Consequently, the central government is playing a leading role in reducing its influence on market activities. Although China is a vast country with many provinces, they are all responsible to a single central government (Chen & Gao, 2012), which acts as their direct supervisor. Relative to sub-provincial local governments, provincial governments are expected to carry out central government policies more faithfully and strictly.
Thus, distinguishing between different levels of economic incentives, we hypothesize the following: H2. Relative to central economic incentives, regional economic incentives have a stronger impact on underwriters' market share.
The IPO system is a fundamental aspect of China's capital market. Before the implementation of the Securities Law in 1999, stock issuance was conducted using an approval system, in line with the completely planned economy. This approval system, which took two forms (i.e. quota management and indicator management), contributed to the development of China's nascent capital market. However, it also suffered from many problems, including low resource allocation efficiency and slow market construction and proved to be inadequate for the capital market's long-term development (Deng, 2017). Consequently, following the promulgation of the Securities Law in July 1999 and in line with the development needs of China's capital market, a verification system was adopted in 2000 that included both the channel system and the sponsorship system. When the channel system was initially adopted, it restricted competition between underwriters and therefore was not conducive to the market's healthy development. Therefore, in December 2003, the CSRC issued the Interim Measures for the Stock Issuance and Listing Sponsorship System, which contains detailed provisions for the responsibility and practice qualifications expected of underwriters. Its introduction marked the formal adoption of a sponsorship system for securities issuance and listing. The sponsorship system was officially implemented in February 2004. Compared with the former IPO systems, the sponsorship system has distinctive market characteristics and is more effective in stimulating underwriters' due diligence and in monitoring IPO firms (Chen et al., 2017). Under the existing sponsorship system, prior to the IPO, the underwriter must recommend and coach the firm planning an IPO and must assume joint and several liabilities for the authenticity, accuracy and completeness of the application materials. After taking the firm public, the underwriter must continue to supervise the firm and must assume joint and several liabilities for any breaches committed by the firm during the supervision period. Therefore, the underwriter's integrity will determine not only the quality of the listed firms but also the effectiveness of China's sponsorship system.
As the to-be-listed firm's sponsor, the underwriter is the overall coordinator and leader of the whole listing process. Thus, the underwriter's integrity and understanding of the law can determine the extent to which they engage in ethical behavior and due diligence when doing business. However, as business competition between underwriters becomes increasingly fierce, the fight for market share has intensified. Some underwriters have adopted more lenient client selection criteria, lowered risk assessment standards, or turned a blind eye to predictable risk factors. This has led to many regulatory breaches during the IPO application process, irresponsible underwriting and other problems. An example is Guilin Sanjin, for which the underwriter was found to have provided untruthful disclosures in the IPO prospectus. In another, during the IPO process of Wanfu Biotechnology, the sponsorship document provided by the underwriter contained some false statements. In the case of Dalian Kemian Wood Industry Co., Ltd., the firm saw a dramatic fall in its post-IPO performance, which severely depressed the stock price. These are not isolated cases and such breaches by the underwriters have not only cast serious doubt on underwriters' reputation and integrity but also led to the existence of some "bad apples" among listed firms and to severe damages to investors.
In addition, under the influence of regional economic incentives, underwriters are more likely to lower selection standards when choosing potential IPO firms even when they do not meet the CSRC's eligibility criteria for IPO application. Thus, underwriters driven by regional economic incentives are more likely to select IPO firms that have not yet met the listing criteria. This results in premature listings and a significant decline in post-IPO performance. In contrast, because central administrators are subject to rigorous supervision by their superiors, they could expect huge losses to their reputation and future career if they risk pushing unqualified firms through an IPO. As a result, underwriters driven by central economic incentives tend to adopt more stringent risk assessment standards when selecting IPO firms. They are also more likely to follow standard business procedures and to handle potential breaches more carefully. Consequently, IPO firms with underwriters driven by central economic incentives do not normally see their performance worsen significantly post-IPO.
Thus, we formulate our third hypothesis as follows: H3. The post-IPO performance of firms with underwriters driven by local economic incentives worsens significantly, whereas the post-IPO performance of firms with underwriters driven by central economic incentives does not.

Research design 3.1 Sample data and source
Our primary sample consists of all IPO firms in China during the period 2006-2016. We remove observations with missing data for the board chairman or general manager of the underwriter, the actual controller of the underwriter, the pass rate, the market share, firms with missing financial data and firms with co-underwriters. The final sample contains 1,596 firms. We start from 2006 because this is the first year for which information about each underwriter's market share is available on the official website of the Securities Association of China (from where we download the data). Information about the actual controller of the underwriter, the pass rate and the market share comes from WIND. We manually collect the employment information of the underwriter's board chair and general manager from their annual reports, supplemented by information collected from personnel biographies in WIND, hexun.com, ifeng.com, sina. com, the official websites of the underwriters, baidu.com and google.com. Information about the IPO firms' names and IPO date during 2006-2016 comes from the CSRC Issuance and Examination Committee's list of approval results for IPO firms reported in WIND. Information about the IPO firm's underwriter, the employment history of the top executives and the financial variables is taken from CSMAR. Information about the actual controller is obtained from the IPO prospectus.
3.2 Variable construction Construction of regional economic incentives, the key explanatory variable in this paper, is based on Chen et al. (2017). This involves determining the chairman and general manager's past or current work experience at various levels of government. As provincial government officials are directly supervised by the central government, for the purpose of this paper and based on the theory behind H2, we code Central Economic Incentives as 1 if the underwriter's general manager or board chairman is currently working or once worked at or above the deputy provincial level and 0 otherwise. We code Regional Economic Incentives as 1 if the underwriter's general manager or board chairman is currently working or once worked in positions ranked between the bureau level and deputy bureau level and 0 otherwise.
The control variables include whether the underwriter is state-owned (CenGov), the pass rate of the underwriter in year t À 1 (lpassrate), whether the underwriter's IPO revenue ranks among the top ten underwriters in t À 1 (Rank10), firm age (FirmAge), firm size (SIZE), current ratio (CR), inventory ratio (INV), operating cash flow (OCF), return to equity (ROE), earnings management (EarnMgmt), financial leverage (LEV), whether the IPO firm is a stateowned enterprise (SOE) and the IPO firm's associated level of regional economic incentives (FirBgd). These control variables are selected following prior studies (Chen et al., 2017;Liu, Tang, & Tian, 2013;Piotroski & Zhang, 2014: Lu et al., 2015Cai et al., 2017). Table 1 contains the detailed variable definitions and summary statistics. Table 2 presents the descriptive statistics for the sample data. The average for the underwriter's revenue market share is 4.0825% and the average for the underwriter's deal number market share is 4.0424%, which is similar. There is a large difference between the minimum and maximum values of the underwriters' market shares, indicating significant differences in individual underwriters' relative industry positions. Considering that China is still a transition economy where the government continues to control huge amounts of public resources and exercises discretion over their allocation and use, regional economic incentives are valuable intangible assets that can help an underwriter develop its business. For this reason, we use this institutional setting to study how regional economic incentives affect underwriters. Further analysis indicates that the IPO firms experience average CAR values of 22.77, 17.43 and 15.69% in the 6, 12 and 18 months post-IPO, respectively. This steady decline in performance is remarkable. In the remainder of this paper, we examine the reasons for the post-IPO decline in performance and the mechanisms through which different levels of economic incentives affect economic activities differently in China.

Empirical models
First, to test H1 (whether regional economic incentives increase underwriters' market share), we estimate regression model (1), where the underwriter's IPO revenue (or deal number) market share is the dependent variable and regional economic incentives (proxied by the work experience of the underwriter's general manager or board chairman) is the key explanatory variable.
Second, to test H2 (whether the positive impact of local economic incentives on underwriters' market share is stronger than that of central economic incentives), we estimate Equations (2) through (4), where the underwriter's IPO revenue (or deal number) market share is again the dependent variable. In Equation (2), the key explanatory variable is central economic incentives, proxied by whether the underwriter's general manager or board chairman is currently working or once worked at or above the deputy provincial level. We use this to test how central economic incentives impact underwriters' market share. In Equation (3), the key explanatory variable is local economic incentives, proxied by whether the underwriter's general manager or board chairman is currently working or once worked in positions ranked between the bureau level and deputy bureau level. We use this to test how local economic incentives impact underwriters' market share. In Equation (4), we examine how the difference between local and central economic incentives affects underwriters' market share. The control variables are those from Equation (1).

IPO Rev Market ShareðNUMshareÞ
Next, to test H3 (whether IPO firms with underwriters driven by local economic incentives suffer a significant decline in performance, whereas those with underwriters driven by central economic incentives do not), we estimate Equations (5) through (7), where the dependent variable is cumulative abnormal returns (CAR) in the 6, 12 and 18 months post-IPO, respectively. In Equation (5), we examine the impact of central economic incentives on post-IPO performance. In Equation (6), we examine the impact of local economic incentives on post-IPO performance. In Equation (7), we examine the impact of level difference in local and central economic incentives on post-IPO performance. The control variables are those from Equation (1).
4. Empirical results 4.1 Regional economic incentives and underwriters' market share Table 3 reports the regression results for Equations (1) through (4). We use an underwriter's share of all underwriters' IPO underwriting revenues in a year as a proxy for the market share of the specific underwriter. Table 3, Column 1 shows that when we control for underwriters' and listed firms' characteristics as well as industry and year fixed effects, regional economic incentives have a positive and significant impact on underwriters' market share. In terms of economic significance, underwriters subject to the influence of regional economic incentives increase their market share by 0.448% on average. This indicates that regional economic incentives act as the government's helping hand (by facilitating communication with government agencies, among other things) and can increase underwriters' market share. Thus, H1 is supported. We also explore the impact of different levels of regional economic incentives on underwriters' market share. Diff in loc cen Economic Incentives is our proxy for the level difference between local and central economic incentives. To determine which type of regional economic incentives has a stronger impact on underwriters' market share, we estimate Equation (4), using Diff in loc_cen Economic Incentives as the key explanatory variable. As shown in Table 3, Column 4, when we control for underwriters' and listed firms' characteristics as well as for industry and year fixed effects, the coefficient on Diff in loc cen Economic Incentives is positive and statistically significant at the 1% level. This indicates that relative to central economic incentives, local economic incentives have a stronger positive impact on underwriters' market share. Thus, H2 is supported.
Next, we separately examine the impact of central and local economic incentives on underwriters' market share. As shown in Table 3, Column 3, when we control for various underwriter and listed firm characteristics as well as for industry and year fixed effects, the coefficient on local economic incentives is positive and statistically associated with underwriters' market share. In terms of economic significance, local economic incentives raise the underwriters' market share by 2.12% on average. Thus, local economic incentives driven underwriters have an apparent competitive advantage in their business development. Table 3, Column 2 indicates that central economic incentives do not have a positive impact on underwriters' market share; in fact, we find the opposite. The intuition behind this finding is that at a specific point in time, there is a fixed demand for seeking an IPO, yet there is an excess supply of underwriters capable of handling IPO listings. Therefore, this is a zero-sum game and competition for underwriting is fierce. When underwriters driven by local economic incentives have a competitive advantage and succeed in attracting more client firms, this will inevitably affect the market share of other types of underwriters such that the market share of underwriters driven by central economic incentives is negatively affected. In addition, we note that both the historical pass rate and whether the historical revenue ranked among the top ten have a positive impact on underwriters' market share. In terms of economic significance, a 1% increase in historical pass rate increases the underwriters' market share by 0.004% on average. In addition, being ranked in the top ten in terms of historical revenue, increases the underwriters' market share by about 4% on average. These findings are as expected: a high historical pass rate increases the probability that the underwriter is chosen by a firm seeking an IPO, as having such an underwriter as the sponsor would increase the chance of successful approval by the CSRC's Issuance and Examination Committee. Therefore, historical pass rate is positively associated with underwriters' market share. In addition, ranking among the top ten in terms of IPO underwriting revenue means that the specific underwriter is a leading player with high market recognition. Such brand recognition makes the underwriter more attractive to potential client firms, which increases its ability to secure more IPO deals and hence its market share.
Considering the size differences in underwriters' client base and to ensure the completeness and robustness of the results, we re-estimate Equations (1) through (4) Table 1 for the variable definitions. The intercept is included but not tabulated. T-statistics are in brackets. Standard errors are clustered at the industry level. *, ** and *** indicate significance at the 10, 5 and 1% levels, respectively Table 3. Regional economic incentives and underwriters' market share Regional economic incentives both the share of an underwriter's IPO revenues in the industry's total IPO revenues in a given year and the share of an underwriter's deal number in the industry's total number of IPO deals in a given year as proxies for underwriters' market share. Table 4 reports the results. As demonstrated, when we control for various underwriter and listed firm characteristics as well as industry and year fixed effects, the coefficient on regional economic incentives ðβÞ remains positive (although statistically insignificant). The lack of statistical significance is attributed to the conflation of the different market share impacts of local and central economic incentives. Furthermore, local economic incentives have a stronger positive impact than central economic incentives on underwriters' market share. The rest of the results are consistent with those in Table 3. Our results indicate that in China's transition economy, local economic incentives give rise to more severe interference with economic activities, as administrative measures are used to influence public resource allocation. This also implies that although the economic system reforms in China have achieved a certain degree of success, they have not been fully effective.
4.2 Impact of different levels of regional economic incentives on firm future performance We next examine the following two questions.
(1) Under the dual pressure of a local fiscal deficit and a tournament for political promotion, will local governments and their officials have stronger incentives (relative to their central counterparts) to interfere with economic activities by pushing companies within their jurisdiction to seek an IPO prematurely? (2) When faced with keen market competition and given the huge financial gains from securing IPO projects, will some underwriters disregard predictable risks and lower their selection standards such that companies that are illegible for an IPO are brought to the market prematurely? It is expected that underwriters driven by local economic incentives are more likely to relax their selection standards for target clients: they may help some unqualified companies meet the CSRC's qualifications for IPOs and prematurely take them public. For this reason, IPO firms with underwriters driven by regional economic incentives may experience a significant decline in performance post-IPO. To test these conjectures, we use the   Table 1 for the variable definitions. The intercept is included but not tabulated. T-statistics are given in brackets. Standard errors are clustered at the industry level. *, ** and *** indicate significance at the 10, 5 and 1% levels, respectively. The control variables in Table 4 are the same as those in Table 3 and are untabulated for brevity Table 4. Regional economic incentives and underwriters' market share: additional test CAR during the 6, 12 and 18 months after the IPO date as a proxy for firms' post-IPO performance and estimate Equations (5) through (7). Table 5 reports the results.
From Columns 1, 4 and 7, we note that for IPO firms with underwriters driven by central economic incentives, their performance 6, 12 and 18 months post-IPO does not change significantly. In contrast, Columns 2, 5 and 8 show that the performance of IPOs with underwriters driven by local economic incentives deteriorates significantly during the same period. In terms of economic significance, these latter IPO firms see their CAR decline by 9, 10.3 and 11.7%, respectively, during the 6, 12 and 18 months post-IPO. It thus appears that the performance of IPO firms with underwriters driven by local economic incentives worsens over time. We also investigate whether the difference in local and central economic incentives significantly affects firms' post-IPO performance. As shown in Table 5, Columns 3, 6 and 9, firms with underwriters driven by local economic incentives suffer a significant decline in their post-IPO performance. We note that the difference in regional economic incentives has a significantly negative impact on firm performance in the 12 and 18 months post-IPO. Our results indicate that relative to central economic incentives, local economic incentives lead to IPO firm underperformance for the following reasons. First, relative to their central counterparts, local governments and their officials tend to interfere more with economic activities because the tournament-style assessment for political promotion puts them under pressure and as the number of IPOs is used as an important performance indicator for the promotion of local government officials, they have strong incentives to push firms within their jurisdiction toward IPOs. Second, underwriters subject to different levels of economic incentives adopt different client selection criteria, risk management attitudes and practice standards, with underwriters driven by local economic incentives tending to be less prudent in their business practices. For the above reasons, IPO firms come in diverse initial conditions and over time, problems begin to surface. Eventually, the difference in post-IPO performance becomes greater, depending on the level of economic incentives that have driven firms' public listing.
In summary, underwriters driven by central economic incentives risk losing more highpotential clients and attracting regulators' attention, which increases the likelihood of a regulatory audit if they breach any rules. In addition, any breaches would attract media attention and expose the top leaders of the underwriter to serious risks of reputational damage, which may severely undermine their career development. Therefore, given the high costs associated with breaching the rules, underwriters driven by central economic incentives tend to adopt more stringent risk assessment policies and practice standards and to have lower motivation to take an under qualified client firm public. In contrast, because underwriters driven by local economic incentives may find it easier to secure local government's support and cooperation in their IPO underwriting business, to pursue huge financial gains, they may lower client selection criteria and provide underwriting for firms that do not meet listing requirements. This is also why underwriters driven by local economic incentives are well received in the market. Our findings also suggest that the difference in the economic incentives that drive underwriters to adopt different compliance standards is one of the reasons for the very different qualities that we see among listed firms. The difference in the extent to which central and local economic incentives interfere with economic activities leads to variation in the impact of different levels of economic incentives on economic activities. Underwriters that are subject to the influence of different levels of economic incentives adopt different risk assessment approaches and different client selection criteria and they have different probabilities of regulatory breaches in the course of their business practice. Thus, the IPO firms they handle demonstrate diverse post-IPO performance: firms with underwriters driven by central economic incentives do not suffer a significant decline in performance, whereas those with underwriters driven by local economic incentives see a significant decline in performance. Thus, H3 is supported.  Note(s): See Table 1 for the variable definitions. The intercept is included but not tabulated. T-statistics are given in brackets. Standard errors are clustered at the industry level. *, ** and *** indicate significance at the 10, 5 and 1% levels, respectively. The control variables in Table 5 are the same as in Table 3 and are untabulated for brevity Table 5.
Regional economic incentives and firm future performance 5. Robustness tests 5.1 Different regional economic incentives and IPO waiting periods We use the number of days between the listing date and the signing date of the IPO prospectus as a proxy for the IPO waiting period and investigate the impact of regional economic incentives on the efficiency of the underwriter's IPO business. We delete three firms with missing data for the waiting period and estimate the regressions using the remaining observations. The results are reported in Table 6, Column 1. As shown, the coefficient on Regional Economic Incentives is negative and statistically significant at the 1% level. This indicates that regional economic incentives can enhance the efficiency of the underwriter's IPO business. In terms of economic significance, underwriters driven by regional economic incentives are associated with a waiting period that is 3 days shorter on average. This is also one reason why the market prefers to use the service of underwriters driven by regional economic incentives: they are more likely than underwriters not driven by regional economic incentives to provide their clients with more efficient services by shortening the administrative approval process and IPO waiting period.
The results in Table 6, Columns 2, 3 and 4 show that the coefficient on Central Economic Incentives is negative but statistically insignificant. Thus, underwriters driven by central economic incentives are unable to shorten the IPO waiting period. The coefficient on Local Economic Incentives is negative and significant at the 1% level, suggesting that underwriters driven by local economic incentives can shorten the client firm's IPO waiting period and offer more efficient underwriting services. In terms of economic significance, underwriters driven by local economic incentives are associated with an IPO waiting period that is 5 days shorter on average. This is consistent with our earlier finding. Thus, local economic incentives shorten firms' IPO waiting period. In addition, the results in Table 6, Column 4 indicate that the impact on shortening the IPO waiting period is significantly stronger for underwriters driven by local economic incentives than for those driven by central economic incentives. Therefore, our results for the different impacts on the IPO waiting period for different levels of economic incentives lend further support to our H1, H2 and H3. The results also indicate that in China, local economic incentives interfere more with economic activities than do central economic incentives.

IPOwait
IPOwait IPOwait IPOwait (1) (2) (3) (4) Regional Economic Incentives À2.792*** (À5.74) Central Economic Incentives À1.027 (À1.47) Local Economic Incentives À4.990*** (À7.79) Diff in loc_cen Economic Incentives 0.0949 0.0922 0.0964 0.0924 N 1,593 1,593 1,593 1,593 Note(s): See Table 1 for the variable definitions. The intercept is included but not tabulated. T-statistics are given in brackets. Standard errors are clustered at the industry level. *, ** and *** indicate significance at the 10, 5 and 1% levels, respectively. The control variables in Table 6 are the same as in Table 3 and are untabulated for brevity Table 6. Regional economic incentives and IPO waiting period 5.2 Regional economic incentives and pass rate Next, we examine the impact of regional economic incentives on underwriters' pass rate. The results are reported in Table 7. Column 1 shows that the coefficient on Regional Economic Incentives is positive and significant at the 5% level. This suggests that regional economic incentives can significantly increase underwriters' pass rate, consistent with the literature (Liu et al., 2013). In terms of economic significance, the pass rate of underwriters subject to regional economic incentives (relative to those not subject to such incentives) is 0.81% higher on average. As previous, we further examine the impact of level difference on underwriters' IPO pass rate. The results are significant at the 10% level for underwriters driven by central economic incentives, which are associated with higher pass rates, but insignificant for the types of underwriters. The reason for such a difference is that approval of IPO applications depends on the CSRC. Underwriters driven by central economic incentives are able to increase their IPO pass rates because of their better connections with central government agencies and their control over scarce resources. In contrast, local economic incentives have more limited influences and underwriters driven by central economic incentives adopt stricter risk assessments, have higher standards for selecting IPO firms and follow more formal business procedures. Therefore, the IPO firms that they underwrite are typically well-performing firms that are of higher quality and more likely to meet the CSRC's listing requirements and pass the IPO examination. As a result, relative to underwriters driven by local economic incentives, underwriters driven by central economic incentives are associated with significantly higher pass rates.
Nevertheless, on the whole, the difference in local and central economic incentives does not lead to a significant difference in their IPO pass rates: as shown in Table 7, Column 4, the coefficient on Diff in loc_cen Economic Incentives is statistically insignificant. This is also why companies choose underwriters driven by regional economic incentives even though they do not have significantly higher pass rates. In addition to shortening the IPO waiting period, these underwriters also help foster closer connections with local governments, which have huge influences over firms' day-to-day operations and can provide them with various forms of support. Such connections create new resources and business opportunities for the firms' future development. Therefore, although underwriters driven by local economic incentives do not have higher pass rates than underwriters driven by central economic incentives, IPO firms are still more inclined to choose the former in the hopes of obtaining other types of support.   Table 1 for the variable definitions. The intercept is included but not tabulated. T-statistics are given in brackets. Standard errors are clustered at the industry level. *, ** and *** indicate significance at the 10, 5 and 1% levels, respectively. The control variables in Table 7 are the same as in Table 3 and are untabulated for brevity Table 7. Regional economic incentives and pass rate

Addressing the influence of firm size
To mitigate the influence of firm size on the conclusions, our primary sample excludes mega enterprises jointly underwritten by a syndicate of underwriters. Such companies generally have a long history and sufficient cash flows and they tend to be large industry leaders. Their public listing is often handled by the central government and thus has little to do with the regional economic incentives that drive some underwriters. In addition, these companies, having advantageous circumstances, are the prime target clients of many underwriters, so they are free to choose their underwriters and are not bound by regional economic incentives.
To check the robustness of our results, we further exclude 15 of the largest companies (with total assets ranking in the top 1 percentile) and re-estimate Equations (1) through (4). The results, reported in Table 8, are qualitatively the same as before.

Addressing the influences of suspending and restarting the IPO program
With consideration of the development needs of the secondary market and other system reform issues, the CSRC has temporarily suspended and restarted the IPO issuance arrangements a number of times. These events also impacted the capital market to varying degrees. To eliminate the effects of the IPO suspension and restarting events on our conclusions, we delete all firm observations during the IPO suspension and restarting years. These include 444 observations during the years 2006, 2009, 2014 and 2015. We then re-estimate Equations (1) through (4). The results, shown in Table 9, are qualitatively unchanged.

Endogeneity
Our research question centers on the impacts of regional economic incentives on underwriters' market share and the heterogeneity in the impacts of different levels of economic incentives on underwriters' market share. For this reason, our results are not susceptible to reverse causality. Nevertheless, to further mitigate the influence of endogeneity and ensure the reliability of our results, we have controlled for a variety of underwriter and   Table 1 for the variable definitions. The intercept is included but not tabulated. T-statistics are given in brackets. Standard errors are clustered at the industry level. *, ** and *** indicate significance at the 10, 5 and 1% levels, respectively. The control variables in Table 8 are the same as in Table 3 and are  untabulated for brevity   Table 8. Eliminating the influence of firm size Regional economic incentives firm characteristics, as well as industry and year fixed effects. Therefore, any remaining endogeneity issues may come from omitted variable bias. To address this concern, we add additional control variables. One such variable is the IPO firm's net proceeds. Because it is related to underwriters' IPO revenues, it may also affect their market share. We include the logarithm of IPO firms' net proceeds and re-estimate Equations (1) through (4). Three observations have missing data for this variable and are thus deleted. The results, reported in Table 10, are qualitatively the same as before. Thus, our results are robust to endogeneity arising from omitted variable bias.

Summary and conclusions
Since the founding of the PRC, government power has shifted from being concentrated to being shared. During the post-1978 power-sharing reforms, local governments in China enlarged their fiscal powers (Chen & Gao, 2012). Local governments have since acquired increased economic management power and fiscal autonomy (Pan, Xia, & Yu, 2008). At present, China is still a transition economy. Its laws and regulations have yet to be perfected and governments and officials still have control and discretion over huge amounts of public resources (Yang, 2011). Against this backdrop, IPO underwriting remains heavily government-influenced. To ensure a smooth IPO process, effective communications and interactions with the government are indispensable. These institutional realities and backgrounds and the related literature provide the theoretical basis and empirical setting for our paper.
Our paper empirically examines the impacts of regional economic incentives on underwriters' market share and the heterogeneity in the impacts of different levels of regional economic incentives on underwriters' market share. We use an underwriter's share of IPO underwriting revenues as a percentage of the industry's total IPO underwriting revenues (alternatively, the share of IPO underwriting deals as a percentage of the industry's total IPO underwriting deals) as a proxy for its market share. To capture the influence of   Table 1 for the variable definitions. The intercept is included but not tabulated. T-statistics are given in brackets. Standard errors are clustered at the industry level. *, ** and *** indicate significance at the 10, 5 and 1% levels, respectively. The control variables in Table 9 are the same as in Table 3 and are untabulated for brevity  Note(s): See Table 1 for the variable definitions. The intercept is included but not tabulated. T-statistics are given in brackets. Standard errors are clustered at the industry level. *, ** and *** indicate significance at the 10, 5 and 1% levels, respectively. The control variables in Table 10 are the same as in Table 3 and are untabulated   for brevity   Table 10.

Regional economic incentives and market
share controlling for net proceeds Regional economic incentives regional economic incentives on an underwriter, we use the current or past work experience of the board chairman or general manager of the underwriter in the central government, local governments, the CSRC, the Army and status as a current or past representative of the NPC or member of the CPPCC, as a proxy. On the basis of different levels of regional economic incentives, we further distinguish between central and local economic incentives and we examine the impacts of different levels of regional economic incentives on underwriters' market share. We find that relative to central economic incentives, regional economic incentives have stronger impacts on underwriters' market share. We then explore the reasons for this finding by examining the companies' performance in the 6, 12 and 18 months post-IPO. We find that IPO firms with underwriters driven by regional economic incentives tend to have worse post-IPO performance. We offer the following reasons for these stylized findings. First, local government officials face the dual pressure of balancing the fiscal deficit and engaging in a performance-based tournament for political promotions. Therefore, they have stronger incentives than their central government counterparts to interfere with economic activities in their jurisdictions. The number of IPO firms in a jurisdiction is often used as an important performance indicator that affects officials' performance assessments. Therefore, local government officials have strong incentives to push firms in their jurisdiction to seek an IPO, even if prematurely, to support their personal promotions.
From the firm's perspective, an IPO not only enhances its market recognition, reputation and development opportunities but is also an important source of financing. Therefore, firms are enthusiastic about IPOs. During the IPO process, local government support and cooperation are crucial and effective communication with the government becomes a key focus for firms. For this reason, firms are willing to use underwriters driven by regional economic incentives to strengthen connections with the local government and secure more competitive resources to support their future development. We find that one advantage of using underwriters driven by local economic incentives is that they can significantly shorten the firm's IPO waiting period compared to using underwriters driven by central economic incentives.
Underwriters driven by central economic incentives pay a much higher penalty for regulatory breaches than those driven by local economic incentives. Therefore, underwriters driven by central economic incentives tend to adopt higher practice standards. In contrast, underwriters driven by local economic incentives facing fierce competition tend to lower their client selection criteria, adopt more lenient practice standards and take ineligible firms public prematurely.
By providing an in-depth study of how different levels of economic incentives impact underwriters' incentives, our paper informs progressive policy-making in China's capital market. Our paper also highlights problems (i.e. a lack of integrity, awareness of the rule of law and due diligence on the part of some underwriters) in the current sponsorship system. Our findings are pertinent to the policy debates on how to further reform China's financial market system to optimize resource allocation and how to effectively monitor underwriters so that they can function effectively as gatekeepers in China's capital market. Our findings may serve as a point of reference for policy-making aimed at strengthening financial market regulation and penalizing regulatory breaches by underwriters.
Note 1. Because of the lack of well-accepted measures of underwriters' economic incentives, we follow the approach of Chen et al. (2017). The literature also uses the political experience of top executives (mainly the board chairman, the general manager, or both) as a proxy for firms' political connection (Xu, Jiang, Yi, & Yuan, 2013;Liu, Zhang, Wang, & Wu, 2010;Shen, Yang, & Pan, 2014;Fan, Wong, & Zhang, 2007;Yu, Wang, & Jin, 2012). The literature on political connections is based on theories about the role of the government (i.e. the helping hand vs the grabbing hand), social capital and entrepreneurs' participation in politics (Yu, Yang, & Song, 2017;Pan & Yu, 2011;Li, Qiu, & Yan, 2010). Different from developed countries with a mature capital market system, China is still transitioning from a planned economy to a market-oriented economy (Zhou & Qiu, 2013;Deng & Zeng, 2009). Given their control and discretion over huge amounts of public resources (Tian & Zhang, 2013), governments and officials at various levels tend to interfere with the market through various administrative systems and regulations (Piotroski & Zhang, 2014). In addition, when assessing the performance of local government officials, the central government has increasingly turned to some easily observable and quantifiable indicators, both to avoid erroneous assessments arising from the use of ambiguous performance criteria and to better incentivize local government officials and promote local economic development. This has led to the adoption of a tournament-style appraisal method for political promotions (Zhou, 2007). As an important indicator of capital market development, the number of IPO firms within an official's jurisdiction has become an indicator of performance (Piotroski & Zhang, 2014). Consequently, when faced with the dual pressures of balancing fiscal revenues and expenses and pursuing political promotions, governments and officials at various levels have strong incentives to interfere with the economic activities of enterprises and financial intermediaries within their jurisdiction. For practical and theoretical reasons, we use the work experience of the general manager or board chairman of the IPO underwriter as a proxy for regional economic incentives.