Bulletin – March 2018 Global Economy Non-bank Financing in China


The rise of ‘shadow banking’ activities over the past decade has provided a range of benefits to the Chinese economy and financial system. Yet, it has also raised financial stability risks, prompting Chinese authorities to make many changes to the regulatory and supervisory framework. This article examines the nature and complexity of shadow banking in China and how policy makers have responded to the challenges it poses.


Non-bank financial institutions (NBFIs) often provide some bank-like services, including extending credit by connecting savers with borrowers. Similar to banks, this generally involves the transformation of short-term liquid liabilities (such as deposit-like products) into long-term illiquid assets (such as loans). When these activities are conducted by NBFIs, the activities are typically referred to as ‘shadow banking’. Shadow banking activity is common in developed financial systems and can deliver a range of benefits, including: more flexible and innovative finance; greater access to funding; more competition for banks; and diversification of risks for investors.

The global financial crisis (GFC) illustrated, however, that shadow banking activities can also pose financial stability risks and amplify shocks in several ways (Adrian and Jones, 2018; Pozsar et al, 2013). Shadow banking is subject to less stringent regulation and supervisory oversight, which can result in weaker lending standards, facilitate an excessive build-up of leverage and reduce capital and liquidity buffers within the financial system. In addition, entities outside the formal banking sector generally do not have access to official sector backstops such as central bank liquidity facilities, which makes them less resilient to liquidity squeezes. The activities of shadow banks are generally linked to those of banks, for example through funding channels, contingent credit lines, and implicit guarantees. Risks emanating from shadow banks can therefore spill over to the rest of the financial system in episodes of financial instability.

In China, shadow banking activity has grown rapidly over the past decade. The Financial Stability Board (FSB) estimates that China's shadow banking system – measured as the total financial assets (or in the case of funds, assets under management) of NBFIs relative to GDP – is now similar in size to that in many advanced economies, and is large relative to that in most other emerging market economies (Graph 1). However, there are different definitions and measures of shadow banking activity. The stock of shadow financing – defined as the amount of non-bank credit intermediation to the real economy (excluding bonds) – is a subset of the shadow banking system. We estimate shadow financing accounts for around 25 per cent of total non-financial sector debt in China, up from 5 per cent a decade ago (Graph 2). Shadow financing has grown at an annual average rate of around 40 per cent since 2009 to be around US$7 trillion by September 2017, equivalent to about 60 per cent of GDP. For more details about these estimates see Appendix A. Shadow finance has become significantly more important as a source of finance to the real economy and, as later discussed, has also been closely connected to the traditional banking system.

The rapid growth in China's shadow banking system has occurred largely because the traditional banking system could not meet the strong growth in demand for funding after the GFC because of regulatory constraints. The state retains substantial influence over banks' allocation and pricing of credit, for instance through loan quotas, directed lending, single borrower credit limits, and the effective setting of interest rates. The stimulus package implemented by the Chinese authorities during the GFC led to strong demand for credit, particularly from local governments and state-owned enterprises (SOEs) who had primary responsibility for executing the desired large increase in spending on infrastructure and construction projects. Regulatory constraints and tighter bank credit conditions from late 2009 meant that Chinese banks were not able to meet the increase in demand in full.

Graph 1
Graph 1 Shadow Banking Activity
Graph 2
Graph 2 China – Non-financial Sector Debt

To help fund the large expansion in credit, banks and NBFIs created a variety of investment products, which were sold to households, corporations and other financial institutions. The returns on these investments were attractive compared with bank deposit rates, which were capped by regulators. These developments also spawned a large asset management industry in China.

Over the years, the Chinese authorities have attempted to address risks stemming from the expansion of shadow banking activity by boosting their oversight, improving their coordination and tightening regulations. However, rapid financial innovation, especially where new products and entities traverse the responsibilities of multiple regulators, has made this difficult. This is a familiar challenge that has confronted policy makers in other economies in the past. So far, the measures imposed by the Chinese government have had tangible effects on shadow banking activity over the past year or so, with the overall pace of shadow financing growth slowing significantly.

The Nature of Shadow Banking in China

The entities and products that constitute shadow banking differ in form from one economy to another. Figure 1 sets out a simplified depiction of the range of entities and products involved in China's shadow banking sector.

Types of Shadow Credit

There are five broad types of shadow financing in China (Graph 3):

Graph 3
Graph 3 China – Shadow Financing

Trust loans

In China, trust companies are financial firms that manage assets and make investments (including writing loans) on behalf of clients.[1] They are the largest type of NBFI in China and are the main non-bank financial entity licensed to issue loans directly. Regulation of trust companies over the past decade has intermittently attempted to limit banks channelling funds through trusts.[2]

Other shadow debt

Other shadow debt captures a range of illiquid and opaque debt assets that are difficult to separately identify. This includes bank-originated loans that have been subsequently sold, letters of credit, accounts receivable or other non-standard types of debt assets. It is likely that both banks and NBFIs are involved in the origination of these assets, which are then typically on-sold to other NBFIs. NBFIs usually pool these and other assets together to back various investment products sold to external investors. Other shadow debt has grown rapidly over recent years, with limited regulatory scrutiny, partly because no single regulator has responsibility for the oversight of such claims.

Entrusted loans

Entrusted loans are inter-company loans facilitated by a financial institution (typically a bank) for a fee. This arrangement exists because direct borrowing and lending between non-financial companies is restricted. Theoretically the credit risk is borne entirely by the lender in the arrangement.[3] While entrusted lending usefully brings together companies with surplus funds and those with financing needs, it can also be used to circumvent regulations, such as restrictions on lending to certain industries.

Bank-accepted bills

Bank-accepted bills are short-term tradeable debt instruments issued by banks to facilitate inter-company payments. A bank will lend a bank-accepted bill to a company that then uses it to pay another company for goods and services. In effect, the bank-accepted bill allows the bank to provide a guarantee for the payment of trade credit. It also allows the bank to record the exposure off-balance sheet. The use of bank-accepted bills has declined in recent years due to changes in the regulatory framework that have minimised their usefulness.[4]

Alternative financing

Alternative financing covers four types of lending by relatively smaller and nascent financial players: peer-to-peer lenders (P2P), microfinance companies, pawnshops and financial leasing companies. These lending activities have not typically been licensed by the financial regulators.

  • P2P is a form of direct lending where groups of lenders and borrowers (both of which can be either corporations or individuals) are matched using an online platform (also known as marketplace lending).
  • Microfinance captures most forms of small loan intermediation, including some online lending platforms.
  • Pawnshop financing refers to small loans collateralised with personal items.
  • Financial leasing is the practice of companies letting out business or household assets in return for periodic payments.

Borrowers of Shadow Finance

The vast majority of shadow financing in China supports the activities of the corporate sector and local governments, with a relatively small share flowing to individuals. While shadow financing can give some creditworthy borrowers access to finance, it can also result in lending to more marginal and riskier borrowers that would not normally be financed through the prudentially regulated banking system.

Some indication of the types of firms that access shadow financing can be inferred from trust company investment data (Graph 4). Industrial and commercial enterprises account for the largest share of trust investments. Within this category, the industries with the largest shares of trust financing are business services, construction and ‘environment and public facilities management’. Infrastructure investments account for another large share of trust company assets. This category largely comprises trust loans to local government financing vehicles (LGFVs), which are corporate entities used by sub-national governments to raise debt for purposes such as infrastructure spending. Real estate is another clearly identifiable sector receiving financing from trust companies. Authorities have often instructed banks to restrict lending to property developers as part of broader efforts to control the extent of real estate construction and speculation. As a result, borrowers have often sought finance from these shadow banks instead.

Graph 4
Graph 4 China – Trust Company Investments

Banks, NBFIs and Shadow Financing

The relationship between banks and NBFIs is central to the generation of shadow financing and risks to financial stability. Banks' claims on NBFIs (through the purchase of their investment products and through direct lending) are now a major asset class for banks, exposing banks to losses experienced by NBFIs. Banks' claims on NBFIs (which are recorded on-balance sheet) have grown rapidly over the past five years or so, to account for around 5 and 15 per cent of total assets at large and smaller banks respectively (Graph 5). This understates banks' total exposure to NBFIs as they also invest in NBFIs using off-balance sheet funds raised via wealth management products (WMPs) and sometimes have direct ownership of NBFIs.

Graph 5
Graph 5 Banks' Claims on NBFIs

The process of directing funds through NBFIs to issue shadow financing or make other investments is called ‘channel investing’ (see Box A for an example). Banks direct funds through NBFIs by purchasing their investment products or by making direct loans to NBFIs. Banks' purchases of the investment products sold by NBFIs are recorded on their balance sheet as ‘investment receivables’, which can attract significantly lower risk weights, loss provisioning and capital provisioning than traditional loans. It has also enabled banks to make loans to borrowers that would otherwise not have been permitted (for example, due to loan quotas), because the loans do not appear on their balance sheet as loans (or at all). Loans provided through shadow banking also gave banks the flexibility to charge a higher interest rate as compensation for at least some of the added risk when lending to higher-risk borrowers.

The investment products purchased by banks from NBFIs are typically referred to as asset management products (or AMPs).[5] An AMP is a tradeable asset that gives the holder the right to an income stream coming from a pool of assets. A range of assets can underlie AMPs, including all types of shadow financing as well as other financial assets (such as corporate bonds and equities). The process of pooling a range of different assets to form a single tradeable asset should provide diversification benefits for investors.

Banks also have linkages to NBFIs that are not reflected on their balance sheets. A lot of the funding that banks channel to NBFIs is raised off-balance sheet by issuing a type of AMP called a wealth management product (WMP). WMPs have grown rapidly since 2012, although the value of the assets under management has levelled off over the past 18 months (Graph 6). WMPs are short-term investment products offering fixed rates of return well above regulated bank deposit rates. They have been sold by banks directly to households and corporate investors, typically through bank branches and online platforms. While investors technically bear the credit risk associated with WMPs, investors are likely to expect banks to cover losses to avoid reputational damage (partly because banks have covered losses in the past; Perry and Weltewitz 2015).

Graph 6
Graph 6 China – Wealth Management Products

Smaller banks are particularly exposed to NBFIs (see Graph 5). The recent increase in this exposure has been funded in large part by borrowing from the larger banks – including by accepting their deposits or engaging in repurchase agreements (repo) with them (Graph 7). The large banks have been able to provide this funding to the smaller banks because of their larger deposit base.[6]

Graph 7
Graph 7 Net Outstanding Repo Lending Balance

Over the past year, the growth of banks' funds extended to NBFIs, including through WMPs, has slowed sharply. This is because regulations on banks' shadow banking activities (including channel investing and interbank financing) have been tightened to address the build-up of systemic risks that have accompanied the rapid growth in this sector (see below).[7]

Box A
Shadow Banking Complexity in China

Chinese shadow banking has become increasingly complex over time as banks and NBFIs have developed structures to circumvent authorities' attempts to control their activities. While these structures can vary considerably, there are common features among the more popular structures. In this box, we outline a stylised shadow banking structure, where a bank channels funds through an NBFI, to provide an illustration of the interconnections between financial institutions involved in Chinese shadow banking.

A Stylised Example

If a Chinese bank wants to extend a new loan to a company, but cannot do so due to a lending restriction (such as a loan quota), one way the bank can still extend credit to this borrower is by asking or directing a trust company to make the loan (on the understanding that the bank will ultimately provide the funding; Figure A1). The loan is then sold to a special purpose vehicle (SPV), which is a separate legal entity set up by the trust company to administer such loans. The SPV funds its purchase of the loan by selling notes to the bank and other investors, which give investors rights to the cash flows from the underlying loan. This process can involve assigning notes different rights to the loan's cash flows (sometimes referred to as ‘tranching’). In this example, the bank provides most of the funding by purchasing the highly rated senior notes (or ‘tranches’). This is recorded as an investment rather than a loan on the bank's balance sheet, enabling the bank to bypass the loan quota. Purchasing just the highly rated senior tranche may also enable the bank to hold less capital and lower loss provisions (than if it had made a traditional loan). The junior notes are packaged into the bank's off-balance sheet WMPs and the trust company's AMPs, but are ultimately purchased by non-bank investors, including households and corporations.

The Reality is More Complex

In practice, structures can be much more complex than the example provided above. For example, WMPs and AMPs can invest in other WMPs and AMPs. The asset pool of the SPV can include a wide range of assets, such as bank-accepted bills, corporate bonds, repos, equities and other AMPs. Guarantee companies can also play a role in the structure, particularly where the ultimate borrowers are small- and medium-sized businesses (which can lack long credit histories and high-quality collateral, making it difficult for them to borrow from a bank). Some analysts suggest that there can also be informal agreements between intermediaries, for example, where the senior tranche of an SPV is purchased by another bank, and the organising bank promises to buy it back in the event of default (Bedford 2017).[8] These activities reduce the credit risk borne by the end investors, which can allow the borrower to gain credit they may not have otherwise obtained, but it also means that credit risk may be transferred to entities in a way that is not readily identifiable.

Figure A1
Figure A1: Stylised example of a shadow banking structure

View larger image


The rapid expansion in Chinese shadow banking activity has delivered a range of benefits. However, the risks associated with this activity have also risen. At an aggregate level, the rise in shadow financing has facilitated much of the post-GFC run-up in Chinese corporate debt, which Chinese authorities have flagged as an issue. Shadow banking activity has also challenged the effectiveness of prudential safeguards by allowing financial institutions to circumvent regulations designed to contain risks, often through complex and opaque structures with strong links back to the banking system (including through banks' ‘implicit guarantees’ of WMPs and AMPs). This has resulted in lower loss-absorbing capacity and liquidity buffers and, partly as a by-product, higher credit, liquidity and contagion risks.

The nature and extent of risks vary considerably across different shadow banking activities, and not all shadow banking activities are high risk. For example, some entrusted lending will simply be for cash management purposes within corporate groups and, more generally, the intermediating banks for such loans do not bear credit risk. Bank-accepted bills are a common credit enhancement product offered by banks globally. Alternative finance would seem to pose low systemic risk currently given its small size (notwithstanding its rapid expansion and regulators' concerns about fraud, misuse of client money, money laundering and illegal fundraising).[9] Moreover, because of restrictions on Chinese banks' lending activities in an environment of rapid economic development, some shadow banking activities will have supplied funds to creditworthy borrowers that may otherwise have faced difficulty obtaining credit.

Nevertheless, other shadow banking activities appear to contain higher risks. For example, banks' channel investing through NBFIs poses higher risks for several reasons. First, channel investments enable banks to circumvent restrictions on their lending to riskier borrowers, thereby increasing credit risks. Second, some banks have also reportedly used channel investments to extend credit to poor-performing firms, as well as to hide non-performing loans (NPLs) by selling them to an NBFI and buying them back in the form of an AMP, which can obscure the quality of banks' assets (Bedford 2017).[10] Third, capital, loss provisioning and impairment recognition requirements are generally lower for assets reported as investments rather than loans.[11] The increased complexity that arises from channel investing relationships also increases contagion risks, by raising the number of interconnections between financial institutions and obscuring the extent and risks of individual institution exposures. Increased complexity also increases credit risks, because a larger distance between the ultimate borrower and lender can lead to lower lending standards at origination and less monitoring of borrowers.[12]

A particular concern Chinese authorities have been trying to address is the increase in channel investments by smaller banks, which has coincided with an increase in these banks' short-term wholesale funding. This raises liquidity and interest rate risks because the maturity of assets underlying AMPs tends to be longer than that of banks' short-term wholesale funding. If lenders were to cease rolling over their short-term funding, smaller banks could be forced to liquidate some of their asset holdings to repay their loans, which could trigger ‘fire sales’ with potentially adverse spillover effects on other parts of the financial system. Short-term wholesale funding has also increased contagion risks because the large banks are important suppliers of this funding to the smaller banks. As such, weaker asset performance or liquidity pressures at the smaller banks could directly spill over to the large banks. However, these concerns could be partially mitigated if the People's Bank of China (PBC) were to provide short-term liquidity support to financial institutions that experience difficulty rolling over their short-term funding.

There are also risks associated with WMPs and AMPs sold by banks and NBFIs (respectively) which are purchased directly by households and non-financial corporations. But while these vehicles also invest in shadow financing products, they appear less risky than banks' channel investments in WMPs and AMPs for two main reasons. First, household and corporate investors do not have high leverage compared with banks, so the returns they earn are not as sensitive to changes in the prices of the assets underlying the WMPs and AMPs. Second, bank and NBFI sponsors typically bear no direct credit risk, as this is borne by the investors (though some products have explicit principal guarantees). However, in practice, many investors assume that the sponsor will repay the unguaranteed principal and/or make good on the returns promised, even when asset performance is inadequate.[13] These implicit guarantees mean that WMPs and many AMPs are treated by household and corporate investors as de facto deposit substitutes, resulting in a contingent liability for the sponsor. Implicit guarantees can also extend to cross-holdings of products by banks, WMPs and AMPs (see Box A for details).

Implicit guarantees increase financial risks because they discourage investors from scrutinising the quality of the underlying assets. Sponsors may also have limited capital to honour their implicit guarantees: until recently, banks' capital and liquidity requirements were mostly unaffected by the size of their WMP contingent liabilities, while NBFI sponsors often have little equity funding relative to their assets under management and limited fundraising capacity. In addition, many WMPs and NBFI AMPs have limited disclosure of their underlying assets and most do not report their net asset value; some products have also been managed on a pooled basis, whereby funds raised through separately issued products by the same sponsor are combined into a single pool of funds and invested as one portfolio (enabling the sponsor to repay investors in maturing products using other products' resources).

Together, these characteristics raise the likelihood that investors could suddenly reassess the risks of their investments following an adverse shock and stop rolling them over. This could cause liquidity problems for WMPs and AMPs, and trigger broader contagion, because product maturities are often much shorter (typically less than three months, with some products offering daily redemptions) than maturities of the underlying debt assets (typically greater than one year). Although WMPs and AMPs invest substantially in liquid financial securities, such as equities and government bonds, widespread asset sales in response to liquidity pressures could trigger fire sales. Some products fund their investments with leverage obtained through short-term repos, which could add to selling pressure if repo lenders ceased rolling over these agreements. The reduced flow of shadow financing to the corporate sector would also weigh on economic activity and exacerbate the adverse shock.

Regulatory Response

Since 2010, Chinese regulators have announced a series of policies to mitigate the risks posed by rapidly expanding shadow banking activities (see Appendix B). However, banks and NBFIs have been tenacious in their efforts to sidestep new regulations. In particular, regulations were circumvented by operating through different entities or creating new ones.

Over the past two years, regulators have ramped up their efforts to address shadow banking risks. A new Financial Stability and Development Committee was established under the State Council to boost coordination between the five main Chinese financial regulators and increase their authority, which should help to reduce regulatory leakage. New regulations include a sweeping set of uniform standards for all types of asset managers and their products, including for leverage, liquidity, information disclosure and investment scope.

Other key policies include proposed caps on banks' short-term borrowing and lending, limiting bank WMP investments in NBFI AMPs, reiterating banks' obligation to treat loan-like assets – such as some AMP investments – as loans for capital and provisioning purposes, and extending the PBC's macroprudential assessment framework to include credit extended through banks’ WMPs. Regulators have also announced plans for many unlicensed alternative forms of finance to be brought under the supervision of the banking regulator, the China Banking Regulatory Commission (CBRC).

The PBC has facilitated an increase in the level and volatility of money market rates, which has made it more difficult for smaller banks to fund their channel investments with short-term wholesale debt (Graph 8). Policymakers have also allowed more instances of corporate bond defaults.

Graph 8
Graph 8 Chinese Financial Markets

While not all of the recently announced policies have been implemented yet, there has already been a tangible effect on shadow banking activity and financial markets more broadly. The overall pace of shadow financing growth has slowed significantly over the past year or so, driven by a reduction in banks' new channel investments and issuance of WMPs (see Graphs 5 and 7); banks' short-term debt issuance has also slowed. The reduced flow of shadow banking activity is likely to have contributed to the significant rise in Chinese sovereign and corporate bond yields and the slowdown in corporate bond issuance, since AMPs also invest significantly in bonds (Graph 9). It has also weighed on the balance sheet growth and profitability of smaller banks, which are especially active in shadow banking activities.

Graph 9
Graph 9 Chinese Bond Markets


Lending outside the formal banking system has expanded rapidly in China since the onset of the GFC, with non-bank financing now accounting for a sizeable share of total debt. This expansion has provided a range of benefits to the Chinese economy, in particular by opening an alternative avenue for credit supply and increasing the availability of higher-yielding investment products to Chinese households. However, these benefits have come at the cost of increased financial stability risks. While the risks vary across the different types of non-bank financing, some elements involve much higher risk, with limited prudential safeguards and transparency. Non-bank financing in China has facilitated higher leverage – both within and outside the financial sector – significant liquidity and maturity mismatches, lending to risky borrowers and an array of complex interconnections with strong links back to the banking system. These risks are not unique to China as many economies have managed the financial stability risks stemming from an increase in shadow banking activity, albeit with varying degrees of success.

Chinese regulators have been trying to mitigate these risks for some time, but it has been a challenge to design regulations that address these risks and are not easily circumvented. If these reforms are implemented effectively, the risks should continue to be addressed and may even decline. Non-bank financing is often beneficial to growth and development; retaining these benefits while containing systemic risks is likely to remain a challenge.

Appendix A – Data Sources and Compilation

Our estimate of shadow financing uses a range of assumptions and data sources. In particular, it gauges the flow of credit to the real economy and the aggregate level of indebtedness (when combined with official credit statistics). Our estimate of shadow financing is conceptually different to the FSB's measure of shadow banking activity. Graph 1 shows the FSB's narrow measure of shadow banking, which measures the total financial assets (or in the case of funds, assets under management) of NBFIs that: (i) engage in credit intermediation; (ii) are not bank affiliated; and (iii) involve liquidity and/or maturity transformation and leverage (FSB, 2018). In some ways, our measure is narrower than the FSB measure. The FSB focuses on the size of the entities involved in non-bank credit intermediation, measured as their financial assets or assets under management (FSB, 2018). This includes all the NBFIs' assets, not just shadow financing, and will include cross-holdings of other NBFIs' products. In other ways, our measure is broader than the FSB measure. This is because it includes a range of non-bank financing that is bank-affiliated, which is not included in the FSB measure.

The data has been sourced from CEIC Data and WIND Information, but can also be accessed directly from the PBC, the Asset Management Association of China, the China Insurance Regulatory Commission and the China Trust Association. The data on shadow financing products have varying degrees of back history; data lacking a reliable time series have been excluded. Graph 3 indicates the series breaks where data on each shadow financing product first become available.

All data are presented as reported except for other shadow debt.[14] Other shadow debt was estimated using assets under management data and investment allocation ratios for trust companies, securities companies, insurance companies and subsidiaries of fund management companies. In the few cases where the investment allocation ratios did not separately identify other shadow debt we assume that the residual (other investments) falls within the other shadow debt category. This assumption means our estimate of other shadow debt is likely to be an overestimate.

Appendix B – Regulatory Changes

Table B1: Selected New Chinese Financial Regulations
Date Regulator(s) Main target of regulation Key contents
August 2010 CBRC
(Document 72)
Direct bank-trust cooperation Banks required to make capital and loss provisions for off-balance sheet business with trust companies; trust companies prohibited from engaging in channel business
December 2010 CBRC
(Document 102)
Channel investments Credit asset transfers must be authentic – including being clear from any repos, whether explicit or implicit – and cover all outstanding principal and interest payments
September 2011 CBRC
(Document 91)
Bank WMPs Banks required to boost WMP information disclosures, ensure standalone risk and return calculation for each WMP and minimise regulatory arbitrage
March 2013 CBRC
(Document 8)
Bank WMPs Cap on NSDA investments at the lesser of 35 per cent of total WMPs and 4 per cent of the bank's balance sheet assets; WMPs cannot be managed using asset pools; full disclosure of investments in NSDAs to investors, including the underlying borrower, maturity and structure of the transaction (e.g. counterparties); banks prohibited from providing any explicit or implicit guarantees or repurchase commitments for NSDAs[15]
April 2014 CBRC
(Document 99)
Trust companies Trust companies prohibited from managing new AMPs using asset pools and investing in NSDAs; new AMPs must be reported to regulators at least 10 days before issuance
May 2014 CBRC, PBC, CSRC, CIRC, SAFE (Circular 127) Interbank activities (incl. borrowing, lending and repos) Banks' interbank borrowings must not exceed one-third of total liabilities; standardised accounting and capital requirements for interbank business; stronger oversight of interbank activities
July 2014 CBRC
(Document 35)
Bank WMPs More stringent requirements for banks' wealth management operations, including: separate accounting, risk isolation and centralised management of WMP business by a specialised department
March 2016 CBRC
(Document 82)
Investment receivables Tighter supervision of credit assets transferred off-balance sheet, including restricting retail WMP fund investments into NSDAs and requiring banks to treat loan-like assets as loans for capital and provisioning; banks encouraged to register transferred credit assets on a centralised platform
December 2016 PBC Bank WMPs Banks' off-balance sheet WMPs to be included in the PBC's macro-prudential assessments (starting from March 2017). Penalties to be issued for non-compliance
Early-mid 2017 CBRC
(Document 45, 46)
Regulatory arbitrage Banks required to review and monitor their existing channel investments and to correct any under-reporting or misreporting of capital, provisioning and non-performing loans; stricter enforcement of existing regulations and penalties for violations
November 2017 CBRC, PBC, CSRC, CIRC & SAFE (Draft) All AMPs (including WMPs) Unified rules covering all asset management products: asset manager sponsors prohibited from promising guaranteed returns and required to set aside 10 per cent of management fees for provisioning (up to 1 per cent of AUM); NAV should be regularly reported to investors; limits on leverage; restrictions on investing in other AMPs; explicit guarantees banned
January 2018 CBRC (Draft) Entrusted loans Clarification that entrusted loans can only be facilitated by banks and that banks cannot assume any credit risk; ban on entrusted loans extended by asset management plans (such as NBFI AMPs); restrictions on use of entrusted fund proceeds; banks required to strengthen entrusted loan risk management

Sources: Analyst reports; CBRC; PBC; RBA


This work was completed while the authors were in International Department, Financial Stability Department and Economic Group respectively. [*]

This differs from the British law concept of a trust, which administers fiduciary obligations accorded to a trustee. [1]

For example, in 2010 the authorities required banks to bring all their trust investments onto their balance sheets, increasing the cost to banks of lending through these firms. Growth in trust lending initially slowed but then rebounded as banks began using bridge companies to channel funds to trusts, thereby circumventing the tighter regulations. [2]

Entrusted lending can also be between companies within a corporate group, between a company and a housing provident fund (a government provider of affordable housing finance), or between a number of companies in a syndicated loan (Chen et al (2016)). [3]

Bank-accepted bills improved the loan-to-deposit ratio (LDR) of banks while the 75 per cent cap on the LDR was in place, allowing them to grant more loans all else being equal. [4]

AMPs can also be purchased directly by households, non-financial companies and other NBFIs. Insurance companies provide a small source of funding for shadow financing. Some insurance companies invest in other NBFI AMPs using funds raised from issuing ‘universal life insurance’ products, which are essentially high-yielding asset management products. [5]

For more information about small banks in China, see RBA (2016), ‘Box A: Recent Growth of Small and Medium-sized Chinese Banks’, Financial Stability Review, October, pp 14 – 16. [6]

Further information can also be found in RBA (2017), ‘Box B: Recent Development in Chinese Financial Regulations’, Statement of Monetary Policy, August, pp 27–30. [7]

This is called a repurchase, or ‘repo’, agreement. The informal repo market in China is also known as the dai chi market. See Kendall and Lees (2017) for details. [8]

Regulators have also been concerned that alternative financing (particularly P2P lending) could be adding to risks in the housing market because it has been used to finance housing deposits. [9]

Banks can also disguise NPLs in a variety of other ways, including dai chi agreements (see footnote 8). [10]

For example, bank loans to corporations generally attract a risk weight of 100 per cent, whereas investment assets can have risk weights as low as 20 per cent. Additionally, investments are only required to have individual provisions determined by management discretion, whereas loans have pre-set collective provisioning requirements. Banks usually disclose more granular information about their loan portfolios than their investment portfolios in their financial statements. [11]

Similar risks in the US contributed to the GFC: mortgage lending standards were undermined by lenders making loans which were on-sold to investors, resulting in widespread mortgage defaults and losses on investment products backed by the loans. [12]

These perceptions have been underpinned by sponsors marketing their products as ‘deposit-like’ with very limited risk and performance disclosures. There have also been few instances of investors incurring losses on these products, partly due to sponsors repaying unguaranteed principal when asset performance has been inadequate, including reportedly at the direction of the Chinese government. For example, see RBA (2015) ‘Box A: The "Credit Equals Gold #1" Collective Trust Product Default’, RBA Bulletin, June, pp 64. [13]

Our estimate of ‘other shadow debt’ broadly aligns with the China Banking Regulatory Commission (CBRC) concept of ‘non-standardised debt assets’ (NSDA), once separately identified components of NSDA (trust loans, entrusted loans and bank-accepted bills) have been removed. NSDA is an intentionally broad term used by the CBRC to describe any packages of debt assets that are not traded in a liquid market (such as the interbank market), such as letters of credit, accounts receivable, securitised bank loans or other non-standard forms of debt (CBRC, 2013). [14]

See Appendix A for the definition of NSDAs [15]


Adrian T and B Jones (2018), ‘Shadow Banking and Market Based Finance’, International Monetary Fund, Departmental Paper, forthcoming.

Andrews S, H Fan and J Zuo (2017), ‘Financial Deleveraging Impact Series 3 – Mounting Capital Pressure’, Deutsche Bank Markets Research Report, 1 May.

Bedford J (2017), ‘Are the Shadow Loan Books Hitting a Tipping Point?’, UBS Research Report, 23 August.

CBRC (2013), ‘银监发[2013]8号 [Decree No 8 (2013)]’, 25 March. Available at <http://www.cbrc.gov.cn/govView_2B22741AFBC446CF890636DACAB71166.html>

Ehlers T, S Kong and F Zhu (2018), ‘Mapping Shadow Banking in China: Structure and Dynamics’, Bank for International Settlements Working Paper, February

FSB (2018) ‘Global Shadow Banking Monitoring Report 2017’, Financial Stability Board Report, 5 March. Available at: <http://www.fsb.org/wp-content/uploads/P050318-1.pdf>

Kendall R and J Lees (2017), ‘The Chinese Interbank Repo Market’, RBA Bulletin, June, pp 75–86.

Perry E and F Weltewitz (2015), ‘Wealth Management Products in China’, RBA Bulletin, June, pp 59–67.

Pozsar Z, T Adrian, A Ashcraft, H Boesky (2013), ‘Shadow Banking’, Federal Reserve Bank of New York Economic Policy Review, 19(2), pp 1–16.

RBA (Reserve Bank of Australia) (2016), ‘Box A: Recent Growth of Small and Medium-sized Chinese Banks’, Financial Stability Review, October, pp 14–16.

RBA (Reserve Bank of Australia) (2017), ‘Box B: Recent Development in Chinese Financial Regulations’, Statement of Monetary Policy, August, pp 27–30.