CHINA'S FINANCIAL SYSTEM AND ECONOMY: A REVIEW

China's financial system has been integral to its spectacular economic growth over the past 40 years. We review the recent literature on China's financial system and its connections to the Chinese economy based on the categories of Aggregate Financing to the Real Economy (AFRE), a broad measure of the nation's yearly flow of liquidity accounting for unique features of China's financial system. While early work on China's financial system emphasizes the state-owned enterprise (SOE) reform, the recent literature explores other more market-based financing channels—including shadow banking—that grew rapidly after 2010 and have become important components of AFRE. These new financing channels are not only intertwined with each other, but more importantly often ultimately tied back to the dominant banking sector in China. Understanding the mechanisms behind these channels and their intrinsic connections is crucial to alleviate capital allocation distortion and mitigate potential systemic financial risk in China.


Introduction
China's financial system, including its various markets and financial intermediaries, has been essential to China's spectacular economic growth over the past four decades (Song et al., 2011), and will likely become even more important for mobilizing capital across society in the next stage of China's economic development.To oversimplify, China's financial system-built around a colossal banking sector, an attention-grabbing stock market, a fast-growing bond market, and an increasingly complex and multi-faceted shadow banking sector-is large yet underdeveloped.However, as Allen et al. (2005) and Allen et al. (2019b) emphasize, interesting and deep heterogeneities underpin China's financial markets; thoroughly understanding its development has important implications for the broader literatures of law, institutions, finance, and growth.
By reviewing recent academic articles on China's financial system, this chapter aims to provide a comprehensive survey that inspires its connections to China's impressive growth.Most papers reviewed here were published in leading academic journals, though some still circulate as working papers; most are up-to-date. 1 We aim to summarize intellectual reflections on recent developments in China's financial system, which has in the past two decades undergone and is still undergoing dramatic and structural changes.
Unlike many excellent review articles on similar topics (Song and Xiong, 2018;Allen et al., 2017Allen et al., , 2019a)), we organize this chapter around the concept of Aggregate Financing to the Real Economy (社会融资规模, AFRE).By accounting for unique features of China's financial system, AFRE is a broad measure of the nation's yearly flow of liquidity, described by officials as "indicating total funds the real economy obtained from the financial system over a certain period of time."(Elliott and Yan, 2013) Including bank credits, equity and bond issuances, and various shadow banking items like trust loans, AFRE is perhaps the most closely monitored and analyzed statistic released by China's central bank for macroeconomic management today.Although few academic works have studied this concept in the western academic community (as it does not directly compare to other

INTRODUCTION
3 behind the Chinese central bank's move to introduce AFRE: Given rapid developments in China's financial system over the past two decades, especially since the 2009 four-trillion stimulus program, traditional credit measures like bank loans increasingly do not reflect the extent of financing to the real economy.This point is a recurring theme of this review.
Section 3, we cover the stock market.Although cumulative equity issuance only accounted for 3% of AFRE by 2021,2 the stock market's development, i.e., the launch of exchanges in both Shanghai and Shenzhen in the early 1990s, marked a milestone event in the history of SOE reform.During that period, many Chinese SOEs conducted share issuance privatizations (SIPs).We emphasize that China's privatization program during the 1990s, essentially a step toward broader SOE reforms, differs fundamentally from western-style privatizations; Beijing, never using the terms "privatization" or "partial privatization," introduced private investors to SOEs but never intended to transfer state control.More directly, ultimate privatization was never the goal of SOE reform, even for the significant 2005-2007 stock-split share reform, which converted all nontradable shares to tradable shares.
Beijing hoped these reform efforts could subject SOEs to the stock market's discipline, as in a well-functioning financial market share prices reflect information on managerial decisions and firm performance.As we review in Section 3.1, the empirical evidence is mixed, which is largely expected given the SIP program's nature, plus China's underdeveloped legal and regulatory environment typified by frequent regulatory interventions on IPO approvals (Section 3.2) and poor corporate governance practices (Section 3.3).Finally, we cover other channels whereby China's stock market supports its real economy, including the market feedback effect and stock market-banking system interactions (e.g., share pledging).
Section 4, which starts with a brief overview of Chinese banking reform history, reviews the literature on the banking and shadow banking sectors in China; these two AFRE categories together accounted for about 67% of AFRE by 2021.We combine these two lines of literature because shadow banking in China, unlike in other developed countries, can always be traced back to its mammoth banking system; as Amstad and He (2019) observe, "Chinese shadow banking is literally just the 'shadow' of commercial banks."The other unique feature of the Chinese shadow banking is the pervasive "implicit guarantee," highlighted by Zhu (2016) and Allen et al. (2021a).Most articles reviewed in this section reflect these two features.
The last important AFRE category is the bond market, accounting for about 10% of AFRE by 2021.China has taken enormous strides to develop its bond markets, an integral step in financial reforms (Amstad and He, 2019).Our review focuses on two points: Uniquely in China, two welldeveloped but segmented bond markets coexist; corporate bonds defaults have risen since 2014, generating increased attention.We highlight intrinsic connections between the corporate bond market and the (shadow) banking system.
Finally, Section 6 covers fintech development in China.Fintech, which played an important role when the shadow banking sector reached its last peak around 2017, is partially included in AFRE but not counted distinctly.As Section 6 explains, the remarkable growth of fintech in China is likely driven by both underdeveloped traditional banking (Chen, 2016) and regulators' unusually friendly approach toward the burgeoning fintech industry in the early 2010s (Allen et al., 2019a).The section covers a wide range of fintech topics, including fintech lending driven by big data technology, broad economic implications of fintech disruptions, and the rise and fall of the peer-to-peer (P2P) industry.
We emphasize that this review article does not aim to cover the complete literature on China's financial system.Political risk and corruption are two particular areas receiving significant attention, but are not specifically covered here (see, Lin et al., 2016, Liu et al., 2017and Piotroski et al., 2022).We also have not covered the interaction between local government incentives and financial market (Xiong, 2018) and China's real estate market (see Song and Xiong (2018)'s and Liu and Xiong (2019)'s excellent reviews).

Aggregate Financing to the Real Economy
Aggregate Financing to the Real Economy (社会融资规模, AFRE), also known as "Total Social Financing," was introduced by the People's Bank of China (PBC, China's central bank) in 2011 to reflect the magnitude of financial support from the entire financial system to the real economy.This section explains PBC's underlying motivation and the detailed structure of the AFRE measure, around which we organize the existing literature on China's financial system.
In this review, we mention important literature on each category.However, we do not elaborate on financial instruments that have not been studied sufficiently (e.g., government bonds, acceptance bills, asset-backed securities, and finance leases), notwithstanding their increasing contributions to the Chinese economy.

What Is Aggregate Financing to the Real Economy?
As a broad credit measure, AFRE accounts for unique features of China's financial system.
Without a clear international counterpart, this indicator is now closely monitored and analyzed for macroeconomic management and is also widely followed by the private sector and members of the general public interested in the Chinese economy.AFRE refers to the total amount of financing that the real economy can access via the financial sector during a given period.As a "flow" measure, AFRE was first released on a quarterly basis in 2011, while AFRE "level" data-defined as outstanding stock of funding provided by the domestic financial system at the reference period's end-have been available since February 2015. 3The real economy refers to nonfinancial corporations and households; the financial system refers to the entire domestic financial system, covering financial institutions (commercial banks, financial leasing firms, etc.), security markets (bond and stock markets), and other intermediary markets (e.g., banks' offbalance sheet items).More specifically, AFRE includes the following four categories.

Category 1: Bank loans in RMB and in Foreign Currencies
Category 1 includes loans made by banking institutions, mainly comprised of short-, medium-, and long-term loans and bill financing, which are on the balance sheet of the formal banking system.Bill financing (票据融资) is the PBC term for acceptance bills4 exchanged to banks often at a discount from their face value.Bank loans also include finance leases from financial leasing firms, which are classified as banking institutions in China (Chang et al., 2022).The level data show that, by 2021, outstanding bank loans accounted for 61.68% of AFRE.

Category 2: Banks' off-balance sheet items
Category 2 includes trust loans, entrust loans, and undiscounted bankers' acceptances, which are closely connected to shadow banking activities in China.Trust loans (1.39% of AFRE) are loans made by trust companies on behalf of individuals and institutions, while entrusted loans (3.46% of AFRE) are firm-to-firm loans for which trustees-either trust companies or commercial banksserve as a passive facilitator, given regulations prohibiting nonfinancial firms from making loans.
For trust loans, trust companies can decide on fund usage per prespecified guidelines, whereas for entrusted loans trustees must follow the lender's request.These two types of loans could go to various sectors, including real estate and local governments with restricted access to formal bank credit.Finally, undiscounted bankers' acceptances (0.96% of AFRE) are off-balance-sheet guarantees by banks on behalf of depositors, providing relatively weak support to the real economy (An and Yu, 2018).

Category 3: Market-based direct financing
Category 3 covers net corporate bond issuance and equity financing in China's domestic markets by nonfinancial enterprises, often called market-based direct financing as opposed to indirect financing (Categories 1 and 2).Cumulative corporate bond financing and equity financing accounted for 9.53% and 3.02% of AFRE, respectively.Corporate bond financing covers the fixedincome securities issued by nonfinancial firms.5 Equity financing includes initial public offerings and seasoned equity offerings by nonfinancial firms on the Shanghai and Shenzhen Stock Exchanges.In its "level" measure, AFRE counts the cumulative "book" equity issuance, which excludes capital gains.So far, AFRE only covers equity financing in public markets, leaving private equity financing uncounted due to data limitations.

Category 4: Other forms of financing
Category 4 includes other forms of recent development in AFRE.The largest component of this category is government bonds (16.89% of AFRE), including Treasury and municipal bonds.
Treasury bonds are issued by the Ministry of Finance (MOF) and backed by the Chinese central government, functioning as a key PBC tool for implementing monetary policy through open market operations (Fan et al., 2012;Sun, 2020;Hu et al., 2021).Municipal bonds, which were introduced during the 2014 fiscal reform to rein in the ever-growing local government debt problem, are directly issued by the local governments; these take two forms: "special-purpose" and "general-purpose," corresponding to "revenue" and "general obligation" municipal bonds in the U.S. (Lam and Wang, 2018;Chen et al., 2020b;He, 2020;Liu et al., 2021a).
The second largest component includes ABS issued by deposit-taking institutions (0.69% of AFRE) and loan write-offs (2.01% of AFRE), introduced as part of AFRE in July 2018.Other funding sources in Category 4 (including micro-lending, investment real estate, etc.) accounted for 0.37% of AFRE.Fintech loans made by micro-lending companies are essentially micro-lending loans and therefore counted in AFRE. 6

Why Aggregate Financing to the Real Economy?
For a long time, bank loans represented the most important funding source for the Chinese economy.In the past two decades, however, rapid development has diversified channels of financing for the real sector, so traditional bank credit measures have become less useful for reflecting the scope of financing to the real sector.This trend challenges the appropriateness of bank loans for representing aggregate funding, especially after the 2009 stimulus plan.
Figure 1 illustrates this point, by showing that the ratio of bank credit over GDP in China jumped from 97% in 2008 to 118% in 2009, but has remained stable since.However, using the AFRE measure, the AFRE/GDP ratio has continued to climb, rising from 119% in 2008 to 170% in 2012; the ratio reached 275% in 2021.Crucially, rapid growth in nonbank financing has caused concerns over financial stability from top policymakers in Beijing.
There are several important caveats for the AFRE measure.First, as emphasized by Zhu et al. (2013), the AFRE concept covering much broader financing sources than on-balance-sheet bank loans, does not distinguish how much various components support the economy. 7The second point 6 The two lucrative fintech lending businesses, Huabei ("just spend it"; which functions like a virtual credit card) and Jiebei ("just borrow it"; which functions like a short-term consumer loan provider), are operated by two Ant Financial micro-lending subsidiaries.
7 For instance, similar to a letter of credit, an undiscounted bankers' acceptance is merely a type of commitment that banks provide and hence represents weak support to real sectors: it is not a loan unless banks need to pay or transfer to other banks in the secondary market, nor a shadow loan until the acceptance holder obtains liquidity via is more relevant to researchers.Although closely related, AFRE is a separate concept from shadow banking in China, which mainly consists of trust and entrusted loans, wealth management products (WMPs), and underground lending (say, P2P platforms).AFRE is defined based on the asset side of banks' and financial institutions' balance sheets, focusing on supporting real economic activities.
By contrast, shadow banking focuses on various nonbank financial businesses; some components could overlap given its often vague definition, like WMPs, which could be funding sources-i.e., liability-for trust funds, implying that simply adding trust loans to WMPs will overstate the size of shadow banking.

Stock Market
The inception of China's stock market in 1990, including the two domestic stock markets (in Shanghai and Shenzhen), is the most significant event for China's financial system in the 1990s.By the end of 2021, there were 4,615 listed firms on China's stock market; its total market capitalization underground lending (say, pledging it on P2P platforms).
reached RMB 91.61 trillion, second only to that of the U.S. Despite developing rapidly, the stock market's contribution to AFRE-only 3.02% by 2021-is surprisingly small (in contrast, corporate bonds contribute 9.53%).As discussed in Section 3.4, this is partly because the accounting method of AFRE excludes capital gains and partly due to frequent regulatory interventions in IPO approvals.The latter reflects the unpleasant fact that China's stock market is less efficient than its western peers'.
After reviewing early articles from the angle of SOE privatization, we cover IPOs in Section 3.2, highlighting problematic IPO and delisting processes in China.The inefficiency of China's stock market is also reflected in ineffective corporate governance, discussed in Section 3.3.We conclude this section by discussing other important channels whereby China's stock markets connect to the real economy.

Stock Market and Privatization
To promote market forces, in the 1980s Beijing launched a series of programs that decentralized the managerial decision rights of SOEs from the government to firms.SOEs were partially "privatized" by issuing some minority shares to individual investors, who could trade on the newly established stock exchanges in Shanghai and Shenzhen.Per the literature, we call these share issuance privatizations (SIPs).
We emphasize, however, that China's privatization program differs fundamentally from other countries', say, in Eastern Europe.The top Chinese authorities never used the terms "privatization" or "partial privatization," which, by explicitly acknowledging private ownership, would intimate an ultimate goal of full privatization.This has never been the intention.Instead, Beijing officially calls this program "share ownership reform" (股份制改造), only introducing private investors to SOEs but not transferring control from the state to private investors.Besides the obvious reason that SIPs permit raising capital (Jiang et al., 2009), the main motivation behind SIPs was to subject the corresponding SOEs to market discipline, as share prices reflect information on managerial decisions and firm performance.As put by Wang et al. (2004), "China's share issue corporatization aims to transform an SOE into a modern-form corporation that features both state and non-state institutional shareholders in addition to small individual shareholders."Nevertheless, our article still uses "privatization" to be consistent with the literature.
Although the SIP program became an important measure taken by Beijing in the 1990s to reform SOEs, many scholars, including Xu and Wang (1997), question its effectiveness, arguing that it amounts to "nothing but a logo" or just "old wine in new bottles."We will thoroughly review this point shortly.
Our SIP program review relates to another literature on how SOE reform affected the performance of Chinese SOEs in early years, with mixed evidence.On the positive side, Groves et al. (1994) show that introducing certain elementary incentives brought a significant increase in productivity for Chinese SOEs.Lin et al. (1998), though concur with Jefferson and Rawski (1994) who praise China's partial and gradual reform agenda, yet argue that the reform was far from successful due to SOEs' policy burdens and resulting soft budget constraints.More recently, He et al. (2022a) study government-led incentive systems by examining a staggered reform in the SOE performance evaluation policy, and shed light on challenges faced by economic reforms in China.Finally, although the SIP program was a milestone for SOE reform, many other privatization methods were widely used.Gan et al. (2018) report that the most popular method of privatization between the late 1990s and mid-2000s was direct sales to insiders and/or outside private owners, accounting for about 69% of all privatization programs; public offering-the SIP program here-made up only 1%.

The impact of SIP program on firm performance
In an early study, Sun and Tong (2003) track 634 SOEs that went through the SIP program from 1994 to 1998.By comparing performance three years before and after privatization, the authors find encouraging signs of improvement in "total" output measures like earnings and sales, but not in "per-unit" output measures like return on sales; the second negative impact was confirmed by Wang et al. (2004).Sun and Tong (2003) conclude that China's privatization has achieved some success, though limited in comparison to other countries' privatization programs.
Like many papers, Sun and Tong (2003) blame partial privatization for the disappointing outcome, as state ownership is negatively related to firm performance after SIPs.An important caveat, though, is "legal entity ownership," essentially shareholding of other SOEs.In the context of Chinese corporate ownership structure, "legal entities" (法人股) are typically business agencies or enterprises with significant resources to be used.The positive association between performance and legal entity ownership documented in Sun and Tong (2003) is consistent with Xu and Wang (1997), who argue that legal entities-typically large blockholders that serve the same role as institutional investors in developed economies-can be more effective at monitoring.Many papers followed Sun and Tong (2003).Jiang et al. (2009) highlight that China's SIP program is a primary offering process wherein the SOE-turned listed company issues new shares directly to private investors and keeps the offering proceeds, rather than a secondary offering process as in other countries (see Bolton and Roland (1992) for central and eastern European countries) wherein the government (which sells its equity stake in an SOE) receives the offering proceeds.
The chosen "privatization" method, involving no sales of government stake, is consistent with the notion mentioned in Section 3.1 that SIPs allow the Chinese government to raise capital for SOEs without any intention of ultimate privatization.More importantly, Jiang et al. (2009) argue that since primary offerings cause a commensurate increase in firms' assets, size-related measures (i.e., "total" output like sales) used by Sun and Tong (2003) and Wang et al. (2004)  Methodology-wise, Jiang et al. (2009) use a matching sample method by identifying matched SOEs that were in the same industry and had similar characteristics but had not gone through the SIP process; this partially addresses the selection bias in Sun and Tong (2003).Jiang et al. (2009) show that SIP firms continued to experience negative post-SIP profitability changes, and this "difference" result is consistent with Sun and Tong (2003).However, the performance decline of these SIP firms was significantly less than that of matched non-SIP SOEs, and this "differencein-differences" result suggests a positive SIP treatment effect during 1999-2002 period.The same "difference-in-differences" approach is used in Fang et al. (2017) to show that corporate innovation increases after SOE privatizations.Huyghebaert et al. (2014) examine the post-listing financing decisions of 221 Chinese SOEs from 1994 to 1999 and ask whether stock market quotations alone can discipline them.Although offering an overall negative answer, Huyghebaert et al. (2014) point to the stock market's positive role by documenting that SOEs with increasing accounts receivable-i.e., trade credit-are more likely to take out bank loans and to issue new shares.Arguably, these listed SOEs used their better access to external finance to ease their trading partners' financial constraints, potentially resulting in more efficient capital allocation.In this regard, this paper is connected to Cull et al. (2009), who study how Chinese firms with privileged access to bank loans redistribute their loans to their trading partners-presumably without such access-via trade credit; see Section 4.2.1.
Ownership structure and firm performance is another active research topic.Unlike Sun and Tong (2003), Wang (2005) find that neither state ownership nor concentration of ownership is associated with negative post-listing operating performance, while legal entity ownership affects performance nonlinearly.Wei et al. (2005) give a nice description of the six different categories of shares (state, legal entity, foreign, insider, employee, and individual A-shares), and show that state and legal entity shares are significantly negatively related to Tobin's Q, while foreign shares are positively related, suggesting that agency conflicts in Chinese listed SOEs are exacerbated (alleviated) by state (foreign) ownership.

Split-share structure reform and its implications
Here we review the literature on China's split-share structure reform, largely enacted between 2005 and 2007.

Background of split-share structure reform
The SIP program beginning in the 1990s left behind an awkward legacy problem: the split-share structure.The split-share structure involves two classes (tradable and nontradable) of domestic A-shares with otherwise identical rights coexisting in a listed firm.All shares that existed before SIPs, owned by the state, legal entities, and natural persons, could not be traded on the exchanges and therefore were called nontradable shares.Only new shares issued in SIPs or seasoned equity offerings and those derived from tradable shares in rights offerings and stock splits were tradable.
The split-share structure's legacy hindered how China's financial markets functioned, as the interests of tradable and nontradable shareholders diverged naturally given different pricing mechanisms.Even worse, Liao et al. (2014) report the median of nontradable shares, as a fraction of total shares outstanding, was 61.85% before the reform, implying that concerns about split-share structure are especially acute because controlling shareholders typically held nontradable shares, which lack the capital gains benefits.Controlling shareholders would push listed firms to relentlessly raise money through seasoned offerings, only to exploit minority shareholders through either related-party transactions (Jian and Wong, 2010) or, even more bluntly, by asking listed firms to make personal loans to controlling shareholders (Jiang et al., 2010).Simply put, neither control dilution nor adverse market reactions concerned controlling shareholders, who had absolute control but did not benefit from capital gains.
In 2005, Beijing launched the split-share structure reform, or simply share reform, aiming to convert all nontradable shares into legitimate tradable shares by paying tradable shareholders certain negotiated considerations.The consideration package was market-driven without government imposed pricing schedules; often it was sweetened with some legally binding promises from controlling nontradable shareholders on future dividend payouts and/or asset injections.The proposals then circulated for tradable shareholder feedback and negotiation.By 2010, 98.58% of listed firms had completed the reform; nontradable shares were history.8 Share reform: Does it only matter for SOEs?Although most papers view the share reform as part of SOE "privatization," the deep economic issue of nontradable shares is also relevant to non-SOEs.Liao et al. (2014) report that about 40% of listed firms are non-SOEs; more importantly, the fraction of nontradable shares of SOEs and non-SOEs are remarkably similar (60.15% and 62.51%, respectively).As Section 3.3 discusses, this echoes Jiang and Kim (2020), who observe that "the dominant agency problem in China is the horizontal agency conflict between controlling and minority shareholders arising from concentrated ownership structure," regardless of whether the controlling shareholders are government agencies (for SOEs) or founders (for non-SOEs).
The impact of share reform Based on 633 SOEs and 399 non-SOEs completing share reform from 2005 to 2007, Liao et al. (2014) study the effect of "privatization" (Megginson et al., 1994;Megginson and Netter, 2001) by exploiting the fact that the reform was carried out on both SOEs and non-SOEs.For non-SOEs, the reform dismantled the split-share structure only, while for SOEs the reform additionally removed legal barriers to in-depth privatization.Liao et al. (2014) find significant improvement in "total" output measures (e.g., output, profit, employment) for both groups after the reform, and much more so for SOEs.On "per-unit" efficiency measures (e.g., ROE) and corporate governance measures (e.g., related-party transactions), both SOEs and non-SOEs performed much better after the reform, with insignificant between-group differences.
Concerning narrowly with the "privatization" effect, Liao et al. (2014) conclude that share reform did have this effect, which "quickly boosted SOE output, profits, and employment, but did not change their operating efficiency and corporate governance."However, their empirical results are also consistent with an interpretation that a successful share reform improved firms' efficiency and governance.First, the additional "total" output boost in SOEs can be driven by state-shareholders having more resources for asset injections as part of the consideration package.
Second, by removing nontradable shares, controlling shareholders' incentives were better aligned with minority shareholders', leading to a significant improvement in firms' "per-unit" measures regardless of state ownership.9 This alternative view echoes Beijing's long-standing position that the ultimate goal of SOE reform is not full privatization.In fact, after share reform allowing controlling shareholders to sell, state-shareholders had sold only 2.95% of their now-tradable shares to public investors, as of October 2011 (Liao et al., 2014).10Today, although the split-share structure no longer exists, controlling shareholders' shares remain illiquid for various reasons including insider trading regulations (Lian et al., 2022), maintaining control rights, and implicit trading restrictions.
Taking the angle of corporate governance and cash holdings, Chen et al. (2012) study the distinct impacts of share reform between SOEs and non-SOEs, each subject to different agency conflicts.The SOEs' agency conflict is shaped by the controlling government-agency's objective function, typically involving nonpecuniary considerations; the government-agency itself is often subject to its own internal control system (Lin et al., 1998).Non-SOEs' controlling shareholders would not have social welfare objectives, but may view their listed firms' cash as a means to satisfy their own needs.Corroborating this perspective, Chen et al. (2012) report that cash holdings reduction in response to the share reform is larger in non-SOEs.
Through which economic channel do pre-share reform agency conflicts operate, free cash-flows or financial constraints?Chen et al. (2012) show that after the share reform, non-SOEs increased dividend payouts but not capital investment, indicating the free cash flow channel, whereas SOEs increase capital investment and short-term borrowings, suggesting the financial constraint channel.
These findings suggest that the ability of corporate insiders to make personal use of corporate cash was more constrained in SOEs, a perspective consistent with Lin et al. (2022), who study the impact of the State Capital Operation Program initiated in 2007.
Compensation amidst the share reform As explained, the share reform involves a market mechanism wherein nontradable shareholders offer negotiated compensation-typically additional shares-to tradable shareholders.Li et al. (2011) report that average compensation from nontradable shareholders is a 30% increase in the number of shares held by tradable shareholders.They further show that compensation size is positively associated with nontradable shareholders' gain from risk-sharing.
Highlighting ownership structure, Firth et al. (2010) show that the compensation ratio decreases with the tradable-share ownership of mutual funds, suggesting that mutual funds bow to political pressure to facilitate the reform.A broader takeaway is that minority shareholders cannot count on institutional investors in China to monitor effectively (Jiang and Kim, 2020).

Initial Public Offerings (IPO)
In the SIP program, IPOs and privatization are two sides of the same coin.Several papers explore connections between IPO underpricing and long-run underperformance from the perspective of political economy.Chen et al. (2015) find that SOEs underprice their IPOs more than non-SOEs, and SOEs controlled by the central government underprice even more compared to local government peers.In Fan et al. (2007), politically connected IPO firms underperform other IPO firms in the long-run.Piotroski and Zhang (2014) argue that IPOs-often considered a barometer for economic growth-can be politically motivated; indeed, local IPO volume increases before an impending provincial-level political promotion event, but underperforms in the long-run.
Finally, the lengthy IPO approval process and regulatory-driven suspensions are unique features in China's IPO markets.A firm seeking to conduct its IPO must navigate an elaborate process administered by the China Securities Regulatory Commission (CSRC); regulators have on multiple occasions suspended all IPO activities.11In normal, nonsuspension times, firms may need two to three years to obtain IPO approval.Because firms cannot time the IPO market given the multiyear time frame from application to listing, market-wide IPO suspensions come largely as a surprise for IPO-seeking firms.Exploiting suspension-induced delay as a plausibly exogenous shock, Cong and Howell (2021) show that IPO delay has a long-lasting negative effect on corporate innovation, while the negative effect on tangible investment is only temporary.Tight IPO regulations fuel a high demand for reverse mergers, leading to a significant "shell" value of under-performing public firms and a near-zero delisting rate (Lee et al., 2022).

Corporate Governance
China's corporate governance problem is rooted in its concentrated ownership structure, featuring severe conflicts of interest between controlling shareholders and minority shareholders.In a thorough and insightful review paper, Jiang and Kim (2020) highlight several distinct features of controlling shareholders in China.First, there are two types of controlling shareholders, the state and the family; the latter maximizes profit while the former also carries political responsibilities.12Second, controlling shareholders, not CEOs or chairs of the board, are the primary decision-makers.

Controlling shareholders and related party transactions
Before 2006, a common practice whereby controlling shareholders expropriated minority shareholders was intercorporate loans, a form of related party transactions.Typically reported as "other receivables," these intercorporate loans averaged about 8% of total assets, and were rarely repaid (Jiang et al., 2010).Gul et al. (2010) argue that when large shareholders in China own more than 50% of the firm, their significant cash-flow rights make expropriation costly; indeed, intercorporate loans are more severe when controlling shareholders have fewer shares, as shown in Jiang et al. (2010).This point not only echoes the agency conflict between controlling and minority shareholders as a major drawback of the split-share structure (Section 3.1.2)),but also suggests a benefit for controlling shareholders once they have sufficient cash-flow rights.This hypothesis is confirmed in Jiang et al.
(2017): In China, listed firms with a controlling shareholder who owns less than 50% of the firm have the worst performance.
Controlling shareholders may also benefit minority shareholders when they conduct related party transactions on favorable terms to listed firms, including buying firms' assets at high prices, lending at favorable interest rates, or propping up earnings.However, Peng et al. (2011) show that propping only temporarily benefits minority shareholders, and in some cases cash is transferred back to the controlling shareholder eventually.

Boards of directors
The controlling shareholders also nominate and appoint a board of directors, rendering the monitoring role of the board largely ineffective (Jiang and Kim, 2015).Nevertheless, some directors still conduct partially effective monitoring; Jiang et al. (2016) show that independent directors' dissent improves corporate governance by conveying value-relevant information to the market.However, in Jiang et al. (2016), only 6% of highly reputed independent directors dissent even once; most dissents occur during the directors' last terms when they are less beholden to controlling shareholders.Giannetti et al. (2015) study directors having experience with superior corporate governance practices from abroad.Starting in the late 1990s, many provinces have implemented policies to attract skilled emigrants to return to China.Giannetti et al. (2015) show that directors with foreign experience facilitate the adoption of strong corporate governance practices.

The Real Role of China's Stock Market
The literature on China's stock markets is much wider than on its financial market reform.
The topics we review here typically fall in the mainstream finance literature; they have been chosen not only for their significance but also for their close connection to the real economy.
As we have mentioned, China's stock market, despite its rapid development in the last three decades, contributes negligibly (3%) to the AFRE.This is partly due to AFRE's accounting method, which measures the cumulative contribution of the stock market's financing support to the real economy but excludes capital gains.By total market capitalization, China's stock market accounted for 23.12% of AFRE. 13  China's stock market supports its real economy by at least three other channels.First, stock prices formed in a well-functioning market provide valuable information to the economy (Section 3.4.1).Second, share pledging allows listed firms' shareholders to use their shares as collateral to obtain credit (Section 3.4.2).Finally, though not covered here, the Chinese economy's internationalization continues to accelerate as the capital market further opens up.Although its capital account remains tightly controlled, China's stock market has adopted several innovative schemes to ease foreign access restrictions, including the B-share market, the Qualified Foreign Institutional Investor (QFII) scheme, and the Stock Connect program. 1413 Of course, China-as an emerging economy with a bank-dominated financial system-still lags behind the U.S. in terms of stock market capitalization.By 2021, the total market capitalization to GDP is 232% for the U.S., but only 80% for China (110% if we include the Hong Kong Stock Exchange alongside the two exchanges in Shanghai and Shenzhen).
14 See Ma et al. (2021) for more institutional details of the opening-up process of China's stock market.Among these initiatives, the Stock Connect program is a milestone event.In November 2014, the Shanghai-Hong Kong Stock Connect was launched for establishing mutual stock market access between Mainland China and Hong Kong; Shenzhen followed suit in December 2016.The Stock Connect quickly surpassed QFII as the main access route for foreign investors, whose success directly prompted the opening-up of other securities, including the Bond Connect launched in 2017.Chen et al. (2020a) document a strong predictability of foreign flows (Northbound trading) on Mainland stock returns; however, He et al. (2022c) demonstrate that this "informativeness" is partly driven by mainland insiders who evade the see-through surveillance by round-tripping via the Stock Connect.

Price informativeness
In an influential paper, Morck et al. (2000) run a CAPM-style regression for each individual stock, and use the resulting R 2 as an inverse measure of firm-specific information content in stock prices.The greater the R 2 , the higher the synchronicity across all stocks, and hence the less the price-informativeness.Morck et al. (2000) document that emerging markets typically have a higher synchronicity and note that China is an example with especially high synchronicity.Consistent with Morck et al. (2000), Gul et al. (2010) show stock price synchronicity in China is positively related to whether the largest shareholder is government related.Finally, Carpenter et al. (2021) measure stock price informativeness as the predicted variation in a cross-sectional regression of future earnings on past market valuations.Surprisingly, the authors find that stock prices in China are as informative about future profits as they are in the U.S., especially after financial reforms in the first decade of the 21st century.

Share pledging
Share pledging, wherein shareholders obtain loans with their shares of listed companies as collateral, is a unique financing vehicle at the intersection of the stock market and banking system.
It is a significant component of China's financial system; at the market's peak in 2017, more than 95% of A-share listed firms had at least one shareholder pledged, with the total value of pledged shares amounting to 6.15 trillion RMB (> 10% of total market capitalization).
Most studies show that share pledging destroys firm value, since pledging shareholders who face margin calls are likely to alter their decisions15 -e.g., cutting corporate innovation (Pang and Wang, 2020), conducting related party transactions (Li et al., 2020b).The exception is Guo et al.
(2020) who document a positive shareholder wealth effect on listed firms.
One widely held belief in this literature is that share pledging funds circle back to the listed firms (e.g., Meng et al., 2019;Pang and Wang, 2020).He et al. (2022b)

Risk, returns, and factors
Finally we briefly mention the extensive literature on the risk-return relation in China's stock market.Hu et al. (2021) report that from 1993 to 2020, the Chinese stock market yielded an average annual return of 13%, with a volatility of 43%.In contrast, the U.S. stock market during the same period delivered an average annual return of 10% and volatility of 19%, suggesting that China's stock market underperforms (once risk adjusted; see also Allen et al. (2021b)).
Considering cross-sections, many papers have studied the "standard" U.S.-based factors (say Fama-French three factors); Hu et al. (2021) offer a comprehensive review of this literature.We highlight that a few papers take into account the China-specific institutional settings.As we discussed in Section 3.2, under-performing listed companies (which tend to be small) have a significant "shell" value because of reverse mergers analyzed in (Lee et al., 2022).Liu et al. (2019b) therefore construct size factors in China differently from the classic Fama and French (1993) by excluding the smallest 30% of firms, which could be valued significantly as potential shells.

Banks and Shadow Banking
As has been observed by many leading scholars (Allen et al., 2019a(Allen et al., , 2017), China's financial system is dominated by a large banking system compared with other economies.In this section we first offer a brief overview of Chinese banking reform history, drawing heavily from two books written by former top banking officials in China (Liu, 2009;Jiang and Zhan, 2019).We then review two key topics for understanding how the Chinese banking system functions: how bank credit affects China's economic growth and shadow banking.

A Brief History of Banking Reform in China
Before 1978, China had a mono-bank system in which the PBC operated simultaneously as the central bank and sole commercial bank.In 1978, the state embarked on a major economic reform program that unfolded in a gradual and evolutionary fashion.

The first stage: 1978-1993
In the first stage, the banking system was expanded and diversified.The government split off PBC's commercial banking function into four newly established or reestablished state-owned specialized banks, also known as the Big Four: the Agricultural Bank of China (ABC), the Bank of China (BOC), China Construction Bank (CCB), and the Industrial and Commercial Bank of China (ICBC).It then formally promulgated that the PBC would function as China's central bank.
Beijing also gradually introduced competition into the banking system by allowing the entry of new financial institutions, including credit cooperatives, postal savings institutions, and joint-stock banks.However, the banking sector still operated on a credit plan system controlled by the PBC, 16   and the Big Four dominated the lending market, functioning mainly as policy lending conduits.

The second stage: 1993-2003
In 1993, the State Council issued "Decision on Financial System Reform," preluding the second stage featuring the commercialization of the banking system.Three policy banks were established in 1994 to take over policy lending deals from the Big Four; China Development Bank (CDB) is the largest one. 17In 1995, the Big Four were officially named "state-owned commercial banks" by the Commercial Bank Law.Rural credit cooperatives left the ABC's administrative control; urban credit cooperatives were transformed into city commercial banks.
In 1998, the PBC abolished the credit plan system and required banks to improve the assetliability management.An important rule of asset-liability management, which Beijing removed in 2015, was the 75% loan-to-deposit cap: Banks were prohibited from lending more than 75% of the value of their deposits.Although the PBC could still intervene, banks gained more independence to 16 The credit plan system is a mandatory quota system under which the PBC sets the lower limit on new loans and their allocation to specific sectors on an annual basis. 17The other two are the Export-Import Bank of China and the Agricultural Development Bank of China.
manage their credit allocation and pursue commercial objectives.However, the separation between policy-lending and commercial-lending functions was far from complete.With SOE reform ongoing, the Big Four shouldered a growing burden to extend loans to loss-making SOEs, resulting in a high proportion of non-performing loans (NPLs).
Under the "Grasp the Large, Let Go of the Small (抓大放小)" strategy in late 1990s (Hsieh and Song, 2015), the Big Four's loan quality severely deteriorated.In 1998, MOF issued RMB 270 respective AMCs (about 11% of total book assets), evident from the NPL ratio evolution in Figure 2, several international financial institutions including Goldman Sachs joined as strategic investors promoting their expansion (Gao et al., 2021b), though city commercial banks and rural financial institutions were classified as regional banks under strict geographical restrictions (Zhang et al., 2012).In 2014, Beijing started opening its banking sector to private investors; some private commercial banks operate as online banks and have substantially advanced the development of fintech, as discussed in Section 6.1.Figure 2

Bank incentives and capital allocations
Using a sample of Chinese SOEs from 1980 to 1994, Cull and Xu (2003) provide perhaps the first evidence that bank finance was positively linked to both profitability and reform effort (e.g., giving SOE managers discretion to set wages); Firth et al. (2009)  properly incentivized to allocate their funds to stronger firms?Cull and Xu (2003) argue that the banking reform launched in 1978 improved bank employees' incentives to evaluate credit risk.
Based on an interview with a branch-level executive, the authors reported that the compensation package of bank employees was linked to loan-portfolio quality.Supervised by high-level offices, bank branches-though often subject to certain credit plans specifying the total amount of loans to various sectors-had discretion to determine which firms receive loans based on factors such as profitability. 19f course, the financial system could work even without credit allocation to better firms.For instance, Cull et al. (2009) document that firms with privileged access to bank loans redistribute their loans to trading partners-who presumably lack such access-via trade credit.Cull et al. (2009) highlight the difference between SOEs and non-SOEs; based on a sample of firms from 1998 to 2003, the authors show that poorly performing SOEs were more likely to extend trade credit, while non-SOEs' probability is positively correlated with trade credit.Bailey et al. (2011) paint a different picture than Cull and Xu (2003).Bank loans could serve as a positive signal for borrower's quality, or could be made for policy goals.Indeed, based on loan announcements for all listed companies from 1999 to 2004, Bailey et al. (2011) document significant declines in borrowers' stock prices around loan announcements.The negative effect is stronger for firms with more frequent related-party transactions, suggesting that state-owned banks do not serve as monitors as in developed countries.
The two leading papers reviewed above present a puzzle for the finance-growth nexus literature, which has shown that financial development is an important factor in promoting economic growth.20 On the one hand, as Allen et al. (2005) argue, China seems to be a counterexample to the financegrowth nexus hypothesis; on the other hand, Ayyagari et al. (2010) document a positive correlation between bank loans and economic growth in China.We return to this topic in Section 4.2.3.
Although the 1994 tax sharing reform prohibited local governments from borrowing or running deficits, to circumvent these restrictions, in 2009 local governments were allowed to borrow heavily via LGFVs to fund stimulus spending.In a typical arrangement, a local government transfers ownership of a plot of land to an LGFV, which then borrows from banks using the land as collateral.Bai et al. (2016) estimate that about 90% of local government investment was financed via bank loans in 2009, consistent with the sudden dramatic increase of newly issued bank loans in 2009 recorded in the AFRE data (Figure 1).
This one-time massive on-balance-sheet credit expansion has had long-lasting and unintended consequences; it not only planted the seeds for the shadow banking surge several years later (Chen et al., 2020b), but also caused capital misallocation as emphasized by Bai et al. (2016) 2012) and Song and Xiong (2018), though some others have challenged this view by looking at the private sector's shares in the economy (Kroeber, 2020;Lardy, 2014).

Formal vs informal finance
Conventional wisdom, e.g., LaPorta et al. (1998), suggests that well-developed financial institutions play a critical role in economic growth.In an influential paper, Allen et al. (2005) argue that China presents a counterexample to the law and finance literature, as "neither its legal nor financial system is well developed, yet it has one of the fastest growing economies."Allen et al.

Recent banking research based on loan-level data
Haoyu Gao and Xiaoguang Yang, with the access to an administrative loan-level dataset,22 have produced a series of interesting papers alongside various coauthors.Focusing on the informational role of interfirm ownership networks in bank lending, Gao et al. (2021c) show that banks' internal loan ratings at issuance predict subsequent delinquent events more accurately when borrowers are providing direct evidence that SOEs selectively avoid defaulting on loans extended by the largest policy bank CDB, which has a higher political rank than the Big Five.
As explained in Section 4.1, the CDB, unlike other commercial banks putting profit a primary performance target, operates under the policy mandate to grant subsidized credit to SOEs in strategic industries.Ru (2018) show that CDB loans to SOEs crowd out private investment in the same industry but crowd in private investment in downstream industries, not only shedding light on the important question of how government-directed credit affects the economy, but also reconciling the ambivalence from prior studies (that rely on aggregate data only either in China (Banerjee et al., 2020) or other countries (Gale, 1991)).

Shadow Banking
According to the PBC, shadow banking refers to "credit intermediation involving entities and activities outside the regular banking system that serves to provide liquidity and credit transformation and which could potentially be a source of systemic risk or regulatory arbitrage."Shadow banking in China takes a multifaceted form and should be generally defined as markets and/or institutions operating partially outside the traditional banking sector with lighter regulations.
We first explain commonalities and distinctive features of shadow banking in China when compared to those in other developed countries.We then review several articles specific to certain shadow banking products and discuss their connections to the real economy.

Shadow banking in China: What is different?
China's shadow banking sector has two themes in common with shadow banking sectors worldwide, but also two unique features.First, as Chen et al. (2020b) argue, the upsurge of shadow banking is typically linked to rising financing demands from certain real sectors, typified by the popularity of state-chartered trust companies in late-19th century America associated with largescale railroad construction.China is no exception: Zhu (2021) argues that shadow banking activities initiated by banks prior to 1996 helped direct credits to the more productive nonstate sector whose growth craved funding.
Chinese shadow banking, like the rest of the world, also features regulatory arbitrage.As highlighted in Hachem (2018), shadow banking is credit intermediation occurring outside the regulated sphere in response to stringent regulations; blatant regulatory arbitrage takes place both when large U.S. banks actively set up asset-backed commercial papers conduits before the 2007-2009 financial crisis (Acharya et al., 2013), and when Chinese non-Big-Five banks made off-balance-sheet loans via trust companies after 2010 (Chen et al., 2020b).
Shadow banking in China is also quite unique.First, unlike in other developed countries, almost every aspect of shadow banking activities in China-made manifest by all the papers reviewed here-can be traced back to its mammoth banking system.As put by Amstad and He (2019), "Chinese shadow banking is literally just the 'shadow' of commercial banks," a view echoed in Ehlers et al. (2018) and Sun (2019).This stands in stark contrast to its U.S. counterpart where secularization and market-based instruments play a prominent role.
Another central feature of China's shadow banking sector is the prevalence of implicit guarantees whereby investors expect returns on risky investments.Unlike the fixed-income products offered in developed markets, "bond-like" investments in China, including trust products and WMPs, carry implicit guarantees from their issuers, underwriters, regulators, and even different levels of governments.As Zhu (2016) observe, "many investors believe that, as long as the (issuers and) financial institutions are concerned with their reputations, as long as the regulators are concerned with career advancement, and as long as the Chinese government is concerned with social stability, they will take care of the risks that investors themselves should bear when investing in such products."Although implicit guarantees are not necessarily unique to China, as the case of Fannie Mae and Freddie Mac in the U.S. illustrates, the scope of implicit guarantees-or investors' perception of them-is rooted in soft budget problems (Kornai, 1979(Kornai, , 1980) ) and is likely more severe in China.

Shadow banking activities
In this section we review the literature on certain shadow banking products and their connections to the real sector.state-owned bank, driven by its president's personal preferences, was particularly willing to make stimulus-related loans, raising deposits aggressively to finance these loans.The authors then instrument deposit availability with banks' exposure to competition from this large state-owned bank, and show that their issuance of WMPs exposed them to future rollover risk (He and Xiong, 2012).
Heightened rollover risk is further studied in Huang et al. (2020a), who show that banks perceived to be riskier strategically provide stronger implicit guarantees to their issued WMPs to build reputation and reduce rollover costs.For each WMP, Huang et al. (2020a) construct its "shortfall," indicating whether this WMP's realized return falls short of its highest target return the issuer bank promised.By directly measuring (the absence of) implicit guarantees, Huang et al.
(2020a)'s method is an important improvement over Allen et al. (2019c)'s, which can only measure expected implicit guarantee indirectly.

Shadow banking, real economy, and policy implications
The intersection between shadow banking and the Chinese economy has received growing attention in the literature.We review two papers here; for a comprehensive review, see Hachem (2020b) focus on municipal corporate bonds (城投债, MCB), bonds issued by LGFVs to support infrastructure investment (Section 4.2.2).Adopting an instrumental variable motivated by the literature on local governors' career incentives (Zhou, 2004;Xiong, 2018), the authors use crosssectional variation to show that MCBs rose quickly around 2014 as a way to refinance stimulus-era loans and provide aggregate evidence that WMPs have been investing heavily in MCBs to support local government infrastructure investment.We return to this paper in Section 5.2, covering Chinese bond markets.

Bond Market
Over the past twenty years, China has taken enormous strides to develop its bond markets, as an integral step in implementing financial reforms.This section covers two major topics of the Chinese bond market: the co-existence of two segmented bond markets, and corporate bonds.For readers who are interested in the historical background and institutional details, see Amstad and He (2019) and Hu and Wang (2022). 23Two additional articles are worth mentioning here: Huang and Zhu (2009) provide a detailed account of the Chinese bond market's history since 1911, including the post-1978 reform era and the early development stage between 1990 and the early 2000s; while Mo and Subrahmanyam (2020) focus on 2010 to 2019, arguing that government policies and foreign investors have significantly impacted the liquidity of Chinese bond markets.

Segmented Markets
As Amstad and He (2019) explain, there are two distinct and largely segmented markets in today's Chinese bond markets: the quote-driven, over-the-counter, "wholesale" interbank market; and the order-driven, centralized, "retail" exchange market.The exchange bond market resides within the Shanghai and Shenzhen stock exchanges; its participants include nonbank financial institutions, corporate investors, and retail investors (mainly high net-worth individuals).In contrast, the interbank market was established by the PBC in June 1997; by the end of 2021 it hosted 4,158 financial institutions, including commercial banks, urban and rural credit cooperatives, and nonbank financial institutions (e.g., mutual funds, insurance companies, securities firms).
The exchange market has aggressively competed with the interbank market, resulting in dual-  Based on the same structural framework of credit risk modeling (Merton, 1974) (2022) take the framework of He and Xiong (2012) and show that liquidity is much more important than default risk for explaining variation in the spreads of short-term commercial papers.

Corporate Bonds
As emphasized in Section 4.3.1,corporate bonds in China may carry implicit government guarantees, especially when issued by SOEs.As Geng and Pan (2021) show, amid a wave of default events after 2018, non-SOE issuers experienced a much larger increase in credit spreads, relative to SOEs.However, the implicit guarantee is not "guaranteed" and the opposite was true during the first large SOE default in 2015 (Jin et al., 2022).

Corporate bond issuances
Several recent papers focus on corporate bond issuance.Ding et al. (2022) discover robust overpricing in China's primary interbank market; driven by rebates and self-purchases by underwriters, issuance overpricing stands in sharp contrast to issuance underpricing in developed financial markets (Ritter and Welch, 2002;Cai et al., 2007).
Expanding the scope to offshore corporate bond markets, Huang et al. (2021b) show that offshore USD-denominated Chinese corporate bond issuance is driven by the wedge between domestic and foreign interest rates.From a different angle, Flannery et al. (2022) show that the issuance of two USD-denominated sovereign bonds issued by the Chinese government in 2017 helped reduce yield spreads, bid-ask spreads, and volatility of offshore dollar-denominated Chinese corporate bonds.

Fintechs
Globally, the financial industry has undergone radical transformation wherein incumbent banks face increasing competition from the rising fintech sector.The rapid development and astonishing growth in China's fintech sector are particularly remarkable, so we cover this topic separately although fintech in only partially included in AFRE and not counted distinctly.24 As Chen (2016) mention, due to underdeveloped traditional banking, especially small/medium enterprises' and consumers' lack of credit access, the large unfulfilled demand facilitates the entry of fintech firms.But perhaps more importantly, regulators took an unusually friendly approach to the burgeoning fintech industry in the early 2010s.For instance, privately owned companies faced no stifling restrictions to entering financial service industries (Allen et al., 2019a).
Ant Financial provides a good case study of success and many papers reviewed here use data from Ant Financial and its parent company Alibaba.Its business model rests on enabling direct interactions among mass users on digital platforms, with a large stock of user data as an essential byproduct.Ant Financial uses these data as an input to offer a range of services that exploit network effects, generating further user activity and facilitating further financial inclusion (Luohan Academy, 2019).However, fintech's development in China is not without potential concerns.Pushing this idea of "data as a new type of collateral for fintech lending business" further, Gambacorta et al. (2022) compare the lending practices of Ant-backed MYbank, a private depository institution providing online banking services (Section 4.1.3),to those of traditional banks typically relying on physical collateral. 25The authors show that fintech loans do not correlate with local business conditions and house prices, but react strongly to changes in firm characteristics, like transaction volumes and network scores (important inputs of firms' credit ratings developed by Ant Financial).In sharp contrast, bank credit is positively correlated with local house prices, consistent with the classic "collateral channel" a la Kiyotaki and Moore (1997).By showing that fintech credit is less reliant on traditional collateral, this paper delineates potential transmission mechanisms of an effective monetary policy in China.

Financial Disruption and Economic Implications
By challenging incumbent financial institutions with new business models, the rising fintech sector could potentially disrupt the traditional financial system, with profound implications for both macro-and microeconomic policies in China.among those who won coupons to those who just lost and find that coupons generate an immediate increase in weekly consumption among winners by RMB 3-6 out-of-pocket spending for every RMB in government subsidy, suggesting that the coupon program, which delivered a much larger marginal propensity to consume (MPC) than traditional policies in the U.S. (e.g., Johnson et al. 2006), can be a useful addition to policymakers' stimulus toolbox.

Peer-to-Peer Lending Platform
Peer-to-peer (P2P) lending allows individuals to bypass traditional financial intermediaries and raise funds directly from investors, potentially providing alternative funding to Chinese small businesses.Outstanding P2P loans in China went from almost nothing in 2007 to a peak of RMB 1.3 trillion in June 2018, only to collapse due to pyramid-scheme scandals and absentee bosses, sparking public anger and a broader government crackdown.
He and Li (2021) provide a comprehensive analysis for why P2P platforms failed in China.
The original P2P platform business model is to charge a service fee for connecting borrowers and lenders; lenders cannot appeal to platforms for payment if a borrower defaults.However, since Beijing initially took a light regulatory touch from 2015 to2017 as mentioned in Section 6, most Chinese P2P platforms guaranteed their investors a return.Moreover, platforms themselves issued short-term WMPs that promised high returns (Li et al., 2020a).P2P platforms, essentially engaging in credit and maturity transformation like traditional banks, became a breeding ground for risky lending and fraud (see Ezubao case in Li et al. (2020a)).In November 2019, Chinese authorities mandated that all existing P2P lending platforms be converted to micro-lending companies within two years, essentially eliminating the Chinese P2P industry.
P2P platforms themselves need financing.Jiang et al. (2021) show that P2P platforms with SOE affiliations have higher trading volumes, attract more investors, and offer lower interest rates.
Using a clever identification strategy of fake SOE affiliations, the authors demonstrate that the SOE affiliation itself serves as a positive signal.Relatedly, Li et al. (2020a) show that the number of lenders increase significantly immediately after P2P platforms obtain venture capital investment, supporting the certification effect of venture capital on the P2P industry.

Conclusion
China's financial system has become increasingly intertwined with the Chinese economy's astonishing growth over the past four decades.Unlike most existing surveys, this review is structured around Aggregate Financing to the Real Economy (AFRE), a broad PBC credit measure including bank credits, equity and bond issuances, and various shadow banking items like trust and entrusted loans.Besides contextualizing the development of China's financial markets within SOE reform from the 1990s to mid-2000s, our review highlights how non-bank credit, representing new marketbased financing channels, plays an increasing role in supporting the real economy.Furthermore, all these new financing channels were somewhat affected by the 2009 four-trillion stimulus plan's hangover and are often ultimately tied back to the dominant banking sector with "implicit guarantees." Therefore, understanding the mechanisms behind these financial channels and their intrinsic connections is crucial to alleviate capital allocation distortion and mitigate potential systemic financial risk in China.

Figure 1 :
Figure 1: Aggregate Financing to the Real Economy.This figure shows the ratio of AFRE level over GDP from 2002 to 2021, which is decomposed into four categories, bank loans, off-balance-sheet financing, market-based direct financing, and other forms of financing.
cannot capture efficiency improvement and hence are inappropriate in the Chinese context.Revisiting Sun and Tong (2003)'s research question, Jiang et al. (2009) demonstrate a positive impact of SIPs on Chinese SOEs from 1999 to 2002, three years after Sun and Tong (2003)'s sample period.The authors attribute it to the improved institutional environment in the Chinese stock market since 1999.
Qian et al. (2021)'s great survey covers many important aspects of the IPO market in China-the policy evolution, IPO pricing, bidding and allocation practices, and aftermarket trading-from 1990 to 2018.Besides extremely severe IPO underpricing,Qian et al. (2021) present another interesting empirical fact: during aftermarket trading, IPO institutional bidders, whether or not they received an allocation, rarely buy the stock on the open market; those who receive allocations sell the first week they are allowed.This implies that Chinese investors-including institutional ones-pay little attention to firm fundamentals given their ultra-short investment horizons.
clarify that pledging shareholders have discretion over pledging funds, which could be personal consumption, etc.Based on He et al. (2022b)'s survey, a majority of shareholders (67.3%) used pledging funds outside the listed firms.By linking firm registration data with share pledging data, He et al. (2022b) offer a more positive perspective on share pledging, arguing that shareholders' pledging transactions are related to their entrepreneurial activities.Utilizing a policy shock favoring share pledging by natural person shareholders (over legal entity shareholders), He et al. (2022b) show natural person shareholders increased entrepreneurial activities significantly in response to the policy shock.
billion of special Treasury bonds to recapitalize the Big Four, and four newly created state-owned asset management companies (AMCs) acquired nearly RMB 1.4 trillion of NPLs at book value from the Big Four.Meanwhile, China's accession to the World Trade Organization (WTO) in 2001 yielded a timetable for integrating China into the global economy.Specifically, the banking sector was committed to opening up to foreign competition within five years, heavily pressuring domestic banks.The third stage started in 2003 and focused on "share ownership reform."In October 2003, Beijing officially approved the state-owned commercial banks' share ownership reform plan.Besides the Big Four, the Bank of Communications (BOCOM), which the China Banking Regulatory Commission (CBRC) 18 officially classified as a state-owned commercial bank in 2007, was also included, turning the "Big Four" into the "Big Five."The share ownership reform involves three steps: recapitalization, the disposal of NPLs through AMCs, and the introduction of foreign strategic investors.In December 2003, Central Huijin Investment Company was established as a wholly state-owned company, authorized by the State Council to exercise rights and obligations as a shareholder in state-owned financial institutions.From 2003 to 2008, Central Huijin injected capital totalling USD 79 billion funded from foreign exchange reserves into the Big Four, and RMB 3 billion capital funded from PBC relending to BOCOM.Meanwhile, after the Big Five transferred a combined RMB 2 trillion of NPLs to their

Figure 2 :
Figure 2: Evolution of Banking Sector Composition and NPL Ratio.This figure shows the composition of banking sector in terms of total assets and NPL ratio by composition from 2003 to 2021.Others include policy banks and non-bank financial institutions.
also confirm this positive relationship with non-SOE firms covered in the 2003 World Bank survey.But are Chinese state-owned banks

( 2005 )
examined three real sectors during the past four decades: the listed sector (covered in Section 3), the (non-listed) SOE sector, and the private sector including all other firms with private and local government ownership.Consistent with the law-finance-growth nexus theory(LaPorta et al., 1998),Allen et al. (2005) find slow growth in the first two sectors due to poor legal protections and weak financial markets.However, the astonishing growth of the third sector, which has received credits from nonbank and nonmarket sources, challenges this classic theory.Allen et al. (2005) argue that there exist effective, alternative financing channels and corporate governance mechanisms to support China's growth.This view is challenged by Ayyagari et al. (2010) who study a comprehensive firm-level dataset from the 2013 World Bank Investment Climate survey.Ayyagari et al. (2010) show that bank financing is associated with faster growth; and all firms, irrespective of size, benefit from access to formal financing.The authors hence conclude that reputation and relationship-based informal financing is likely limited and unlikely to substitute for formal mechanisms.What drives the discrepancy between Allen et al. (2005) and Ayyagari et al. (2010)?Allen et al.(2019b), as a response toAyyagari et al. (2010), provides an answer.Even though the 2013 survey constructed by the World Bank covers many financing sources, the largest financing component for Chinese firms is "others."Allen et al. (2005) classify unidentified "others" as informal financing, whileAyyagari et al. (2010) consider it internal financing, which largely reconciles the seemingly contradictory results(Allen et al., 2019b).21 connected to banks' existing customers via ownership networks.And Gao et al. (2021b) study how the 2009 deregulation of bank entry altered China's local banking industrial organization and its economic consequences.Finally, Gao et al. (2021a) paint a politics-finance nexus picture by

(
2018).Chen et al. (2018) provide an excellent overview on the institutional background of China's quantity-based monetary system, its commercial bank regulations, and the relationship between shadow and traditional banking.Using a dataset on entrusted loans similar toAllen et al. (2019c),Chen et al. (2018) show that contractionary monetary policy from 2009 to 2015 stimulated shadow banking and encouraged banks to bring shadow banking products onto their balance sheets, hampering the effectiveness of monetary policy.This is consistent with the PBC's underlying motivation for introducing AFRE in 2011 (see Section 2.2).Chen et al. (2020b) argue that the accelerated growth of China's shadow banking after 2012 is tightly linked to its 2009 stimulus plan reviewed in Section 4.2.2.Although the dawn of shadow banking in China around 2008 can be attributed to other factors(Hachem and Song, 2021), only after 2012 did its growth begin accelerating: WMPs grew by a total of RMB 3.5 trillion during the three-year period from 2008 to 2011 but increased by RMB 2.5 and 3.1 trillion in the single years 2012 and 2013, respectively; a similar pattern holds for trust and entrusted loans.Chen et al.
listed bonds.Due to market segmentation documented inChen et al. (2022), dual-listed bonds could show different prices despite the same fundamentals.Liu et al. (2019a) show that dual-listed bonds with stronger retail demand are traded at significantly higher prices on the exchange market than those traded by institutional investors on the interbank market.Taking advantage of an institutional feature wherein the two markets adopt different rules for repurchase agreement (repo) transactions,Chen et al. (2022) identify the value of asset pledgeability, a challenging task given the endogeneity of asset pledgeability itself.They use a policy shock in December 2014 that rendered a class of AA+ and AA bonds ineligible for repo on the exchange market to study the value of pledgeability in bond pricing.
, Geng and Pan (2021) study long-term corporate bonds and find that credit spreads in China become informative only after the first corporate bond default in 2014.Focusing on short-term bonds, Huang et al.

Fintech
lenders, beyond automated services, have revolutionized the collection and use of credit information.Ant Financial and Alibaba's success in the past decade drives a burgeoning literature on fintech lending in China.Using data from Ant Financial,Hau et al. (2019) show that big data technology helps fintech promote inclusive financing, in that fintech firms have advantages in making loans to vendors who either have low credit scores or had been excluded from bank credit.Hau et al. (2019) extend this result and argue financial inclusion matters most for underdeveloped and largely segmented credit markets in emerging markets.Hau et al. (2021) analyze automated credit lines by Ant Financial to vendors on Alibaba's online retail platform (Taobao).Using a discontinuity in the credit decision algorithm,Hau et al. (2021) show that a vendor's credit approval boosts its sales and transaction growth.Interestingly,Hau et al. (2021) report that proximity to Big Five Bank branches correlates positively with the take-up of fintech credit, implying that vendors face greater credit frictions when closer to Big Five Banks, consistent with traditional banking system in China hurting financial inclusion.Screening borrowers is only one of many critical steps wherein fintech lenders utilize big data technology.Using a large sample of personal loans from a fintech lender in China,Dai et al. (2020) show that information acquired by the lender through borrowers' digital footprints can significantly increase repayment likelihood on delinquent loans.Digital footprints serve as a new type of collateral and can enhance financial inclusion by facilitating delinquent loan collection.
Buchak et al. (2021) argue that the fintech sector is an important economic force driving interest rate liberalization in China.Using data from Ant Financial,Buchak et al. (2021) study fintech's role in ending "financial repression" in China via the introduction of a money market fund with deposit-like features.Cities and banks whose depositor bases are more exposed to fintech see greater deposit outflows and respond to fintech competition by offering competing products with market interest rates.AsBuchak et al. (2021) observe, fintech thus facilitates a "bottom-up" interest rate liberalization, with direct monetary policy implications for China.Indeed, by studying loan-level data from a P2P platform,Huang et al. (2021a) find that monetary policy easing is associated with a higher probability of fintech lenders, who serve as a lightly regulated financial intermediary, granting loans to risky borrowers.Braggion et al. (2021) analyze P2P lending around a 2013 tightening of mortgage loan-to-value caps in several Chinese cities and show that home buyers borrow from P2P platforms to bypass the tighter loan-to-value cap.The fintech industry underwent tremendous growth during the global COVID-19 crisis.Bao and Huang (2021) compare fintechs and banks during the COVID-19 outbreak in China and find that fintechs are more likely to expand credit access to financially constrained borrowers.But the increased credit provision may not be sustainable, as the delinquency rate of fintech loans tripled after the outbreak (but not for banks).On the government policy side, during 2020 summer several Chinese local governments launched a consumption stimulus program, leveraging mobile payment platforms to dispense small-value, use-it-this-week-or-lose-it coupons.Exploiting participants' rush to first-come-first-served digital portals,Xing et al. (2021) andLiu et al. (2021c) compare spending Buchak et al. (2021)ion of China's banking sector shows that the relative proportion of state-owned commercial banks by total assets decreased from 54.93% in 2003 to 39.27% in 2021.Embarking on interest rate liberalization reforms in the late 1970s, China made significant progress in the 2010s.The PBC scrapped bank lending rates and deposit rate ceilings in 2013 and 2015, respectively.In 2019, the PBC introduced the market-driven loan prime rate (LPR) as the benchmark lending rate and reformed how participating banks submit quotations.In 2015, PBC launched an official deposit insurance scheme that covered more than 99% of depositors up to RMB 500,000 each, paving the way for banks to compete without putting depositors at risk.Replacing the implicit guarantee (Section 4.3.1)withanexplicitguarantee,this is an important step for interest rate liberalization, studied inLiu et al. (2021d)andBuchak et al. (2021).
Chen et al. (2018)021)gulation, the CBRC tightened enforcement of the 75% loan-to-deposit cap after 2008.Hachem and Song (2021), discussed in Section 4.2.1, argue that in the presence of interbank market power, a regulatory tightening of liquidity rules can trigger unintended credit booms funded by shadow banking, echoingChen et al. (2018), who show that shadow banking hampers the effectiveness of monetary policy in China.Although the 75% cap for loan-to-deposit ratios was removed in 2015, banking regulators are still monitoring banks' loan-to-deposit ratio as a liquidity indicator.

The 2009 four-trillion RMB stimulus plan and its impact
focus on interest rate liberalization; their general equilibrium model features both an inefficient SOE sector and an efficient non-SOE sector.Although interest rate liberalization improves capital allocations within each sector, it exacerbates misallocations across sectors-in general equilibrium, a higher deposit rate encourages more savings flowing disproportionately to less efficient SOEs.Taking a different angle at funding competition in the interbank market, Hachem and Song (2021) construct a model in which, somewhat surprisingly, liquidity regulation can trigger unintended credit booms in the presence of interbank market power.After liquidity tightening, smaller price-taking banks respond with more aggressive shadow banking activities, causing a funding reallocation away from the big price-setting bank and potentially a credit boom.Both papers demonstrate that regulatory actions often lead to unintended yet profound consequences, especially given the Chinese financial markets' unique institutional features.Right after the global 2007/08 financial crisis hit the export-driven Chinese economy, the State Council announced the four-trillion RMB fiscal plan in November 2008.Most of the massive 2009 stimulus package was implemented through China's local governments, who finance infrastructure investment by bank loans through off-balance-sheet local government financing vehicles (LGFVs) Acharya et al. (2021) largest nonbank financial industry in China since 2012, trust companies are non-deposit-taking institutions-hence less regulated than traditional banks-that raise funds through the issuance of investment products to investors.Typically, these "trust" products are risky, but Chinese investors expect implicit guarantees, presumably provided by trust companies or their selling banks, as Section 4.3.1 notes.Allen et al. (2021a)analyze a comprehensive dataset of trust products from 2002 to 2015 and document that the pricing of trust products (i.e., the promised yield) depends on both the underlying investment risk and the implicit guarantee's strength.For instance, yield spreads are lower if the trust firm is larger or controlled by an SOE, or if the Big Five sells the product.And, following the first high-profile near-default of a trust product in 2014, which shocked perceptions of implicit guarantees, spreads increased but the effect was concentrated among products with lower-strength guarantees.Wealth management products.The Big-Five were instructed to extend loans to LGFVs in 2009 (Section 4.2.2).Using a proprietary dataset that covers the largest 25 commercial banks,Acharya et al. (2021)study the role of WMPs from 2007 to 2014.They argue that one large Like the banking system and stock market, the corporate bond market plays a prominent role by channeling household savings toward the real sector in China.At the end of 2021, the outstanding corporate bond balance in AFRE was RMB 29.93 trillion, comparing quite favorably to RMB 9.48 trillion in the stock market yet significantly behind RMB 193.77 trillion in bank loans.spreads depend not only on the financial conditions of the issuing LGFVs but also on the fiscal conditions of their backing municipalities.As reviewed in 4.3.3,Chen et al. (2020b) connect the dramatic 2012-2015 growth of MCBs to the 2009 stimulus plan, and provide further evidence that WMPs invest heavily in MCBs.Another rapidly growing literature concerns rising corporate default in China.Li and Ponticelli (2022) study the recent progress of bankruptcy resolution in China, based on a case-level dataset from 2011 to 2020.They compare bankruptcy cases handled by specialized versus traditional civil courts and argue court specialization increases efficiency via selection of better-trained judges and greater judicial independence from local politicians.