The Roots of Shadow Banking

By Enrico Perotti Jan 18, 2014, 12:16 PM 

The ‘shadow banking’ sector is a loose title given to the financial sector that exists outside the regulatory perimeter but mimics some structures and functions of banks. This column introduces a new CEPR Policy Insight that looks into what we have learned about shadow banking since the Global Crisis.

Shadow banking operates outside the regulatory perimeter, but it replicates the structure of banking in many ways. As such, its prudential standards should be brought into alignment to avoid further regulatory arbitrage. Shadow banking assets at the time of the crisis had grown larger than the banking sector proper. The rapid withdrawal of funding to this segment played a major role in the credit crunch of 2007-2008.

My new CEPR Policy Insight No. 69 updates and fleshes out the analysis in my June 2012 Vox column (reproduced in full below). The Policy Insight reviews recent work that is finally shedding some light on this ill-defined and poorly understood segment of the financial system.

The Policy Insight offers a structural definition of shadow banking activities, showing that even proper banks use them to avoid stricter capital requirements. The decision by the Basel committee in these days to accept a relaxed definition of the leverage ratio, for which banks lobbied fiercely, appears a serious setback.

Column as posted on 21 June 2012

The ‘shadow banking’ sector is an ill-defined financial segment that expands and contracts credit outside the regulatory perimeter.

  • It was critical in the build up and demise of the credit boom.

While much reduced since 2008, in the US its size still exceeded bank assets in 2011.

  • What have we learned since the crisis on shadow banking?

Defining shadow banking by function

By definition, shadow banking is not a precise category. In this column, I focus on financial intermediaries that take credit risk.

Banks acquire illiquid risky assets, funding them with inexpensive, demandable debt.

  • Most investors prefer safe, short term and liquid assets, so banks can use cheap funding, by promising liquidity on demand.
  • This is made credible by deposit insurance and access to central bank refinancing.

Confidence on immediacy ensures that demandable debt is routinely rolled over, thus supporting long-term lending and high leverage.1

As bank credit volume is constrained by capital ratios and deposit base, financial markets have thought of new ways to carry risky assets on inexpensive funding. Shadow banking requires creating a variant of demandable debt, credibly backed by a direct claim on liquidity.

But the dominant funding channel is issuing collateralised financial credit, such as repos or derivative-based claims.2  This is the source of shadow banks’ very short-term, inexpensive funding source.

How can these liabilities deliver investors credible liquidity upon demand?

Jump the running queue: Superior bankruptcy rights

Security-pledging (the collateral part of collateralised financial credit) grants access to easy and cheap funding thanks to the steady expansion in the EU and US of “safe harbour status” – the so-called bankruptcy privileges for lenders secured on financial collateral.

Critically, lenders in this collateralised financial credit transaction can immediately repossess and resell pledged collateral. They also escape most other bankruptcy restrictions such as cross default, netting, eve-of-bankruptcy and preference rules (see Perotti 2010).

  • These privileges ensure immediacy for their holders.
  • Unfortunately, they do so by undermining orderly liquidation, the foundation of bankruptcy law.

The consequences became visible upon Lehmann’s default, when its massive stock of repo and derivative collateral was taken and resold within hours. This produced a shock wave of fire sales of ABS holdings by safe harbour lenders. While these lenders broke even,3  their rapid sales spread losses to all others, forcing public intervention.

When the safe harbour provisions were massively expanded in a coordinated legislative push in the US and EU (Perotti 2011),4 they supported an extraordinary expansion of shadow banking credit and mortgage risk taking. The guaranteed ease of escape fed the final burst in maturity and liquidity mismatch in the 2004-2007 subprime boom, where credit standards fell through the floor.

Borrowing securities to generate collateralised financial credit

While shadow banks expanded with securitisation, they can also rely on the liquidity of assets they do not own. They do this by borrowing securities from insurers, pension and mutual funds, custodians, and collateral reinvestment programmes.5 In exchange, the beneficial (i.e. real) owners of such securities receive fees for lending the assets and they book these as yield enhancement. Borrowed securities are then pledged to ‘repo lenders’ (the short name for credit grantors in a collateralised financial credit transaction) or posted as margins on derivative transactions.

Experienced asset managers who lend securities in this way protect themselves via collateral swaps, i.e. a related transaction where the security borrower pledges collateral of lower liquidity. This so-called ‘liquidity risk transformation chain’ (which transforms illiquid assets into short term credit) may have more links.

The financial logic behind the liquidity risk transformation chain is clear. Security pledging activates the liquidity value of assets from long-term holders who do not need it. Such extraction of unused liquidity value may be seen as enhancing “financial productivity”. It certainly increase asset liquidity, and boosts securitisation. Yet this can be an illusory gain, flattering market depth in normal times, at the cost of greater illiquidity at times of distress.

The repo lenders and security lenders typically require a more than one-to-one exchange to protect themselves against the possibility of the collateral losing value. These ratios are called ‘haircuts’ since each $1 of collateral generates less than $1 of credit.

Shadow banking runs: Rising haircuts

A jump in market haircuts, and ultimately a refusal to roll over security loans or repos, is the classic shadow bank run.

  • As a security borrower cannot raise as much funding from its own illiquid assets, it is forced to deleverage fast or goes bust.

In both cases this triggers fire sales.

  • Once repo lenders seize collateral, they have all reasons to wish to sell fast.

First, they are not natural holders. Second, they do not suffer from a fire sale as long as the price drop is less than their haircut. Third, they are aware that others are repossessing similar collateral at the same time, so they have an incentive to front sell.

  • In addition, real money investors that lost their original holdings are likely to sell the repossessed, less liquid collateral.

First, they may wish to re-establish their portfolio profile. More critically, they legally need to sell within days to be able to claim any shortfall in bankruptcy court.

  • This leads to a dramatic acceleration of sales for assets originally committed to a long holding period.

While central banks are not in charge of shadow banks, they do come under pressure to stop fire sales and create outside liquidity. This completes the banking analogy.

The safe harbour debate

It is now evident that shadow banks need the safe harbour privileges to replicate banking. No financial innovation to secure escape from distress can match the proprietary rights granted by the safe harbour status, which ensure immediate access to sellable assets. Traditional unsecured lenders have taken notice, and now request more collateral, squeezing bank funding capacity and limiting future flexibility.

Many attentive observers find such an unconditional assignment of superpriority to repo and derivative claimants excessive.6 Duffie and Skeel (2012) discuss in an excellent summary the merit of safe harbour. In their words:

“Safe harbours could potentially raise social costs through five channels: (1) lowering the incentives of counterparties to monitor the firm; (2) increasing the ability of, or incentive for, the firm to become too big to fail; (3) inefficient substitution away from more traditional forms of financing; (4) increasing the market impact of collateral fire sales; and (5) lowering the incentives of a distressed firm to file for bankruptcy in a timely manner.”

All these arguments demand attention. Repo lenders and derivative counterparties enjoy not just immediacy in default, but also reset margins daily. By construction, this produces a unique safe claim. Just as insured depositors, these claimants can afford to neglect credit risk, and perform no monitoring role.

Collateral lending, by splitting up liquidity transformation, lengthens credit chains and expands the number of connections among intermediaries, contributing to systemic risk (Gai et al. 2011).

What should happen to the safe harbour privileges?

The main proposals aim at restricting eligibility. Tuckman (2010) suggested only cleared derivatives should enjoy the status. Duffie and Skeel argue it may be limited to appropriately liquid collateral (thus not ABS!) and only transparent uses (derivatives listed on proper clearing exchanges).

In recent research (Perotti 2011), I suggest that claims be publicly registered (just as secured real credit is) as a precondition for safe harbour status. This will ensure proper disclosure, essential to macro prudential regulators, and avoids unauthorised or misunderstood (re)hypothecation.

Investors who wish to claim superpriority in distress seek a scarce resource. They should be paying for the privilege, and for any risk externality it creates. In normal times, a low charge should be levied on registered claims. Charges should be adjusted countercyclically, lowered in difficult times, and raised when aggregate liquidity risk builds up, to brake an otherwise uncontrollable expansion.

Other approaches involve limiting the stock of safe harbour claims directly (Stein 2012) by a cap and trade model, which a registry receiving fees could support.

A critical issue is the treatment of collateral posted for central bank refinancing. For central banks to operate as ultimate liquidity providers, their claims should not be undermined. A specific privilege for eligible collateral is justified, as central banks are by definition not likely to create fire sales.


Thanks to the safe harbour rules, a shadow bank can hold risky illiquid assets and earn full risk premia with funding at the overnight repo rate. In what is essentially a synthetic bank, repo and collateral swap haircuts act as market-defined capital ratios.

Liquidity transformation across states and entities has procyclical effects.

It enhances credit and asset liquidity in normal or boom times, at the cost of accelerating fire sales in distress. While any reform to the shadow banking funding model should take into account its favourable effects on asset liquidity and credit in normal times, the associated contingent liquidity risk is not at present controllable (nor is it well measured!). There is an academic consensus that a balance has to be struck (Acharya et al. 2011; Brunnermeier et al. 2011; Gorton and Metrick 2010; Shin 2010).

Appropriate tools are also necessary to align capital and risk incentives in banks and shadow banks (Haldane 2010). Security lending may also undermine Basel III liquidity (LCR) rules.

At a time when all lenders seek security, questioning the logic of safe harbour provisions may seem unwise. Yet at the system level, it is simply impossible to promise security and liquidity to all. Uncertainty on the stock of pledged assets may create a self-reinforcing effect, feeding a frenzy among lenders to all seek ever-higher priority. This is already taking place, and is ultimately unsustainable at the individual and aggregate level.

Finally, it is questionable whether the highest level of protection should be granted to collateralised lenders, and to shadow bank funding, over all other investors. For all these reasons, regulators and the wider society need to make an informed decision.


•Acharya, Viral, Arvind Krishnamurthy, and Enrico Perotti (2011), “A consensus view on liquidity risk”,, 14 September.
•Acharya, Viral and Sabri Öncü (2012), “A proposal
for the resolution of systemically important assets and liabilities: the case of the repo market
”, CEPR DP 8927, April.
•Brunnermeier, Markus, Gary Gorton, and Arvind Krishnamurthy (2011), “Risk Topography”, NBER Macroannual 2011.
•Duffie, Darrell and David Skeel (2012), A Dialogue on the Costs and Benefits of Automatic Stays for Derivatives and Repurchase Agreements, Stanford University, March.
•Haldane, Andrew (2010), “The $100 Billion Question”, Bank of England, March.
•Gai, Prasanna, Andrew Haldane, and Sujit Kapadia (2011), “Complexity, Concentration and Contagion”, Bank of England discussion paper.
•Gorton, Gary, and Andrew Metrick (2010), “Regulating the Shadow Banking System”, Brookings Papers on Economic Activity, (2):261-297.
•Perotti, Enrico (2010), “Systemic liquidity risk and bankruptcy exceptions”,, 13 October.
•Perotti, Enrico (2011), “Targeting the Systemic Effect of Bankruptcy Exceptions”, CEPR Policy Insight No. 52 and Journal of International Banking and Financial Law (2011)
•Shin, Hyun Song (2010), “Macroprudential Policies Beyond Basel III”, Policy memo.
•Stein, Jeremy (2010), “Monetary Policy as Financial-Stability Regulation”, Quarterly Journal of Economics, 127(1):57-95.
•Tucker, Paul (2012), “Shadow Banking: Thoughts for a Reform Agenda”, Speech at the European Commission High Level Conference, 27 April, Brussels.
•Tuckman, Bruce (2010), Amending Safe Harbors to Reduce Systemic Risk in OTC Derivatives Markets, Centre for Financial Stability, New York.

1 Historically, confidence was supported by high capital, reputation and limited competition. As competition increased and capital fell, central banks’ emergency liquidity transformation and deposit insurance allowed steadily higher credit and bank leverage.

2 Trivially, shadow banks may also access bank credit lines (as SIVs did).

3 The rest of the creditors had to wait years to get less than 20 cents on the dollar.

4 Limited safe harbour status was granted as exceptions in the 1978 US Bankruptcy code, limited to Treasury repos and margins on futures exchanges for qualifying intermediaries. They were broadened progressively to include margins on OTC swaps. The massive changes took place in 2004, when any financial collateral pledged under repo or derivative contracts, whether OTC or listed, by any financial counterparty, came to enjoy the bankruptcy privileges (Perotti 2011).

5 According to Poszar and Singh (2011): “At the end of 2010.. about $5.8 trillion in off-balance sheet items of banks related to the mining and re-use of source collateral… down from about $10 trillion at year end-2007”.

6 Creation of new proprietary rights is exceedingly rare. Limited liability is the last main case.

Be the first to comment

Leave a Reply

Your email address will not be published.


This site uses Akismet to reduce spam. Learn how your comment data is processed.