VoxEU Column International Finance

Can shadow banking be addressed without the balance sheet of the sovereign?

The shadow banking system is vast; but why did it arise? Some view it as regulatory arbitrage while others view it as the market fulfilling investors’ demand for ‘riskless’ assets. This column explains the issues and discusses policy options.

The shadow banking system  – the name given to the financial infrastructure that exists outside the regulator’s remit – has been much in the news since the global financial crisis (see recent Vox contributions Acharya 2011 and Acharya et al 2011). Much thought has gone into what the best ways are to reform the shadow banking system. Recently the Financial Stability Board has issued a set of recommendations on the topic. But, as with other financial sector issues, it is important to have a good diagnosis as to what makes the shadow banking system of systemic importance and how these causes relate to the best necessary reforms.

There are two approaches to understanding the shadow banking system, a phenomenon of systemic importance largely, but not exclusively, for the US financial system.

  • One approach is to look at the shadow banking system from the supply side.

This view examines how modern banks conducted securitisation-based credit intermediation through the globally interlinked webs of subsidiaries, vehicles, and funds that constituted the shadow banking system (see Pozsar et al 2010). It concludes that this mode of credit intermediation was primarily a story of regulatory arbitrage.

  • Another approach is to look at the shadow banking system from the demand side.

This view examines why institutional cash investors chose to fund modern banks not via deposits, but via wholesale funding markets and instruments. This view concludes that as a byproduct of credit intermediation, modern banks’ webs of subsidiaries, vehicles, and funds were also about the creation of safe, short-term, liquid instruments to cater to the safety preferences of institutional cash investors and a deficit of US Treasury bills in general (see Pozsar 2011).

The intersection of these two approaches implies that shadow banking is very much about what goes on within modern banks. More specifically, shadow banking refers to how modern banks use the web of subsidiaries, vehicles, and funds under their holding companies’ umbrella to tap wholesale funding markets via the issuance of safe, short-term, liquid instruments (as alternatives to US Treasury bills) to institutional cash investors.

In pursuit of the issuance of these riskless instruments, the credit intermediation process of modern banks should perhaps more appropriately be called risk intermediation, or risk-stripping.

Shadow banking from the risk-intermediation perspective

Understanding shadow banking from a risk intermediation perspective must begin with the definition and profiling of institutional cash investors, and their pre-crisis menu of safe instruments. Institutional cash investors manage institutional cash pools. The rise of these cash pools was mainly driven by the rise of the modern ‘asset-management complex’ and its day-to-day investment management activities. These include securities lending, derivatives-based investing, and the liquidity management of long-term funds.

The volume of institutional cash pools rose from an insignificant $100 billion in 1990 to a systemic $3.5 trillion in 2010 (see Figure 1). Of this, the asset-management complex accounts for about two thirds, with the bulk of the rest accounted for by corporations. The size of individual cash pools is at least $1 billion and as large as $100 billion, with an average of $10 billion.

Figure 1. The secular rise of institutional cash pools

Source: CapitalIQ, RMA, ICI, BIS, The Economist, Pozsar (2011)

Safety first

Institutional cash pools prioritise principal safety and are highly methodical about obtaining it. However, prior to the financial crisis, the cash pools were denied safety via insured deposits on the scale they needed it, so obtaining principal safety was challenging.

To seek safety, institutional cash pools have three types of money market instruments to choose from.

  • Government guaranteed,
  • Privately guaranteed; and
  • Unguaranteed instruments.

Government-guaranteed instruments include Treasury bills, agency discount notes, and insured deposits; privately guaranteed instruments include (among others) repurchase agreements and asset-backed commercial paper; and unguaranteed instruments include uninsured deposits and commercial paper.

These types of money are liabilities of the US government, the shadow banking system, and the traditional banking system, respectively.

The supply of and demand for these instruments is not evenly distributed, however. For example, government-guaranteed instruments are in limited supply.

  • First, there are an insufficient number of banks for institutional cash pools to spread their cash balances across in insured increments.
  • Second, in the three years preceding the financial crisis, there was a structural deficit of $1.5 to $2.0 trillion in short-term Treasury bills and agency discount notes to fill institutional cash pools’ safety needs (see Pozsar 2011).

Unguaranteed instruments were not an option to fill this safe instrument vacuum as institutional cash pools’ managers were bound by risk mandates limiting too much unsecured exposure to banks via uninsured deposits (see Gorton 2009).

Shadow banking filling the gap

The shadow banking system arose to fill this vacuum. This vacuum is best thought of as a result of a limited supply of government-guaranteed money market instruments on the one hand (that is, inelastic supply), and an aversion to unguaranteed exposures to the traditional banking system via uninsured deposits on the other hand (that is, inelastic demand).

The vacuum was filled with privately guaranteed instruments such as repos and asset-backed commercial paper issued by the shadow banking system. These instruments were secured by collateral (both government-guaranteed and private collateral), liquidity enhanced by insured banks, and par value enhanced by money funds (most of them also affiliated with insured banks).

Due to these layers of protection, privately guaranteed money instruments issued by the shadow banking system were perceived safer than unsecured deposits and yielded less (see Figure 2).

Figure 2. Yield difference between 3-month CDs and other instruments

Source: Haver, Pozsar (2011)

Creating riskless assets: Risk stripping

The creation of privately guaranteed money involves ‘risk stripping’. This involves – in this order – credit-, maturity-, and liquidity-transformation. Take for example a pool of mortgages that go into the production of privately guaranteed money market instruments for institutional cash pools.

  • This process starts with credit-transformation.

Pools of mortgages were routinely placed into securitisation trusts, where their cash flows are credit-tranched into equity, mezzanine, and AAA tranches.

  • Next came the maturity-transformation.

The AAA tranches were then purchased by levered maturity transformation vehicles that issued short-term liabilities to fund them.

The rollover risks embedded in such structures were eased by liquidity guarantees from banks.

  • Finally came the liquidity-transformation.

The short-term money market instruments issued by maturity transformation vehicles ended up in money funds and similar funds that intermediated institutional cash pools’ funds. Money funds turned tradable money-market instruments into checkable instruments, with implicit par value guarantees from sponsors.

Such chains of transactions (see Figure 3) were integral parts of the credit intermediation process. The FSB (2011) and Pozsar et al (2010) refer to such chains as ‘credit intermediation chains’. However, from the perspective of institutional cash pools that ultimately hold the riskless end-products of these chains (that is, privately guaranteed money), such chains of activities rather resemble a process of risk-stripping whereby underlying pools of long-term, risky loans are stripped of their component risks and turned into safe, short-term liquid instruments, or money.

Figure 3. Money creation via risk-stripping

Credit intermediation and risk-stripping

The concepts of credit intermediation and risk-stripping chains are not contradictory, but complementary. In some sense, traditional banks also do both – they make loans and fund them with riskless deposits. However, in the case of traditional banks, the risk-stripping process is simple – systemic risks are offloaded to the sovereign via deposit insurance and other parts of the public safety net. Credit intermediation and risk-stripping in the case of modern banks is a more complex affair, however.

Inside modern banks, finance company subsidiaries originate loans, broker-dealer subsidiaries process and distribute securitised loans, and asset-management subsidiaries hold securitised credit exposures ranging from very safe to very risky via funds ranging from money funds to hedge funds, respectively. In conjunction with this process, modern banks’ bank subsidiaries internalise large volumes of systemic risk. In essence, the risks that traditional banks offload to the sovereign via deposit insurance, the shadow banking system offloads to traditional banks via liquidity puts and derivatives. The type of risks offloaded and the instruments they are offloaded with are depicted in Figure 3 above. Traditional banks’ internalising the systemic risks involved in private money creation is one way of interpreting Raghuram Rajan’s 2005 article “Has Financial Development Made the World Riskier?”.

The right policy response

Seeing the shadow banking system from the perspective of the safe asset demands of institutional cash investors is crucial for drafting the right policies for shadow banking.

From this view, shadow banking arose in response to a shortage of government-guaranteed money market instruments (see Figure 4). To fill this shortage, the shadow banking system issued privately guaranteed instruments. These in turn were guaranteed by insured banks and hence ultimately the sovereign (whether via central bank backstops or equity injections).

Figure 4. Filling the vacuum of short-term government-guaranteed debt

Source: Pozsar (forthcoming)

Thus, the balance sheet of the sovereign plays into both the emergence of shadow banking and the integrity of the money claims issued by it. This implies that solutions to managing or regulating the shadow banking system must also involve the balance sheet of the sovereign.

Create more T-bills?

One such solution would be for the US Treasury Department to increase its supply of bills. As argued above, and as also shown by Krishnamurthy and Vissing-Jorgensen (2010) and Greenwood et al (2010), Treasury bills serve as money for institutional cash investors and are substitutes to money claims issued by banks. This solution to shadow banking is not without costs, however, as it would increase rollover risks and the variability of interest expenses for the US Treasury. Still, these externalities associated with unlevered public money creation by the US Treasury are smaller than the externalities associated with levered and privately guaranteed money creation via the shadow banking system. A recent proposal by the Treasury Borrowing Advisory Committee to introduce floating-rate notes is also similar in spirit to the idea of increased supply of bills to serve as money for institutional cash pools.

Regulatory approaches to design a smaller and less run-prone banking system can complement this solution. However, besides passively monitoring the shadow banking system, more hands-on approaches could include taking proactive steps to engineer a migration of institutional cash pools away from wholesale funding markets and toward short-term, sovereign claims. Full accommodation of this shift would require the US Treasury to think of itself as providing monetary services to institutional cash pools, and in line with this, also issue a more granular set of instruments including overnight, and one-, two-, or three-week bills. Bills are not the only way, however. Alternative solutions could include the introduction of central bank bills and lender-of-last-resort access to broker-dealers and money funds in exchange for regulation.

Regardless of the approach, addressing the shadow banking system requires a view on the fundamental question of whether it is the banking system or the US Treasury that has a comparative advantage in intermediating the world’s uninsured, institutional dollar liquidity.

Author's Note: The author wishes to thank the IMF’s Research Department for its hospitality when writing the working paper on which this summary is based and the beneficial discussions held with James Aitken, George Akerlof, Olivier Blanchard, Elie Canetti, Stijn Claessens, Julia Kiraly, Elena Liapkova-Pozsar, Laura Kodres, Phil Prince, Paul McCulley, Perry Mehrling, Dan Neilson, Lev Ratnovski, Hyun Song Shin, Manmohan Singh, and James Sweeney.


Acharya, Viral, Thomas Cooley, Richardson Matthew, and Ingo Walter (2011), “Dodd-Frank: One year on…”, VoxEU.org, 12 July.

Acharya, Viral (2011), “Measuring Systemic Risk and the dismal failure of the Basel risk weights”, VoxEU.org, 17 June.

FSB (2011): “Shadow Banking: Strengthening Oversight and Regulation: Recommendations of the Financial Stability Board”, 27 October.

Gorton, Gary (2010), Slapped by the Invisible Hand, Oxford University Press.

Greenwood, Robin, Samuel Hanson, and Jeremy Stein (2011), “A Comparative-Advantage Approach to Government Debt Maturity”, HBS Working paper No. 11-035.

Krishnamurthy, Arvind and Annette Vissing-Jorgensen (2010), “The Aggregate Demand for Treasury Debt”, NBER Working Paper No. 12881.

McCulley, Paul (2010), “Statement of Paul A. McCulley, Managing Director, PIMCO Before the Financial Crisis Inquiry Commission”, 6 May.

Mehrling, Perry (2010), The New Lombard Street: How the Fed Became the Dealer of Last Resort, Princeton University Press.

Mehrling, Perry (2011), “The New Lombard Street: Financial Globalisation and the Future of the Fed”, Unpublished Mimeo.

Pozsar, Zoltan (2008), “The Rise and Fall of the Shadow Banking System”, Moodys’s Economy.com.

Pozsar, Zoltan, Tobias Adrian, Adam Ashcraft and Hayley Boesky (2010), “Shadow Banking”, Federal Reserve Bank of New York Staff Report No. 458.

Pozsar, Zoltan (forthcoming), “Institutional Cash Pools and the Triffin Dilemma of the US Banking System”, IMF Working Paper no. 11/190.

TBAC (2011), “US Treasury Financial Innovation: Opportunities and Challenges”, Treasury Borrowing Advisory Committee presentation to the US Treasury, 1 February.

5,774 Reads