As required by section 38(k) of the Federal Deposit Insurance Act (FDI Act), the Office of Inspector General (OIG) conducted a material loss¹ review of the failure of Franklin Bank, S.S.B. (Franklin), Houston, Texas. On November 7, 2008, the Texas Department of Savings and Mortgage Lending (DSML) closed the institution and named the FDIC as receiver. On November 28, 2008, the FDIC notified the OIG that Franklin’s total assets at closing were $4.9 billion, and the material loss to the Deposit Insurance Fund (DIF) was $1.5 billion. As of May 31, 2009, the estimated loss to the DIF decreased to $1.4 billion.

When the DIF incurs a material loss with respect to an insured depository institution for which the FDIC is appointed receiver, the FDI Act states that the Inspector General of the appropriate federal banking agency shall make a written report to that agency which reviews the agency’s supervision of the institution, including the agency’s implementation of FDI Act section 38, Prompt Corrective Action (PCA); ascertains why the institution’s problems resulted in a material loss to the DIF; and makes recommendations to prevent future losses.

The audit objectives were to: (1) determine the causes of the financial institution’s failure and resulting material loss to the DIF and (2) evaluate the FDIC’s supervision² of the institution, including implementation of the PCA provisions of section 38 of the FDI Act. Appendix 1 contains details on our objectives, scope, and methodology;

¹ As defined by section 38(k)(2)(B) of the FDI Act, a loss is material if it exceeds the greater of $25 million or 2 percent of an institution’s total assets at the time the FDIC was appointed receiver.
² The FDIC’s supervision program promotes the safety and soundness of FDIC-supervised institutions, protects consumers’ rights, and promotes community investment initiatives by FDIC-supervised insured depository institutions. The FDIC’s Division of Supervision and Consumer Protection (DSC) (1) performs examinations of FDIC-supervised institutions to assess their overall financial condition, management policies and practices, including internal control systems; and compliance with applicable laws and regulations; and (2) issues related guidance to institutions and examiners.

This report presents the FDIC OIG’s analysis of Franklin’s failure and the FDIC’s efforts to ensure Franklin’s management operated the bank in a safe and sound manner. The FDIC OIG plans to issue a series of summary reports on our observations on the major causes, trends, and common characteristics of financial institution failures resulting in a material loss to the DIF. Recommendations in the summary reports will address the FDIC’s supervision of the institutions, including implementation of the PCA provisions of section 38.

BACKGROUND

• had 46 full-service branches in Texas, commercial loan offices in 6 states, and 42 mortgage-banking offices in 22 states;
• had a 2-tier holding company structure, 2 wholly-owned subsidiaries, and 4 non-bank affiliates;
• provided traditional banking activities within its marketplace; and
• specialized in residential and commercial real estate (CRE) lending, with concentrations in 1-4 family residential and acquisition, development, and construction (ADC) loans. In addition, the bank was highly dependent on Federal Home Loan Bank (FHLB) borrowings and brokered deposits for its funding.

REASON FOR FAILURE AND MATERIAL LOSS

Overall, Franklin failed due to bank management’s high-risk business strategy. The strategy focused on asset growth concentrated in 1-4 family residential and ADC loans funded with wholesale funding, including potentially high-cost and volatile deposits and borrowings. Coupled with weak risk management practices and controls, this business strategy left the bank unprepared and unable to effectively manage operations in a declining economic environment. Franklin’s asset quality deteriorated significantly as the real estate market and economy slowed. For example, adverse loan classifications increased from $178.5 million reported in the October 2007 ROE to $783.7 million reported in the July 2008 ROE. Franklin’s adverse classifications, including loan losses, resulted primarily from its portfolio of 1-4 family residential loans and ADC loans. As adverse loan classifications increased, earnings eroded, liquidity became strained, and Franklin’s capital became increasingly deficient.

High-Risk Business Strategy

Franklin’s management employed a high-risk business strategy in which it concentrated assets in 1-4 family residential and ADC loans and funded its loan growth with wholesale funding, including higher-cost and volatile deposits and borrowings, without sufficient mitigating controls. Franklin’s loan portfolio grew over 50 percent between December 2004 and December 2006, peaking at almost $4.7 billion. This high-risk strategy was a significant contributing factor to the failure of Franklin. In particular, the following concerns were noted.

1-4 Family Residential and ADC Concentrations. Franklin’s asset quality problems were exacerbated by its emphasis in high-growth markets and concentrations in 1-4 family residential loans (that contained a significant volume of nontraditional and subprime mortgages) and ADC loans, which, as of September 2008 totaled 937 percent and 736 percent of total capital, respectively. In particular, Franklin’s management allowed significant loan concentrations to exist without adequate risk identification, measurement, monitoring, and controls. As shown in Table 2, the bank’s concentration in 1-4 family residential loans began to grow significantly in 2002 and remained a major product segment into 2008.

• 1-4 Family Residential Loans: From December 2003 through September 2008, Franklin maintained a significant concentration in 1-4 family residential loans, which it originated through a retail network of 55 loan production offices in 24 states as well as purchased through wholesale origination channels. Franklin’s concentrations in such loans peaked at about 1,077 percent of total capital as of December 2004 but remained high throughout the almost 6-year period. Although 1-4 family residential loans are typically considered a less-risky type loan, there were factors in Franklin’s 1-4 family residential loan

portfolio that increased that risk. Specifically, within its 1-4 family residential loan portfolio, Franklin originated, purchased, and sold an array of mortgage products that included nontraditional and subprime mortgages.³ The bank’s nontraditional mortgage lending program included the following underwriting characteristics:
• interest-only loans;
• no documentation, limited documentation, and stated income loans;
• payment option adjustable rate mortgages;
• simultaneous second-lien loans (not held by the bank);
• high combined loan-to-value ratios, high combined debt-to-income ratios, and loans to borrowers with low credit scores;
• purchased loan pools serviced by others; and
• multiple risk layers.

As of July 2008, 82 percent of the 1-4 family residential loans held on the bank’s books had been originated under reduced documentation or stated income loan programs. In addition, 16 percent of the residential loans were considered subprime loans, and 72 percent of the bank’s residential loans were purchased from and serviced by others. The purchased loans were typically collateralized with first-lien positions; however, many of the homes that collateralized the bank’s loans also had second liens in place. As a result, the borrowers had limited equity positions in the homes they purchased or refinanced. Further, these loans were concentrated in markets that had experienced a significant level of appreciation and then deterioration.

Due to the collapse of the subprime mortgage market and the tightening of the mortgage credit market, bank management halted the bank’s nontraditional mortgage and subprime operations and, in the first quarter of 2007, began to limit the types of 1-4 family residential loan products that it originated to only conforming high-quality loans. However, Franklin’s curtailment of its nontraditional mortgage and subprime operations was not sufficient to improve the overall performance of its loan portfolio.

• ADC Loans: In 2005, Franklin management began to change its loan mix by increasing its emphasis in ADC loans. Although never quite exceeding the concentrations maintained in its 1-4 family residential loan portfolio, Franklin’s ADC loan concentration reached 736 percent of total capital, as of September 2008. Additionally, as of March 2006, Franklin began to purchase unsecured ADC loan participations, which added an additional element of risk to its loan portfolio. It is important to note that, according to the

³ According to DSC, Franklin did not have a subprime lending program as defined by interagency guidance. Nonetheless, Franklin had a significant volume of loans with subprime characteristics. Based on data provided within the October 2007 and July 2008 ROEs, subprime loans represented approximately 67 percent and 171 percent, respectively, of Tier 1 Capital.

FDIC, Franklin’s ADC loan growth in 2007 and 2008 was restricted primarily to the bank’s funding of existing loan commitments.

An additional element of risk in Franklin’s ADC loan portfolio was that Franklin underwrote ADC loans with corresponding interest reserve loan provisions, which allowed borrowers to fund their interest payments through a borrowing line with the bank. Although the use of interest reserves is common in certain forms of ADC lending, based on our review of the bank’s loan policies, as retained within the examination workpapers, the bank did not have interest reserve loan policies or standards for the acceptability of, and limits on, the use of interest reserves. Furthermore, Franklin had not established a system to measure, monitor, and control the volume of loans underwritten with interest reserves. Franklin reported to the FDIC that as of May 2008 approximately $410 million (30 percent) of the bank’s ADC loan portfolio (based on dollar volume) was comprised of loans with interest reserve funding provisions. Based on our review of the 2007 and 2008 ROEs and related guidance, we identified instances where Franklin may have inappropriately used interest reserve loans to (1) bring delinquent loans current; (2) modify loans on projects that were experiencing construction delays, funding shortfalls, and deteriorating collateral values/positions; and (3) fund raw land loans.

The FDIC has taken recent action addressing the issue of interest reserves. Specifically, on March 17, 2008, the FDIC issued a Financial Institution Letter (FIL) titled, Managing Commercial Real Estate Concentrations in a Challenging Environment (2008 CRE FIL), which, in part, articulated the FDIC’s concern about the use of interest reserves for ADC loans. The guidance stated that the FDIC has noted the inappropriate use of interest reserves when the underlying real estate projects are not performing as expected.

Volatile Wholesale Funding Sources. Franklin’s management employed a funding structure that centered on high-cost volatile funds to fund its growth, which we believe was a significant contributing factor leading to the failure of the institution. The bank’s funding structure relied on wholesale funding sources, including FHLB borrowings, brokered deposits, time deposits of $100,000 or greater, and high-rate core deposits to fund asset growth. As stated in the DSC Risk Management Manual of Examination Policies (Examination Manual), a heavy reliance on potentially volatile liabilities to fund asset growth is a risky business strategy because the availability of and access to these funds may be limited in the event of deteriorating financial or economic conditions, and assets may need to be sold at a loss in order to fund deposit withdrawals and other liquidity needs. Management did not establish policies or controls that adequately limited or mitigated the level of risk related to these activities.

A bank’s net non-core dependency ratio indicates the degree to which the bank is relying on non-core/volatile liabilities to fund long-term earning assets. Generally, a lower ratio reflects less risk exposure, whereas higher ratios indicate greater risk exposure and a reliance on funding sources that may not be available in times of

Based on the July 2008 ROE, the bank’s FHLB borrowing lines were restricted (and a hold was placed on the bank’s FHLB deposits), its securities portfolio was fully pledged, and other borrowing lines at correspondent banks were cancelled. In addition, the FDIC notified Franklin’s board of directors (BOD) of the bank’s change in PCA category to significantly undercapitalized in October 2008, subjecting the bank to brokered deposit and deposit rate restrictions. As indicated in Table 3, brokered deposits were Franklin’s primary funding source in September 2008, and these restrictions impacted the institution’s liquidity. Franklin was unable to raise additional capital or sell off its assets without incurring significant losses, resulting in the bank’s inability to meet liquidity needs and its ultimate failure.

Weak Risk Management Practices

Franklin’s BOD allowed bank management to pursue a high-risk business strategy without adequate risk management practices and controls. In addition, management failed to effectively implement audit and examination recommendations or to ensure that, as the bank grew, the sophistication of the bank’s risk identification and monitoring systems also expanded to effectively identify, measure, monitor, and control bank operations and risks. Franklin’s management did not ensure the accuracy of financial reporting and soundness

Internal Audit. In the ROEs from September 2003 through July 2008, the FDIC identified weaknesses in the structure and independence of Franklin’s internal audit program. The FDIC also noted that the scope and frequency of internal audit coverage was not fully adequate. In the October 2006 ROE, the FDIC cited Franklin’s contravention of the Interagency Policy Statement on the Internal Audit Function and Its Outsourcing, dated December 22, 1997. The policy statement identifies key characteristics and sound practices for the internal audit function and management of internal audit outsourcing arrangements. In the July 2008 ROE, the FDIC noted that one Franklin vice president, who was not independent of management, was involved in developing risk assessments, testing programs, and selecting audit samples. The lack of independence was a contravention of the policy statement.

Due Diligence. Franklin’s BOD did not implement an adequate due diligence process for purchased pools of 1-4 family residential loans. Specifically, it became apparent upon our review of the ROEs and discussions with FDIC examiners that Franklin purchased such loan pools without a complete understanding of what is was purchasing. Most notable, Franklin was not aware that loans it purchased contained second-lien positions that, in hindsight, made the loans far less attractive and valuable. In our opinion, the lack of adequate due diligence on these loan pools indicates that Franklin’s BOD did not ensure that as the bank grew, the sophistication of the bank’s risk identification and monitoring systems expanded to effectively identify, measure, monitor, and control bank operations and risk.

Internal Control and Financial Reporting. Franklin’s management did not ensure the accuracy of financial reporting and soundness of related accounting controls. As a result, management was unable to ensure the timely and accurate reporting of the bank’s financial condition, and the bank’s financial deterioration was masked to a certain degree. As the result of a “whistleblower” complaint, Franklin’s BOD arranged for an independent investigation of the bank to be conducted by Baker Botts, Limited Liability Partners, during the first and second quarters of 2008. The investigation revealed significant accounting errors, inappropriate accounting entries, a lack of internal controls, and significant questions regarding the competency of management.

In the July 2008 ROE, the FDIC noted that management and the BOD did not provide for internal controls and information systems that would ensure timely and accurate financial reporting. According to the FDIC, Franklin’s management disclosed that financial reporting since December 2006 could not be relied on. The FDIC also noted that management made multiple amendments to the September 2007 through March 2008 Reports of Condition and Income (Call Report). These amendments were the result of major accounting and internal

Allowance for Loan and Lease Losses (ALLL) Methodology. Franklin’s management did not establish a sufficient ALLL or an adequate ALLL methodology. Specifically, management did not develop an ALLL methodology that fully complied with the Interagency Policy Statement on the Allowance for Loan and Lease Losses (FIL-105- 2006), dated December 13, 2006. According to FIL-105-2006, each institution must analyze the collectibility of its loans and maintain an ALLL at a level that is appropriate and determined to be in accordance with Generally Accepted Accounting Principles (GAAP).⁴ An appropriate ALLL covers estimated loan losses on individually evaluated loans that are determined to be impaired as well as estimated loan losses inherent in the remainder of the loan and lease portfolio. As previously discussed, Franklin’s adverse loan classifications increased significantly between 2007 and 2008. An effective loan review system and controls (including loan classification and credit grading system) helps ensure asset quality problems are identified and an appropriate ALLL is established.

Beginning with the September 2004 examination, the FDIC repeatedly reported concerns with the bank’s ALLL policies and/or methodology for calculating the ALLL, including the need for management to consider and document adjusting qualitative factors (such as industry, geographic, and economic factors) to the industry’s loss rates, and to implement and document a methodology for measuring loans for impairment. In the October 2007 ROE, the FDIC reported that Franklin did not provide documented support that showed how the ADC loan portfolio’s historical loss rates had been determined. Further, the FDIC stated that Franklin did not consider the impact of current environmental factors in its analysis despite the rapidly deteriorating economic conditions in the bank’s primary markets. As Franklin’s assets deteriorated, it became apparent that its ALLL was insufficient to absorb loan losses.

Historically, as shown in Table 4, which follows, from 2002 through 2006, Franklin maintained the bank’s ratio of ALLL to total loans and leases at levels that were consistently well below its peer group average – ranging from the 3rd to 10th percentile. From 2004 until the bank closed, Franklin also maintained its capital levels below peer.

⁴ Interagency Policy Statement on the Allowance for Loan and Lease Losses (FIL-105-2006), dated December 13, 2006, reiterates key concepts and requirements pertaining to the ALLL included in GAAP and existing supervisory guidance. In addition, the policy describes the nature and purpose of the ALLL; the responsibilities of BODs, management, and examiners; factors to be considered in the estimation of the ALLL; and the objectives and elements of an effective loan review system, including a sound loan grading system.

Implementation of Examiner Recommendations. Franklin management did not effectively implement certain recommendations that were repeatedly made in the FDIC’s ROEs. Such recommendations included: (1) identification and monitoring of loan concentrations, (2) establishment of liquidity risk limits and contingency liquidity plans (CLP),5 and (3) enhancement of the internal audit function. In addition, management did not implement corrective actions in a timely manner to adequately address risk management control deficiencies identified by the FDIC in relation to ADC concentrations, internal auditing, accounting, and financial reporting.

Other Matters. Additionally, in our opinion, Franklin’s management did not implement a systematic assessment of the economic environment appropriate for the bank’s size, complexity, and risk profile, nor did Franklin implement an adequate stress testing model to identify, measure, monitor, and control risk. With respect to the assessment of the economic environment, Franklin did not perform a systematic economic review that utilized key market indicators and was tied to specific strategic action plans in case of deteriorating market conditions. Regarding stress testing, in the October 2007 ROE, the FDIC recommended that Franklin perform a portfolio-level stress test or sensitivity analysis to quantify the impact of changing economic conditions on asset quality, earnings, and capital. In response, Franklin management agreed to implement stress testing by products and geographies. However, Franklin reviewed loans individually and did not implement a portfolio-level stress test, as recommended by the FDIC.

⁵ The FDIC uses the terms CLP, liquidity contingency plan, and contingency funding plan interchangeably. For purposes of this report, we use CLP.

ASSESSMENT OF FDIC SUPERVISION

Over the life of Franklin, the FDIC provided supervisory oversight in many areas, including risk management examinations, visitations, and offsite monitoring. Although such supervision was extensive, we concluded that the FDIC could have performed additional analysis, exercised greater supervisory concern, and taken additional action to help prevent the bank’s failure and/or to mitigate the potential level of losses incurred.

Historical Snapshot of FDIC Supervision

The FDIC and DSML performed timely joint safety and soundness examinations of Franklin, conducting six examinations beginning September 2003 through July 2008. Franklin’s composite ratings remained at 2 until the October 2007 ROE when the bank’s composite rating was downgraded to 3, indicating increasing risk. As a result of the July 2008 ROE, Franklin’s composite rating was downgraded to 5, indicating extremely unsafe and unsound practices or conditions, critically deficient performance, and inadequate risk management practices.

In addition to providing composite and component CAMELS ratings for each ROE, the FDIC took other supervisory actions. In particular, within the ROEs, the FDIC made many specific recommendations to Franklin related to areas of its operations where improvements were appropriate. These areas included Franklin’s identification and monitoring of loan concentrations, establishment of liquidity risk limits and CLPs, and enhancement of the internal audit function. (A more complete mention of these areas is in Appendix 3, which provides an overview of examiner comments and recommendations regarding management, asset quality, and liquidity.) Another supervisory action included an August 2007 visitation in which the FDIC performed a targeted review of the bank’s mortgage banking operations to assess the potential impact on Franklin from the secondary mortgage market liquidity crisis.

Further, to address examiner concerns documented in the October 2007 ROE, including apparent violations of laws and regulations, inadequate risk management and internal controls, and other safety and soundness issues, the FDIC and DSML requested Franklin to adopt a Bank Board Resolution (BBR), which the bank’s BOD adopted on March 31, 2008. Among its provisions, the BBR contained a provision that addressed the bank’s management, stating:

Within 30 days, the Bank, with the Board’s approval, will have employed and retained an experienced and qualified Chief Credit Officer and an experienced and qualified senior executive responsible for establishing and directing an internal loan review program.¹

¹ According to the July 2008 ROE, a Chief Credit Officer was elected on December 19, 2007, and a Chief Risk Officer was elected on February 12, 2008, which fulfilled this BBR requirement.

Additionally, Dallas Regional Office officials stated that subsequent to the October 2007 examination, they had provided a continual on-site presence at Franklin and were actively investigating accounting issues.

Further, after downgrading Franklin to a composite 5 on September 5, 2008, the FDIC performed daily liquidity monitoring, which ultimately led to the decision that Franklin should be closed. The FDIC’s final supervisory action was the issuance of a Cease and Desist Order (C&D) on November 4, 2008, 3 days before the bank was closed.

OIG Assessment of FDIC Supervision

Based on our review, we concluded that the FDIC could have performed additional analysis, exercised greater supervisory concern, and taken additional action to help prevent the bank’s failure and/or to mitigate the potential level of losses incurred. Specifically, in the 2006 ROE, the FDIC could have better identified and analyzed risk to ensure that Franklin established and appropriately implemented controls and risk limitation and mitigation strategies. For example, the FDIC did not clearly identify in the 2006 ROE the risk posed by Franklin’s 1-4 family loan portfolio. The FDIC also did not identify ADC loan administration weaknesses on a timely basis. As a result, the bank’s risk profile and asset quality weaknesses became evident only after the real estate market deteriorated and significant delinquencies and losses occurred, starting in 2007.

In our opinion, there are two broad areas where the FDIC’s supervision of Franklin could have been enhanced. These areas are: (1) risk identification and analysis and (2) actions to address risks.

Risk Identification and Analysis. Overall, the FDIC could have better identified and analyzed risk to ensure that Franklin established and appropriately implemented controls and risk limitation and mitigation strategies. Examples where we believe the FDIC’s risk identification and analysis could have been enhanced are discussed in more detail below.

• Franklin’s 1-4 Family Residential Loan Portfolio: In ROEs prior to 2007, the FDIC could have better identified the risk posed by Franklin’s 1-4 family residential loan portfolio. Specifically, the FDIC described the overall credit quality of the bank’s 1-4 family residential loan portfolio as strong due to the portfolio’s weighted-average credit scores, weighted-average loan-to-value percentage, and estimated average total debt service-to-income ratios; however, the FDIC did not identify the bank’s failure to stratify or segment its mortgage loan portfolio by risk factors and risk layers. As a result, the FDIC did not identify as a potential concern Franklin’s lack of risk identification, measurement, monitoring, and control of its nontraditional and subprime loan portfolio.

  Although the FDIC had issued examination guidance in October 2006 titled, Interagency Guidance on Nontraditional Mortgage Product Risk, the FDIC did not use this guidance during the October 2006

examination, nor does it appear that the guidance was used at subsequent examinations. For example, in the July 2008 ROE, the FDIC stated that the 1-4 family residential loan portfolio “should be segmented into groups of loans with similar risk characteristics; for example, nontraditional mortgages should be separated from traditional mortgages.” However, the FDIC did not emphasize the need for Franklin to establish adequate policies and controls that considered risk-layered limits and risk-mitigation practices and strategies. Emphasis on risk layering was included in the October 2006 guidance.

In our opinion, had the FDIC encouraged Franklin to adequately identify, measure, monitor, and control its nontraditional and subprime loan portfolio, the level of loss incurred by the bank due to the economic decline could have potentially been reduced. In addition, both bank management and examiners could have more effectively assessed and managed/supervised the risk associated with the bank’s nontraditional and subprime mortgage loan products.

• Due Diligence for Purchased Loan Pools: The FDIC did not adequately assess Franklin’s due diligence related to its purchased loan portfolio. The July 2008 ROE states that the primary risk factors leading to performance problems in the bank’s mortgage loan portfolio related to the existence of second liens (not held by the bank), limited documentation or no documentation of income, and geographic location. The existence of second liens was unknown to both bank management and the FDIC prior to the October 2007 examination due, in part, to Franklin’s failure to perform an adequate level of due diligence for third-party loan originations and purchased loan pools. The FDIC stated that it did not identify the bank’s lack of due diligence in these areas due, in part, to the decision to not include a detailed review of that area in the scope of the risk management examination. In our opinion, had the FDIC encouraged Franklin to perform a more thorough due diligence review, the increased risk associated with the bank’s purchased loans may have been more quickly identified and actions may have been taken to limit or mitigate the level of risk assumed.

  Interest Reserve Policies: According to information Franklin provided to the FDIC, as of May 2008, 30 percent of the bank’s ADC loan portfolio (based on dollar volume) was comprised of loans with interest reserve funding provisions. However, we found that the FDIC did not review for, or express adequate concern about, Franklin’s lack of loan policies regarding the use of interest reserves. Appendix A to Part 365—Interagency Guidelines for Real Estate Lending Policies requires that each insured bank adopt and maintain a written policy that establishes appropriate limits and standards for all extensions of credit that are secured by liens on, or interests in, real estate or are made for the purpose of financing the construction of buildings or other improvements. The guidance also states that policies should address “Standards for the acceptability of and limits on the use of interest reserves.” Although required, our review of the FDIC’s ROEs and examination documentation indicated that Franklin’s loan policies

contained no such guidelines and limitations. Accordingly, although the FDIC was aware that Franklin was employing interest reserves, it did not address Franklin’s failure to establish the required standards on the use of interest reserves.

Further, although we recognize that the use of interest reserves is common in certain forms of ADC lending, we found that in some cases, Franklin appeared to be using interest reserves inappropriately. Specifically, based on our review of the 2007 and 2008 ROEs, we noted interest reserves were provided for projects that had experienced development or construction delays, cost overruns, sales or leasing shortages, or otherwise were not performing according to the original loan agreement and that had some level of collateral impairment. Interest reserves were also provided for raw land acquisitions. Further, some loans were modified by Franklin with increased interest reserve lines and/or the capitalization of past-due interest.

In our opinion, had the FDIC required Franklin to establish appropriate policies that addressed standards for the acceptability of and limits on the use of interest reserves, Franklin could have improved its ability to mitigate associated risk.

• Earnings Performance: The FDIC could have enhanced its assessment of the quality of the bank’s earnings performance by analyzing the level of interest income that had been derived from interest reserve loans. Although the bank may have accounted correctly for the interest income, the borrowers had not “made” interest payments. Rather, the interest is capitalized as part of the loan and recognized as income over time by the institution. As a result, the bank’s receipt of such interest, through the borrowers’ payoff of the loans, may not be assured. Examiners could have assessed the bank’s earnings performance based on the amount of interest income that was actually received and then measured the potential risk to earnings. In our opinion, had the FDIC performed this analysis, both the FDIC and Franklin could have had a better understanding of the bank’s risk profile (associated with ADC loan concentrations), degree of reliance on interest reserves to support earnings performance, and potential earnings exposure in a deteriorating economic environment. Based on this understanding, improved risk management controls and limits could have been implemented.
• Volatile Liability Dependence: Beginning with the September 2003 ROE, the FDIC repeatedly identified the bank’s reliance on non-core funding sources; however, the FDIC also reported that the bank’s strong asset quality and liquid loan portfolio mitigated any significant concern over the bank’s volatile liability dependence. In our opinion, this position was not supported by the higher-risk nature of the bank’s concentrations in nontraditional 1-4 family residential and ADC loans. Instead, we believe the FDIC’s assessment of liquidity overrelied on the strength of asset quality to mitigate the bank’s volatile liability dependence. The FDIC did not identify the potential level of risk until the October 2007 ROE, when asset quality began to significantly deteriorate and the marketability of the bank’s mortgage portfolio declined. In addition, based on the October 2007 ROE, the FDIC first recommended that the bank establish a

reasonable risk limit for its volatile liability dependence. Had the FDIC not overrelied on the strength of asset quality, more conservative operating parameters and/or mitigating risk factors (such as increased capital levels) could have been established. Based on these actions, risk associated with the bank’s significant asset growth and loss exposure to adverse economic events could have been reduced.


• Economic Risk Management: Examiner comments and analysis concerning the bank’s economic risk management practices were lacking. In the September 2003 and September 2004 ROEs, examiners routinely addressed (and responded favorably to) the ROE’s Risk Management Assessment question, “Are risk management processes adequate in relation to economic conditions and asset concentrations?” In the September 2004 ROE, examiners also noted that the BOD and management closely monitored economic conditions. However, the examiners’ responses typically did not address the bank’s economic environment or address how the bank planned to respond to a potential deterioration in its key market areas. The ROEs after the September 2004 examination did not include the risk management assessment pages.
• ALLL Analysis: The October 2006 through July 2008 ROEs and examination workpapers showed that the FDIC did not ensure that the bank had established its ALLL for the estimated credit losses inherent in the bank’s nontraditional mortgage loan portfolio consistent with regulatory requirements, nor did the FDIC ensure that management considered the higher risk of loss posed by layered risks when establishing the ALLL. According to the Interagency Guidance on Nontraditional Mortgage Product Risk, dated October 2006, banks should segment their nontraditional mortgage loan portfolios into pools of loans with similar risk characteristics, but Franklin had not done so. The basic segments typically include collateral and loan characteristics, geographic concentrations, and borrower qualifying attributes. In our opinion, had the FDIC performed or encouraged Franklin to perform this analysis, both the FDIC and Franklin could have had a better understanding of the risk profile of the bank’s loan portfolio, and of the amount of ALLL and/or capital needed to reserve for expected and unexpected losses; thereby, mitigating, to a certain degree, the bank’s risk profile.

FDIC Actions to Address Risks. The FDIC could have better ensured that it made recommendations and took corrective actions that were effective and timely in addressing Franklin’s risk.

• 1-4 Family Residential and ADC Lending: From the September 2003 ROE until the final ROE in July 2008, the FDIC noted Franklin’s concentrations in California 1-4 family residential lending. Since the November 2005 ROE, the FDIC also noted Franklin’s concentrations in ADC lending. These concentrations were consistently mentioned in the ROEs; however, examiner recommendations were limited to asking Franklin to improve concentration monitoring or to adopt combined concentration limits for ADC loans

and builder lines. In our opinion, these recommendations were too limited given the known risks associated with Franklin’s loan portfolio. Not until the October 2007 and July 2008 ROEs did the FDIC make recommendations related to limiting the bank’s loan concentrations or mitigating the bank’s risk by requiring Franklin to increase its capital levels.


• Internal Audit Function: Although the FDIC noted concerns regarding the bank’s internal audit function in the September 2003, September 2004, November 2005, October 2006, and October 2007 ROEs, the FDIC did not highlight inadequacies in the internal audit function as a repeated area of concern. Further, only the July 2008 ROE identified the apparent contraventions of Appendix A to Part 364, Interagency Guidelines Establishing Standards for Safety and Soundness, because the institution had not established internal controls, information systems, and an internal audit system appropriate to the size of the institution and the nature, scope, and risk of activities. Due to the significance and importance of this function, in our opinion, greater supervisory emphasis could have been placed on the bank’s failure to ensure full compliance with this regulatory requirement.
• Volatile Liability Limits and an Adequate CLP: In our opinion, the FDIC could have expressed greater supervisory concern over Franklin’s repeated failures to adequately implement examiner recommendations regarding improving the bank’s liquidity policies and establishing a CLP and volatile liability risk limits. The ROEs and interviews with FDIC examiners indicate that Franklin had not established volatile liability limits or an adequate CLP. Specifically, in the November 2005 through October 2007 ROEs, the FDIC repeatedly recommended that management set liquidity risk limits and formulate and/or document a CLP. However, bank management repeatedly failed to fully implement the FDIC’s recommendations. Further, the FDIC examiners we interviewed stated that bank management did not fully understand the need for contingency liquidity and crises plans and that management considered the bank’s CLP to be its ability to market and sell the 1-4 family residential loan portfolio. Examiners also stated that the bank’s failure to ensure that its loans were marketable was a critical management error that ultimately contributed to the bank’s liquidity crisis and failure.

IMPLEMENTATION OF PCA

The purpose of PCA is to resolve problems of insured depository institutions at the least possible long-term cost to the DIF. PCA establishes a system of restrictions and mandatory and discretionary supervisory actions that are to be triggered depending on an institution’s capital levels. Part 325 of the FDIC’s Rules and Regulations implements PCA requirements by establishing a framework for taking prompt corrective action against insured nonmember banks that are not adequately capitalized. Franklin was categorized as significantly undercapitalized just prior to its failure. As a result, the FDIC issued a C&D that contained a capital provision that directed Franklin

In its response, DSC stated its view that substantial and ongoing supervisory concern had been demonstrated by examiners since 2003 and quarterly offsite monitoring that was conducted because of Franklin’s rapid-growth strategy. DSC mentioned that, from 2003 to 2006, examiners consistently noted (in the ROEs) Franklin’s dependence on high-risk, volatile funding and its concentrations in 1-4 family residential and ADC loans. DSC further stated that “. . . because asset quality appeared strong and capital adequacy and liquidity sufficient, Franklin received composite ‘2’ examination ratings consistent with the FDIC’s Risk Management Manual of Examination Policies.” DSC also stated that at the October 2007 examination, Franklin was downgraded to a composite 3 rating, and the BOD adopted a comprehensive resolution with 14 specific provisions to be addressed by management. In March 2008, DSC became aware of significant errors and possible intentional falsification of Franklin’s Call Reports. Subsequently, in April 2008, DSC decided to accelerate the next scheduled examination to July 2008, which ultimately resulted in the downgrading of Franklin to a composite 5 rating. Of particular note, DSC pointed out that examiners recognized Franklin’s use of poorly underwritten non-traditional mortgage products, growth that outpaced its internal control structure, and due diligence on purchased loan pools that became problematic.

Our findings in this MLR and DSC’s statement related to asset quality, capital adequacy, and liquidity underscore one of the more difficult challenges facing FDIC examiners – assessing risk management by banks when their reported financial condition and ratios make them appear to be safe and sound. This issue and others will be discussed in our future MLR summary reports.

Objectives

We performed this audit in accordance with section 38(k) of the FDI Act, which provides that if a deposit insurance fund incurs a material loss with respect to an insured depository institution, on or after July 1, 1993, the Inspector General of the appropriate federal banking agency shall prepare a report to that agency reviewing the agency’s supervision of the institution. The FDI Act requires that the report be completed within 6 months after it becomes apparent that a material loss has been incurred.

Our audit objectives were to (1) determine the causes of the financial institution’s failure and resulting material loss to the DIF and (2) evaluate the FDIC’s supervision of the institution, including implementation of the PCA provisions of section 38.

We conducted the audit from November 2008 to May 2009 in accordance with generally accepted government auditing standards. However, due to the limited scope and objectives established for material loss reviews, which are generally applied to just one financial institution, it may not have been feasible to address certain aspects of the standards, as described on the next page.

Scope and Methodology

The scope of this audit included an analysis of Franklin’s operations from December 31, 1998 until its failure on November 4, 2008. Our review also entailed an evaluation of the regulatory supervision of the institution from 2003 to 2008.

To achieve the objectives, we performed the following procedures and techniques:

• Analyzed examination and visitation reports prepared by the FDIC and DSML the FDIC from 2003 to 2008.
• Reviewed the following:
  • Bank data and correspondence maintained at DSC’s Dallas Regional Office and Houston Field Office.
  • Reports prepared by the Division of Resolutions and Receiverships (DRR) and DSC relating to the bank’s closure.
  • Records of the bank’s external auditor, Deloitte & Touche, LLP, Houston, Texas, as made available through the external auditor’s counsel Latham & Watkins, LLP, San Francisco, California.

• Pertinent DSC policies and procedures.
• Interviewed the following FDIC officials:
  • DSC management in Washington, D.C.; Dallas, Texas; and Houston, Texas.
  • FDIC examiners from the DSC Houston Field Office who participated in Franklin examinations.
• Met with officials from the DSML of Austin, Texas, to discuss their historical perspective of the institution, its examinations, state banking laws, and other activities regarding the DSML’s supervision of the bank.

We performed the audit field work at the DSC offices in Houston and Dallas, Texas.

Internal Control, Reliance on Computer-processed Information, Performance Measurement, and Compliance With Laws and Regulations

Due to the limited nature of the audit objectives, we did not assess DSC’s overall internal control or management control structure. We performed a limited review of Franklin’s management controls pertaining to its operations as discussed in the body of this report.

For purposes of the audit, we did not rely on computer-processed data to support our significant findings and conclusions. Our review centered on interviews, ROEs and correspondence, and other evidence to support our audit.

The Government Performance and Results Act of 1993 (the Results Act) directs Executive Branch agencies to develop a customer-focused strategic plan, align agency programs and activities with concrete missions and goals, and prepare and report on annual performance plans. For this material loss review, we did not assess the strengths and weaknesses of DSC’s annual performance plan in meeting the requirements of the Results Act because such an assessment is not part of the audit objectives. DSC’s compliance with the Results Act is reviewed in program audits of DSC operations.

Regarding compliance with laws and regulations, we performed tests to determine whether the FDIC had complied with provisions of PCA and limited tests to determine compliance with certain aspects of the FDI Act. The results of our tests were discussed, where appropriate, in the report. Additionally, we assessed the risk of fraud and abuse related to our objectives in the course of evaluating audit evidence.

APPENDIX 3

EXAMINER COMMENTS AND RECOMMENDATIONS

---

APPENDIX 3

APPENDIX 3

APPENDIX 4

ACRONYMS IN THE REPORT

---