A financial crisis with Chinese Characteristics

Recently there has been a lot of media attention on the increased probability of an imminent financial crisis in China because of the continual built-up in financial sector vulnerabilities. With the recent memory of a financial crisis in the US, many commentators are drawing parallels and arrive at the inevitable conclusion that China too must experience something similar. Before diving into the specific mechanics of how a financial crisis can actually unfold in China, it is important to appreciate the differences in the two systems. These differences will ultimately determine how the mechanics will play out, the response of policymakers and the ultimate conclusion. The main differences I can easily recall are:

China (2016) US (2007 – Pre-crisis)
Main source of financing to the real economy Bank loans Lightly regulated money markets
Financing of major financial institutions Deposits (stable) Money markets (unstable)
Legislative authority to act Largely unconstrained, the Chinese authorities can lend to whoever they want, whenever they want with no limits on quantity (i.e. equity market rescue in Aug 2015) Legally and ideologically constrained. The Fed had no authority to lend to investment banks or money market vehicles that was at the heart of the crisis.
Source of financial vulnerability Corporate sector debt mostly linked to unproductive SOEs Household sector linked to subprime mortgages
Financial capacity of government Nominally strong, but uncertain once all the liabilities of local governments and SOEs are included. Relatively strong, federal debt at 40% of GDP.
Monetary policy Constrained due to the trade-offs of maintaining a fixed exchange rate. Independent
Predominant government ideology Social stability Market discipline and prevention of moral hazard

What are the sources of vulnerabilities in China that can start a financial crisis?

It is almost impossible ex ante to identify the exact sparks that would start a financial crisis. (i.e. In 2006, the US Treasury conducted an exercise to identify a potential source of a future financial crisis, no one predicted the crisis would have come from subprime mortgages[1]). However, we do know several points of vulnerabilities that can exacerbate into a crisis. Whether these vulnerabilities will materialize into shocks to the financial system will depend on the reaction of market participants and policymakers.

  • High levels of debt in the corporate sector, mostly issued by unproductive SOES that operate in sectors that are grinding to a halt if not in outright contraction.
  • Large amount of outstanding local government debt issued by LGFVs with many provinces clearly insolvent and unable to pay off their stock of debt by any realistic forecast of cash flows.
  • A rapidly developing and evolving shadowing banking system that is competing away deposits from the traditional banking system to earn high rates of return through opaque channels[2].
  • A growing number of small banks who are competing for deposits and market share and are willing to enter into business the dominant banks have shied away from[3].
  • Large capital outflows that can lead to a drying up of liquidity in the interbank market if not properly managed.

What would the mechanism of a Chinese financial crisis be?

Despite the uncertainty regarding the asset side of the Chinese financial system, one point is clear, the wealth management products (WMP), is likely the weakest link on the liability-side. Unlike deposits of the banks that are now guaranteed up to 500,000 RMB by a deposit insurance fund, the WMPs are very short-term investments (usually 3 months) that are advertised to be liquid and riskless. The WMPs now represent a major part of lending to local governments and firms operating in the real economy. As of June 2014, the Reserve Bank of Australia (RBA) estimates that the stock of financing provided WMPs stood at 17.2T, or 15% of the stock of total social financial during that time[4]. Despite the relatively small share this investment vehicle plays compared to overall financing, due to the liability structure of these funds, this source of funding can disappear in a relatively short time should there ever be doubts among investors regarding the quality of the underlying assets and the extent of implicit guarantee provided by the Chinese authorities. The ultimate problem is that WMP engages in maturity transforming but without the proper safety precautions to safeguard against insolvency or panic runs sparked by worried investors.

In addition, these shadow banking conduits poses a real risk towards the actual banking system because of the implicit relations between the two. Despite being a legally separate entity, investors often assume all WMPs sold by commercial banks are guaranteed by the selling bank. So far the market results have proven the investors’ instinct to be correct. Every episode of a near default has the government scrambling to provide a solution to protect investors from loss despite the contract clearly stipulates many of these WMPs are not in any way guaranteed. The typical chain of WMP can be view in the following stylized way.

Anatomy of WMP

Source: (Hachem and Song, 2016)

Similar to the SPVs set up in the US before the Global Financial Crisis, the idea was that these vehicles can take risk independent of the sponsor and any economic consequence will be restricted to the SPV. However, as the case with Bear Sterns clearly shown, the reputational risk of being labelled as unable to honor the commitments of their SPV, will link the risk back to the issuing firm regardless of the legal structure. Once Bear Sterns began dedicating its own capital and liquidity to protect their SPVs, this brought unforeseen risk on to its balance sheet and a run by its creditors. When greed turns to fear, what seemed to be a very liquid environment can turn into a desert almost overnight. With the market suspecting the asset quality of Bear Sterns, the fear became a self-fulfilling prophesy; creditors demanded greater collateral and withdrew deposits. Within the matter of a week, Bear Stern went from being the fifth largest investment bank in the US to being acquired at $2 a share in a government sponsored bail-out simply because liquidity evaporated.

This chain of event is unlikely to occur a large bank in China because of their access to near unlimited liquidity from a generous central bank and widespread recognition in the Chinese market of a large bank’s systematic importance.  However, a Bear Sterns-like type of crisis, where an off-balance vehicle had to be placed on balance sheet, is not an impossible situation to imagine for a smaller bank. Smaller banks were responsible for the most aggressive promotion of WMPs because of the binding limits on the amount loans they could have made from a small deposit base[5]. Hachem and Song (2016) estimated that between 2008 and 2014, small banks were responsible for 73% of all new WMP issuances. It is not unforeseeable to see that as the real economy continues to slow, some of the assets of the more aggressive WMP begins to fail as a result of some exogenous shock. Given that much of the funds in WMPs are invested in troubled sectors such as real-estate and construction, a default by a major firm in this sector can be enough to trigger a loss of confidence and commence a run. The biggest single factor in this scenario is the response to policymakers: will they communicate that the government will essentially honor the liabilities of the entire financial system (or at least the parts that are in most stress)? If not, the moment doubt is instilled into the minds of investors, a cascading effect of contagion was quickly spread to other asset classes[6].  However, if the government does communicate it will protect the liabilities of the entire financial system, then the market will to begin to question the solvency of the state and whether such a commitment is credible (i.e. the case with the response of the Irish and Spanish government).

investment of AUM

In order to stem a panic, the issuing bank (in this case a smaller bank) will honour the liabilities of this WMP. Given the opaque nature of the WMP, deposits and creditors of this bank may begin to pull their deposits from the sponsoring bank starting a classic bank run[7]. In a system that values stability above all else, it is most likely a small bank would be bail-out either in an acquisition by a larger bank, or by perpetual assistance from the state. But despite the protection guaranteed by a larger institution, trust in WMPs as a viable savings instrument will be affected, leaving this source of financing to be unreliable going forward.

What would a financial in crisis in China look like?

In a system whose basic logic is to preserve stability, it is very improbable that China will face a Lehman-like moment where a major financial institution is allowed to fail. Instead, given the similarities in development-model, China would face a crisis similar to what Japan experienced in the 1990s: a long drawn out process in which growth grinds to a halt and the economy is plagued by ‘zombie’ firms that is dependent on constant injection of fresh credit to remain alive and distort factors in the real economy for productive firms.

A paper published in 2008[8] details the consequences of lending to zombie firms had serious consequences in terms of productivity gains in the Japanese real economy. Perpetual rollovers of noneconomic loans allowed Japanese banks and authority to play the game of ‘extend-and pretend’.  This game prevented Japanese banks from recognizing bad loans in orders to meet regulatory requirements and avoid the sharp reduction in profits that would have accompanied recognizing NPLs. By keeping zombie borrowers alive, these unproductive firms distorted factor markets by artificially keeping wages high and the prices of outputs low as unprofitable firms continue to produce. These two distortions were a two-pronged attack on productive sectors of the economy: revenue was artificially suppressed while labour cost were higher than otherwise. These unfavourable factors discouraged productive investments from taking place, prolonging Japan’s economic slump.

These dynamics can easily play out in China with the political connections of many SOEs and the concentration of SOE activity in the less-dynamic regions of the country. The biggest threat to the Chinese financial system is not a cataclysmic collapse, but the failure to finance the full potential of a booming economy because of the inability to let go past legacy assets that are not representative of China’s future.


By solely observing the asset side, it is impossible ex-ante to identify the exact source of vulnerability that will lead to a crisis. However, if we monitor the liability side, we can observe severe deficiencies in WMPs that can be potentially be a source of stress for the system in the future. Given the basic differences in the two systems, a financial crisis will look very different in China than it did in the US. However, this difference does not subtract from the potential severity of the event or its consequences. By observing the experience of Japan, this episode serves as a baseline model for the consequences of a financial crisis and the implications it may have for the real economy.




[1] https://www.minneapolisfed.org/news-and-events/presidents-speeches/lessons-from-the-crisis-ending-too-big-to-fail

[2] Hachem and Song (2016) described the parallels between WMPs and asset-backed commercial paper (ABCP), a tool of contagion for the GFC: “In this regard, there is a notable similarity between unguaranteed WMPs and the asset-backed commercial paper vehicles that collapsed during the 2007-09 financial crisis. Another similarity is the use of implicit guarantees and/or back-up credit lines by the sponsoring bank to allay investor concerns about the riskiness of off-balance-sheet products. As a result, the products are only off-balance-sheet for accounting purposes.”

[3] To reflect the growing importance of small banks to the system, in 1995 the Big Four Chinese commercial banks held 80% of all deposits. By 2005, this fell to 60%, and in the present, it is estimated to be around only 50%.

[4] For a more complete discussion behind WMPs please see http://www.rba.gov.au/publications/bulletin/2015/jun/pdf/bu-0615-7.pdf

[5] Hachem, Kinda and Song, Zheng Michael, Liquidity Regulation and Unintended Financial Transformation in China (January 2016). NBER Working Paper No. w21880. Available at SSRN: http://ssrn.com/abstract=2717291

[6] One of the lessons of the GFC was that despite the relatively small size subprime mortgages were compared to the entire system, the entire system was paralyzed with fear and doubt. During a financial crisis, it mattered less whether a firm had good assets, but more so on how they were financed. Firms that had nothing to do with subprime found themselves in a credit crunch and in need of a government bail-out (i.e. Northern Rock) because they were very dependent on having access to very short-term and liquid debt.

[7] Having a deposit insurance fund does not guarantee bank runs will not happen. Firstly, deposits are only guaranteed up to 500,000 RMB, any surplus will be at risk of loss giving depositors with excess deposits very high incentive to withdraw their funds at any suspicion of a bank’s assets. Secondly, during the height of the GFC, Washington Mutual experienced severe run on its deposits despite being an FDIC-insured bank. Given the novelty of deposit insurance in China, it is not imaginable to see Chinese depositors lining up once news of a crisis breaks.

[8] Ricardo J. Caballero & Takeo Hoshi & Anil K. Kashyap, 2008. “Zombie Lending and Depressed Restructuring in Japan,” American Economic Review, American Economic Association, vol. 98(5), pages 1943-77, December

Aside | This entry was posted in Uncategorized and tagged , , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s