- Carla Vecchio
Contents (Jump to)
1.0 Light Touch Regulation and the Global Financial Crisis
2.0 Stress Scenarios and Credit Modelling
1.0 Light Touch Regulation and the Global Financial Crisis
In October 2008, almost three years after stepping down as Chairman of the U.S. Federal Reserve Bank, Alan Greenspan, also known as the Maestro, admitted in a Congressional hearing that he had been “partially” wrong in assuming that lending institutions would act in the best interest of their shareholders (Greenspan 2009), thus deflecting blame for the Global Financial Crisis (GFC) that led to the collapse of dozens of major financial institutions and millions of mortgage defaults, costing the global community trillions of dollars in savings and millions of jobs. Yet it was he, an ex-director of JP Morgan, first appointed by Ronald Reagan, who was instrumental in creating the conditions that made it possible.
Greenspan approved bank consolidation, pushed financial deregulation, advocated a reduction in bank capital reserves and blocked efforts to stop abusive subprime lending (Pearlstein 2013). Finally, when presented with warning signs of an impending disaster by fellow Federal Reserve Board members just before the GFC, he dismissed them and instead drew conclusions best described by Lord Adair Turner’s words (2010) “Panglossian, that is blindly or naively optimistic”. Greenspan did however concede during his congressional grilling that there should have been greater regulatory oversight of financial institutions, and it is now universally accepted that this “light touch regulation” leading up to the GFC materially contributed to the crisis. For people like Lord Adair, who now have the benefit of hindsight, the “major cause of the crisis” was the fact that “over several decades prior to 2008, private credit grew faster than GDP in most advanced economies” and with it, leverage (The Institute for New Economic Thinking 2014). He agrees that the regulators failed, and proposes a new set of policies to “constrain the growth of private credit” and more importantly to “influence the type of credit extended” (INET2014).
The generally accepted ultimate causes of the GFC are deregulation of the financial markets; financial innovations; executive compensation; low interest rates; sub-prime loans; and speculation (The Global Financial Crisis 2012, 141).Whilst the changes to the financial sector were driven by ideology, the motivating force behind them was for the most part greed. Neoliberal theories “advocated policies that aided the accumulation” of wealth in fewer hands arguing that it would create jobs causing wealth to eventually trickle down to all (Beder 2009, 3). They also maintained that “government intervention in the management of the economy is unnecessary” because the market is self-correcting, an idea attractive to government because it absolved it of responsibility (Beder 2009, 3).
The rapid growth of international trade in the 1980’s facilitated global financial liberalisation which made it easier for American banks to argue for deregulation to make them more competitive against foreign banks (The Global Financial Crisis 2012, 141). They found an ally in President Ronald Reagan who had been elected on a platform of limiting the role of government, and they embarked on an unprecedented and expensive lobbying campaign to convince other politicians of the benefits of financial deregulation (Johnson 2012).
The first burst of deregulatory bravado came in 1982 with the ushering in of the Garn-St. Germain Depository Institutions Act. Key provisions of the Act like raising the “allowable ceiling on direct investments by savings institutions in non-residential assets from 20% to 40%” set the scene for the savings and loan crisis of the 1980’s and would later be blamed for thousands of bank failures (Gilani 2009).
The ultimate price however, was the undoing of the Glass-Steagall Act of 1933, also known as the Banking Act of 1933. Among other things, the Act governed banks’ domestic operations; separated commercial and investment banks; and established the Federal Deposit Insurance Corporation (FDIC), thus ensuring bank deposits and giving the Federal Reserve greater control (The Chronology of Bank Deregulation n.d.).
In 1987, Alan Greenspan took over the chairmanship of the Federal Reserve Board, and his free-market philosophies would champion the deregulatory movement (Gilani 2009).
A year later, in 1988, the “Basel Accord established international risk-based capital requirements for deposit taking banks” that would require lenders to set aside reserves (Gilani 2009). Conversely, marketable securities would only require minimal reserves, which allowed unscrupulous banks to free up reserves by shifting from “originating and holding mortgages to packaging them and holding the mortgage assets in a now-securitized form”, thus severing the link between asset quality considerations and asset liquidity considerations (Gilani 2009).
Greenspan asserted that bank deregulation was necessary for banks to become global financial powers, and by using his own powers, Greenspan set out to dismantle the Glass-Steagall Act firstly by allowing banks to deal in debt and equity securities, and finally by allowing banks to own securities firms (The Chronology of Bank Deregulation n.d.).
The final demise of the Glass-Steagall Act came when Citibank was bought by Travelers, a deal which under the Act was illegal. It was then made legal when the Gramm-Leach-Billey Financial Services Modernisation Act, bulldosed through by Senator Gramm, was signed into law by Bill Clinton and at once doing away with the Glass-Steagall Act (Gilani 2009).
Senator Gramm who was an economist and free market ideologist, further used his position of power to espouse the virtues of subprime lending by famously declaring “I look at subprime lending and I see the American Dream in action” (Gilani 2009).
Subprime lending or lending to people who would ordinarily have little hope of obtaining a loan, thus came to be looked upon favourably by politicians as it allowed record numbers of consumers to purchase a home (The Financial Crisis and the Great Recession n.d., 337). A direct outgrowth of easy lending, its roots can be traced to the Technology Bubble of the late 1990’s which had been encouraged by the loose money policies of the Federal Reserve under Alan Greenspan (Bello 2008).When the bubble burst and sent the U.S. into recession, Greenspan tried to counter it by lowering the prime interest rate to a historical low of 1% (The Financial Crisis and the Great Recessionn.d., 338). This in turn encouraged another bubble: the real estate bubble (Bello 2008). “Driving the demand for subprime loans was the development of a culture of entitlement” and the U.S. government’s push of home ownership as an inalienable right (The Global Financial Crisis 2012, 145). This combination of low interest rates and high levels of liquidity facilitated higher risk taking and speculation. Caution was often equated to lack of optimism so even the most cautious were driven by herd mentality into the market, even in the face of continuously rising house prices (The Global Financial Crisis 2012, 146).
Financial wizards were in the meantime designing ever more complex financial products. Initially meant to “manage risk and make capital less expensive and more available”, they ironically ultimately led to the GFC (The Global Financial Crisis 2012, 142). Loans were bundled in a process called securitization, and sold globally to others who had no direct interest in the customers’ ability to repay the loans. In the process, vast amounts of money were made available to borrowers leading to more loans and further driving house prices up. Credit Derivatives, essentially bets on the credit worthiness of a particular company were used to transfer risk away from the banks leading to even more risk taking on the part of the banks (The Global Financial Crisis 2012, 143). Huge executive salaries and compensation packages played a major role in the creation of the GFC. Tied to short-term performance, they further encouraged risk taking, relaxing of lending criteria (The Financial Crisis and the Great Recessionn.d.,340) and even accounting fraud as in the case of Enron, Global Crossing and WorldCom (The Global Financial Crisis 2012, 144).
The proliferation of subprime lending was responsible for doubling the mortgage borrowing in the U.S. from an annual average of $500 billion in 1998 to over $1 trillion in the 2003-6 period (The Financial Crisis and the Great Recession n.d., 341). When mortgage repayments became more difficult in 2006, a wave of subprime foreclosures ensued creating a glut in the market and dramatic drop in house prices (The Financial Crisis and the Great Recession n.d., 341). The rest is history. Banks deemed “too big to fail” failed, and trillions of dollars were lost. In the U.S. alone, 3 million homes were foreclosed and 9 million people lost their jobs.
In his congressional testimony, Alan Greenspan basically testified that he thought he could trust bankers and credit rating agencies to do the right thing by their shareholders and price risks accordingly, but with such huge profits to be made, it appears that greed triumphed. Considerable evidence has in fact mounted since the GFC to show that his vision of the markets and organisations “is not only oversimplified, but utopic” (Turner 2010). Critics and many economists now blame Greenspan for the crisis. Indeed, it is clear that notwithstanding his faith in others, it was incumbent upon him as the steward of the world’s largest economy to be vigilant. “You had the authority to prevent irresponsible lending practices that led to the subprime mortgage crisis. You were advised to do so by many others,” said Representative Henry A. Waxman of California, chairman of the committee (Andrews 2008). “Do you feel your ideology pushed you to make decisions that you wish you had not made?” Mr. Greenspan conceded: “Yes I’ve found a flaw” (Andrews 2008).
2.0 Stress Scenarios and Credit Modelling
When credit is extended by a lender to a borrower, there is a certain risk that the borrower may default on its payment, thus causing a loss to the lender. If the losses are large enough, the lender may be forced to default on its own obligations to others, as seen during the most recent Global Financial Crisis (GFC) which saw a number of large banks file for bankruptcy.
To minimise the probability of the borrower defaulting, banks adopt lending practices and ratios, and conduct a review of the borrower’s ability to repay the loan. Competitive pressure from other banks to make credit more affordable means that banks have to try as best they can, to estimate the probability of defaults and the size and nature of possible losses, and make provision for them. The banks’ credit is mathematically modelled, which is then used to estimate the likely outcomes produced by different hypothetical but realistic and potential scenarios.
Credit models can be divided into two groups: credit risk models and credit growth models. A credit risk model is used to predict the main credit risk parameters, particularly the probability of default. Conversely, a credit growth model is used to estimate the growth in bank portfolios and to estimate the growth of the bank’s risk-weighted assets, that is, the bank’s off-balance-sheet exposures weighted according to risk and hence the capital requirement as explained in the Basel Accord. A bank’s ability to withstand the most adverse conditions is tested by carrying out a so called “stress test”, whereby extreme values for certain variables are used in the bank’s credit model to predict the outcome. Most commonly, the stress test is applied to credit risk as this carries the most important and most serious consequences for a bank.
The Banking Committee on Banking Supervision (BCBS) (2009, p.1) states that stress testing is a vital risk management tool employed by many banks “as part of their internal risk management and, through the Basel II capital adequacy framework, is promoted by supervisors”. The rigorous stress testing program requires management to adopt a forward thinking mentality and create “what if” scenarios that are extreme, but plausible. Thus the purpose of such an exercise is to assess a bank’s resilience to potential adverse shocks in the financial economic environment that may have a catastrophic effect on the institution or financial system as a whole. For example, a modeller may measure the effect rising interest rates has on home loan defaults. A mathematical formula can then be derived to link the two factors. As well as being a supplementary tool for other risk management approaches, stress testing provides management with an indication of the “appropriate level of capital necessary to endure deteriorating economic conditions” (BCBS 2009, p. 1).
Since the GFC, stress testing of banking systems has been used more extensively and in a broader variety of contexts. The internal risk management exercises within the Basel II capital adequacy framework has led many financial institutions and supervisors to focus attention on stress tests in relation to credit risk as an additional way to test the reliability of the internal models they adopt (Schechtman and Gaglianone 2011). The increasing need for financial stability within today’s economic environment, and its role as a policy goal of central banks, has also promoted interest in macroeconomic stress testing and the link it has to credit risk (Schechtman and Gaglianone 2011).
There are two broad types of stress scenarios: The reduced-form stress scenario and the sharply contrasting structural stress scenario. According to Roger Stein from Moody’s Research Labs (2011), these terms have been adopted from the credit modelling literature, and are the two main approaches used to model credit risk. A structural scenario possesses a “causal, economically intuitive relationship” (Stein 2011) between a firm’s asset and the probability of it defaulting, that is, it has a clear and logical economic rationale for the effect of a particular factor on a portfolio. It is focused on the state of the economy, as described by the macroeconomic factors involved and requires a definite link between asset behaviour and the stress factor. Because different asset classes within the portfolio are dependent on the same common factors, there is also a very high coherence or consistency of results between them. Unlike other models, the structural model can explain why a company or a bank, for example, is likely to default. Although intuitive, structural stress scenarios make high demands on the testers because not only do changes in the economic factors have to be consistent throughout the different asset classes, but the resultant asset behaviour must also be fully described by the mathematical function linking the assets to the economic factors. Because of this, few structural models for stress testing have been developed so far. One such model is at the core of the Bank of England’s stress testing agenda. Generally, structural models are useful from a central bank’s perspective as they assume a linear relationship between macroeconomic factors and credit risk, hence providing a way of estimating financial stability risks.
In contrast, a reduced-form scenario focuses on the state of the assets and treats default events as “surprises”. It does not provide an economic cause for the resulting state of the assets, thus only requiring the stress tester to define the asset behaviours themselves (Stein 2011). Modelling credit risk under this approach requires no assumptions to be made concerning why defaults occur. Instead, the dynamics of default are directly linked to the default rate. Default in the feduced form stress model is an unforeseeable event which will always have a positive default probability.
A relationship between assets is not required in the reduced form, nor is a logical reason given for a certain observed effect. Because of their less rigorous demand, reduced form stress models are used predominantly in the financial industry. For example, the stress testing approach used even by the Bank of France is based on a reduced form of the credit risk model wherein a borrower’s ability to repay his loan is found as the difference between the value of the assets and that of his loan, and default occurs when the value of the debt exceeds the value of the loan.
Credit risk is one of the most important areas for stress testing since it ultimately affects a bank’s profits and even its solvency. When used in conjunction with credit models, both the structural and reduced-form stress scenario approaches assist management in providing a means of mitigating “risk by enabling intuitive interpretations of states of the world that may cause a portfolio or organisation to experience high losses” (Stein 2011). Stein (2011) states that it is indeed this “intuitiveness that makes stress testing useful in evaluating a credit model’s behaviour in general, and the appropriateness of a model’s linking functions in particular”.
Both the Structural and the Reduced-form Stress Test models have found an important and useful role in the financial industry. Modellers will use either one depending on what is being tested and what is known or can be quantified. As it is, even though the credit models used by their very nature do not perfectly represent the real world, the stress tests applied to them, “still provide a measure of intuition that is generally otherwise not feasible” (Stein 2011). This is because both structural and reduced-form stress scenarios induce a connection to both the credit models and the risks in the portfolio which provides management with insights into both the model’s behaviour and also the drivers of the portfolio’s credit risk (Stein 2011).
- Ricardo Schechtman and Wagner Piazza Gaglianone 2011 – Macro Stress Testing of Credit Risk Focused on the Tails – http://www.bcb.gov.br/pec/wps/ingl/wps241.pdf
- STEIN 2011 – The Role of Stress Testing in Credit Risk Management
- BCBS 2009 – Basel Committee on Banking Supervision
- Breuer, Jandacka, Rheinberger & Summer 2009 – How to find plausible, severe and useful stress scenarios