VoxEU Column Financial Markets

Subprime crisis and credit risk transfer: something amiss

Securitisation transferred credit risk from bank’s balance sheets to the market. The subprime problem became a crisis when some of this risk landed back on banks. Regulators need to find a way to deal with the off-balance sheet operations of banks that made this possible and to improve transparency concerning banks’ effective exposure to risk.

By now everyone in Europe knows all about American subprime loans – ranging from “Alt-A” to the “ninja” variety (granted with “no verification of income, job status or assets”). Still, it is not obvious why an even pronounced increase in delinquency rates on such loans, with the attendant losses on mortgage exposures, should have sparked a financial crisis that touched all classes of assets globally, even those relatively immune from credit risk. True, the share of the less safe loans in the issuance of mortgage-backed securities had almost doubled in the past few years. But an estimate of the direct losses of the actual and expected defaults ranges between 100 and 200 billion dollars – relatively little, considering the valued of aggregate of financial assets (and also in comparison with the 5 trillion dollars lost in the dot.com crisis).

We know how the crisis has unfolded1. After a sharp drop in the prices and market liquidity of all mortgage-backed securities, an equally sharp increase in the price of risk and in spreads, and a drying-up of the issuance of all asset-backed securities, contagion extended to the short-term end of the financial market – first to a wide class of commercial paper and then to the money and interbank markets. As uncertainty and mutual mistrust spread to counterparties (even banking counterparties), overnight interest rates jumped and, as they say, cash became king. The repeated injections of liquidity on the part of various monetary authorities have so far provided only limited solace to this state of affairs. All this is clear, but the question is: Why should a surge of subprime defaults affect (though not disruptively for the moment) the banking system and (more worryingly) general credit conditions?

The question arises because the subprime mortgage-backed securities that sparked the crisis represent an extreme version of the credit risk transfer process in which the core banks have been engaged for a long time pursuing the “originate and distribute” business model. The banks originate the loans and then distribute the underlying risk to a myriad of outside investors. This made credit “something that is largely bought and sold on the markets, rather than held … on the balance sheets of financial intermediaries”2. Among the undisputable merits of this model (more complete markets, a wider range of instruments available to investors, enhanced liquidity, improved allocation of resources) is that the transfer of credit risk away from banking intermediaries would make the system more resilient to financial shocks. The fragmentation of risk and its distribution to non-bank players providing liquidity in several markets would alleviate the systemic consequences and allow an easier absorption of such shocks. This, however, is not what has happened. Though the credit underlying all kinds of asset-backed securities and of credit derivatives should no longer be on the balance sheet of the originating banks, the collapse of one segment of those securities has affected and is affecting the banking system. Why is that? The answer is that part of the credit risk flowed back to some banks, though not on to their books.

This has mostly happened through the growing diffusion of “conduits” and structured investment vehicles, widely know as SIVs3. These are entities, off the banks’ balance sheet, that invest long-term, largely in high-yield asset-backed securities, and raise short-term finance by issuing correspondingly collateralized commercial paper (so-called asset-backed commercial paper). The banks provide such entities with financial guarantees that only appear below-the-line in their balance sheet, playing the role of last-resort liquidity providers if and when difficulties of refinancing arise. The precise extent of such commitments in the aggregate and for individual banks is unknown. According to market estimates reported by the BIS outstanding asset-backed commercial paper reached a sum of $1.5 trillion last March, of which some $300 billion was based on mortgage-backed assets. According to another estimate , European banks have more than $500 billion invested in asset-backed-commercial-paper conduits with German banks holding a quarter of this sum.

As the prices and market liquidity of the collateral collapsed, refinancing by rolling over the outstanding commercial paper has become almost impossible. This is why the committed banks and financial institutions have been required to provide emergency liquidity. In this way, some of the credit risk that was transferred to the market by the banking system has re-emerged on the banks’ books, straining capital requirements to an extent depending on the size of the commitment relative to the assets of the bank. Two small German financial institutions (IKB and SachsenLB) found themselves exposed to their vehicles to such an extent, that they had to be bailed out with emergency rescues. The position of bigger banks, with more solid capital positions, has obviously not reached anything near a critical stage, but the suspicion of hidden difficulties infects the market. The problem is compounded by the fact that when the credit risk transfer mechanism froze in August, some core banks which arranged bridge financing for important private equity operations found it impossible to pass on the risk as usual. They are now unable to securitize and transfer the credit that they granted on the eve of the crisis (to the tune of over 200 billion).

The next question is: Were regulators aware of this roundtrip of credit risk (from banks to market and back to banks)? Hardly, one is tempted to answer. First, the official literature (central banks’ financial stability reports, Bank of International Settlements reports, papers of the Financial Stability Forum) shows that regulators were indeed worried by the vulnerability of the system following a change in risk appetite and in market liquidity conditions, but also that most of their attention was addressed to hedge funds as the banks’ riskiest counterparties. This is that source of the widespread recommendations that banks should obtain greater information on hedge-fund positions and improve counterparty risk measurement and management when dealing with them. I may be wrong, but I could find no mention of the potential problems arising from the existence of off-balance sheet vehicles sponsored and guaranteed by the banks. Second, only in the past few weeks have supervisory bodies been hastily inquiring on the exposure of individual banks to such vehicles. Supervisors have puzzled over the problem of where the credit risk ended up. Little did they know, apparently, that a chunk of it had returned where it came from.
This crisis, however it ends, is likely to prompt ill-conceived regulatory proposals. But, if there is one field where something ought to be done, even before damning the sins of rating agencies, it is to find a way to deal with the off-balance sheet operations of banks and achieve greater transparency of their effective exposure to risk.


1 A useful rendition of the events since June and up to August is in Bank of International Settlements, BIS Quarterly Review, September 2007. Also see the Vox columns on the subprime crisis.
2 Mario Draghi, “Monetary Policy and New Financial Instruments”, Central Bank of Argentina – 2007 Money and Banking Conference.
On which see BIS Quarterly Review, cit and Stephen Day and Agnes Molnar, “Investment vehicles: The advantages of flexibility”, International Financial Law Review, November 2006
3 Ivar Simensen and Ralph Atkins, “’Not uncritical’: Subprime exposure drags down German banks”, Financial Times, August 22, 2007,

525 Reads