This article describes a concept which could impact a variety of companies, countries or industries. To see what companies and articles reference this concept page, click here.

TOP CONTRIBUTORS

Vikram Parsani

A stress test is an assessment or evaluation of a bank's balance sheet to determine if it is viable as a business or likely to go bankrupt instead. It is a colloquial term often used in relation to the Supervisory Capital Assessment Program (SCAP), a forward-looking appraisal of capital conducted by the Federal Reserve and thrift supervisors to determine if the 19 largest financial institutions in the U.S. have sufficient capital buffers to withstand the recession and financial crisis. Regulators chose Tier 1 common capital as the yardstick to assess capital levels, although it was originally thought that they would use tangible common equity.^[1][2]

Scope

The assessment was limited to bank holding companies with assets exceeding $100 billion. In order of decreasing assets, the 19 institutions tested were:

• Bank of America
• J P Morgan Chase
• Citigroup
• Wells Fargo
• Goldman Sachs
• Morgan Stanley
• Metropolitan Life Insurance Company
• PNC Financial Services
• U.S. Bancorp
• The Bank of New York Mellon
• GMAC
• SunTrust Banks
• Capital One Financial Corporation
• BB&T
• Regions Financial Corporation
• State Street Corporation
• American Express Company
• Fifth Third Bancorp
• KeyCorp

The financial institutions included in this exercise represented two-thirds of aggregate U.S bank holding company assets and more than one-half of loans U.S. banking system.

Framework

Each financial institution was asked to estimate two figures over a two-year time horizon starting in 2009 and ending in 2010. These figures were: a) potential losses from loan, securities and investment portfolios, including off‐balance sheet commitments and contingent liabilities and exposures; and b) the resources available to absorb these potential losses, including pre‐provision net revenue (PPNR) and allowance for loan and lease losses (ALLL).

These estimates were arrived at under the assumption that one of two macroeconomic scenarios would play itself out over the next two years. These scenarios were: a) the baseline scenario, which reflected the consensus expectation of professional forecasters in February 2009 on the depth and duration of the recession; and b) the more adverse scenario, which was designed to reflect an recession that is longer and more severe than the consensus expection.

Baseline Scenario

As mentioned earlier, the baseline scenario was intended to represent a consensus view about the depth and duration of the recession. As such, the baseline assumptions for GDP growth and the unemployment rate in 2009 and 2010 were estimated to be equal to the average of forecasts published in February 2009 by Consensus Forecasts, the Blue Chip survey and the Survey of Professional Forecasters. In addition, the baseline assumptions for house prices were estimated to be consistent with the path implied by futures prices for the Case‐Shiller 10‐City Composite Index in February 2009 and the average response to a special question on house prices in the Blue Chip survey.

More Adverse Scenario

The more adverse scenario was intended to reflect the possibility that the economy could turn out to be appreciably weaker than expected under the baseline outlook. As such, the more adverse assumptions for GDP growth and the unemployment rate in 2009 and 2010 reflect a deeper and longer recession than the baseline assumptions, and were constructed from the historical track record of private forecasters as well as their current assessments of uncertainty.^[3] In addition, the more adverse assumptions for house prirces were estimated to be about 10% lower at the end of 2010 relative to their level in the baseline scenario.^[4] Note that the more adverse scenario is not intended to be a worst case scenario, but to reflect conditions that are severe yet plausible.

Results

The bank stress test results were released on 7 May 2009. They indicated that, if the economy were to track the more adverse scenario, losses in 2009 and 2010 at the 19 financial institutions assessed could be $600 billion. Of this amount, approximately $455 billion would be losses from accrual loan portfolios, particularly from residential mortgages and other consumer-related loans, while the remaining $135 billion would be losses from trading-related exposures and securities held in investment portfolios, specifically from charge-offs and write‐downs on the values of securities.

Besides estimating potential losses at these firms in 2009 and 2010, the stress tests also assessed potential resources available to absorb these losses. At the end of 2008, capital ratios at all 19 bank holding companies exceeded minimum regulatory requirements and Tier 1 capital at these firms totaled $835 billion. Nonetheless, 10 of the 19 assessed banks were ordered to raise $75 billion in additional capital buffers, most of which would need to be Tier 1 common capital. Regulators had originally required that the banks raise $185 billion in additional capital buffers, but noted that asset sales and capital restructuring had reduced this amount by $110 billion.

Loan Loss Rates

In the more adverse scenario, the estimated two-year cumulative loss rate on total loans equaled 9.1%, especially high by historical standards. As shown in the chart, this loss rate was higher than the two-year loss rates observed for U.S. commercial banks from 1920 to 2008.

Losses from Securities (AFS and HTM)

To evaluate losses for securities in available-for-sale (AFS) and held-to-maturity (HTM) portfolios, regulators focused on securities subject to credit risk. At the end of 2008, the 19 assessed banks held $1.5 trillion worth of securities. More than one‐half of these securities were Treasury, agencies or sovereign securities, or high-grade municipal debt, and so subject to limited or no credit risk. About $200 billion was in non‐agency mortgage‐backed securities (MBS). For securitized assets, regulators assessed if the security would become impaired during its lifetime and, if so, the security was written down to fair value with a corresponding other than temporary impairment (OTTI) charge equal to the difference between book and market value. These OTTI charges equaled $35 billion in the more adverse scenario, with almost one‐half of the estimated losses coming from the non‐agency mortgage-backed securities.

Trading‐Related Market and Counterparty Losses

Firms with trading assets of $100 billion or more were asked to estimate potential trading‐related market and counterparty credit losses under a market stress scenario based on the severe market shocks that occurred in the second half of 2008. The estimated losses from trading‐related exposures were substantial, close to $100 billion across the five firms to which it was applied. The primary drivers of potential stress losses were private equity holdings, other credit‐sensitive trading positions and possible losses stemming from counterparty credit exposures to over‐the‐counter (OTC) derivatives trading counterparties.

Criticisms

On 3 May 2009, renowned investor Warren Buffet criticized the stress tests being conducted on the 19 largest U.S. banks on the grounds that "a one-size-fits-all attempt to identify capital shortfalls was inappropriate".^[5] He argued that stress tests were not necessary for 15 of the 19 banks being evaluated because they were not too big to fail and did not pose that kind of risk. He also added that all but 4 of the banks being tested could easily be sold with the assistance of the Federal Deposit Insurance Corporation.

See Also

• The Supervisory Capital Assessment Program - Design and Implementation, The Federal Reserve
• The Supervisory Capital Assessment Program - FAQs, The Federal Reserve
• The Supervisory Capital Assessment Program - Overview of Results, The Federal Reserve
• Comparing the 19 Banks that were Stress Tested, The Wall Street Journal

References

1. ↑ Banks won concessions on tests, The Wall Street Journal
2. ↑ U.S. looks to tangible common equity in stress tests, Reuters
3. ↑ Based on the historical accuracy of Blue Chip forecasts made since the late 1970s, the likelihood that the average unemployment rate in 2010 could be at least as high as in the alternative more adverse scenario is roughly 10%. In addition, the subjective probability assessments provided by participants in the January Consensus Forecasts survey and the February Survey of Professional Forecasters imply a roughly 15% chance that real GDP growth could be at least as low, and unemployment at least as high, as assumed in the more adverse scenario.
4. ↑ Based on the year‐to‐year variability in house prices since 1900, and controlling for macroeconomic factors, there is roughly a 10% probability that house prices will be 10% lower than in the baseline by 2010.
5. ↑ Buffett hits out at stress tests, Financial Times