One of the programs created under the TARP is the Capital Assistance Program (CAP). The CAP was designed to promote confidence in the financial system by ensuring that the nation’s largest banks have sufficient capital cushion against larger than expected future losses. A stress test, called the Supervisory Capital Assessment Program (SCAP), was crafted and implemented by the Federal Reserve and the Treasury to aid them in assessing said capital cushion. Details of the stress test were released to the public by the Federal Reserve on April 24th, through a 21-page white paper. I present here a synopsis of the white paper.
All domestic bank holding companies (BHCs) with year-end 2008 assets exceeding $100 billion were required to participate in the SCAP. According to ProPublica, 19 firms fell under this requirement. They are listed in the second table below. These 19 firms hold 66% of the assets and 50% of the loans in the US banking system. They were asked to project their losses, and available resources for absorbing these losses, for 2009 and 2010 based on two economic scenarios — a baseline scenario and an adverse alternative. The table below lists the components of the economic scenarios and the effect of the baseline and adverse conditions on each of them. The supervisors, then, assessed whether their capital was adequate for them to function during this period.
Step 1: Loss Projections. BHCs were asked to project losses for 2009 and 2010 for 12 separate categories of loans held in the accrual book, for loans and securities held in the available-for-sale (AFS) and held-to-maturity (HTM) portfolios, and in some cases for positions held in the trading account. The losses were to be consistent with the economic outlooks in the baseline and more adverse scenarios. The BHCs were instructed to estimate forward-looking, undiscounted credit losses, that is, losses due to failure to pay obligations (“cash flow losses”) rather than discounts related to mark‐to‐market values. The required assessments were broadly classified as:
- First and Second Lien Mortgages: institutions provided detailed descriptions of their residential mortgage portfolio risk characteristics – type of product, loan-to-value (LTV) ratio, FICO score, geography, level of documentation, year of origination, etc.
- Credit Cards and Other Consumer Loans (e.g., auto, personal, student): portfolio information included FICO scores, payment rates, utilization rates, and geographic concentrations.
- Commercial and Industrial Loans: based on the distribution of exposures by industry
- Commercial Real Estate Loans: included loans for construction and land development, multi-family property, and non-farm non-residential projects. Information such as property type, loan-to-value ratios, debt service coverage ratios, geography, and loan maturities was provided.
- Other Loans: farmland lending, loans to depository institutions, loans to governments, etc.
- Securities in AFS and HTM Portfolios: majority are public-sector securities such as Treasury securities, government agency securities, sovereign debt, and high-grade municipal securities. Private-sector securities include corporate bonds, equities, asset-backed securities, commercial mortgage-backed securities (CMBS), and non-agency residential mortgage-backed securities (RMBS). Supervisors focused on evaluating the private-sector securities. Loss estimates were based on an examination of 100,000 of these securities. Loss estimate, and subsequent “write-down” to fair value, for each security was determined based on credit loss rates on the underlying assets, consistent with loss rates for unsecuritized loans listed above.
- Trading Portfolio Losses: estimated by applying market stress factors to the firm’s trading portfolio based on actual market movements that occurred between June 30 and December 31, 2008.
- Counterparty Credit Risk: the risk that an organization is unable to pay out on a credit-related contract when it is supposed to, which directly impacts a firm’s earnings and the value of its assets. The action taken by the firm to account for this risk is referred to as credit valuation adjustment (CVA). Supervisors focused specifically on a firm’s loss estimates for mark-to-market losses stemming from CVA associated with market shocks applied to assets in trading books.
Step 2: Resources to Absorb Losses. Institutions were also instructed to provide projections of resources available to absorb losses under the two economic scenarios. These include the pre-provision net revenue (PPNR) and the allowance for loan losses over the two-year horizon.
- PPNR is the income after non-credit-related expenses that would flow into the firms before they take provisions or other write-downs or losses.
- BHCs supposedly had some allowance for loan and lease losses at the end of 2008. They were required to estimate what portion of this allowance would be required to absorb potential future credit losses on their loan portfolio under each economic scenario. This calculation could either result in depletion of the year-end 2008 reserves (if there is adequate allowance) or indicate the need for building the reserves (if the allowance is inadequate).
Step 3: Determination of Necessary Capital Buffer. Supervisors examined two main elements as indicators of capital adequacy – pro forma equity capital and Tier 1 capital.
- Pro forma equity capital was estimated by rolling tax-adjusted net income (PPNR minus credit losses minus reserve builds) for the two-year horizon through equity capital.
- Tier 1 capital is composed of common and non-common equity, with the dominant component being common stockholder’s equity.
The initial assessment of the capital adequacy, or lack thereof, was conveyed to the BHCs in late April and is expected to be released to the public on May 4th, 2009. As yet it is uncertain whether the publicized results will reveal much about the banks.