Funny Money or Legitimate Finance? Shadow Banking and Policy Impacts in China

Abstract

Government officials in China have taken different views regarding shadow banking. Some have seen the industry as overly risky, potentially undermining the formal financial system, while others have asserted that it is an increasingly important part of the financial system, filling a gap in finance provision to particular sectors and smaller firms. Do their views matter? Regulators have striven to crack down on the riskiest practices in shadow banking, but are the policies effective? In this article, we analyze the impact of government attitudes and actions on the shadow banking sector. Using a unique data set based on information collected from various sources in a difference-in-difference model, we find that shadow banking regulation plays a strong role in China’s financial sector, while contradictory government views (in the form of commentary in the People’s Daily) on shadow banking do not. This reveals that shadow banking is strongly affected by political authority when it is codified into regulation. Only some aspects of shadow banking can be legitimized through regulation, while the remainder of China’s financial system remains constrained due to state dominance over the financial sector. This underscores the “funny” nature of shadow banking’s money flows. This article is one of the first to study the effects of government views and regulations on the shadow banking system.

Keywords

shadow banking institutional formation regulation money flows financial system

Introduction

Is Chinese shadow banking “funny money” or legitimate finance? From the perspective of government officials, there is no clear answer. Shadow banking in China has been viewed by some officials and industry experts as illegitimate finance but as a key means of financing by others. For the former, the industry has been seen as overly risky, potentially undermining the formal financial system. The latter see shadow banking as an increasingly important part of the financial system, filling a gap in finance provision to particular sectors and smaller firms.

The divergent views on shadow banking relate to its dual impact on the financial system, particularly on commercial banks. The size of the shadow banking sector in China positively affect banks’ profitability but weakens stability in the short-run (Cao & Wang, 2019). These different impacts on the financial system have resulted in either opposition toward shadow banking or opposition toward regulation. The government itself has taken both views over the development trajectory in this industry, sometimes at the same time. Whether these views matter more or less than policy itself in reducing shadow banking activity is the question.

Before we analyze the effect of divergent views on shadow banking, we answer three key questions to better understand the nature of Chinese shadow banking. First, what is the importance of shadow banking in China? As China’s financial system is inhibited by expectations that banks will fund government policies and entities first and foremost, shadow banking came to play a couple of conspicuously absent roles in China that were badly needed: lender to riskier entities and provider of higher yields to investors. Some of the riskier borrowing entities were state-related firms building up infrastructure under the poststimulus policy directive. In contrast, others, particularly small- and medium-sized enterprises (SMEs), have provided and continue to provide the bulk of gross domestic product (GDP) production. This bifurcation in types of borrowers is particular to the Chinese economy, which has promoted the state sector to carry out policies and has permitted the private sector, in the form of private, often family-run SMEs that are not encouraged to grow too large.

Shadow banking’s promise of higher returns was attractive to households with increasingly higher incomes. This is because China’s banking system, the main source of finance, has not provided significant returns on deposits. Even after deposit interest rates were liberalized, they remained close to the central bank’s official benchmark rates. Other investments, such as stocks and bonds, were not well developed and often incurred high risks. Wealth management products promised far higher returns than bank deposits and appeared to enjoy implicit repayment guarantees.

Shadow banking provided alternative investment forms, especially when underlying assets, such as junk bonds, risky loans, stocks, and a panoply of higher yield investments, were bundled together. The rise of trust and entrusted loans, bankers’ acceptance notes,¹ wealth management products, and peer-to-peer (P2P) lending platforms imitated a better developed financial market that offers various instruments with different risk levels. China’s shadow banking system was not, in reality, a developed financial market, since most of the shadow banking channels were simply outgrowths of the existing bank-centric structure. The effect, however, was the same: funds could be made available to riskier borrowers, and buyers could purchase the instruments as investments.

Trust loans consisted of loans from trust companies, nonbank financial institutions. Entrusted loans are loans between two (usually state-owned) companies that use a bank as an intermediary. Bankers’ acceptance notes were used between buyers and sellers to defer payment on purchases, while the bank intermediary is entrusted to make the payment on a particular day. Wealth management products were issued by banks, trusts, and securities firms and bundled underlying assets to provide higher returns than depository accounts. P2P lending platforms connected borrowers with lenders, cutting out bank intermediaries in the lending process.

Second, how is China’s shadow banking system different from that in the West? China’s shadow banking is markedly different from the U.S. shadow banking system. A major reason for this is that Chinese financial institutions and local and central governments have provided implicit guarantees on shadow banking products (Dang et al., 2014). This guarantee is based on the assumption that if a product fails, the government will step in to bail out the failed product and repay investors. Shadow banking entities have been active in issuing guarantees; for example, when banks are viewed as riskier, the bank provides even greater guarantees on shadow banking products in order to maintain its reputation (Huang et al., 2020).

This maintains stability in the financial system while creating moral hazard, allowing investors to purchase risky products while placing the burden of the government’s potential product failure. Besides, the Chinese shadow banking system is bank-focused, unlike the U.S. system, which runs parallel to the banking system. As shadow banks cannot accept depository funds, they are closely tied to the banking system. Banks are closely regulated, and the larger state-owned banks are often relied on to carry out government policy. Initially, shadow banks took on channeling funds to the riskier sectors that banks were not permitted to lend to. They also provided an alternative investment channel for retail investors who had few profitable options outside of the stock market.

The bank-oriented nature of China’s shadow banking system has both reduced investor risk and increased systemic financial risks. Banks issued over 702,000 wealth management products between January 2007 and February 2019.

Despite the variety of shadow banking institutions that arose, many products were loan-centric and did not incorporate direct financing methods that might have better-aligned investors and borrower incentives. Trust loans, entrusted loans, and even P2P loans dominated the industry and comprised securitization. The need for short-term loan repayments frequently led to Ponzi-like financing to repay investors when loans wobbled. Thus, these alternative financial channels’ structure was not a strong complement to bank loans and were possibly even poor substitutes.

Chinese shadow banking also differs from that in the West since many of those holding risky investments were retail, rather than institutional investors, such as pension funds or life insurance companies. Another difference is that the underlying Chinese assets often were associated with fewer covenants that restricted debtor activities, such as taking out many additional loans.

The implications of the first are that investors, mainly individuals, did not understand risks. It didn’t help that products sold to these individuals often lacked transparency about the underlying assets. Because investors did not understand the risks of their investments and would protest when and if they failed, causing social instability, the government frequently stepped in to cover the losses. This meant that there were really two investors–households, which enjoyed the upside of shadow banking products. The government (often local governments) absorbed the downside of shadow banking products. This resulted in excessive investment in risky products due to investors’ intentional or unintentional awareness of moral hazard.

Due to a lack of restrictions on debtor activities, borrowers became overleveraged. This led to the excessive buildup of debt throughout China’s economy. For state-owned companies, moral hazard arose again since financial firms were amenable to lending to institutions protected by the state. The state would likely assist state-owned enterprises in trouble.

Third, how does shadow banking play an important role in money creation? In China, money is actively created outside of the formal banking system by both banks and other entities that lend in order to profit, going against China’s established creed of lending to government policy-upholding institutions. This active creation of money runs contrary to economic orthodoxy, which views money as passive, and upholds the Keynesian view that money is an active force in social life.

Although banks play a central role in China’s shadow banking system, banks’ shadow activity consists of loans taking the form of other types of assets, which may not be covered by financial regulation or monetary policy (Sun, 2019). This reduces the purview of regulators within the shadow banking sector. In fact, much shadow banking activity grew in order to escape regulation, particularly restrictions on the traditional financial sector that reduced funding to key economic entities. Strict regulations on lending to the real estate industry and to local government financing platforms in particular led to the rise of unregulated or less-regulated shadow banking finance (Allen et al., 2020).

Scholars have argued that regulators do not understand the nature of money, and therefore do not understand shadow banking sufficiently in order to regulate this (Seru, 2019; Tymoigne, 2006). In China, however, this is not the case. Regulators have been aware of risks associated with the shadow banking system, but due to the unique constraints in China’s formal financial system and overall political economy, with preference given to state-owned enterprises and banks, they have faced major challenges in cracking down on shadow banking due to its ability to generate growth in favorable circumstances.

Other scholars have argued that regulators and financiers collude to reduce regulation in profitable areas in order to mutually benefit (Gabor, 2018; Skidelsky, 2018). This is also not the case in China, as regulators do not generally profit from shadow banking. Perhaps this is true for local government officials, whose jobs depend on increasing local GDP growth, do benefit from shadow banking, but such economic indicators do not determine the job security of regulatory authorities.

However, while regulators do not benefit from reducing regulation, shadow banking entities win or lose by becoming legitimized through regulation, which either quashes or endorses existing activities. Investors may also win by obtaining another outlet in which to save or invest, reducing the “existential uncertainty” of the future for people living in regimes of self-responsibility (Wray, 2012, pp. 12-13).

Now that we have described the major characteristics of shadow banking, we discuss the views and policies related to shadow banking. We then perform a series of analyses on shadow banking’s entrusted loans, looking at the impact of government policy and media hype surrounding the sector. The strengthening of regulatory policies in the fourth quarter of 2017 provides us with a quasi-natural experiment for determining the extent to which policies were effective in dampening shadow banking. We use provincial panel data within a DID model to analyze the impact of government officials’ views and shadow banking regulatory policies on the size of entrusted loans. We end with a discussion about our unique findings on government influence on shadow making activities.

We note that very little has been written on this topic. Some papers discuss monetary policy’s impact on the shadow banking sector, but not on government regulation. One finding within the monetary policy/shadow banking literature shows that contractionary monetary policy, policy that reduces funds in the economy, brought about shadow banking after the global crisis (K. Chen et al., 2016). Our article and findings are therefore unique and represent one of the first attempts to understand the impact of Chinese policies and views on its financial sector.

Officials’ Views of Shadow Banking

Shadow banking arose after the global financial crisis in 2008, mainly to serve the needs of local governments and corporations attempting to fulfill government policy to stimulate the economy. As bank finance was constrained, particularly toward riskier sectors, alternative means of financing became prominent. In some cases, shadow banking was even endorsed by the government.

Government officials’ view of shadow banking as “funny money” versus legitimate finance varied greatly. For example, Zhuo Chen et al. (2017) assert that the central government encouraged local governments to borrow through local government financing vehicles. The government allowed other shadow banking channels, such as trust and entrusted loans, to rise without excessive regulation, as they were necessary to bolster China’s growth trajectory. At the same time, excessive risks were condemned.

Shadow banking was accepted in the beginning. In 2012, the Governor of the PBC, Zhou Xiaochuan, stated, “shadow banking is inevitable when banks are developing their business . . . but there are fewer problems here than the shadow banking sector in some developed countries that have been hit by the global financial crisis.”

Hu Xiaolian, Vice Governor, People’s Bank of China, stated on September 26, 2014, “we can’t just say ‘no’ to shadow banks, because to some extent, they satisfy some financing demands of a diversified economy.”

Some officials, however, voiced their concerns over shadow banking. Xiao Gang, Chairman of the Bank of China, stated in November 2012, “because the number is so large and difficult to manage, China’s shadow banking sector has become a potential source of systemic financial risks in the coming years” (Zhao, 2012).

Differences in regulators’ viewpoints on shadow banking were made public in 2013, as banks fought against Regulation 9, a rule that was to be issued by the CBRC (Wei & Davis, 2014). The rule limited banks’ off-balance-sheet lending activities, preventing commercial banks from using the interbank market to make corporate loans. After a meeting between regulators and bankers in October 2013, the CBRC altered its plans to accommodate bankers. Thus, the People’s Bank of China became concerned that the CBRC would protect the interests of banks over the rest of the financial system and regulate too lightly. The CBRC itself believed that regulators should balance bank profitability and stability.

In general, however, regulators understood that, at the minimum, rules needed to be implemented in step with rising risks. For example, in August 2010, banks were restricted in bundling loans off-balance sheet through trust companies into wealth management products. Wealth management funds were restricted regarding how much could be invested in nonstandardized debt assets as of March 2013. In December 2013, banks were required to separate wealth management product funds from their own funds. In April 2014, trust companies were forbidden from carrying out business with shadow banking characteristics, and interbank business was standardized. Entrusted loan rules were laid out in January 2015. Internet guidelines were implemented in July 2015. In July 2016, the market value of securities held by wealth management products was restricted.

After the entrance of Guo Shuqing in February 2017 as head of the China Banking and Insurance Regulatory Commission, shadow banking was more strongly constrained to reduce risks. Guo had an order to audit the risky assets of China’s banks. He stated, “rules aren’t being followed. Those who were caught before have been repeating their offenses” (Yap, 2017). From the start, Guo uncovered violations of Chinese regulations, including those by China Minsheng Bank, which had obtained 1.65 billion yuan ($239 million) illegally from investors for nonexistent wealth management products.

Other officials agreed with the need to regulate shadow banking. In July 2017, Wang Zhijun, an official with the Office of the Central Leading Group on Finance and Economic Affairs, stated that China’s financial grey rhinos include shadow banking, property bubbles, state, and local government debt, and rampant illegal fundraising. Yi Huiman, the Chairman of the Industrial and Commercial Bank of China, stated in March 2017, “if we do not deal correctly with shadow banking, the risks could be huge” (Bradsher, 2017).

Indeed, since 2017, the supervision of the regulatory authorities has been greatly strengthened. The CBRC has carried out special rectification and issued multiple documents to strengthen risk management and control, make up for regulatory shortcomings, and provide collateral management guidance. The content covers most of the problems in bank operations, including financial management, banks’ regulatory adjustment (such as fraud), banks’ disguise of bad loans, and compliance with various rules and regulations in daily business activities. The year 2017 was called “the most stringent” regulatory year in the country, and regulatory orders were issued intensively.

In November 2017, the People’s Bank of China, together with the China Banking Regulatory Commission, the China Securities Regulatory Commission, the China Insurance Regulatory Commission, and the Foreign Exchange Bureau, drafted the “Guiding Opinions on Regulating the Asset Management Business of Financial Institutions (Draft for Comment).” The “Guiding Opinions” contains a total of 29 articles, including guidelines to: establish classification standards for asset management products; reduce the risks of shadow banking; reduce liquidity risk; break payment guarantees; control the leverage level of asset management products; restrain multilayer nesting and channel services; strengthen supervision and coordination; and set a reasonable transition period.

On December 22, 2017, the China Banking Regulatory Commission issued the “Notice on Regulating Banking and Credit Services.” This is the first regulatory rule issued after implementing the new regulatory documents for large asset management. The reform measures require banks to: clarify the definition of bank and trust business and channels; use supervision that is based on substance over form and penetration principles; and regulate the behavior of trust companies.

In July 2018, the CBIRC announced draft measures for commercial banks’ asset management business, while the People’s Bank of China announced a notice regarding the guidance of the financial asset management business of financial institutions.

The size of trust and entrusted loans declined in 2018 due to tightened regulations. Guo’s crackdown on the shadow banking industry reduced the risk of contagion and increased the potential for financial stability.

Importantly, the aim of regulation has not been to quash shadow banking altogether. Some officials have continued to voice their support for shadow banking as long as risks are controlled. According to Wang Zhaoxing, vice-chairman of the China Banking and Insurance Regulatory Commission, regulators are trying to distinguish between good and bad shadow banking to balance benefits and risks (Zhou, 2019).

Figures 1 and 2 show the rise and fall of shadow banking, first as a percentage of RMB bank loans, and second in terms of entrusted loans for 2012 through 2019. One can see the impact of regulations on the shadow banking sector taking effect in 2017 through 2019, particularly in Figure 2.

Figure 1.

Shadow banking to RMB bank loan ratio.

Figure 2.

Trust and entrusted loans.

Next, we analyze the impact of government views and regulations on the shadow banking system, and discuss the implications.

Hypothesis Development

The Chinese government has significant power in controlling shadow banking activity by making known its risk sentiment; that is whether government risk fears are high or low, whether the government wishes to allow shadow banking to run its course or not, is rapidly transmitted to the shadow banking industry as well as to the financial industry overall.

There are two major mechanisms in which this works. The first mechanism is through policy curbing shadow banking activities. This has an immediate effect on the shadow banking industry. Government policy may be a direct command given to institutions, but this is most commonly reserved for banks. For example, in February 2019, a circular issued by the Communist Party Central Committee and the State Council required banks to lend to private firms (Tang, 2019). For other institutions, failure to comply with regulations can mean dissolution or severe punishment. In Hunan province in October 2019, P2P lenders were entirely banned due to their failure to comply with new regulations (Liu, 2019).

The second is through the media. State media represents government views on shadow banking risks, which the public reads and acts upon. The state media may incorporate contradictory views on shadow banking, making this channel less effective in impacting shadow banking activity itself.

In terms of the former mechanism, one major reason that the government risk sentiment is transmitted so effectively through policy, as we will see in our empirical analysis, is because the state, through three major financial regulators as well as the central bank, has the ultimate power in the financial sector to either quash or encourage activity. All institutions are ultimately subordinate to governmental power. Entrusted loans in particular have usually been made between state-owned enterprises, normally held responsible for carrying out government policy, so they are an excellent subject to study as we concern ourselves with the legitimate nature of shadow banking.

The empirical analysis seeks to prove the following main research hypotheses:

Hypothesis 1: The regulatory policy on the shadow banking system will reduce the scale of entrusted loans.

Hypothesis 2: To a lesser extent, we also explore whether media mention of shadow banking risks results in a decline in the scale of entrusted loans.

Data and Methodology

Data Source

We first test whether government opinion on shadow banking risks affects entrusted loans’ holdings to find out whether changing government opinions affect investors’ actions.

Data for the base case is from January 1, 2010, through December 31, 2019. Policy data is taken from the main financial government regulatory bodies: the China Banking and Insurance Regulatory Commission, the China Securities Regulatory Commission, and the People’s Bank of China. Data collected on state media views of shadow banking is taken from the People’s Daily, considered the Communist Party’s mouthpiece. We tabulate whether the phrase “shadow banking risk” is mentioned in the daily news, inputting a “1” if the phrase is found and “0” if it is not.² Some article contexts for “shadow banking risk” include discussion of shadow banking risk contagion, trust product defaults, and the necessity of external supervision rather than self-discipline for shadow banking risks in the banking industry.

We first test whether regulation on shadow banking resulted in a decline in entrusted loans. Data used in the provincial DID model is from 2013 to 2019. We include an indicator for shadow banking policy, denoting 1 if a shadow banking policy was announced on a particular day or 0 if it was not. Shadow banking policies include those made on entrusted or trust loans, wealth management products, bankers’ acceptance notes, and others. Other macro-level variables come from the CEInet Economic Statistical database.

Empirical Model

Government Policy of Shadow Banking System on Entrusted Loans

Our baseline tests examine the relationship between government shadow banking policy and mention of shadow banking in the People’s Daily on entrusted loans. The primary regression specification is a standard OLS regression as Model 1. The explained variable is entrusted loans represented by EL. The explanatory variable is government policy on the shadow banking system at period t represented by $G o v P_{t}$ . If a government proposes a regulatory policy on the shadow banking system, then the government policy is represented by RuleP, with an indicator of 1 if a rule was announced on the date and 0 if no rule was announced. If there is any mention of “shadow banking risk” found in the People’s Daily, China’s government mouthpiece, this is reflected in the variable RiskP. $X_{t}$ represents control variables.

Specifically, we include the following control variables:

GDP growth, measured by the current year-on-year actual growth rate of GDP.

The degree of financial development, measured by the ratio of the current domestic and foreign currency loans of financial institutions to GDP.

Proportion of real estate development investment, measured by the ratio of current real estate development investment to GDP.

Financial autonomy, measured by the ratio of per capita general public budget revenue to general public budget expenditure in the current period.

Consumer price index (corresponding period of last year = 100).

Investment in fixed assets year-on-year growth rate.

E L_{t} = β_{0} + β_{1} G o v P_{t} + r X_{t} + ε_{t}

(1)

A Quasi-Natural Experiment: Shadow Banking Supervision and Entrusted Loans

The strengthening of regulatory policies in the fourth quarter of 2017, as described above, provides a quasi-natural experiment for our research. This article will use provincial panel data and adopt the DID method to analyze the impact of shadow banking regulatory policies on the size of entrusted loans. Considering that shadow banking supervision policies have a stronger inhibitory effect on regions with a larger shadow credit scale, we divide the data into a processing and control group according to whether the size of regional shadow credit is larger than the median level. Specifically, if the sum of entrusted loans, trust loans, and undiscounted bills (used to describe shadow banking) in a region is higher than the annual median level, it will be regarded as a region strongly affected by the shadow banking regulatory policy; otherwise, it will be regarded as regions weakly affected by shadow banking regulatory policies. The empirical model is shown below, as in Formula (2).

\begin{array}{l} E L_{i t} = δ_{0} + δ_{1} T r e a t e d_{i} * P o s t_{t} + δ_{2} T r e a t e d_{i} + δ_{3} P o s t_{t} \\ + r X_{i t} + T i m e_{f i x e d - e f f e c t} + P r o v i n c e_{f i x e d - e f f e c t} + ε_{i t} \end{array}

(2)

$E L_{i t}$ represents the size of the entrusted loans of the area i in the t period. $T r e a t e d_{i}$ is a dummy variable of the experimental group, in which the processing group takes a value of 1, and the control group takes a value of 0. $P o s t_{t}$ is a policy identification variable, taking the value 0 before the fourth quarter of 2017, and 1 thereafter. $X_{i t}$ represents the control variable. The selection and calculation method of the control variable is the same as in Model 1. The model further controls for time fixed effects $T i m e_{f i x e d - e f f e c t}$ and provincial fixed effects $P r o v i n c e_{f i x e d - e f f e c t}$ . This article mainly observes the coefficient of $δ_{1}$ . If the $δ_{1}$ coefficient is significantly negative, it means that the proposed shadow banking supervision policy will restrain the scale of entrusted loans.

Empirical Results

Government Policy of Shadow Banking System on Entrusted Loans

Considering that the data may have autocorrelation problems, this will affect the regression results. We control the fixed effects of year, month, and quarter to reduce autocorrelation issues’ impact on empirical results. To further ensure that the model does not have serious autocorrelation problems, we first calculate the residual error and its lag value, then draw a scatter plot of the residual error and the residual lag term (Figure 3). The results show no linear relationship between the residual and the first-order lag of the residual term; that is, there is no autocorrelation problem in the model setting.

Figure 3.

Residual autocorrelation test.

Table 1 reports the results on the estimation of government regulation on entrusted loans. Panel A shows the regression result of governments’ regulatory policy of shadow banking on entrusted loans. In columns I and II of Panel A, we have controlled the year and month fixed effects and respectively added no control variables and all control variables. In Columns III and IV, we further control the quarterly fixed effect. The empirical results show that the government’s regulatory policy of shadow banking harms the scale entrusted loans, proving Hypothesis 1. However, the results also show that the impact of mention of shadow banking risks in the People’s Daily has no significant impact on the scale of entrusted loans, disproving Hypothesis Two.

Table 1.

Government Policy Effect on Entrusted Loans.

Dependent variable	I	II	III	IV
Dependent variable	Entrust loans
Panel A
RuleP	−0.1132* (0.066)	−0.1193** (0.059)	−0.1132* (0.066)	−0.1193** (0.059)
Controls	N	Y	N	Y
Year FE	Y	Y	Y	Y
Month FE	Y	Y	Y	Y
Quarter FE	N	N	Y	Y
Observations	120	109	120	109
R ²	.805	.841	.805	.841
Panel B
RiskP	−0.0413 (0.027)	−0.0312 (0.023)	−0.0413 (0.027)	−0.0312 (0.023)
Controls	N	Y	N	Y
Year FE	Y	Y	Y	Y
Month FE	Y	Y	Y	Y
Quarter FE	Y	Y	Y	Y
Observations	120	109	120	109
R ²	.805	.839	.805	.839

Note. Standard errors in brackets. FE = fixed effect.

p < .01. **p < .05. ***p < .10.

Table 2 reports the difference-in-difference regression results of shadow banking regulatory policies on entrusted loans. Columns I and II in Table 2 add no control variables and then add all control variables. The first two columns control provincial fixed effects, year fixed effects, quarterly fixed effects, and province * year fixed effects. The results show that the Treated * Post coefficient of the interactive item is negative at the 5% statistical level, indicating that after the shadow banking supervision policy is strengthened, the size of the entrusted loans in the regions affected by the policy has decreased more than those in the regions less affected by the policy (they have lower levels of shadow banking overall). Columns III and IV in the table further control the province * quarterly fixed effect. The results show that after controlling all the information, the interactive item Treated * Post coefficient is −0.6811, which is significantly negative at the 5% statistical level, and the conclusion still has not changed.

Table 2.

Difference in difference model.

Dependent variable	I	II	III	IV
Dependent variable	Entrust loans
Treated * post	−0.5058** (0.214)	−0.6018*** (0.217)	−0.6109*** (0.220)	−0.6811** (0.257)
Treated	0.8802*** (0.195)	0.9338*** (0.193)	0.8570*** (0.183)	0.9110*** (0.184)
Post	−0.3247** (0.151)	−0.0628 (0.203)	−0.2479 (0.167)	−0.0380 (0.210)
Controls	N	Y	N	Y
Province FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
Quarter FE	Y	Y	Y	Y
Province * Year FE	Y	Y	Y	Y
Province * Quarter FE	N	N	Y	Y
Observations	738	714	738	713
R ²	.788	.792	.835	.836

Note. Standard errors in brackets. FE = fixed effect.

p < .01. **p < .05. ***p < .10.

The basic assumption of the difference-in-difference model is the parallel trend hypothesis, which requires that the difference between the changing trend of the treatment group and the control group can only occur after the time point under examination. However, the coefficient may be driven by some factors before the policy time point, rather than the policy itself. Therefore, it is necessary to verify that in the fourth quarter of 2017, before the enhancement of the shadow banking regulatory policy, the processing group, and control group regions have similar trends in the size of entrusted loans, and the trend began to change in the fourth quarter of 2017.

In the parallel trend test, this article defines 14 dummy variables, Before8, Before7, Before6, Before5, Before4, Before3, Before2, Before1, if the time is respectively in the 8th quarter, 7th quarter, 6 before the fourth quarter of 2017 Quarters, 5 quarters, 4 quarters, 3 quarters, 2 quarters, 1 quarter, and Treated is equal to 1, the dummy variable is 1; otherwise, it is 0. If the meeting time is also the fourth quarter of 2017 and Treated = 1, then Current = 1. Otherwise, Current is 0. After1, After2, After3, After4, the value of 4 dummy variables is 1; if the time is in the first quarter after the fourth quarter of 2017, 2 quarters, 3 quarters, and 4 quarters, and Treated = 1.

The first two columns in Table 3 show whether there is a difference in the trend of the change in the size of the entrusted loans between the processing and control groups, which would occur before the strong supervision policy for shadow banking in the fourth quarter of 2017 took place. The results show no significant difference in the size of the entrusted loans between the processing and control groups before the shadow banking regulatory policy was strengthened at the end of 2017. The current coefficient is significantly negative at the statistical level of 10%, indicating that the parallel trend hypothesis test has passed. Columns III and IV in Table 3 set the time dummy variable of the first quarter of 2016 and before. The interactive item with Treated to Before8_ shows that the coefficients of Before8_, Before7, Before6, Before5, Before4, Before3, Before2, Before1 are not significant, indicating that the parallel trend assumption is still true.

Table 3.

Parallel Trend Test.

Dependent variable	I	II	III	IV
Dependent variable	Entrust loans
Before8	0.8329* (0.447)	0.5296 (0.433)
Before8_			−0.3843 (0.402)	−0.2052 (0.383)
Before7	0.1492 (0.513)	0.1873 (0.443)	−0.1446 (0.652)	0.0342 (0.573)
Before6	−0.1120 (0.511)	−0.0021 (0.502)	−0.4354 (0.673)	−0.1751 (0.632)
Before5	−0.2322 (0.426)	−0.0536 (0.421)	−0.5571 (0.565)	−0.2249 (0.509)
Before4	0.5507 (0.463)	0.5101 (0.480)	0.1323 (0.547)	0.2744 (0.468)
Before3	0.5480 (0.474)	0.6608 (0.469)	0.2666 (0.596)	0.5054 (0.603)
Before2	−0.4159 (0.358)	−0.4570 (0.365)	−0.7165 (0.523)	−0.6199 (0.541)
Before1	−0.8094 (0.690)	−0.7086 (0.624)	−1.1248 (0.841)	−0.8735 (0.750)
Current	−0.6947* (0.373)	−0.6305* (0.369)	−1.1107** (0.451)	−0.8596** (0.414)
After1	1.1266*** (0.402)	1.2268*** (0.409)	0.8802** (0.360)	1.0946*** (0.376)
After2	1.7196*** (0.459)	1.5275*** (0.378)	1.4558*** (0.417)	1.3917*** (0.380)
After3	1.5221*** (0.444)	1.3887*** (0.371)	1.2543*** (0.405)	1.2487*** (0.351)
After4	1.1921** (0.505)	1.2210** (0.448)	0.8259* (0.477)	1.0108** (0.421)
Controls	N	Y	N	Y
Province FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
Quarter FE	Y	Y	Y	Y
Observations	738	714	738	714
R ²	.495	.538	.494	.538

Note. Standard errors in brackets. FE = fixed effect.

p < .01. **p < .05. ***p < .10.

To ensure that the shadow banking regulatory policy brings about the change in the size of the entrusted loans and is not caused by other policies, this article further uses the placebo test to replace the double differential discontinuity point (Table 4). Assuming that the shadow banking supervision policy’s proposed time is one year ahead of schedule, we use the fourth quarter of 2016 as the policy split point. The interactive term Treated * Post2 of the pseudo dummy variable of the constructed shadow banking supervision policy and the experimental group’s dummy variable will be added to the model Rereturn in the middle. The results show that Treated * Post2 is not significant, indicating that the change in the size of the entrusted loans occurred after enhancing the shadow banking supervision policy, indicating the robustness of the empirical results of this article.

Table 4.

Robustness Test: Placebo Test.

Dependent variable	I	II	III	IV
Dependent variable	Entrust loans
Treated * Post2	0.0016 (0.201)	−0.1100 (0.211)	−0.0479 (0.192)	−0.1562 (0.204)
Treated	0.6922*** (0.176)	0.7656*** (0.179)	0.6611*** (0.176)	0.7455*** (0.174)
Post2	0.3479 (0.340)	0.4151 (0.351)	0.3949 (0.329)	0.4013 (0.299)
Controls	N	Y	N	Y
Province FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
Quarter FE	Y	Y	Y	Y
Province *Year FE	Y	Y	Y	Y
Province * Quarter FE	N	N	Y	Y
Observations	738	714	738	713
R ²	.784	.789	.831	.833

Note. Standard errors in brackets. FE = fixed effect.

p < .01. **p < .05. ***p < .10.

In the robustness test, this article divides the strong and weak areas affected by the shadow banking regulatory policy according to the entrusted loan size above the 3/4 quantile level and below the 1/4 quantile level. After changing the processing group’s division method and the control group, the DID double-difference results are shown in Table 5. The results show that, regardless of whether the provinces * quarterly fixed effects are controlled or whether control variables are added, the Treated2 * Post interaction term’s coefficient is significantly negative at the 5% level. That is, the conclusion of this article is stable.

Table 5.

Robustness Test: Different Division of Treatment and Control Group.

Dependent variable	I	II	III	IV
Dependent variable	Entrust loans
Treated2 * Post	−1.4613*** (0.431)	−1.3999*** (0.436)	−1.5833*** (0.460)	−1.7174** (0.751)
Treated2	1.7080*** (0.429)	1.7003*** (0.464)	1.9806*** (0.522)	2.0460*** (0.602)
Post	0.1639 (0.368)	0.5506 (0.482)	−0.1223 (0.575)	0.5842 (0.846)
Controls	N	Y	N	Y
Province FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
Quarter FE	Y	Y	Y	Y
Province * Year FE	Y	Y	Y	Y
Province * Quarter FE	N	N	Y	Y
Observations	359	343	323	308
R ²	.820	.828	.872	.882

Note. Standard errors in brackets. FE = fixed effect.

p < .01. **p < .05. ***p < .10.

Discussion

This article has confirmed that government policy curbs entrusted loans, a key component of China’s shadow banking system. Talk of “shadow banking risks” in the People’s Daily does not seem to play an important role in these industries’ real activity. We note that mention of “shadow banking risks” in the government newspaper is not necessarily stated by government officials, so this may play a role in reducing the impact of this factor on shadow banking activity. In addition, however, as we can see above from the qualitative evidence, officials themselves have held mixed views on shadow banking; this may be a reason that mention of risks in the newspaper is not a deciding factor in determining shadow banking activity.

From this, we can gather that, while officials may support or oppose shadow banking, as noted above, it is regulation itself that lenders and borrowers pay attention to. This may be why Chinese government officials are able to provide both positive and negative comments about shadow banking; the statements themselves do not move markets, which would create financial instability. That would render commentary politically unacceptable.

The 2017 crackdown on shadow banking was necessary because the Chinese financial system is deeply flawed, and shadow banking has failed to fix these flaws. Even though shadow banking has filled a gap in both financial supply and demand, the risks that have arisen for both suppliers and consumers of shadow banking finance have often produced negative effects that outweigh the benefits since proper credit assessment was often not applied. In fact, by 2017, the risks created by shadow banking had become a political liability, with President Xi Jinping declaring financial security to be vital to national security and Premier Li Keqiang stating that the country’s financial sector was vulnerable to shadow banking.

Furthermore, the problem of excessive debt had also become a major political issue by 2017. Shadow banking exacerbated both financial risks as well as excessive debt. Top economic officials widely noted this. The rest of the world had become well aware of these twin risks. Western analysts frequently commented on the fact that shadow banking risks compounded the fragility of debt-ridden firms.

Policy advice was provided in the hopes that China would move further toward market-based, less risky finance. It was suggested that regulators should attempt to encourage direct and market-based finance to improve the efficiency and reach of the financial sector. The reasoning was that, even though officials have, to some small extent, promoted direct finance, the capital markets comprised of the corporate bond sector, stock market, and private equity sector remain underdeveloped and insufficiently market-based. Also, increased coverage of equity financing could extend more funding to some firms. As of April 2015, equity markets only accounted for 11% of the M2 money supply, a broad measure of money, in China compared with 250% in the United States (Shepard, 2015). U.S. companies are heavily dependent on equity financing, but Chinese corporations rely much more heavily on bank loans and retained earnings. In addition, under 20% of China’s direct financing is based on debt financing, which implied that China has room to expand and improve bonds’ issuance, particularly of corporate bonds.

The fact that the crackdown on shadow banking was spearheaded by a single-minded regulator, Guo Shuqing, meant that there would be no backtracking on the policy once implemented. Guo Shuqing himself took one of the most hawkish stances toward shadow banking among government officials. He wrote in August 2020 that

at present, shadow banking has undergone unremitting governance and risks have been controlled to a certain extent, but the soil for its survival has not been completely eradicated. A slight deregulation may lead to a full resurgence and abandonment of all previous efforts. It is necessary to maintain strategic determination and maintain high pressure on high-risk businesses. (Guo, 2020)

The results of our analysis reveal that, in fact, Guo’s crackdown on shadow banking was highly successful. This underscores the widely held understanding that the Chinese government is strongly effective in implementing policies where necessary and can maintain financial stability in its economy at will. No other country in the world can purport to have the same policy effectiveness, essentially aligning political intent with political action. Therefore, if we circle back to the title of this article, shadow banking is considered funny money or legitimate finance when the government designates it as such. Government officials may wobble around this view in their media statements, but the ultimate authority lies in the hands of policy makers themselves to legitimize or delegitimize shadow banking.

For those shadow banking entities that do survive regulation, the process serves to legitimize these institutions or products, at least in the short run. For example, despite the extensive regulation imposed on wealth management products in recent years, they remain an important part of China’s financial system, and have been permitted to continue being sold. Wealth management products are now sold by banks’ wealth management subsidiaries, and there are plans to expand the products to other financial institutions (“China Issues Draft,” 2020). As we have seen from the model above, the process delegitimizes institutions or products that do not survive regulations. Entrusted loans, which have been delegitimized to some extent due to regulation, have experienced negative year-on-year growth since the regulations took hold.

Conclusion

We have discussed how government officials viewed shadow banking, with some viewing the sector as overly risky and others viewing it as an important means of diversifying the financial system. We then analyzed the impact of government views and government regulation on the shadow banking sectors. Using a unique data set on shadow banking policies in an OLS regression and provincial panel data in a Difference-in-Difference method model, we find that shadow banking regulation plays a strong role in curbing Chinese entrusted loans. The strengthening of regulatory policies in the fourth quarter of 2017 provides a quasi-natural experiment for our research. The DID regression results show a significantly negative impact on the shadow banking regulatory crackdown in 2017 on entrusted loans. We implement robustness tests to ensure that 2017 is a meaningful breakpoint for the data series and that our conclusions remain strong. We follow this with suggestions as to how regulators could lead the way in improving direct and market-based finance.

Shadow banking has provided lenders, borrowers, and investors with new channels for finance, and its regulation has served to legitimize the highest-performing practices and delegitimize the worst practices. However, shadow banking does not entirely serve China’s financial needs due to constraints built into China’s bank-centered policy lending regime. The large impact of regulations on the shadow banking sector reveals that this type of finance cannot easily survive without creating unwanted risks, and its wholesale legitimization is not the answer to China’s funding shortages. Indeed, market-based finance is the only true resolution to reducing financial constraints.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Sara Hsu is a Visiting Scholar at Fudan University. She was previously an Associate Professor of Economics at the State University of New York at New Paltz. Dr. Hsu specializes in Chinese shadow banking and fintech.

Xun Han is an Assistant Professor at Beijing International Studies University, Beijing, China. She holds a PhD from the Central University of Finance and Economics in Beijing and specializes in research on the Chinese financial system.

References

Allen

Qian

(2020). Implicit guarantees and the rise of shadow banking: The case of trust products. Imperial College London. http://cfrc.pbcsf.tsinghua.edu.cn/Public/Uploads/upload/CFRC2019_1864.pdf

Bradsher

(2017, March 18). Shadow lending threatens China’s economy, officials warn. New York Times. https://www.nytimes.com/2017/03/18/business/china-economy-debt-shadow-banking.html

Cao

Wang

(2019). The effects of shadow banking on stability and profitability of China’s commercial banks: Evidence from panel VAR. General and Social Science: Current Trends.

Chen

Ren

Zha

(2016, January). What we learn from china’s rising shadow banking? Exploring the nexus of monetary tightening and banks’ role in entrusted lending (NBER Working Paper No. 21890). http://www.nber.org/papers/w21890

Chen

Liu

(2017). The financing of local government in China: Stimulus loan wanes and shadow banking waxes. http://www.zhiguohe.com/uploads/1/0/6/9/106923057/p7_zc_draft_20170825.pdf

China issues draft rules on bank units’ sales of wealth management products. (2020, December 25). Reuters. https://www.reuters.com/article/us-china-banks-wealth/china-issues-draft-rules-on-bank-units-sales-of-wealth-management-products-idUSKBN28Z0GT

Dang

T. V.

Wang

Aidan Yao

(2014). Chinese shadow banking: Bank-centric misperceptions. http://www.columbia.edu/~td2332/Presentation_Banking.pdf

Gabor

(2018, November 27). Why shadow banking is bigger than ever. Jacobin Magazine https://jacobinmag.com/2018/11/why-shadow-banking-is-bigger-than-ever

Guo

(2020, August 16). Guo Shuqing: Dispose of non-performing assets as soon as possible to prevent high-risk shadow banks from rebounding. Sina Finance. https://finance.sina.com.cn/china/gncj/2020-08-16/doc-iivhvpwy1287721.shtml

10.

Huang

Shao

(2020). The rise of implicit guarantees: Evidence from shadow banks in China. https://policyconf.pbcsf.tsinghua.edu.cn/Public/Uploads/upload/The%20Risk%20of%20Implicit%20Guarantees%20Evidence%20from%20the%20Shadow%20Interbank%20Market%20in%20China.pdf

11.

Liu

(2019, October 17). China’s Hunan province imposes total ban on P2P lenders after operators fail to comply with regulations. South China Morning Post. https://amp.scmp.com/business/china-business/article/3033378/chinas-hunan-province-imposes-total-ban-p2p-lenders-after

12.

Seru

(2019, June 30). Regulating banks in the era of shadow banks [YouTube]. Bank for International Settlements. https://www.youtube.com/watch?v=TTzsMJKV81c

13.

Shepard

(2015, August 11). When the stock bubble burst, did it take the rest of China with it? Reuters. https://www.reuters.com/article/idUK250613536420150811

14.

Skidelsky

(2018). Money and government: The past and future of economics. Yale University Press.

15.

Sun

(2019). China’s shadow banking: Bank’s shadow and traditional shadow banking (BIS Working Papers, 822). https://www.bis.org/publ/work822.pdf

16.

Tang

(2019, February 15). China’s government tells banks to ‘increase financial support’ for private firms to aid slowing economy. South China Morning Post. https://www.scmp.com/economy/china-economy/article/2186309/chinas-government-tells-banks-increase-financial-support/

17.

Tymoigne

. (2006, November). An inquiry into the nature of money: An alternative to the functional approach (Working Paper 481). The Levy Economics Institute.

18.

Wei

Davis

(2014, January 14). Regulators at odds on reining In China’s shadow lending. Wall Street Journal. https://www.wsj.com/articles/SB10001424052702303819704579320054102965402

19.

Wray

L. R.

(2012, May). Introduction to an alternative history of money (Working Paper No. 717). Levy Economics Institute of Bard College.

20.

Yap

C.-W.

(2017, May 14). Chinese banks rattled by regulatory blitz. Wall Street Journal. https://www.wsj.com/articles/chinese-banks-rattled-by-regulatory-blitz-1494766803

21.

Zhao

. (2012, November 21). Hot spotlight: China’s “shadow bank” will not risk losing control. People’s Daily. http://theory.people.com.cn/GB/n/2012/1121/c49155-19650540.html

22.

Zhou

(2019, March 10). China to “strike a balance” between good and bad shadow bankers to halt fall in growth. South China Morning Post. https://www.scmp.com/economy/china-economy/article/2189394/china-mulls-reprieve-some-shadow-bankers-stop-fall-growth