The Basel 2 Accord of January 2007 considerably enhanced the credit risk regulations. The schedule of successive Accords, from Basel 1 to Basel 2, is summarized as follows:
2 Basel Committee on Banking Supervision (1996, updated 2005), Amendment to the capital accord to incorporate market risk, [19].
The approaches of Basel 2 for credit and the update for market risk were published in June 2006 in “International Convergence of Capital Measurement and Capital Standards – A Revised Framework, Comprehensive Version” [21].
The goals of the new Accord were:
• To promote stronger management practices;
• To promote more risk-sensitive capital requirements through a greater use of banks' own assessment of the credit standing of the borrowers;
• To provide a range of approaches for determining the capital charges for credit risk by allowing banks and supervisors to select the options that were most appropriate for their operations.
The Accord also introduced new capital requirements for operational risks.
For making the capital charge risk sensitive, the Accord provides incentives to use the “internal ratings-based” (IRB) approach, using as inputs for the risk weights the internal ratings, or credit risk assessments, of banks. When not applicable, the banks can rely instead on a “standardized approach” where the risk weights are regulatorily defined.
2.3 Some Lessons of the Financial Crisis
The new waves of regulations, known under Basel 2.5 and Basel 3, are inspired by the lessons of the crisis, which are briefly summarized hereafter. The crisis raised a number of issues, with respect to liquidity, fair value accounting or solvency, as they interact and result in contagion and pro-cyclicality. Contagion refers to the waves of failures triggered from individual failures to the system as a whole. Pro-cyclicality refers to the mechanisms that amplify the cycles of the financial system.
The crisis was characterized by the liquidity crunch that plagued the financial system in 2008.5 A lack of liquidity can emerge from the risk aversion of lenders in stressed times. Players refrain from providing liquidity in a context of failures as no one knows who is next to fail. The “who is going to be next to lose” issue makes potential lenders reduce their exposures to others, for fear that they would suffer unexpected losses of undetermined magnitudes.
Other mechanisms contributed to the liquidity squeeze. The overreliance on short-term funds, a characteristic of the system at the time of the crisis, exacerbates the effect of a liquidity crunch. Once liquidity dried up, all liquidity commitments of banks, whereby banks commit to lend within limits to borrowers, were triggered and translated in a great deal of “involuntary lending”. Financial players had to comply with their commitments precisely when they had insufficient liquidity for themselves. Involuntary lending was a source of liquidity for some, but it made liquidity even scarcer as financial firms started to hoard liquidity.
The liquidity crunch lasted even after massive injections of liquidity by central banks through various programs of purchases of financial assets from banks. Presumably, banks and financial firms were “hoarding liquidity” as a protection against a lack of liquidity instead of using it for extending credit. Monetary authorities could not prevent the credit crunch that followed.
Fair value is pro-cyclical as it extends the markdowns to all assets accounted for at fair value, traded or not.6 Many assets lost value in inactive and illiquid markets. As a consequence of the magnitude of the downturn, fair values appeared disconnected from the fundamental values of assets. The category of assets subject to model valuation extended to assets that, in normal circumstances, would have been fairly valued from prices. Model prices were subject to a negative perception, since many assets lost perceived value as the confidence in models evaporated.
When markets move down, fair value rules trigger markdowns of portfolios, even if there is no intention to sell them, which translates into losses and erodes the capital base of banks. Fair value rules in severe conditions made losses unavoidable. Moreover, leveraged banking firms tend to reduce their debt through fire sales of assets. When liquidity relies on fire sales of assets under adverse conditions, markdowns bite the capital base. Under stressed conditions, solvency and liquidity become intertwined. Whether illiquidity or solvency is the initial cause does not matter. Once the mechanism triggered, it operates both ways.
For borrowing, financial firms pledge their portfolios of securities to lenders. Such collateral-based financing is subject to loan-to-value ratios, whereby the value of pledged assets should be higher than the debt obligation. In a market downturn, complying with ratios imposes that additional collateral be posted for protecting lenders, or, alternatively, that debt be reduced. In a liquidity and credit crunch, cash is raised from the sales of assets for paying back the debt. Fire sales of assets for reducing debt and bringing back asset value in line with loan-to-value ratios add to the market turmoil.7
The mechanism is pro-cyclical. Fire sales of assets create a downward pressure on prices, which triggers a new round of collateral calls. This new round results in additional sales of assets and starts another cycle of market decline, and so on. An adverse feedback loop between asset prices and system liquidity develops as a result of such interactions. The mechanism is strongly pro-cyclical: if asset values move down, sales of assets amplify the downturn. In a highly leveraged system, the adverse dynamics develop until the system deleverages itself.
Unregulated funds tend to highly leverage their portfolios for enhancing the return to investors. In a favorable environment, asset values are up and extending collaterized credit to funds is easy. In a stressed environment, the deleveraging of funds puts pressure on the entire system.
Regulations are also pro-cyclical because they impose capital buffers that tend to increase in adverse conditions, while simultaneously the losses shrink the capital base. The process results in credit contraction precisely when credit is needed most by firms chasing funds in illiquid markets, and contributes to the contraction of the whole system.
Securitizations refer to the sales of pools of banking book assets to investors in the capital markets by issuing bonds backed by these assets. Securitizations were a key technique in the “originate and distribute” business model of banks, whereby the banks finance their loans in the markets and, simultaneously, free their capital from backing the risk of sold loans. Prior to the crisis, banks off-loaded and distributed massive amounts of their credit risk into the capital markets.8 The so-called “toxic assets”, such as subprime loans, were believed to have found their way into the pools sold in the markets and the risk was perceived as disseminated throughout the whole system.
Rating agencies recognized that they underestimated the risk of asset-backed bonds, many with the highest quality grade, and a wave of downgrades followed. Downgrades command a higher cost of funds, and a higher required return, which translate in a loss of value of the downgraded assets. Investors in asset-backed bonds of securitizations, originally of a high quality, incurred massive losses. The trust in the securitization mechanism disappeared, and with it a major source of funds for the banking system.
The frequency of downgrades by rating agencies increased abruptly by the end of 2007, as rating agencies seemed not to have anticipated the effect of the crisis and tried to catch up with bad news. Lagged downgrades were concentrated in time, instead of gradually measuring the actual credit standing of issues.
All entities were hit by rating downgrades. Among those are insurance companies, or monolines, acting as “credit enhancers”.