The US Supervisory Capital Assessment Program (SCAP) and Capital Assistance Program (CAP)

The Treasury released the Financial Stability Plan on February 10, 2009, to “address the uncertainty, troubled assets and capital constraints of our financial institutions as well as the frozen secondary markets that have been the source of around half of our lending for everything from small business loans to auto loans” (UST PR, 02/10/2009). The plan had several components:

1. The Financial Stability Trust. The comprehensive stress test, which would later become the SCAP, and the Capital Assistance Program (CAP) were core features of this piece of the Plan. All investments made under CAP would have been placed into a separate Financial Stability Trust, which was specifically designed to manage the government’s investments (UST PR, 02/10/2009).
2. Public-Private Investment Fund. This fund was designed to purchase legacy assets from distressed institutions with the goal of soliciting private investors to co-invest with the government. The Treasury aimed for it to include a public-private financing component of $500 billion, with the potential to go up to $1 trillion, and private-sector pricing of assets; however, it was marginally used and Treasury committed only about $18.6 billion to the program, receiving a net positive return of about $3.9 billion (PPIP Quarterly Report: September 2013; UST PR, 02/10/2009).
3. Consumer & Business Lending Initiative. This was a joint venture of the Treasury and Fed designed to unfreeze key secondary markets, specifically by expanding the capacity of the not-yet-implemented Term Asset-Backed Securities Loan Facility (TALF). Treasury and the Fed set the maximum size of TALF at $1 trillion; usage ultimately peaked at $49 billion (FRBNY: TALF 101; UST PR, 02/10/2009).
4. Transparency, Accountability, Monitoring, and Conditions. With this new plan, the government expected recipients of aid to explain how they were going to use their funds and how this new aid would impact their ability to lend. The plan also included limits on executive compensation and on common stock dividends, repurchases, and cash mergers (UST PR, 02/10/2009).
5. Housing Support and Foreclosure Prevention. Treasury pledged $50 billion to help middle-class homeowners reduce their monthly mortgage payments. Additionally, the government committed to “establishing loan modification guidelines and standards for government and private programs,” as well as mandating recipients to participate in foreclosure mitigation plans (UST PR, 02/10/2009).
6. Small Business and Community Lending. Small Business Administration (SBA) lending, which had declined by 57 percent from the first quarter of 2008, was addressed. The administration’s Small Business and Community Bank Lending Initiative financed the purchase of triple-A-rated SBA loans, increased the guarantee of said loans to 90 percent, reduced SBA lending fees, and streamlined the application process to encourage lending (UST PR, 02/10/2009).

The announcement of the plan was the first time the Treasury released details about SCAP and CAP. It released the terms and conditions of the program on February 25 (UST PR, 02/10/2009; UST PR, 02/25/2009). The SCAP was widely publicized, with the goal of reducing uncertainty and restoring confidence in the 19 BHCs that were tested. The CAP was a government backstop to recapitalize BHCs if they were unable to raise funds on private capital markets within six months (SCAP: Design and Implementation).

Legal authority for the Capital Assistance Program came from Section 101 of the Emergency Economic Stabilization Act (EESA) of 2008, which Congress passed on October 3, 2008. Section 101 states that, “The [Secretary of the Treasury] is authorized to establish the Troubled Asset Relief Program (or “TARP”) to purchase, and to make and fund commitments to purchase, troubled assets from any financial institution…”(P.L. 110-343). The Treasury and Federal Reserve ultimately decided to invest directly in the capital structure of banks rather than purchase assets off their balance sheets. Its first TARP program was the Capital Purchase Program, announced in October (SIGTARP 10/05/2009).

The Treasury later used this authority to make more targeted injections in AIG, Bank of America, and Citigroup (SIGTARP 10/05/2009). This same authority formed the legal basis for CAP in February. See Figure 5 for a comparison of capital instruments used in these two facilities.

In its February 10, 2009, press release, the Treasury said that the stress tests would identify the capital buffer banks needed to remain well capitalized in an adverse scenario. It said it would make government funds available through CAP to build that buffer. But it emphasized that banks would “be encouraged to access private markets to raise any additional capital needed to establish [the capital] buffer” rather than using CAP. Treasury framed CAP as a temporary facility, “a bridge to private capital until market conditions normalized”. The government capital instruments would be convertible to common equity “if needed to preserve lending in a worse-than-expected economic environment”. The announcement also stipulated that any investments made under CAP would be “placed in a separate entity set up to manage the government’s investments in US financial institutions” (UST PR, 02/10/2009).

In its more detailed announcement on February 25, Treasury issued (1) a white paper detailing the reasons for establishing such a facility as well as a broad overview of key elements, (2) the official term sheet that bank holding companies would use to apply to the program, and (3) a list of frequently asked questions (FAQs) to help increase understanding of the program (UST PR, 02/25/2009). The Treasury expanded on its description of the program on February 25 in a few distinct ways.

The original plan did not specify what Treasury expected the banks that received CAP capital to do with it, only stating that “the CAP instrument should improve confidence and increase the willingness of financial institutions to lend”. In the February 25 release, however, the expected usage was much clearer, with Treasury requiring any applicant to “submit a plan for how they intend to use this capital to preserve and strengthen their lending capacity”. Specifically, Treasury desired applicants to increase their lending relative to what they would have been able to do without government capital. These plans would be made public by Treasury once government equity under the CAP had been obtained (UST PR, 02/25/2009).

The “separate entity” that would hold the government’s investments was further explained in the white paper that Treasury released. Treasury stated that all investments made under CAP would be put in a separate trust that would manage the government’s investments, and that the trustees’ primary objective was to “protect and create value for the taxpayer as a shareholder over time”. This was done to reinforce Treasury’s goal of keeping any period of government investment “as temporary as possible” (CAP: White Paper).

The names of applicants would not be published. However, for any completed transactions, the name of the bank and the amount invested would be published within 48 hours of the investment. Applicants could request portions of the application be treated confidentially if they could “specifically demonstrate the harm (for example, loss of competitive position, invasion of privacy) that would result from public release of information” (CAP: FAQ).

In the case of the SCAP, the Federal Reserve released a detailed report to the public on April 24, 2009, on the design of the stress test. The Fed said the purpose of the report was to “assist analysts and other interested members of the public in understanding the results of the [SCAP]” (FRB PR, 04/24/2009). This report was released shortly after the test had concluded, and the results were released on May 7 (SCAP: Overview of Results).

In a speech on May 11, 2009, at the Federal Reserve Bank of Atlanta, Fed chairman Ben Bernanke stated that all agencies involved aimed to be as transparent as possible (Bernanke, 05/11/2009). There was, however, much internal debate among officials about how transparent the government ought to be about communicating the results of the stress tests. While there were regulators that wished to make many details of the tests public, doing so “ran counter to decades of a banking supervision philosophy that extolled the virtues of confidentiality”. Some regulators believed that too much transparency could be destabilizing by singling out “weaker” banks, which could lead to the very bank runs the SCAP was trying to prevent. In the end, the Federal Reserve agreed to a policy of transparency so the public could “decide for themselves [if the] stress test was sufficiently rigorous and credible (Clark, Kabaker, and Sachs 2020).

On May 6, the day before the official release, the Fed, Treasury, FDIC, and OCC released a joint statement that previewed the SCAP results. The agencies explained that the tests were not only about the quantity of Tier 1 capital but its quality (FRB PR, 05/06/2009). The SCAP, they said, was “by design” more stringent than a traditional solvency test (SCAP: Overview of Results). As a result, any additional capital that the BHCs would need as a result of the tests did not indicate inadequate capitalization, but was to provide an additional buffer in the event that a more pronounced recession occurred (FRB PR, 05/06/2009). The government also explained that any BHCs that needed additional capital would be required to submit a capital raising plan and would have six months to implement it (SCAP: Overview of Results).

Figure 6: Five-Year CDS Spreads for Six Large Bank Holding Companies (basis points)

Source: Bloomberg.

After the release of the SCAP results on May 7, CDS spreads for the six major bank holding companies, as well as the Libor-OIS spread, fell dramatically. Several of the largest participants in the Capital Purchase Program were also able to repay the government’s investment in June. See Key Design Decision No. 7 for more details on the impacts of the SCAP, and Figure 6 for historical bank CDS spreads during this time.

Investors remained concerned about bank solvency in early 2009, despite the extraordinary efforts prior to the SCAP by the FDIC, Treasury, and Fed to recapitalize banks and restore confidence. The focus of those concerns had shifted. While banks’ early losses during the financial crisis had resulted largely from exposures to subprime mortgages and complex credit products, the concerns in 2009 largely stemmed from more traditional credit risks, such as “rising delinquencies on prime as well as subprime mortgages, unpaid credit card and auto loans, worsening conditions in commercial real estate markets, and increased rates of corporate bankruptcy”. According to Bernanke, the SCAP was designed to measure the impact of high expected loan loss rates and to address the uncertainty around the extent of these expected loss rates in the event of more adverse economic conditions (Bernanke, 5/11/2009).

Section 104 of EESA established the Financial Stability Oversight Board (FSOB) to review the policies implemented under section 101. These policies included reviewing “the appointment of financial agents, the designation of asset classes to be purchased, and plans for the structure of vehicles used to purchase troubled assets”. The FSOB was also responsible for examining the effects that programs made under Section 101 had on “assisting American families in preserving home ownership, stabilizing financial markets, and protecting taxpayers” (P.L. 110-343).

The Board had the following members:

1. The Chairman of the Board of Governors of the Federal Reserve System
2. The Secretary of the Treasury
3. The Director of the Federal Housing Finance Agency (FHFA)
4. The Chairman of the Securities and Exchange Commission (SEC), and
5. The Secretary of Housing and Urban Development (HUD).

Additionally, the Board was required to report to Congress and the Congressional Oversight Panel (COP) at least quarterly to discuss the matters described above (P.L. 110-343).

The Special Inspector General for the Troubled Asset Relief Program, or SIGTARP, was established through Section 121 of EESA. The Special Inspector General was to be appointed by the President and confirmed by the Senate, and was responsible for “coordinat[ing] audits and investigations of the purchase, management, and sale of assets by the Secretary of the Treasury . . .” (P.L. 110-343).  SIGTARP did this by collecting the following information from the Treasury:

1. Descriptions of categories of troubled assets purchased.
2.  Listings of assets that fell into the aforementioned categories.
3. Explanations of the reasons for purchasing said assets.
4. Listings of the financial institutions that the troubled assets were purchased from.
5. Detailed biographical information about the people hired to manage these troubled assets.
6. Estimates on the total number and value of troubled assets, how much remained with the Treasury, how much had been sold, and the profit or loss incurred from said sales (P.L. 110-343).

According to its mission statement, SIGTARP “is a federal law enforcement agency and is an independent watchdog protecting taxpayer dollars that fund TARP” (SIGTARP Annual Report – FY 2018). SIGTARP reported to Congress about its supervisory activities twice a year, audited various TARP programs, and conducted and reported on investigations of misconduct for TARP recipients (SIGTARP – Official Website). SIGTARP’s quarterly reports generally consisted of assessments of various components of the program, such as the CPP, as well auto and housing industry support programs (SIGTARP April 2010). SIGTARP reports included detailed recommendation sections, informed by the audits and investigations it conducted, and also provided detailed tables of the implementation status of each of these recommendations (SIGTARP July 2014).

The COP was another crucial component of TARP oversight. Created on the same day that TARP was passed, COP’s primary functions were to:

1.  Oversee Treasury’s actions;
2.  Assess the impact of spending to stabilize the economy;
3.  Evaluate market transparency;
4.  Ensure effective foreclosure mitigation efforts; and
5. Guarantee that Treasury’s actions are in the best interest of the American people (COP  – About Us).

In its approximately two and a half years of operation, the COP heard testimony from high-level officials and published 30 reports on the impacts of TARP initiatives. The Panel asked questions about the more immediate, stabilizing effects of TARP, as well as longer-term impacts, such as how TARP recipients were structuring their business and policies after receiving aid (COP December 2008). Its reports examined, for example, the government support to specific firms like GMAC and AIG and the effectiveness of the SCAP stress testing methodology (COP – Reports).

The COP was disbanded on April 3, 2011, after releasing its final report. In it, the Panel explained that, while TARP alone was not responsible for the economic recovery, “TARP quelled the immediate panic and helped to avert an even more severe crisis” (COP March 2011). This final report had few comments on SCAP and CAP.

CAP was intended to be an unlimited capital backstop for the financial sector (Clark, Kabaker, and Sachs 2020; UST PR, 02/10/2009). The program was estimated to have a total capacity of roughly $300 billion, which officials believed was sufficient to meet any potential needs (see Funding Source KDD).

Congress authorized a total of $700 billion of TARP funds that could be “outstanding at any one time.” However, Congress originally allowed only $250 billion to be outstanding; it required the president to return to Congress with a report asking to disburse the remaining funds, should they be necessary (P.L. 110-343)^FThe phrase “outstanding at any one time,” which implicitly allowed the government to re-use TARP funds, was removed with the passage of Dodd-Frank in July 2010.. This authority lasted until October 3, 2010, or two years after the date of the passage of EESA. After this point, Treasury was not able to make new commitments under TARP (TARP Two-Year Retrospective). The Dodd-Frank Wall Street Reform and Consumer Protection Act, passed on July 21, 2010, reduced this original commitment from $700 billion to $475 billion (P.L. 111-203). 

However, the unlimited nature of CAP meant there was some uncertainty around how much, if any, capital the government would need to provide. Fed and Treasury officials, in a paper published later, estimated that the program had a total capacity of roughly $300 billion. That sum included about $100 billion in appropriated but still uncommitted TARP funds, plus the roughly $200 billion in existing CPP preferred shares that the Treasury could convert into new CAP preferred shares if needed (Clark, Kabaker, and Sachs 2020).

In hindsight, those officials wrote, the level that the government had committed “was sobering—we were committing the government to buy an indeterminate amount of common stock in the banking system at a fixed price” (Clark, Kabaker, and Sachs 2020). However, even the most extreme scenarios under SCAP determined the capital buffer to be far less than the remaining amount of TARP funds available. TARP would have had sufficient funds to cover the entire $185 billion capital buffer that SCAP had identified if the 10 BHCs that failed the stress test had been unable to raise funds from other sources^FTotal disbursements under TARP were approximately $441.8 billion, which was far less than the initial $700 billion appropriated in the Emergency Economic Stabilization Act (TARP Tracker).. However, it was not known at the time how much TARP money would be available to finance the SCAP buffer, as many of the programs were still being developed and funds had been committed to other programs.

The US government assessed the capital needs of the banking system through the Supervisory Capital Assessment Program, a forward-looking stress test of the 19 largest BHCs under a set of stringent economic scenarios. Each of the BHCs had risk-weighted assets of more than $100 billion^FDespite this being one of the criteria for BHCs under SCAP, one of the firms—State Street—had only $69.6 billion in risk-weighted assets. See Table 1 for more information on SCAP BHCs.. These firms were picked because of their significance in the credit intermediation process,  the fact that they collectively held “two-thirds of the assets and more than one-half of the loans in the US banking system”, and in the case of State Street and Bank of New York Mellon, their significance as the two largest custody banks.  Teams of examiners, led by those at the Federal Reserve, “appl[ied] a consistent and systematic approach across the group to evaluate the projected loss and resource estimates submitted by the firms” (SCAP: Design and Implementation).

The SCAP was deliberately stringent. This allowed the Fed to “[counter] the risk that uncertainty itself exerts contractionary pressures on the banking system and the economy” (SCAP: Overview of Results). The goal of the tests was to ensure that, even in the event of a significant deterioration in the US economy, these critically important BHCs would remain adequately capitalized and able to lend at their normal levels (FRB PR, 05/06/2009). In order to assess these needs, federal regulators estimated losses on a variety of securities and loans for 2009 and 2010. It also calculated loan loss reserves at the end of 2010 at levels that the regulators believed would “captur[e] expected losses in 2011” (SCAP: Design and Implementation).

The Federal Reserve crafted two economic scenarios, the “baseline” and “more adverse” scenarios. In both cases, the Fed projected changes in real GDP, the civilian unemployment rate, and house prices over 2009–10. “The baseline scenario was intended to represent a consensus view about the depth and duration of the recession,” and was based on the average of economic projections published by the February releases of forecasts from three major economic forecasting surveys: Consensus Forecasts, the Blue Chip survey, and the Survey of Professional Forecasters (SCAP: Design and Implementation)^FHouse prices were measured by the fourth-quarter year-over-year change of the Case-Shiller 10-City Composite Index (SCAP: Design and Implementation – p. 6)..

The more adverse scenario did not average the consensus expectations of a variety of forecasters. Rather, it “was constructed from the historical track record of private forecasters as well as their current assessments of uncertainty”. The Fed focused specifically on “subjective probability assessments” from these organizations to gauge the probability that the aforementioned inputs (GDP growth, unemployment, house prices) would be even more affected than the general consensus used in the baseline scenario. Despite the dramatic differences from the baseline, the more adverse scenario was not a “worst-case” scenario, as the Fed argued that the conditions of the stress test ought to be “severe but plausible” (SCAP: Design and Implementation). See Figure 7 for the estimates associated with the baseline and more adverse scenarios.

The BHCs were instructed to estimate losses that were due to failure to pay obligations rather than those due to mark-to-market accounting (SCAP: Design and Implementation). This judgment was based on what some Treasury officials at the time called the theory of special bank relativity. This theory pushed back on the general standard of mark-to-market accounting by suggesting that banks exist through time and shouldn’t be judged based on how they stand during the lowest point in a cycle. Forcing the banks to do this, some argued, would “undermine a key reason [why] they are so important for supporting economic activity in the first place” (Clark, Kabaker, and Sachs 2020).

Each BHC was “asked to estimate their potential losses on loans, securities, and trading positions, as well as pre-provision net revenue (PPNR) and the resources available from the allowance for loan and lease losses”. BHCs with large enough trading accounts ($100 billion in assets or more) were required to provide additional loss estimates in the adverse scenario that factored in counterparty credit risk^FCounterparty credit risk was measured by assessing the risk of a counterparty defaulting, as well as potential credit valuation adjustments that would be made against exposures to counterparties who were more likely to default under the more adverse scenario.. BHCs calculated these loss estimates on their trading books and counterparty exposures using internal stress tests based on market prices on February 20, 2009 (SCAP: Design and Implementation).

Banks projected losses over two years. The Federal Bank of New York (FRBNY) provided loss ranges for 12 different types of loans and securities, such as first- and second-lien mortgages, commercial and industrial (C&I) loans, and credit card loans. Banks could provide alternative loss estimates from what FRBNY gave them, but they were then required to provide “strong supporting evidence, especially if they fell below the range minimum”. To ensure more accuracy, the BHCs were encouraged to provide additional information, such as loss projections on subcategories within the 12 that the Fed selected. The categories of loans were chosen to make the results easier to compare across firms and to compare with what the BHCs had filed with their federal regulator (SCAP: Design and Implementation).  The categories of loans, as well as some of the variables that the examiners used when conducting the evaluations, can be found in Figure 8.

Source: SCAP: Design and Implementation - pp. 12 - 13, 18.

Estimated loss rates were calculated using a variety of methods. For residential mortgages, for example, the agencies built default models based on information about individual and regional loan portfolios and mortgages. For other consumer loans and commercial loans, the agencies conducted regressions of historical default data against several macroeconomic variables, such as the unemployment rate. The regulators also had to address the heterogeneity across firms that would affect their performance in these tests. To this end, they collected additional firm-specific data, such as “past performance, portfolio composition, origination vintage, borrower characteristics, geographic distribution, international operations, and business mix” (SCAP: Design and Implementation). Cumulative loan-loss rates under the more adverse scenario were 9.1 percent, which was higher than the banking industry had ever experienced, even at the height of the Great Depression (SCAP: Overview of Results).  Figure 9 compares the loan-loss rates under the baseline and more adverse scenarios to the actual loss rates the industry experienced in 2009.

Figure 9: Two-Year Loss Rates for SCAP Scenarios and 2009 Actual Loss Rates (percent)

Sources: SCAP: Overview of Results – p. 5; GAO 10-861 – p. 17.

BHCs were also required to provide projections of their resources to absorb losses, including their pre-provision net revenues (PPNR) and their allowance for loan losses over the same horizon. In the case of SCAP, PPNR was defined as “the income after non-credit-related expenses that would flow into firms before they take provisions or other write-downs or losses” (SCAP: Design and Implementation; FRB: Allowance for Loan and Lease Losses). All told, more than 150 federal employees were involved in the creation, organization, and administration of the SCAP (SCAP: Design and Implementation).

When determining a BHC’s capital needs, the regulators looked at both the amount of Tier 1 capital and its composition^FTier 1 capital is composed of qualifying common stockholders’ equity, qualifying noncumulative perpetual preferred stock and senior perpetual preferred stock issued to the Treasury under TARP, certain minority interests and trust preferred securities (SCAP: Design and Implementation). . Tier 1 common equity, the first element of Tier 1 capital to absorb losses, was seen by the Fed as what should be “the dominant component” of Tier 1 capital (SCAP: Design and Implementation). Common equity “gives a BHC greater permanent loss absorption capacity and a greater ability to conserve resources under stress by changing the amount and timing of dividends and other distributions” (SCAP: Overview of Results).

The Fed considered two key questions when evaluating the capital needs of each of the BHCs under the more adverse scenario:

1. How much additional Tier 1 capital would a BHC need today to have a Tier 1 ratio of greater than 6 percent at the end of 2010?
2. How much additional Tier 1 common equity capital would a BHC need today to have a Tier 1 common equity ratio of greater than 4 percent at the end of 2010?

These capital thresholds were known as the “6-4 rule.” The regulators used them to determine the SCAP buffer—the amount of additional capital the 19 BHCs needed to remain well-capitalized even under the more adverse scenario. Nine firms already met the 6-4 rule. The other 10 firms needed a total of $185 billion in additional capital to meet the rule. Their SCAP buffer needs ranged from as high as $92.6 billion for Citigroup and as low as $2.3 billion for PNC. Most of the SCAP buffer needs were due to insufficient common equity. Tier 1 capital numbers for most banks were within acceptable limits under the more adverse scenario. Only two banks—GMAC LLC and Regions Financial Corp.—had insufficient Tier 1 capital. Due to the emphasis on common equity, the Fed concluded that the 10 firms that needed to raise additional equity “had capital structures that [were] too strongly tilted toward capital other than common equity” (SCAP: Overview of Results).

However, the $185 billion figure did not consider capital that firms had raised at the end of 2008, as well as other actions many of the firms had taken to pre-emptively shore up their balance sheets prior to the release of the results. The most noteworthy of these was Citigroup, which reduced its SCAP buffer by $87.1 billion through preferred stock exchanges and other actions^FCiti announced a plan to convert up to $27.5 billion of its existing preferred stock into common stock on February 27, 2009, as well as an agreement with the US government wherein Treasury would match this exchange at up to $25 billion of its own preferred stock. By the time the offers closed at the end of July 2009, Citi had raised approximately $58 billion in capital (Citi PR, 02/27/2009; Citi PR, 07/26/2009; SCAP: Overview of Results). Additional capital actions, characterized as those transactions that were completed or contracted as of the end of 2008, amounted to $29 billion (SCAP: Overview of Results)..  Other banks increased their equity in this period by lesser amounts. By the time the results had been released, the original $185 billion, which was based on the firms’ balance sheets as of December 31, 2008, had been reduced to $74.6 billion (SCAP: Overview of Results). See Figure 10 for the adjusted SCAP requirements of the 10 firms.

BHCs that had SCAP buffer needs were required to submit capital plans to their primary regulator within 30 days. BHCs would have six months to implement these plans and were “encouraged to design capital plans that, wherever possible, actively seek to raise new capital from private sources” (SCAP: Overview of Results). 

Figure 10: SCAP Capital Shortfalls ($ billions)

Source: SCAP: Overview of Result.

The 10 BHCs managed to raise $76.6 billion in Tier 1 common equity by November 9, the end of the six-month deadline (FRB PR, 11/09/2009). The only institution that needed additional capital and failed to raise it from private investors was GMAC. It ultimately used the Automotive Industry Financial Program (AIFP) to help meet the SCAP requirements (UST PR, 11/09/2009).

As a result, no SCAP BHCs drew on the CAP for government capital. See Figure 11 for a breakdown of the Tier 1 ratios of the SCAP BHCs at the end of 2009.

Figure 11: Change in Tier 1 Capital Ratios for SCAP Banks, December 31, 2008 to December 31, 2009

Source: GAO 10-861.

BHCs that had SCAP buffer needs were required to submit capital plans to their primary regulator within 30 days. BHCs would have six months to implement these plans and were “encouraged to design capital plans that, wherever possible, actively seek to raise new capital from private sources”. These could include “restructuring current capital instruments, sales of assets, and restrictions on dividends and stock repurchases” (SCAP: Overview of Results). The plans normally had the following elements:

1. A description of the specific actions the institution would take to increase the level or quality of its capital consistent with the results of the SCAP.
2. A list of steps to address any weaknesses for the BHC’s internal evaluations of its capital needs and capital planning.
3. An outline of the steps the BHC was planning to take to repay all government investments received, as well as to reduce reliance on guaranteed debt issued under the FDIC guarantee program (FSOB Report, 06/30/2009).

BHCs with SCAP buffer needs were able to apply to the CAP but defer issuing convertible preferred stock to the government via the facility for up to six months while they searched for private capital (CAP: White Paper).

Qualifying financial institutions (QFIs) that did not participate in SCAP were eligible to apply voluntarily. They had until May 25, 2009, to apply, though any company applying to become a bank holding company had to do so by January 15, 2009, to be eligible for CAP (CAP: FAQ).

QFIs included any publicly traded US bank or savings association not controlled by a bank holding company or a savings and loan company, top-tier US bank holding companies, and top-tier US savings and loan companies that “[engaged] solely and predominately in activities that are permitted for financial holding companies under relevant law”. Banks or BHCs that were foreign-owned, or subsidiaries of foreign companies, were not allowed to participate. While the term sheet stated that CAP term sheets for privately held banks, S-corporations, and mutual banks were “expected to be made available,” they never were (CAP: Term Sheet).

An institution that was accepted into the program could issue mandatorily convertible preferred stock in an amount no less than 1% of its risk-weighted assets, up to a maximum of 2%.  It could submit a request for capital in excess of this amount to its federal banking regulator, who would then consult with Treasury. If this was successful, the institution would be categorized as needing “exceptional assistance” and would potentially be subject to additional terms and conditions (CAP: Term Sheet). 

Under CAP, the Treasury would purchase mandatorily convertible preferred shares, henceforth referred to as Convertible Preferred, in QFIs. Cash received from the issuance of these shares could be used to redeem shares under other government investment programs, such as the Capital Purchase Program or the Targeted Investment Program (CAP: Term Sheet). Also, government investments made under these programs could be converted into Convertible Preferred. While the Treasury would make all investments under CAP, the investments themselves would be managed by a trust set up “to protect and create value for the taxpayer as a shareholder over time” (CAP: White Paper). 

Stockholder Consent
After issuing the Convertible Preferred, QFIs that did not have enough common stock authorized to reserve for the conversion and exercise of the warrants were required to call a shareholder vote to authorize an increase of the number of common shares so that the conversion and exercise could take place if needed. The terms of the capital could change in several ways (see below) if this authorization was not obtained in a timely manner or at all (CAP: Term Sheet). 

Conversion
The Convertible Preferred had both mandatory and optional conversion clauses as part of its structure. The optional conversion gave the issuer of the equity the right to convert any amount of the preferred equity to common stock at any time, subject to regulatory approval. The holder of the Convertible Preferred, in this case the Trust created by Treasury, could also convert the shares “upon specified corporate events, including certain sales, mergers or changes of control of the QFI”. If a QFI held the Convertible Preferred for longer than seven years, the shares would automatically convert to common stock (CAP: Term Sheet). The price for conversion was unspecified in the initial Financial Stability Plan. The Treasury said then that the price would be “set at a modest discount from the prevailing level of the institution’s stock price as of February 9, 2009” (UST PR, 02/10/2009). The term sheet further explained that the price would be set at 90 percent of the average of the closing price of the QFI’s common stock over the 20 trading day period ending on February 9, 2009. This price, however, would be reduced by 15 percent every six months after the initial injection if the aforementioned stockholder vote to increase the number of common shares did not happen, to a maximum of 45 percent. Conversion would be accompanied by the repayment of any accrued or unpaid dividends, either with cash or common stock (CAP: Term Sheet). 

Mandatory Sale
After the mandatory conversion date, the Treasury was required to “make reasonable efforts” to sell at least 20 percent of its outstanding common stock annually. At any point after either a mandatory or optional conversion, the QFI had the option to repurchase the common stock held by the government by issuing new common stock or drawing on retained earnings. The price of repurchase was the greater of: 

1. The conversion price.
2. The market price, calculated based on the average closing price during the 20 trading day period following the notice of repurchase (CAP: Term Sheet).

The Treasury made it clear that, despite the potential for the government to acquire a large interest in an institution, “US government ownership is not an objective of the CAP” (FSOB Report, 03/31/2009). The mandatory sale clause in the term sheet reflected the intention of the Treasury to keep the period of government investment “as temporary as possible” (CAP: White Paper).

Dividends
The Convertible Preferred paid cumulative dividends of 9 percent, which compounded quarterly. This rate would increase to, and remain at, 20 percent six months after the original issue date if stockholder consent was not given. So long as any Convertible Preferred or Treasury-owned common stock remained outstanding, the dividends on any common stock not held by Treasury could not be greater than $0.01 per share. An additional “dividend stopper” was included in the term sheet. The stopper prevented QFIs from declaring or paying dividends on any shares, whether they were pari passu preferred, junior preferred, or common, if any dividends on the Convertible Preferred remained outstanding (CAP: Term Sheet).  

Repurchases
Repurchases of other securities, such as junior preferred, pari passu preferred shares, or common shares, were affected by the “dividend stopper” and thus prohibited if any dividends on the Convertible Preferred remained outstanding (CAP: Term Sheet).

Warrants
The Convertible Preferred also included 10-year warrants. These warrants, which were immediately exercisable, allowed the Treasury to purchase common stock equal to 20 percent of the amount of Convertible Preferred on the day of initial investment. The price to exercise the warrants was the same as the conversion price. The exercise price would also be reduced in 15 percent increments every six months after the initial injection, to a maximum of 45 percent if shareholder consent is not gained. While the Treasury would gain a larger voting interest in the QFI after exercising the warrants, it pledged not to exercise any voting power with respect to these shares (CAP: Term Sheet). 

Redemption
QFIs were able to redeem the Convertible Preferred “at any time solely with the proceeds of one or more issuances of common stock for cash,” so long as they (1) obtained approval from their federal banking regulator, and (2) the proceeds of the equity issuances equaled at least 25 percent of the issue price of the Convertible Preferred. Redemption was done either at par value (plus accrued and unpaid dividends) if done within the first two years, and at the greater of par value plus accrued and unpaid dividends and the as-converted value if done after two years (CAP: Term Sheet).

This meant that QFIs that issued Convertible Preferred had an implicit time limit on redeeming the shares because, after two years, the cost of redemption could increase dramatically if the QFI’s stock price had increased. The as-converted value of the preferred investment would become greater than the par value of the investment, and the QFI would have to choose between raising enough common equity to satisfy this increased redemption price, continue to pay 9 percent in dividends on the preferred, or convert it to common stock. As noted, repurchases of converted stock could be done by issuing new common equity or drawing on retained earnings (CAP: Term Sheet).   

The Treasury expected institutions with SCAP capital buffer needs to design their own private-sector solution to increase their capital, with CAP funding available as a last resort. These solutions could include a mix of debt and preferred equity conversions, asset sales, and other measures. However, the Treasury did not provide any guidelines for these solutions. In a joint statement the day before they released the SCAP results, the Fed, Treasury, FDIC, and OCC said that “restructuring current capital instruments” was one of the ways that SCAP BHCs could meet their capital needs (FRB PR, 05/06/2009). Treasury officials internally called the program akin to “open bank resolution.” In a typical open bank resolution, existing common equity holders see the value of their shares diluted, in whole or in part; and existing preferred equity and debt investors accept losses or convert their holdings into common equity.

Converting junior debtholders to common equity holders through debt-for-equity exchanges and other tools “was the preferred path before [the BHCs] sought government capital”. Later, Treasury officials said they meant to make clear to junior debtholders that they “should expect to bear losses” (Clark, Kabaker, and Sachs 2020). 

The SCAP BHCs already had about $300 billion in debt outstanding through these instruments, which suggested to regulators that they could meet their SCAP buffer needs without finding new equity investors (Clark, Kabaker, and Sachs 2020). Figure 12 below illustrates the capital actions taken by SCAP banks.

Figure 12: Description of Capital Actions by SCAP Bank Holding Companies (billions)^FNine BHCs—American Express Company, BB&T Corporation, Bank of New York Mellon Corporation, Capital One Financial Corporation, Goldman Sachs Group, Inc., JPMorgan Chase & Co., MetLife, Inc., State Street Corporation, and US Bancorp—were not required to raise SCAP capital because they had sufficient capital to withstand a worse-than-expected economic downturn through the end of 2010 and continue to meet the SCAP capital buffer targets.

Section 111 of EESA outlines executive compensation restrictions and corporate governance standards (P.L. 110-343). Specifically, Section 111(b) describes these standards as they apply to direct purchases or investments in financial institutions. Section 111(b)(2) specified that there would be:

1. “limits on compensation that exclude incentives for senior executive officers of a financial institution to take unnecessary and excessive risks that threaten the value of the financial institution during the period that the Secretary holds an equity or debt position in the financial institution;
2. “a provision for the recovery by the financial institution of any bonus or incentive compensation paid to a senior executive officer based on statements of earnings, gains, or other criteria that are later proven to be materially inaccurate; and
3. “a prohibition on the financial institution making any golden parachute payment to its senior executive officer during the period that the Secretary holds an equity or debt position in the financial institution” (P.L. 110-343).

Senior executive officers (SEOs) were initially defined by EESA as the top five highest-paid executives of companies that took TARP money (P.L. 110-343). However, Treasury passed several interim final rules that provided additional guidance on executive compensation post-EESA, as well (TARP Executive Compensation Rules and Guidance). 

Treasury released additional guidance in an announcement on February 4, 2009, that distinguished between institutions that required “exceptional assistance” and institutions participating in “generally available capital access program[s]” (UST PR, 02/04/2009). Treasury specified that institutions that received exceptional assistance included AIG, Citigroup, Bank of America, and the two automakers that received TARP support, GM and Chrysler (UST PR, 06/10/2009). Institutions that were accepted into the CAP did not automatically fall under this category, but would if they issued Convertible Preferred in excess of the 2 percent limit (CAP: Term Sheet). 

Institutions that required exceptional assistance were subject to stricter requirements. Total compensation was restricted to a maximum of $500,000, with any amount in excess required to be in the form of restricted stock that vested only after the government had been repaid. This was to ensure that the incentives of executives at these institutions were “aligned with both the long-term interests of shareholders as well as minimizing the costs to taxpayers”. Golden parachutes, which were defined as large severance agreements for SEOs, were completely prohibited for the 10 highest-compensated employees, and capped at one year’s compensation instead of three years for the next 25 highest-compensated (UST PR, 02/04/2009; UST Report, 10/20/2008). 

For institutions that participated in generally available capital programs, SEO total yearly compensation was limited to $500,000, though this limit could be waived with “full public disclosure and [a] shareholder vote”. Golden parachute payments for the top five highest-paid employees of these institutions were restricted to one year’s compensation, instead of three (UST PR, 02/04/2009). 

For institutions participating in any of these capital programs, clawback provisions—which enabled the QFI to retake bonuses and incentive payments of SEOs that had earned them by knowingly engaging in deceptive practices—were expanded from the top five SEOs to the next 20 highest-paid employees. The new guidance also required participating institutions to more aggressively police any luxury expenditures and required the CEO’s sign-off on any expenditures that “could be viewed as excessive or luxury items” (UST PR, 02/04/2009). These guidelines, however, did not apply retroactively (UST Report, 06/10/2009).

Treasury published its last Interim Final Rule on June 10, 2009. The final rule implemented and modified much of the expanded restrictions outlined in ARRA and the February 2009, guidance. Treasury further limited bonuses and incentive compensation paid to SEOs, as well as some of the most highly compensated employees, to one third of total compensation, with the number of employees affected by this limit increasing based on the amount of aid given^FInstitutions that received over $500 million in assistance had their top five highest-paid executives, as well as the 20 next highest-paid affected by the one-third of total compensation limit..  The $500,000 limit specified in the February guidance was amended to “link compensation to long-term firm value” by allowing additional compensation over $500,000, provided that it was in the form of long-form, restricted stock. The stock-based compensation was still limited to one third of total compensation. Additionally, golden parachute payment restrictions were broadened to include any payments that were made as a result of a change in control of the company. Institutions were required to exercise their clawback provisions if payments subject to them were found to be based on inaccurate criteria or data, whereas they were merely given the option to do so in the February guidance (UST PR, 06/10/2009). The guidance also mandated that TARP recipients hold an annual shareholder vote to approve executive compensation packages.

The most notable change in the final rule was the creation of a Special Master for TARP Executive Compensation^FA number of the provisions laid out in the June 10 Interim Final Rule, such as the one-thirds limit on total compensation, as well as the number of SEOs and employees affected by executive compensation restrictions based on total TARP aid received, were originally passed via Section 111 of the ARRA. . The Special Master was responsible for reviewing compensation plans at firms receiving “exceptional” assistance to make sure they did so in a way that “maximize(s) long-term shareholder value and protect(s) taxpayer interests”^FIn addition to approving compensation agreements, the Special Master was also responsible for reviewing compensation that was paid prior to the passage of ARRA, as well as negotiating reimbursements for these where relevant (UST PR, 06/10/2009). .  This authority went far beyond the threshold in the earlier guidance. The Special Master was able to review and approve (or disapprove) the compensation structures of SEOs and the next 100 most highly paid employees, as well as executive officers that were not among the most highly paid employees and thus, not subject to the typical bonus and incentive payment restrictions (UST PR, 06/10/2009). Total annual compensation that was less than or equal to $500,000, not including long-term restricted stock, would automatically be approved by the Special Master^FEquity-based compensation was included in this limit by using the fair market value on the day of compensation. Equity-based compensation for prior years would not be included (UST Report, 06/10/2009 - p. 29)..  Some of these responsibilities were originally given to the Secretary of the Treasury as part of the ARRA (P.L. 111-5).

While the February guidelines did not apply retroactively, the Special Master had the power to evaluate payments, bonuses, or compensation made by any TARP recipient before February 17, 2009, “to determine whether any such payments were inconsistent with the purposes of Section 111 of EESA” (UST Report, 06/10/2009). If the Special Master found any issues, firms could negotiate reimbursements to the US government.

If dividends on the Convertible Preferred stock were not paid for six quarters, consecutive or not, then Treasury would have the ability to elect two directors to the board. This right ended after dividends had been paid for four consecutive quarters (CAP: Term Sheet). Treasury would exercise the voting rights under the Convertible Preferred only in the cases of: 

1. an authorization or issuance of shares senior to the Convertible Preferred 
2. any amendment to the rights of the Convertible Preferred, or
3. any merger, exchange or similar transaction which would adversely affect the rights of the Convertible Preferred (CAP: Term Sheet). 

The Treasury retained these voting rights upon conversion to common stock and would receive standard voting rights associated with the QFI’s common stock. In the event of conversion, Treasury would also publish a set of guidelines that explained how it would use these voting rights “prior to closing any transactions” (CAP: Term Sheet).