Each depositor insured to at least $250,000 per insured bank

Federal Deposit Insurance Corporation
Office of Audits
Office of Inspector General
Washington, D.C. 20434

DATE: December 23, 2002

TO: Michael J. Zamorski, Director, Division of Supervision and Consumer Protection

FROM: Russell A. Rau [Electronically produced version; original signed by Russell Rau], Assistant Inspector General for Audits

SUBJECT: Examiner Assessment of High Loan-Growth Institutions (Audit Report No. 03-009)

The Federal Deposit Insurance Corporation’s (FDIC) Office of Inspector General (OIG) has completed work on the second of two objectives in an audit of the Division of Supervision and Consumer Protection’s (DSC) assessment of commercial real estate (CRE) loans in the course of safety and soundness examinations.

The objectives of this audit were to determine whether: (1) the examiners fully assessed appraised value, cash flow, and lending policies in their examination of commercial real estate loans and (2) the examiners’ strategies for assessing a significant level of commercial real estate loan growth were sufficient for identifying increased risk. While our overall audit addressed both objectives, the subject matter and results were distinct enough that we have prepared separate reports to address each objective. This audit report addresses our work with regard to audit objective (2) above and covers our assessment of examiner analysis of institutions that have experienced a significant level of loan growth. A second report will be issued separately to address objective (1).

We reviewed pre-planning memoranda, examination reports, and working papers during fieldwork in the San Francisco and Dallas regions. Our audit sample consisted of a selection of 15 examinations (based on institutions that experienced an annual loan growth rate of 40 percent or greater during the previous year) from the original 35 examinations sampled during the audit of Examiner Assessment of Commercial Real Estate Loans. We discussed matters related to our audit objective with selected DSC Washington senior management and San Francisco and Dallas regional management, field office supervisors, and examiners to:

1. supplement our review of examination documentation,
2. determine management’s interpretations of relevant DSC examination guidance, and
3. clarify management’s expectation of examiners in applying the guidance.

We have included their views, as appropriate, in pertinent sections of our report. Additional details on our objective, scope, and methodology are contained in Appendix I.

BACKGROUND

Between 1980 and 1994, almost 1,600 banks insured by the FDIC were closed or received FDIC financial assistance. Many of the banks that failed during that time were active participants in the CRE markets. (Note: Based upon the Federal Financial Institutions Examination Council instruction book for the Consolidated Reports of Condition and Income (Call Reports), the OIG defined commercial real estate loans, for purposes of this review, as loans secured by real estate, including real estate loans secured by multifamily residential properties, nonfarm nonresidential properties, and construction and land development loans. Loans secured by or for the construction and development of farmland and one-to-four family residential properties were excluded.) A study prepared by the FDIC’s former Division of Research and Statistics (recently combined with the Division of Insurance to form the Division of Insurance and Research) entitled History of the Eighties–Lessons for the Future was published in December 1997. Volume I, An Examination of the Banking Crises of the 1980s and Early 1990s, revealed that concentrations of real estate loans relative to total assets were higher for institutions that subsequently failed than for banks that did not fail. During this period, large demand for real estate investments produced a boom in commercial real estate construction activity. Generally, bank underwriting standards were loosened. (Note: FDIC Rules and Regulations, 12 C.F.R. Part 365, Appendix A – Interagency Guidelines for Real Estate Lending Policies, indicates that underwriting standards should be the clear and measurable lending policies that reflect the level of risk that is acceptable to the board of directors and that enable an institution’s lending staff to evaluate various loan factors. In particular, "prudently underwritten real estate loans should reflect all relevant credit factors, including: The capacity of the borrower, or income from the underlying property, to adequately service the debt. The value of the mortgaged property. The overall creditworthiness of the borrower. The level of equity invested in the property. Any secondary sources of repayment. Any additional collateral or credit enhancements (such as guarantees, mortgage insurance or takeout commitments)." The Barron’s Business Guides Dictionary of Banking Terms defines bank underwriting as the "detailed credit analysis preceding the granting of a loan, based on credit information furnished by the borrower, such as employment history, salary, and financial statements; publicly available information, such as the borrower’s credit history, which is detailed in a credit report; and the lender’s evaluation of the borrower’s credit needs and ability to pay." Based on this analysis, lenders formulate their loan decision and establish the terms and conditions of the debt.) In addition, overly optimistic appraisals, together with the relaxation of debt service ratios, the reduction in the maximum loan-to-value ratios, and the loosening of other underwriting constraints, often meant that borrowers frequently had little or no equity at stake, and in some cases lenders bore most or all of the risk. (Note: The Barron’s Business Guides Dictionary of Banking Terms defines the term "debt service ratio" as the "financial ratio measuring a borrower’s ability to meet payments on a loan after paying expenses. The ratio, also called the debt coverage ratio, measures the number of times loan principal and interest are covered by net (after tax) income. It is generally applied to income property such as apartment buildings and multi-tenant office buildings." The Barron’s Business Guides Dictionary of Banking Terms defines the term "loan-to-value ratio" as "a percent, between the principal amount of a loan and the appraised value of the asset securing the financing." FDIC Rules and Regulations, 12 C.F.R. Part 365, Appendix A – Interagency Guidelines for Real Estate Lending Policies, defines the term loan-to-value as "the percentage or ratio that is derived at the time of loan origination by dividing an extension of credit by the total value of the property(ies) securing or being improved by the extension of credit plus the amount of any readily marketable collateral and other acceptable collateral that secures the extension of credit…Value means an opinion or estimate, set forth in an appraisal or evaluation…of the market value of the real property…For loans to purchase an existing property, the term "value" means the lesser of the actual acquisition cost or the estimate of value.")

Volume I of the FDIC study also detailed the life cycle of a bank failure. The study recognized that rapid loan growth was identified repeatedly as a precursor to failure. (Note: The FDIC’s study entitled History of the Eighties–Lessons for the Future, stated that the FDIC’s "GMS (growth-monitoring system) was developed during the mid-1980s and was designed to detect the initial stage in the life cycle of failing banks - the rapid-growth stage. The system’s premise is that rapid growth in total assets (or loans) represents a risky activity of which bank supervisors should be aware. Growth-related risk can come in at least two areas, loans and bank management: there may be increased loan concentrations in risky areas, and there may be management lapses such as lowered underwriting standards, increased reliance upon volatile funding, or a general weakening of internal controls in order to facilitate rapid growth. Banks that GMS identifies as rapid-growth institutions in these two areas are flagged for off-site review and may receive increased supervisory attention…Composite GMS scores are evaluated separately for two groups of banks. The first group is composed of banks whose quarterly asset and loan growth rates were 5 percent or more (high-growth banks). For all high-growth banks, composite GMS score percentile rankings are computed. Banks in the highest composite GMS score percentiles – currently the 95^th to 99^th percentiles – are "flagged" for off-site review…The second group is composed of banks with quarterly asset and/or loan growth under 5 percent (low-growth banks). These low-growth banks’ GMS scores and related information are available for review by regional office examiners in the GMS system.") In addition, institutions that failed typically moved through three stages of deterioration. In the first stage, there is rapid loan growth, loan concentrations emerge, and lending is aggressive (internal controls in the growth areas tend to be weak, and underwriting standards are generally more lenient). In the second stage, the institution has rising loan-quality problems, profits decline, and inadequate reserve levels become apparent. (Note: The study implies that when commercial real estate lending underwriting standards were significantly loosened, loan-quality weaknesses existed such as: insufficient income generated by a project to cover the interest and principal payments on borrowed funds; insufficient collateral protection – based on the assumption that real estate values (collateral values) would continue to rise in the future as they had in the recent past. As a result, many banks chose to raise their maximum loan-to-value ratios, and appraised property values frequently provided overly favorable collateral values and/or were based on speculative premises; liberalized repayment schedules – principal payments were repeatedly renewed, unpaid interest was frequently added back to the unpaid principal (capitalized); insufficient repayment capacity of secondary repayment sources – the guarantors of the original loan amount or the recourse often were not actively scrutinized; or inadequate managerial expertise of lending area – many banks chose to lend outside their local areas and often became involved in lending on real estate projects for which they had little or no direct experience, and many institutions became involved in real estate transactions without having had adequate experience in structuring, monitoring, or administering specialized commercial real estate loans.) In the final stage, deteriorating asset quality leads to losses and a depletion of bank capital. (Note: Although the study did not distinguish the difference between loan quality and asset quality, it appears that the textual message implies that inclusive with deterioration in loan quality, deterioration in other assets such as other real estate owned (foreclosed real estate) and loan interest receivables would also exist.) The study also notes that only over time do the effects of growth or risk-taking – whether these effects are good or bad – become apparent. This section of the FDIC’s study is presented in its entirety in Appendix II of this report.

Risk-Focused Examination Program

As a result of the many failures of the 1980s and early 1990s, the FDIC initiated a number of programs to improve the effectiveness of bank examinations. One of these initiatives was the risk-focused examination program. The FDIC, in conjunction with the Federal Reserve Board and the Conference of State Bank Supervisors, implemented a risk-focused examination process in 1997. This process was designed to focus examination resources on bank activities that pose the greatest risk exposure. In addition, the program encourages less regulatory burden by focusing on testing, rather than duplicating, the work of audit and control functions. One management control in particular, the bank’s loan grading system, can be tested. (Note: DSC’s Manual of Examination Policies states that "Credit grading involves an assessment of credit quality, the identification of problem loans, and the assignment of risk ratings. An effective system provides information for use in establishing valuation allowances for specific credits and for the determination of an overall ALLL [allowance for loan and lease losses] level." To ensure consistent application of the risk-focused examination process nationwide, the Division of Supervision and Consumer Protection developed the Examination Documentation modules to provide examiners with a tool to focus on risk management and to establish an appropriate examination scope. The Examination Documentation modules incorporate questions and points of consideration into examination procedures to specifically address a bank’s risk management strategies for each of its major business activities. In particular, the modules are segregated into three categories: Primary Modules, Supplemental Modules, and Loan and Other References. In addition, the format of the primary and supplemental modules is divided into three distinct sections of analysis: Core Analysis, Expanded Analysis, and Impact Analysis. The extent to which an examiner works through each of these three levels of analysis depends upon the conclusions reached regarding the presence of significant concerns or deficiencies. As stated within the Regional Directors Memorandum entitled Guidelines for Examination Workpapers and Discretionary Use of Examination Documentation Modules, dated September 25, 2001, "The use of the ED modules is now discretionary…Although their use is now discretionary, the ED modules are excellent training and reference tools, which provide consistency and standardized procedures." The Loan Portfolio Management and Review: General Examination Documentation module instructs examiners to "Validate the internal loan review system and assigned classification ratings. Also evaluate frequency and timeliness of reviews and updates to the board of directors. (NOTE: Preliminary sampling at this level should be sufficient to judge the accuracy of the internal loan review system without becoming so large that examiners duplicate efforts established by banks with satisfactory internal reviews. An inaccurate internal loan review system will result in expanded loan sampling.)") If the control is determined to be adequate, then the results of that system can be accepted and used by the examiners in evaluating the institution’s loan quality. If management controls are properly designed and effectively applied, examiners can also place greater reliance on the control systems and limit the scope of their review. As an illustration, this guidance would allow examiners to test on a sample basis, the accuracy of a bank’s internal loan grades. If the examiners find that the bank’s loan review methodology and loan grades are acceptable, then examiners can limit further loan review and utilize management’s aggregate loan grades to assess and draw conclusions about the risk and quality of the bank’s loan portfolio. If the bank compiles this same data into various reports that were also tested, then these documents could also be utilized in evaluating the bank’s loan quality. The risk-focused examination program encourages examiners to limit, or in some cases eliminate, traditional examination procedures in low-risk, well managed areas of the institution. This goal has been detailed, in part, in two Regional Directors memoranda as presented below.

The Regional Directors Memorandum entitled, Risk-Focused Examination Process–Program’s Goals and Objectives, dated December 16, 1998, states that

The risk-focused examination process seeks to strike an appropriate balance between evaluating the condition of an institution at a certain point in time and evaluating the soundness of the institution’s processes for managing risk. Moreover, the risk-focused approach attempts to involve less regulatory burden by focusing on testing, rather than duplicating, the work of audit and control functions. Based on the institution’s size, complexity, and risk profile, an examiner can choose to test, evaluate, and accept the results from such controls as internal and external audits, loan policy, loan review, and loan grading systems.

The Regional Directors Memorandum entitled, Risk-Focused Examination Program–Documentation Requirements, dated March 23, 1999, states that

The risk-focused examination program is designed to focus examination resources on those areas that pose the greatest risk to an insured institution… The level of analysis performed largely depends on the examiner’s assessment of management’s ability to identify, measure, monitor and control risks. A key aspect of that assessment is based on the adequacy of management controls such as audit functions, loan policies, loan grading systems, and other similar controls. If management controls are properly designed and effectively applied, examiners can place greater reliance on the control systems and limit the scope of their review.

Loan Review

In addition, the importance of the loan review function has been highlighted in a more recent DSC memorandum. The Regional Directors Memorandum entitled Loan Review, dated September 12, 2001, states that

Recent indicators suggest the potential for an economic downturn. This, coupled with industry loan growth, indicates a need to re-emphasize to examiners the importance of the loan review function and the loan sampling process… A thorough review of a bank’s loan and lease portfolio and other related sources of credit risk is one of the most important elements of the Safety and Soundness examination process. Such credit reviews are a primary means for the examiner to evaluate the effectiveness of internal loan review and credit grading systems, to determine that credit is being extended in compliance with internal lending policies, and to assess the adequacy of capital and the allowance for loan and lease losses. Credit reviews also enable an examiner to ascertain a bank’s compliance with applicable laws and regulations, make an overall judgement as to the safety and soundness of a bank’s lending and credit administration functions, and directly evaluate the quality of a bank’s loan and lease portfolio.

In analyzing and evaluating the quality of a bank’s loan and lease portfolio, the Loan Portfolio Management and Review: General Examination Documentation module instructs examiners to consider existing and developing risk factors such as significant loan growth. Significant loan growth may be generated by lowering underwriting standards or by shifting the mix and focus of new loan originations into lower quality, and therefore higher risk, loans. For example, a bank generally originates loans within various levels of quality, and bank management may assign these loans into different tier structures, such as A, B, and C quality loans. While all three levels are considered acceptable loan quality and deserving of a "Pass" classification for review purposes, each level denotes a distinct level of quality. Specifically, loans assigned to a tier level of "A" might denote borrowers with the highest/strongest loan rating; loans assigned to a tier level of "B" might denote borrowers with modest weaknesses/average loan rating; and loans assigned to a tier level of "C" might denote borrowers with the weakest/below average loan rating. If a bank changes the emphasis of loan origination from "A" quality loans to "C" quality loans, this shift may potentially change the quality of the loan portfolio and the risk profile of the institution. Furthermore, as the loan quality of loan underwriting moves downward, a potentially higher level of growth may be achieved. In addition, this change can impact the adequacy of the bank’s allowance for loan and lease losses and capital levels.

Examination emphasis on the review of new loan originations is important when examining a bank that has undergone significant loan growth, because shifts in the quality of newly originated loans that have not seasoned may cause a disproportionately more severe impact to the bank’s loan portfolio quality and loan losses as the new loans mature and potentially go into default. (Note: Barron’s Business Guides Dictionary of Banking Terms defines a seasoned loan as "a loan that has been on the books for at least a year and has a satisfactory payment record. Mortgage loans that have been on the books for a period longer than a year command a premium over unseasoned loans when sold in the secondary mortgage market.") Conversely, new loan growth (assuming the same level of quality of newly originated loans) in an institution that has not experienced a significant level of loan growth may be absorbed by the profitability of older and more seasoned loans due to the proportionately lower level of loans impacting the bank. Furthermore, examination emphasis on reviewing newly originated loans serves as an early detection of potential loan problems which will allow examiners to promptly identify concerns and allow management to modify portfolio strategies and intensify the supervision of weaker loans in a timely manner.

Report of Examination

The Federal Deposit Insurance Act requires the appropriate federal banking agency to conduct a full-scope on-site examination of each insured depository institution on a regular basis. The Report of Examination conveys the results of the examination process. The report is used, in part, by both supervisory personnel in supporting examination conclusions and recommendations and by bank management in implementing corrective actions. When a full-scope safety and soundness examination is performed, an institution is assigned an overall composite rating based on an evaluation and assignment of six component ratings. (Note: DSC’s Manual of Examination Policies states that composite ratings are based on a careful evaluation of an institution’s managerial, operational, financial, and compliance performance. The six key components used to assess an institution’s financial condition and operations are: capital adequacy, asset quality, management capability, earnings quantity and quality, the adequacy of liquidity, and sensitivity to market risk. The rating scale ranges from 1 to 5, with a rating of 1 indicating the strongest performance and a 5 rating indicating the most critically deficient level of performance. In general, the assessment of asset quality includes the examiners’ review of loans. Similar to the composite ratings, individual components are rated on a scale of 1 to 5.) In particular, the assessment of asset quality is the CAMELS component rating that is most influenced by the loan review process.

Examiners are required to document and support these ratings in the Report of Examination. The DSC Manual of Examination Policies states that "Report comments should clearly support the corresponding component rating. Comments should focus on an assessment, rather than a simple description, of policy, practice, or condition. Comments should explain an examiner’s reasoning for assigning a particular rating and recommendation… Other general concepts to follow include: perform a complete analysis, which formulates a conclusion; identify and assess risks proactively; and use appropriate tone." The Report of Examination also includes a confidential section that documents comments of interest primarily to supervisory agencies. These comments are not provided to bank management for review. However, the comments and data provided assist case managers, other members of FDIC regional and headquarters management, and other regulatory authorities in their case management, applications processing, report review, and general bank supervision duties.

RESULTS OF AUDIT

For the 15 safety and soundness examinations that we reviewed, DSC examiners’ loan review process for institutions that had experienced a significant level of loan growth was not sufficient in identifying risk. Specifically, examiners were not always:

1. targeting new loans for sampling purposes and reporting on the level of new loans reviewed,
2. assessing or commenting on the loan quality of newly originated loans, and
3. assessing the internal loan risk rating process at banks based on a methodology that incorporates a review of non-adversely classified loans. (Note: Barron’s Business Guides Dictionary of Banking Terms describes adversely classified assets as "loans and other assets that are at risk to some degree. Such assets fail to meet acceptable credit standards…" DSC’s Manual of Examination Policies states that "Adversely classified loans are allocated on the basis of risk to three categories: (1) Substandard; (2) Doubtful; and (3) Loss. Other loans of questionable quality, but involving insufficient risk to warrant classification, are designated as Special Mention loans…Loan classifications are expressions of different degrees of a common factor, risk of nonpayment. All loans involve some risk, but the degree varies greatly." Non-adversely classified loans are all loans that are not classified Substandard, Doubtful, or Loss.)

As a result, there was insufficient assurance that examiners were consistently performing a comprehensive review and analysis of newly originated loans in high loan-growth institutions. Furthermore, due to the inconsistent review and analysis performed in this area, regulatory supervision appeared reactive to changing circumstances (such as the identification of loan underwriting and quality concerns after loans begin to default and/or become adversely classified) when proactive measures such as an assessment of the underwriting trends of newly originated and non-adversely classified loans could have been implemented to more promptly assess and mitigate risk.

Sampling and Reporting of New Loans

Examiners did not always target new loans for sampling purposes during the pre-examination planning process, and examiners did not always report on the level of new loans reviewed in the Reports of Examination. (Note: The OIG defined new loans, for purposes of this review, as loans originated since the previous examination’s as of date.) While examiners are instructed to sample loan types that exhibit high rates of growth and new loans, no formal reporting requirements have been established to present this information in the Report of Examination. As a result, there was insufficient assurance for DSC managers that examiners were consistently targeting and reviewing new loans when validating loan classifications and assessing internal loan review practices at high loan-growth institutions.

The Loan Portfolio Management and Review: General Examination Documentation module, Core Analysis procedures, instructs examiners to validate the internal loan review system and assigned classification ratings. Examiners are instructed to evaluate a cross section of loans by type, size, and severity of classification. This includes sampling watch list loans to assess rating accuracy and sampling loans not on the watch list to validate the internal loan review process. (Note: The Barron’s Business Guides Dictionary of Banking Terms defines watch list as "any list of loans or credit exposures compiled by a bank for internal monitoring." The Loan Portfolio Management and Review: General Examination Documentation module instructs examiners, in part, to validate and evaluate the bank’s watch list.) When sampling loans not on the watch list, examiners are directed to consider significant loans originated since the previous examination, new types of loans, and loan types exhibiting high rates of growth. (Note: Based upon the Loan Portfolio Management and Review: General Examination Documentation module, examiners are instructed to "Sample ‘watch list’ loans and assess rating accuracy. When developing the sample consider the following: (1) Credits representing the greatest inherent risk to the bank. (2) Severity of the internal classification. (3) Multiple credit types and categories. (4) Loans to industries or groups affected by adverse economic trends. (5) Loans to facilitate the sale of bank assets or loan collateral… Sample loans NOT on the watch list to validate the internal loan review process, considering the following: (1) Previously classified and Special Mention loans. (2) Significant overdue and nonaccrual loans. (3) Loans to insiders, their related interests, and affiliates. (4) Significant credits originated since the previous examination. (5) New types of loans. (6) Loans originated by each loan officer and loan officers with unusually high loss ratios. (7) Other significant credits as determined by the EIC, including loans to industries or groups affected by adverse economic trends. (8) Loan types and individual borrowers exhibiting high rates of growth. (9) Loans to facilitate the sale of bank assets, insider assets, or loan collateral. (10) Loans at a seasonal low point that could represent large credits when fully drawn.")

As illustrated in Figure 1, in over one-half of the examinations reviewed (9 of 15 examinations), examiners did not initially target new loans for sampling purposes. Based on a review of the pre-examination memoranda, in six instances the examiners stated that the sample would include a selection of newly originated loans. (Note: As stated in the Regional Directors Memorandum entitled Revised Pre-examination Planning Memoranda, dated September 12, 2001, "The primary purpose of the PEP (pre-examination planning) memorandum is to convey and document examiners’ conclusions regarding allocation of examination resources according to perceived risk." Regarding the scope of loan review, "The examiner will comment on the proposed loan scope, with emphasis on risk areas within the portfolio where loan file review will be concentrated. Examiners will, to the extent possible, disclose the target loan penetration percentage.") In the remaining nine cases, no mention was made of including newly originated loans within the loan sample.

Figure 1: Pre-Examination Sampling of Newly Originated Loans

[Note: This image appears in the non-508-compliant version of this audit report.]

Text description of Figure 1: New loans were not targeted in 9 of 15 pre-examination memoranda (60 percent). (Note: In one of these nine cases, the institution was a de novo bank that was undergoing its first full-scope safety and soundness examination, and all loans reviewed could be considered new loans for sampling purposes regardless of the sampling methodology implemented. (Note: As defined in the Regional Directors Memorandum entitled Maximum Efficiency, Risk-focused, Institution Targeted (MERIT) Guidelines, dated March 27, 2002, de novo banks are institutions that have been insured for less than 3 years. Of the six de novo institutions reviewed, one was undergoing its first full-scope safety and soundness examination, four were undergoing their second full-scope safety and soundness examination, and one was undergoing its third full-scope safety and soundness examination.) However, we did not omit the pre-examination memorandum from our sample, because this document did not provide assurance that a more in-depth loan review process would be conducted to address the high-risk characteristics of those newly originated loans.) New loans were targeted in 6 of 15 pre-examination memoranda (40 percent).

Source: OIG Analysis of DSC’s Pre-Examination Memoranda

Additionally, based on a review of the Reports of Examination, as illustrated in Figure 2, in two-thirds of the examination reports reviewed (10 of 15 examination reports), examiners did not note that the scope of the loan review included newly originated loans. In five instances, examiners made comments that indicated that newly originated loans were sampled but did not provide final dollar amounts or percentages.

Figure 2: Examination Reporting of Newly Originated Loans Sampled

[Note: This image appears in the non-508-compliant version of this audit report.]

Text description of Figure 2: New loans sampled were not reported in 10 of 15 reports of examination (67 percent). New loans sampled were reported in 5 of 15 reports of examination (33 percent).

Source: OIG Analysis of DSC’s Reports of Examination

The practice of not identifying and targeting new loans for sampling purposes appears to conflict with DSC’s examination policies and procedures. However, inherent in the sampling process, it is probable that some new loans that exhibit high-risk profiles would be selected for review. (Note: The Regional Directors Memorandum entitled Loan Review, dated September 12, 2001, states that "commercial real estate loans subject to examiner review during an examination should include all known problem loans and all insider loans of significant size. Problem loans comprise past-due loans, nonaccrual loans, loans otherwise impaired as defined in Statement of Financial Accounting Standards No. 114, renegotiated or restructured debt, loans internally criticized or classified by the bank, and loans that were adversely classified at the previous examination. Special Mention loans should also be reviewed…In addition, ‘large’ loans should also be reviewed as needed. Large loans are defined as loans, or aggregations of loans to the same or related borrowers, which exceed a dollar cut-off level established by the examiner-in-charge.") In general, DSC senior managers we spoke with agreed that, for high-growth institutions, examiners should be targeting newly originated loans for sampling purposes. In addition, Dallas Regional Office managers expressed concern over the number of pre-examination memoranda that did not state that new loans would be reviewed. San Francisco Regional Office managers stated that if the banks were de novo institutions, then their assumption would be that new loans would be reviewed.

DSC’s policies and procedures do not specifically require that new loans sampled for loan review purposes be reported in the Report of Examination in the description of the examination’s loan review scope comment. (Note: DSC’s Manual of Examination Policies requires that examiners provide a description of the examination’s loan penetration. The guidance states that the Report of Examination should include the following: Asset review date; Number of relationships reviewed; Dollar volume of credit extensions reviewed/percent of total credit extensions; Dollar volume of non-homogenous credit extensions reviewed/percentage of total non-homogenous credit extensions; and Credit extension cutoff review point (if applicable). The loan penetration comment can also include a breakdown of loans by major loan type, location, officer, or other information, as appropriate. As stated in the Regional Directors Memorandum entitled Loan Review, dated September 12, 2001, "loan scope and sampling should be documented within examination work papers and Page A in the confidential section of the Report of Examination...In particular, examiners should ensure that the reasoning used in determining the composition and volume of the loans reviewed is documented.") In particular, the confidential section of the Report of Examination does not require a breakdown of loans reviewed by age. However, according to DSC managers we interviewed, this concern is mitigated by the age of loans reviewed being documented on loan sheets or being available through the Automated Loan Examination Review Tool (ALERT) program. (Note: Based upon the FDIC’s The Automated Loan Examination Review Tool User’s Guide V2.0, the Automated Loan Examination Review Tool (ALERT) was first introduced in April 1996 as a means of improving the loan review function during bank examinations. Currently, ALERT version 3.0 allows examiners to automatically import and map data received from banks, vendors, or service providers. The ALERT program facilitates the examination’s loan review process by allowing examiners to obtain and query loan data, generate the loan scope and sample, print associated reports, and export related files.) DSC’s policies and procedures also do not define the term "significant loan growth."

Conclusion

For high loan-growth institutions, examiners should target and sample new loans for review purposes, and the pre-examination memoranda should identify that new loans will be targeted for high-growth banks. In accordance with the risk-focused examination program, the sample of new loans to be tested should be formulated based on a risk assessment of management’s ability to identify, measure, monitor, and control risks in this area.

When examiners identify a high-risk indicator, such as significant loan growth, the level of review and analysis performed should be reported to serve as a basis of support for the conclusions reached. In particular, the Report of Examination’s loan scope comment that is located within the confidential section of the report should provide a breakdown of new loans reviewed by dollar amount and/or percentage. By including this information in the confidential section of the Report of Examination, DSC managers would be provided assurance that examiners had considered a sufficient number of new loans to assess the risk associated with significant loan growth of the institution and to formulate conclusions. Furthermore, reliance on the traditional sampling process to capture new loans, no matter how likely it is that new loans would be selected for review, does not ensure that an adequate sample has been formulated in accordance with the risk-focused examination program. This methodology also provides insufficient assurance that examiners will consistently perform a comprehensive review and analysis of newly originated loans in high loan-growth institutions.

Recommendations

We recommend that the Director, DSC:

1. Revise policies and procedures to define the term "significant loan growth" and to require examiners to target and specifically sample new loans for examination when a financial institution has experienced significant loan growth since the last full-scope safety and soundness examination.
2. Revise policies and procedures to require examiners to report on new loans sampled for review purposes when a financial institution has experienced significant loan growth since the last full-scope safety and soundness examination.

Overall Quality Assessment of New Loans

Examiners did not always assess and/or comment on the overall loan quality of newly originated loans within the Reports of Examination. Although examiners are instructed to review and assess individual new loans, existing policies and procedures do not detail how examiners should assess and comment on the loan quality of the portfolio of newly originated loans when high loan growth is a risk factor. As a result, examiners were not always performing a comprehensive review and analysis of newly originated loans. When newly originated loans are sampled, these loans are reviewed to determine each loan’s appropriate loan quality classification (Pass, Special Mention, Substandard, Doubtful, or Loss). However, by not assessing the portfolio of newly originated loans and their dispersion within the bank’s internal rating system, underwriting trends/shifts in loan quality cannot be proactively identified. Furthermore, this analysis will not be available to assess current bank processes, substantiate the quality of loan underwriting, support the assumptions and historical loan loss rates used in the calculation of the allowance for loan and lease losses, and supplement the determination of capital adequacy, one of the CAMELS components. (Note: DSC’s Manual of Examination Policies states that "The quality, type, liquidity and diversification of assets (including off balance sheet activity), with particular reference to assets adversely classified and the adequacy of the ALLL, are necessarily vital factors in determining the adequacy of capital.")

DSC’s Manual of Examination Policies, for safety and soundness examinations, assigns examiners the responsibility for assessing the quality of the loan and lease portfolio, the loan review system, and the adequacy of the allowance for loan and lease losses. Loan portfolio analysis and the determination of loan quality are used to establish and determine the adequacy of the allowance for loan and lease losses. In general, the greater the risk in the loan portfolio, the greater the allowance should be to reserve for and to mitigate the increased risk. The Loan Portfolio Management and Review: General Examination Documentation module, Core Analysis procedures, instructs examiners to assign classifications to loans reviewed, evaluate the internal loan classifications for accuracy, and evaluate the level and trend of classified loans. If the internal grading system is reliable, examiners are to use the bank’s data for preparing the appropriate examination report pages, determining the overall level of classifications, and providing supporting comments regarding the quality of the loan portfolio. When reviewing the allowance for loan and lease losses, examiners are also instructed to determine if management considers any factors that are likely to cause estimated loan losses to differ from historical loss experience. Examiners should consider, in part, changes in the quality of the bank’s loan review system and the degree of oversight by the bank’s board of directors. (Note: DSC’s Manual of Examination Policies states that "Estimated credit losses should reflect consideration of all significant factors that affect collectibility of the portfolio as of the evaluation date. While historical loss experience provides a reasonable starting point, historical losses, or even recent trends in losses, are not by themselves, a sufficient basis to determine an adequate level. Management should also consider any factors that are likely to cause estimated losses to differ from historical loss experience, including but not limited to: Changes in lending policies and procedures, including underwriting, collection, chargeoff, and recovery practices. Changes in local and national economic and business conditions. Changes in the volume or type of credit extended. Changes in the experience, ability, and depth of lending management. Changes in the volume and severity of past due, nonaccrual, restructured, or classified loans. Changes in the quality of an institution’s loan review system or the degree of oversight by the board of directors. The existence of, or changes in the level of, any concentrations of credit." The Loan Portfolio Management and Review: General Examination Documentation module also instructs examiners to determine if management considers any factors that are likely to cause estimated loan losses to differ from historical loss experience. The module details the same factors as noted above, and the module also includes consideration of "The effect of external factors such as competition and legal and regulatory requirements.")

DSC has provided guidance to examiners on assessing the adequacy of a bank’s allowance for loan and lease losses through various sources, including, but not limited to, the Examination Documentation modules, DSC’s Manual of Examination Policies, Regional Directors Memoranda, and FDIC Statements of Policy. However, DSC’s policies and procedures do not specifically detail how examiners should make the determination that estimated loan losses might differ from the bank’s historical loss experience and from the bank’s assumptions utilized within its methodology of determining the allowance for loan and lease losses. When performing the Expanded Analysis procedures of the Loan Portfolio Management and Review: General Examination Documentation module, examiners are instructed to determine, for the portions of the portfolio that have not been adversely classified, estimated loan losses over the upcoming 12 months based on the institution’s average annual rate of net charge-offs experienced over the previous 2 or 3 years on similar loans, adjusted for current conditions and trends.

As presented below, based on a review of the comments presented within the Reports of Examination, further analysis could have been performed on the banks’ loan portfolios, loan review systems, and loan underwriting practices. In addition, studies could have been formulated or reviewed on the banks’ loan migration.

Loan Portfolio Analysis

In all 15 Reports of Examination reviewed, examiners did not fully comment on the overall quality and trend of newly originated loans in the loan portfolio. In particular, examiners did not document an assessment of the loan quality of newly originated loans (loans originated since the last examination) in comparison to loans originated prior to the last examination. Such an assessment could detect shifts in the risk profile of the bank’s loan portfolio and substantiate and support estimated loan losses over the upcoming 12 months for non-adversely classified loans. Furthermore, examiners did not always comment on why the growth was occurring and how the level of loan growth was achieved. In other words, these examination reports did not appear to answer two key questions:

• Did the bank experience significant loan growth by loosening its underwriting standards?
• If so, does the bank have an adequate system in place to identify, measure, monitor, and control this increased risk?

Based on a review of the asset quality comments in the Reports of Examination, only 20 percent (3 of 15) of the comments discussed the cause of the growth beyond the notation that either loan growth or a merger occurred. (Note: The Loan Portfolio Management and Review: General Examination Documentation module states that "If the bank has acquired other institutions or loan portfolios, analyze the effect these purchases have had on the bank’s composition and risk profile.") Also of note, in two instances examiners commented in the Reports of Examination that the low level of adversely classified assets was attributed, in part, to the unseasoned nature of the loan portfolio, which, due to the level of growth achieved, kept adverse classifications to a minimum. While examiners recognized that new loans can make loan quality appear better than it really is, no other compensating analysis was documented, such as an analysis of the stratification of the internal loan grades at origination, an analysis of the loan portfolio growth, and an assessment of how this might impact the allowance for loan and lease losses.

Loan Review System and Loan Underwriting

The examination of asset quality (which encompasses newly originated loans) should also include an analysis of the bank’s loan review system and loan underwriting. (Note: DSC’s Manual of Examination Policies states that "The term loan review system refers to the responsibilities assigned to various areas such as credit underwriting, loan administration, problem loan workout, or other areas. Responsibilities may include assigning initial credit grades, ensuring grade changes are made when needed, or compiling information necessary to assess the adequacy of the ALLL [allowance for loan and lease losses].") While examiners typically commented on the internal loan review process, the comments were in the context of bank management’s ability to accurately identify and grade loans that were adversely classified. No comments were made on the adequacy of the bank’s internal loan ratings of non-adversely classified loans. Examiners also typically commented on the adequacy of loan underwriting, but the examiners’ conclusions had varying levels of support. As illustrated in Figure 3, no underwriting comments were made or supported in two of the reports, while other examiners primarily supported adequacy based on the quality of the bank’s loan analysis at origination, the level of documentation exceptions, the strength of and/or adherence to bank policies, or the aggregate level of loan quality.

Figure 3: Loan Underwriting Analysis

[Note: This image appears in the non-508-compliant version of this audit report.]

Text description of Figure 3: No underwriting comments were made or supported in 2 of 15 reports (13 percent). Underwriting comments were primarily supported based on the bank's loan analysis at origination in 1 of 15 reports (7 percent). Underwriting comments were primarily supported based on the level of documentation exceptions in 3 of 15 reports (20 percent). Underwriting comments were primarily supported based on the adequacy of and/or adherence to policy guidelines in 4 of 15 reports (27 percent). Underwriting comments were primarily supported based on the assessed quality of assets on an aggregated basis in 5 of 15 reports (33 percent).

Source: OIG Analysis of DSC’s Reports of Examination

Loan Migration

A loan migration study was not performed on newly originated loans. Specifically, non-adversely classified loans were not stratified into the bank’s various "Pass" classification categories, and these loans were not compared and contrasted to the bank’s historical loan originations and loss factors for those classification categories. As a result, examiners were not able to identify, track, and measure potential changes in a bank’s risk profile. Had examiners performed a migration study, this analysis could have also been used to assess the adequacy of the bank’s allowance for loan and lease losses.

Subsequent Examinations

From the original 15 examinations reviewed, 11 involved institutions that underwent a subsequent safety and soundness examination during the period covered by our audit fieldwork. We did additional work related to these 11 institutions and found that the composite CAMELS rating at 4 of the institutions was downgraded. As illustrated in Figure 4, the composite ratings were downgraded due, in part, to asset quality concerns. Based on a review of the comments associated with the subsequent examinations, it appears that the new loans originated before the previous examination (during the period of high loan-growth) migrated into more severe classifications and past due status. Furthermore, some of the weaknesses identified during the second examination could have potentially been noted earlier if a more thorough review of the newly originated loans had been performed. If these weaknesses had been noted earlier, deterioration in the institutions’ asset quality could have been identified and corrective action could have been initiated sooner, which would have decreased the risk to the Bank Insurance Fund. (Note: The DSC is responsible for determining the appropriate action based on the severity of the weaknesses identified by the examination. The potential "corrective action" pursued represents a wide range of possible supervisory responses to institution weaknesses, such as examination report comments, board resolutions and memorandums of understanding, and formal enforcement action.)

Figure 4: Subsequent Examination Ratings

[Note: This image appears in the non-508-compliant version of this audit report.]

Text description of Figure 4: The composite rating of the institution remained unchanged in 7 of 15 reports (46 percent). The composite rating of the institution and the component ratings of asset quality and management were downgraded in 4 of 15 reports (27 percent). A subsequent examination had not been performed on 4 of the 15 institutions (27 percent).

Source: OIG Analysis of DSC’s Reports of Examination

One institution dropped from a composite rating of 2 to a composite rating of 4. In this instance, asset quality dropped from a component rating of 1 to a component rating of 4. The first Report of Examination stated that asset quality was strong. Furthermore, the report noted that the bank’s internal watch list and the examiners’ adverse classifications assigned during the examination were identical with one exception. In addition, the report noted that the process for reviewing and grading loans was adequate. Asset quality was deemed strong based, in part, on a ratio of Adversely Classified Assets to Tier 1 Capital and Allowance for Loan and Lease Losses of 5 percent. (Note: DSC’s Manual of Examination Policies provides a definition of Tier 1 Capital as "the sum of: common stockholders' equity (common stock and related surplus, undivided profits, disclosed capital reserves, foreign currency translation adjustments, less net unrealized losses on available-for-sale equity securities with readily determinable fair values); noncumulative perpetual preferred stock; minority interests in consolidated subsidiaries; minus all intangible assets (other than limited amounts of mortgage servicing rights and purchased credit card relationships and certain grandfathered supervisory goodwill); identified losses (to the extent that Tier 1 capital would have been reduced if the appropriate accounting entries to reflect the identified losses had been recorded on the institution's books); investments in securities subsidiaries subject to section 337.4; and deferred tax assets in excess of the limit set forth in section 325.5(g).") No review was evident and no discussion was presented in the report on the quality and trend of newly originated loans or on underwriting in general.

A Problem Bank Memorandum prepared on the following examination that was conducted approximately 13 months later stated that a significant provision to the allowance for loan and lease losses was required as a result of a sharp increase in adversely classified loans. The cause cited was lax administration of the loan portfolio. Specifically, documentation deficiencies were high, past due loans had increased significantly, and the internal loan review and assessment was sub-par. Based on these comments, it appears that the new loans originated during the previous year of the first Report of Examination reviewed (the period of high-loan growth) migrated into more severe classifications and past due status. Furthermore, many of these weaknesses that were identified during the second examination could have potentially been noted earlier if a more thorough review of the newly originated loans had been performed.

In another institution, the composite rating was dropped from a 1 to a 2, and the asset quality component rating dropped from a 1 to a 3. The first Report of Examination noted that asset quality was strong. Furthermore, the report noted that management’s strong underwriting standards and below peer past due ratio was a reflection of sound asset quality. (Note: The term "past due ratio" appears to be in reference to the ratio "Past Due and Nonaccrual Loans and Leases to Gross Loans and Leases" that is presented on the same page of the Report of Examination. In accordance with the Federal Financial Institutions Examination Council’s A User’s Guide for the Uniform Bank Performance Report, dated March 2001, this ratio is calculated by adding all loans and leases past due 30 days or more and still accruing interest with all loans and leases on which interest is no longer being accrued and dividing by gross loans and leases.) The report also noted that the bank’s loan review process and internal grading system were satisfactory, and the bank’s internal watch list identified all loans that were adversely classified by the examiners except for one. Despite the pre-examination memorandum stating that "the review will ensure prudent underwriting standards for newly originated loans," no discussion was presented in the examination report on the quality and trend of newly originated loans.

Based on a review of the Summary Analysis of Examination Report page generated on the subsequent examination report that was conducted approximately 17 months later, the case manager noted that adversely classified assets increased and loan administration and weak underwriting practices were criticized. Technical exceptions were excessive, and many of the apparent violations were loan-related. The internal loan review procedures were inadequate and the allowance for loan and lease losses was also considered inadequate. Similar to our previous example, based on these comments, it appears that the new loans originated during the previous year of the first Report of Examination reviewed (the period of high loan-growth) migrated into more severe classification status. Furthermore, many of these weaknesses that were identified during the second examination could have potentially been noted earlier if a more thorough review of the newly originated loans had been performed.

DSC’s policies and procedures do not specifically detail how examiners should:

• identify potential changes in the quality of the bank’s loan review system,
• identify potential loan quality shifts in the bank’s loan underwriting, and
• make the determination that estimated loan losses may differ from historical loss experience.

However, DSC Washington, Dallas, and San Francisco managers to whom we spoke agreed that an analysis of newly originated loans should be performed at high loan-growth institutions. They expressed the following views and expectations pertinent to analyzing new loans:

• an analysis of new loans and underwriting, if done, would be documented in the examiner’s analysis of the allowance for loan lease losses and in the risk management section of the Report of Examination;
  
  
• conclusions about a bank’s loan portfolio are based on several factors, including a review of the following:
  • the internal risk rating system,
  • the lines of loans (commercial, residential, etc.),
  • the allowance for loan and lease losses,
  • historical loan loss rates, and
  • deviations between the bank’s and the examiner’s loan ratings.
    
    
• a review of the bank’s loan policies would alert examiners to shifts in the bank’s underwriting practices.

Conclusion

Reliance on the traditional review techniques used to assess the loan portfolios of institutions that have not experienced significant loan growth is not sufficient to measure and assess the potential risks to high-growth institutions. As is evidenced by the significant level of downgrades that occurred during subsequent examinations of institutions we reviewed, the following DSC approaches did not go far enough in assessing an institution’s loan growth:

• The formulation of conclusions based on the level and trend of adversely classified assets;
• The reliance on the review of a bank’s loans policies to alert examiners to shifts in loan underwriting trends;
• The reliance on bank management to self-assess the adequacy of the allowance for loan and lease losses without an independent determination of adequacy of the assumptions used; and
• The use of historical loan loss rates without the basis to formulate adjustments for changing conditions and circumstances.

Conclusions primarily based on the above approaches resulted in supervision that was reactive to changing circumstances, and DSC was thus hampered from implementing more proactive measures to address risks. Further, by not performing an analysis on the underlying quality of newly originated loans in institutions that are experiencing high levels of growth, examination resources may not have been focused on those areas that pose the greatest risk to an insured institution.

For high-growth institutions, examiners should review newly originated loans. In particular, examiners should perform a comprehensive review of the bank’s loan portfolio, loan review system, and loan underwriting. In addition, examiners should either perform a loan migration study or review such a study prepared by the bank. Further, examiners should attempt to identify potential shifts in the bank’s loan underwriting, and examiners should make the determination of whether estimated loan losses differ from the bank’s historical loss experience and from the bank’s assumptions in its methodology for determining the allowance for loan and lease losses.

Recommendations

We recommend that the Director, DSC:

3. When loan growth is a high-risk factor, clarify existing policies and procedures to specifically detail how examiners should assess and comment on:
   
   
   1. the loan quality of newly originated loans,
   2. the loan review system, and
   3. loan underwriting.
      
      
4. When loan growth is a high-risk factor, clarify existing policies and procedures to detail how examiners could incorporate a loan migration study into the assessment of loan quality and underwriting.
   
   
5. Re-emphasize to examiners the need to assess and report on management’s processes for controlling risk when potential high-risk indicators are present.

Examiners did not always assess the bank’s internal loan risk rating process based on a methodology that incorporates a review of non-adversely classified loans. While examiners are instructed to validate the bank’s internal loan review system and assigned classification ratings, examiners are not required to validate the appropriateness of the bank’s internal loan grades and definitions for non-adversely classified loans. (Note: DSC’s Manual of Examination Policies states that "A loan review system should, at a minimum, include the following: A formal credit grading system that can be reconciled with the framework used by federal regulatory agencies; An identification of loans or loan pools that warrant special attention; A mechanism for reporting identified loans, and any corrective action taken, to senior management and the board of directors; and Documentation of an institution’s credit loss experience for various components of the loan and lease portfolio." In addition, DSC’s Manual of Examination Policies states that "An effective loan review system is generally designed to address the following objectives: To promptly identify loans with well-defined credit weaknesses so that timely action can be taken to minimize credit loss; To provide essential information for determining the adequacy of the Allowance for Loan and Lease Losses; To identify relevant trends affecting the collectibility of the loan portfolio and isolate potential problem areas; To evaluate the activities of lending personnel; To assess the adequacy of, and adherence to, loan policies and procedures, and to monitor compliance with relevant laws and regulations; To provide the board of directors and senior management with an objective assessment of the overall portfolio quality; and To provide management with information related to credit quality that can be used for financial and regulatory reporting purposes.") Reliance on bank management’s self-assessed ratings of non-adversely classified loans and on bank management’s reports that were prepared from that data may not be appropriate without performing independent verifications of the data’s accuracy.

As stated previously, the Loan Portfolio Management and Review: General Examination Documentation module, Core Analysis procedures, instructs examiners to validate the bank’s internal loan review system and assigned classification rating. Examiners are instructed to evaluate a cross section of loans by type, size, and severity of classification. This includes sampling watch list loans to assess rating accuracy and sampling loans not on the watch list to validate the internal loan review process. When sampling loans not on the watch list, examiners are directed to consider, in part, significant loans originated since the previous examination, new types of loans, and loan types exhibiting high rates of growth. Examiner guidance does not require examiners to validate the appropriateness of the bank’s internal loan grades and definitions for non-adversely classified loans.

From the 103 loans reviewed for the 15 banks that had experienced an annual loan growth rate of 40 percent or greater, the banks’ internal loan risk rating and the examiners’ assessment of loan quality were reconciled in only four instances. Each of these instances involved loans that were adversely classified (Substandard, Doubtful, or Loss). Also, one loan adversely classified and two loans classified as "Special Mention" by examiners were not reconciled with the bank’s internal risk rating. The remaining 96 loans reviewed were assigned a "Pass" classification by FDIC. Furthermore, examiners typically commented in the Reports of Examination on the internal loan review process in the context of bank management’s ability to accurately identify and grade loans that were adversely classified.

The FDIC’s examination guidance does not instruct examiners to validate the accuracy of the bank’s internal loan ratings for non-adversely classified loans. However, a comparison of the bank’s internal loan ratings for non-adversely classified loans with the examiners’ assessment of asset quality, by sampling and testing loans and criteria, could provide an early warning of deteriorating asset quality and increased credit risk. While not applicable to FDIC examinations, the Office of the Comptroller of the Currency’s (OCC’s) examination guidance recognizes within the Comptroller’s Handbook that

Effective risk identification starts with the evaluation of individual credits. Rating the risk of each loan in timely credit evaluations is fundamental to loan portfolio management… These evaluations allow the prompt detection of changes in portfolio quality, enabling management to modify portfolio strategies and intensify the supervision of weaker credits in a timely manner… In grading loans for supervisory purposes, the OCC uses five categories: pass, special mention, substandard, doubtful, and loss. Banks are encouraged to use these regulatory classifications as a foundation for their own risk rating systems. The OCC further encourages banks to expand their risk ratings for "pass" credits. Using multiple ratings to differentiate the risks of "pass" credits facilitates portfolio risk measurement and analysis, pricing for risk, and early warning objectives. The number of additional ratings used will vary from bank to bank and will depend on the bank’s own risk management objectives… After each loan has been risk rated, the ratings of individual credits should be reviewed, and they should be analyzed in the context of the portfolio segment and the entire portfolio. This analysis should ensure that ratings are consistently applied and should consider trends, migration data, and weighted average risk ratings. Risk ratings, when used in conjunction with other information (such as exception levels, past-due trends, and loan growth), can produce an instructive picture of asset quality and credit risk. Risk ratings can help the bank’s portfolio managers in other ways as well – when they set underwriting standards, asset diversification goals, and pricing levels, for example.

(Note: The Office of the Comptroller of the Currency issued the Loan Portfolio Management section of the Comptroller’s Handbook in April 1998. The handbook states that "This booklet, written for the benefit of both examiners and bankers, discusses the elements of an effective loan portfolio management process. It emphasizes that the identification and management of risk among groups of loans may be at least as important as the risk inherent in individual loans.")

The OCC guidance further instructs examiners, in part, to use testing to verify risk ratings and to test the lending function itself. For example, the Comptroller’s Handbook states that

a review of newly underwritten credits should be structured to assess the risk in the new transactions as well as to test the effectiveness of loan approval and other policies and processes that govern credit quality.

When new loan growth is identified as an emerging risk or as an area of "risk of greatest concern," then OCC examiners are expected, in part, to sample, test, and assess new loans and related managerial reports and controls. The FDIC does not have similar guidance.

DSC managers we interviewed did not see the need for examiners to reconcile the bank’s internal loan ratings for loans that were not adversely classified (Pass Loans) and expressed concerns over the lack of time and guidance to perform this type of review. They noted that the internal ratings are the bank’s own definitions, not the FDIC’s, and this type of review would be perceived as micromanaging the bank.

Management’s concerns of overburdening examiners and of examination time constraints would be limited to those institutions that had experienced a significant level of loan-growth. In particular, if a 40 percent annual loan growth factor were used, this analysis would be required during examinations in less than 10 percent of FDIC-supervised institutions. More conservatively, if a 25 percent annual loan-growth factor were used, this analysis would be required in approximately 16 percent of FDIC-supervised institutions. These percentages were based on the annual loan growth of all FDIC-supervised institutions for the year ending 2001. In addition, the percentages used were not selected or determined based on any existing standards or guidance; however, the range provided exceeds the definition established by the FDIC’s growth monitoring system for high-growth banks–banks whose quarterly asset and loan-growth rates were 5 percent or more.

The FDIC’s examination guidance does not encourage examiners to validate the accuracy of the bank’s internal loan ratings for loans that were not adversely classified (Pass Loans) with the examiners’ assessment of asset quality. Nevertheless, when implementing the risk-focused examination program, examination resources should be focused on those areas that pose the greatest risk to an insured institution. In particular, for those institutions that exhibit one or more high-risk indicators, such as significant loan growth and high concentrations in commercial real estate loans, examiners should perform an assessment of management’s ability to identify, measure, monitor, and control these risks. By testing the accuracy of the bank’s internal loan ratings for both adversely and non-adversely classified loans, examiners can validate management’s controls and processes and, if warranted, they can place greater reliance on the control systems in place. Examiners can then also use internal and external bank reports to facilitate examination analysis and conclusions on those factors identified as high-risk indicators, such as new loans in institutions that have experienced significant loan growth.

The fact that the bank’s internal loan ratings for non-adversely classified loans are assigned to the bank’s own loan grade categories and definitions does not negate the FDIC’s responsibility in assessing the quality of and adherence to the bank’s policies, procedures, and controls in high-risk areas. In particular, examiners should be required to validate, on a sample basis, the appropriateness of the bank’s internal loan grades and definitions for loans that are non-adversely classified when significant loan growth is determined to be a high-risk indicator. Furthermore, the adequacy and reliability of management’s controls and processes are a fundamental foundation in evaluating the soundness of the institution’s processes for managing risk. If management controls are not properly designed or effectively applied, then examiners should not place reliance on the bank’s control systems. Alternatively, examiners should consider expanding the scope of their review in these areas.

We recommend that the Director, DSC:

6. Revise existing policies and procedures to require examiners in their review of high loan-growth banks to perform a risk assessment of a bank’s internal loan risk rating process that is based on a methodology that incorporates a review of non-adversely classified loans.

CORPORATION COMMENTS AND OIG EVALUATION

On December 5, 2002, the DSC Director provided a written response to the draft report. The response is presented in Appendix III to this report. Prior to the receipt of DSC’s written response, DSC provided a draft of its written response on November 18, 2002. Based on the draft response, we provided written clarification to DSC on certain aspects of our audit report. We requested DSC to reconsider the draft report and its responses to our recommendations, especially in light of the principles of the risk-focused examination process. In its written response, DSC management did not concur with our recommendations, did not suggest acceptable alternative actions, and did not provide information that would convince us to revise any recommendations. In part, DSC stated in its written response that "Although we share many of the OIG’s underlying desires to identify and correct potentially harmful loan trends at early stages, we believe that existing practices, guidance, and procedures adequately address these issues."

Our audit results are largely driven by the underlying basic tenet of the risk-focused examination process. This process was designed to focus examination resources on bank activities that pose the greatest risk exposure to an institution. The program encourages less regulatory burden by focusing on testing, rather than duplicating, the work of audit and control functions. In particular, the risk-focused examination program encourages examiners to limit, or in some cases eliminate, traditional examination procedures in low-risk, well managed areas of the institution. Conversely, in our opinion, when implementing the risk-focused examination program for those institutions that exhibit one or more high-risk indicators, such as significant loan growth or high concentrations in commercial real estate loans, examiners should perform an assessment of management’s ability to identify, measure, monitor, and control these risks. In particular, when these high-risk indicators are present, more may need to be done than the traditional review process. In addition, these reviews should be documented and the analysis incorporated into the Reports of Examination as support for the examination ratings.

Also of note, the Division of Insurance and Research (DIR) April 2002 semiannual report entitled Economic Conditions and Emerging Risk in Banking notes that one of the four main risks to the Corporation is commercial lending in formerly fast-growing metropolitan areas. Our audit was conducted in four field offices that supervised institutions in five metropolitan areas where the commercial real estate markets were reported by the FDIC as potentially overbuilt. In addition, our sample for the audit work performed for Examiner Assessment of Commercial Real Estate Loans was targeted toward those institutions with concentrations of 300 percent or more of Tier 1 Capital in commercial real estate loans. The sample for this audit was targeted to a subset of those institutions that had also experienced an annual loan growth rate of 40 percent or more during the year prior to the "as of date" of the safety and soundness examinations we sampled. Because of these unusual circumstances, we expected to find evidence that examiners applied additional examination techniques commensurate with the increased risk. However, this was not the case.

Prior to responding to each of the report’s six recommendations, DSC stated that it had a number of significant concerns about the scope of the OIG audit that caused it to question the audit report’s assessments and conclusions. The concerns expressed by DSC are bulleted below, followed by the OIG’s response to those concerns, in italics.

• The audit did not include discussions with examination staff, including the examiner in-charge and examination loan manager.

Discussions were held with Field Office Supervisors and Supervisory Examiners regarding examination samples and how to interpret examination documentation. However, the individuals interviewed did not necessarily work on the specific examinations that were sampled. Nevertheless, they provided their expert opinions and assessment of individual credits, which we felt were adequate under the circumstances. If DSC would like to provide any additional information from the examination staff along with DSC's comments on the final report, we will be glad to consider this information at that time.

• The audit report did not note DSC’s outreach activities, periodic interagency meetings, and internal sessions where commercial real estate and high-growth programs are discussed.

This audit report was one of two reports written for this audit. The other report discussed the Regional Banker Outreach Program of the Dallas Regional Office. However, the objective of this report was to determine whether the examiners’ strategies for assessing a significant level of commercial real estate loan growth were sufficient for identifying increased risk. While outreach activities of the regional office and the periodic interagency meetings of senior management are valuable in achieving the goals of the Corporation, they do not directly relate to the analysis performed by the examiner during an examination.

• The audit sample consisted of only 15 examinations, seven of which were de novo banks (including one bank that recently completed its third year of operations), and three of which were financial institutions whose loan growth was primarily due to a recent merger.

The number of examinations reviewed was based on a targeted sample of high loan-growth institutions. The sample included examinations from four different field offices in two DSC regions.

The context of the source of the loan growth was reviewed by the audit team and discussed in the draft audit report. Furthermore, the diversity of the sources of growth speaks to the strength of the sample generated and to the validity of the audit findings. DSC states that increased supervisory review takes place in de novo banks and banks whose growth was primarily due to recent mergers. While increased supervisory review may take place in these instances, such review does not speak to the issues and concerns discussed in the audit report, and we did not observe any additional examination techniques employed in these situations. Specifically, our concerns focused on what analysis was done by the examiners themselves in assessing high loan growth.

De novo banks represent an increased risk due, in part, to the start-up of operations and are deserving of closer supervision. The presence of this increased risk heightens the significance of the examination process performed by examiners. Loan review is a part this examination process and thus validates the inclusion of these institutions within our sample.

Also, the sample only contained six de novo banks - not seven. As detailed within the audit report, "de novo banks are institutions that have been insured for less than 3 years." The seventh bank identified by DSC was not a de novo bank, because it had been insured for more than 3 years when the examination began.

Mergers also represent an increased risk to the institution due, in part, to the large acquisition of loans at one time. As noted in the audit report, examiners are directed by the Loan Portfolio Management and Review module to perform an analysis of this growth. The loan portfolio module states that "If the bank has acquired other institutions or loan portfolios, analyze the effect these purchases have had on the bank’s composition and risk profile." Similar to de novo banks, the presence of this increased risk heightens the significance of the examination process performed by examiners. Loan review is a part this examination process, and thus validates the inclusion of these institutions within our sample.

• The audit report was critical of the examination loan sampling descriptions in the pre-examination planning memoranda rather than the actual examination activities performed.

To clarify, we were critical of the examination loan sampling descriptions in the pre-examination planning memoranda, the loan review scope comment in the confidential section of the Report of Examination, and the actual examination activities performed, with respect to whether examiners performed a comprehensive review of new loans. We were not critical of the level of new loans actually reviewed, because we expected that new loans would have been captured for review through the normal sampling process. However, the fact that a new loan has been captured by the sampling process does not ensure that the examiners performed a more comprehensive review of these loans as a group. In particular, the report states that "inherent in the sampling process, it is probable that some new loans that exhibit high-risk profiles would be selected for review."

A review of the pre-examination memorandum and the Report of Examination loan scope comment depicts the examiner’s intent and focus of loan review as planned and as executed by the examiner. Thus, these are valid sources of information from which to draw our conclusions. The results of our review show that, despite having new loans in the sample, a more comprehensive review of these loans was not performed.

• Based on DSC’s analysis of the examinations contained in the audit sample, examiners did assess newly originated loans as part of the loans sampled.

We agree that new loans were in the sample; however, our concerns are centered on the analysis that was performed on those loans. The loan line sheets and ALERT records show that examiners sampled newly originated loans; however, those documents do not show that examiners assessed new loans as a group. Our audit found that new loans were not specifically targeted in pre-examination memoranda for 9 of 15 cases, and new loans sampled were not reported in the Reports of Examination for 10 of the 15 cases.

• Of the subsequent examinations discussed in the audit report, DSC noted that of the four institutions that were downgraded, two were de novo banks whose loan samples consisted of new loans, the third bank’s loan sample also contained new loans, and the fourth was a bank whose lending deficiencies were identified at an early stage during the institution’s growth period. DSC also stated that "In each of the four cases cited in the Draft Report as lacking review of newly originated loans, existing examination workpapers show that examiner assessment of newly originated loans was appropriate and thorough."

The audit report does not state that newly originated loans were not reviewed. Rather, the audit report states that the review performed was "lacking." Contrary to DSC’s statement, existing examination workpapers do not show that examiner assessment of newly originated loans was appropriate and thorough.

DSC notes that two of the four banks were de novo banks and the loan samples were exclusively newly originated loans. The fact that these two de novo banks were downgraded in subsequent examinations emphasizes the validity of including de novo banks in the audit sample. This fact also supports our concerns detailed in the draft audit report over the lack of a comprehensive review of these loans. In one of these two cases, asset quality declined from a 1 to a 4.

For the third example, the composite rating dropped from a 1 to a 2; however, asset quality dropped from a 1 to a 3. This is a significant decline. DSC again states that newly originated loans were reviewed. We agree that newly originated loans were reviewed; the report does not state that newly originated loans were not reviewed. Our concern is centered on the lack of analysis performed on those loans. For example, as noted in the audit report, no discussion was presented in the Report of Examination on the quality and trend of newly originated loans. Furthermore, the first examination noted that the bank’s underwriting standards were strong based on the aggregated assessment of asset quality and that the internal grading system was satisfactory based on its agreement with examiner classifications for those loans that were adversely classified. Conversely, the subsequent examination determined that underwriting practices were weak and that the internal loan review procedures were inadequate.

For the fourth example, DSC discussed the bank’s decline from a composite rating of 3 to a 4. To reiterate the bank’s history, this bank was rated a composite 3 for the examination previous to the one selected for our review. When the bank was first rated as a composite 3, it was placed under a Board Resolution. The bank’s growth had been initiated under a branch expansion program, and the bank grew the loan portfolio over 50 percent in 1 year from the examination "as of date." While the Report of Examination we reviewed did discuss why the growth was occurring and how the growth was achieved, the report did not discuss the quality of the newly originated loans. In addition, the bank’s asset quality rating and composite rating remained the same as the previous examination. In short, the lending deficiencies were not identified at an early stage during the institution’s growth period. In the subsequent examination, the composite rating and the component rating for asset quality were downgraded to a 4. The Summary Analysis of Examination Report comment for this subsequent examination states that "The increase in classified assets [is] attributed to further deterioration in credits originated by prior management team…few commercial loans have been booked since the prior examination making complete evaluation of new lending team difficult." This implies that the new loans originated during the previous year of the first Report of Examination reviewed (the period of high loan-growth) migrated into more severe classification status. The weaknesses that were attributed to these credits could have potentially been noted earlier if a more thorough review of the newly originated loans had been performed.

DSC Responses to OIG Recommendations

DSC did not concur with any of our six recommendations. All of these recommendations are considered unresolved, undispositioned, and open. A summary of each recommendation and DSC’s comments follow, along with the OIG’s evaluation of the response.

Recommendation 1: Revise policies and procedures to define the term "significant loan growth" and to require examiners to target and specifically sample new loans for examination when a financial institution has experienced significant loan growth since the last full-scope safety and soundness examination.

DSC did not concur with this recommendation. DSC disagreed that policies and procedures need revision to define significant loan growth and to target the review of such loans. In addition, DSC stated that a strict definition of "Significant Loan Growth" may hinder the identification of risk in individual banks and reduce examiner discretion to risk focus. Furthermore, DSC stated that new loans are a part of the loan scope any time a bank is originating or purchasing new loans, regardless of growth or constriction.

The OIG did not recommend that a "strict" definition should be established. However, guidance does need to be established that provides examiners a basis of reference from which to formulate decisions. In particular, we do recommend that a percentage or a range be utilized that would prompt examiners to consider the risk of significant loan growth in the scope of their examination without eliminating the examiners’ discretion to risk focus their examination as needed. More importantly, as DSC has stated, "With the high number of variables, examiner judgment is a critical component in the assessment of risk and development of an appropriate sample." It is precisely for this reason, the number of variables and the complexity in defining significant loan growth, that guidance should be provided to examiners that discusses how to determine what is significant loan growth. These variables should be detailed and discussed as part of the defining process in order to clarify to examiners the issues that need to be considered, which would allow them to effectively employ "examiner discretion."

As our audit report states, DSC’s policies and procedures do not define the term "significant loan growth," nor do these policies instruct examiners how to determine what constitutes significant loan growth.

Further, the audit report states, "examination reports do not appear to answer two key questions:

• Did the bank experience significant loan growth by loosening its underwriting standards?
• If so, does the bank have an adequate system in place to identify, measure, monitor, and control this increased risk?"

In addition, the audit report states, "Based on a review of the asset quality comments in the Reports of Examination, only 20 percent (3 of 15) of the comments discussed the cause of the growth beyond the notation that either loan growth or a merger occurred."

While DSC’s policies require examiners to sample new loans, our audit shows that in over one-half of the examinations reviewed (9 of 15 examinations), examiners did not initially target new loans for sampling purposes. Based on a review of the pre-examination memoranda, in nine cases no mention was made of including newly originated loans within the loan sample. This indicates that those examiners did not initially target new loans for review purposes, nor were they considering those loans for a more comprehensive review and assessment process than that of a normal loan review.

Recommendation 2: Revise policies and procedures to require examiners to report on new loans sampled for review purposes when a financial institution has experienced significant loan growth since the last full-scope safety and soundness examination.

DSC did not concur with this recommendation. DSC stated that "Examiners routinely review newly originated loans as part of their loan sample and additional documentation of such reviews is unproductive. While policies could be revised to require that the new loan sample size be included in the loan scope comment currently on the A-page of the Report of Examination, regardless of whether significant growth has occurred, we do not find it necessary at this time."

While examiners may routinely review newly originated loans as part of their loan sample, the review performed for high loan-growth institutions that we reviewed was lacking and more needed to be done. In particular, when significant loan growth is a high-risk factor, the consideration provided and the analysis performed by the examiner should be documented. Without this analysis and documentation, there is limited assurance that the institution has been accurately evaluated.

Based on the results of our audit, in two-thirds of the examination reports reviewed (10 of 15 examination reports), examiners did not note that the scope of the loan review targeted newly originated loans.

Recommendation 3: When loan growth is a high-risk factor, clarify existing policies and procedures to specifically detail how examiners should assess and comment on: (a) the loan quality of newly originated loans, (b) the loan review system, and (c) loan underwriting.

DSC did not concur with this recommendation. DSC stated that it believed that its policies and procedures were adequate, examiners were well aware of the risks inherent in high-growth institutions and employ reasonable sampling techniques, and the examiners’ determination of the rating for the Asset Quality component rating requires consideration of existing practices in loan administration and loan underwriting.

As presented in our audit report, "While examiners typically commented on the internal loan review process, the comments were in the context of bank management’s ability to accurately identify and grade loans that were adversely classified. No comments were made on the adequacy of the bank’s internal loan ratings of non-adversely classified loans." This was not a sufficient review of the risk present in a high loan-growth institution.

Further, as presented in our audit report, "Examiners also typically commented on the adequacy of loan underwriting, but the examiners’ conclusions had varying levels of support." No underwriting comments were made or supported in two of the reports (13 percent), three of the reports supported underwriting comments based on the level of documentation exceptions (20 percent), four of the reports supported underwriting comments based on the adequacy of and/or adherence to policy guidelines (27 percent), and five of the reports supported underwriting comments based on the assessed quality of assets on an aggregated basis (33 percent). In only one of the reports was support for underwriting comments based on the bank’s loan analysis at origination (7 percent). This was clearly not a sufficient review of the risk present in a high loan-growth institution.

Most significantly, in all 15 Reports of Examination reviewed, examiners did not fully comment on the overall quality and trend of newly originated loans in the portfolio. In particular, examiners did not document an assessment of the loan quality of newly originated loans (loans originated since the last examination) in comparison to loans originated prior to the last examination. Further, examiners did not always comment on why the growth was occurring and how the level of loan growth was achieved. Based on a review of the asset quality comments in the Reports of Examination, only 20 percent (3 of 15) of the comments discussed the cause of the growth beyond the notation that either loan growth or a merger occurred. Again, this was clearly not a sufficient review of the risk present in a high loan-growth institution.

Recommendation 4: When loan growth is a high-risk factor, clarify existing policies and procedures to detail how examiners could incorporate a loan migration study into the assessment of loan quality and underwriting.

DSC did not concur with this recommendation. DSC stated that "The OIG recommended process, or portions of the recommendation, already exist as part of the Allowance for Loan and Lease Losses (‘ALLL’) analysis, loan underwriting review, and the Loan Underwriting Survey." DSC also stated that "It is not proven and it is not apparent that expanding focus to include gradations of ‘pass’ loans will generate conclusions any more accurate, meaningful, or supportable than those presently derived. Nor is it likely the regulatory response or corrective measures implemented by management will be more effective than actions precipitated by review of ‘Watch List’ and ‘Special Mention’ loans."

We recognized in our audit report that the examiners are directed to assess the adequacy of the allowance for loan and lease losses and that examiners are also instructed to determine if management considers any factors that are likely to cause estimated loan losses to differ from historical loss experience. We also recognized that DSC has provided guidance to examiners on assessing the adequacy of the bank’s allowance for loan and lease losses through various sources, including, but not limited to, the Examination Documentation modules, DSC’s Manual of Examination Policies, Regional Directors Memoranda, and FDIC Statements of Policy. However, DSC’s policies and procedures do not specifically detail how examiners should make the determination that estimated loan losses might differ from the bank’s historical loss experience and from the bank’s assumptions used within its methodology of determining the allowance for loan and lease losses.

While DSC’s policies do mention the use of migration analysis, this guidance is lacking in providing how this analysis should be implemented and, more specifically, how it should be used to assess loan quality and underwriting. DSC states that "It is not proven and it is not apparent that expanding focus to include gradations of 'pass' loans will generate conclusions any more accurate, meaningful, or supportable than those presently derived." We disagree; we presented the FDIC’s own internal study that shows high loan growth is a high-risk factor, and the audit report provided examples to illustrate that potential deterioration was not identified in the examinations that were sampled. Further, the audit report provides examination guidance published by the Office of the Comptroller of the Currency that recognizes the value of this analysis. Based on the results of our audit and the limited analysis performed by examiners on the risk present in high loan-growth institutions, more guidance is needed in this area.

Recommendation 5: Re-emphasize to examiners the need to assess and report on management’s processes for controlling risk when potential high-risk indicators are present.

DSC did not concur with this recommendation. DSC stated that "Examiners appropriately assess and report on bank management’s risk management policies and practices in the report of examination. Much of this recommendation is already covered by the ED modules and other instructions for assessment and reporting of risk areas." In addition, DSC stated that it has numerous other mechanisms to assist in identifying risk factors, and individual Reports of Examination receive review from various supervisory levels which helps ensure that risk areas are appropriately addressed.

Based on our audit, there was insufficient assurance that examiners were consistently performing a comprehensive review and analysis of newly originated loans in high loan-growth institutions. When examiners identify a high-risk indicator, such as significant loan growth, the level of review and analysis performed should be identified in the Report of Examination to support examination conclusions. Further, reliance on traditional review techniques used to assess the loan portfolios of institutions that have not experienced significant loan growth is not sufficient to measure and assess the potential risks to high-growth institutions.

Recommendation 6: Revise existing policies and procedures to require examiners in their review of high loan-growth banks to perform a risk assessment of a bank’s internal loan risk rating process that is based on a methodology that incorporates a review of non-adversely classified loans.

DSC did not concur with this recommendation. DSC stated that "The proposed process is already performed during the loan review in which the vast majority of loans in the sample are non-adversely classified loans." In addition, DSC stated that "Asking examiners to take additional time to decide if a loan fits the bank's definition of high quality or very high quality does not provide meaningful data."

We recognize that DSC is currently sampling non-adversely classified loans for review and that DSC generally validates the bank’s loan grades against regulatory definitions of Substandard, Doubtful, and Loss. However, DSC is not assessing or validating the bank’s assignment of loans into the bank’s various internal loan grades for loans that are not adversely classified. Also, the assessment of the allowance for loan and lease losses does not capture this type of analysis. A comparison of the bank’s internal loan ratings for non-adversely classified loans with the examiners’ assessment of asset quality, by sampling and testing loans and criteria, could provide an early warning of deteriorating asset quality and increased credit risk. Further, by testing the accuracy of the bank’s internal loan ratings for both adversely and non-adversely classified loans, examiners can validate management’s controls and processes and, if warranted, they can place greater reliance on the control systems in place. Examiners can then also use internal and external bank reports to facilitate examination analysis and conclusions on those factors identified as high-risk indicators, such as new loans in institutions that have experienced significant loan growth.

Because all recommendations in this report are unresolved, undispositioned, and open, we have requested DSC to reconsider its response to our report and provide us additional comments.

OBJECTIVES, SCOPE, AND METHODOLOGY

The objectives of this audit were to determine whether: (1) the examiners fully assessed appraised value, cash flow, and lending policies in their examination of CRE loans and (2) the examiners’ strategies for assessing a significant level of CRE loan growth were sufficient for identifying increased risk. While our audit addressed both objectives, the subject matter and results were distinct enough that we have prepared separate reports to address each objective. This audit report addresses our observations with regard to objective (2) above and covers our assessment of examiner analysis of institutions that have experienced a significant level of loan growth. To address this objective, we assessed certain aspects of DSC’s loan review process (such as loan sampling, loan quality assessment, and internal loan risk rating reconciliation) that DSC examiners use during safety and soundness examinations, to evaluate institutions that have experienced a significant level of loan growth.

To address our objective, as discussed in this report, we selected various examinations to review from the original sample generated during the audit of Examiner Assessment of Commercial Real Estate Loans. The original sample consisted of 248 loans and 35 banks that were identified as having CRE portfolios of 300 percent or more of Tier 1 Capital. The banks selected for review were located in Seattle, Phoenix, Las Vegas, Dallas, and Denver–all metropolitan areas that the FDIC had identified as potentially overbuilt in the CRE sector. From this sample, we selected all institutions that had experienced an annual loan growth rate of 40 percent or greater during the year prior to the "as of date" of the safety and soundness examinations that were originally sampled. From this selection, we reviewed 103 loans from 15 banks in the San Francisco and Dallas Regions. Six institutions experienced loan growth through the expansion of the bank’s existing market; six institutions were de novo banks (new entries into the market); and three institutions experienced loan growth through the acquisition of or merger with other banks, branches, or loan portfolios (expansion into new market areas/locations).

The audit was conducted in accordance with generally accepted government auditing standards. We focused our review on examinations that had been performed during the period of September 1999 through April 2001. The audit fieldwork was conducted from April 2001 through July 2002. We performed fieldwork in Washington, D.C., the DSC San Francisco and Dallas regional offices, and four field offices (Seattle, Phoenix, Dallas, and Denver) located in the San Francisco and Dallas regions.

Our fieldwork entailed:

• reviewing pre-planning memoranda, examination working papers, Reports of Examination, and other miscellaneous managerial reports;
• reviewing examiner analysis of loan files to include loan policies, loan line sheets, and report commentary;
• reviewing applicable laws, regulations, and statements of policy;
• reviewing relevant sections of DSC’s Manual of Examination Policies, Regional Directors Memoranda, Examination Documentation modules, the OCC’s Comptroller’s Handbook, and other miscellaneous information resources and studies;
• reviewing Uniform Bank Performance Reports and Consolidated Reports of Condition and Income; and
• interviewing DSC Washington senior management, San Francisco and Dallas regional management, Field Office Supervisors, and examiners. In particular, discussions were held with the Associate Director of Operations, the Regional Directors of the San Francisco and Dallas regions, and the Deputy Regional Director of San Francisco.

The limited nature of the audit objective did not require reviewing performance measures, testing for fraud or illegal acts, testing for compliance with laws and regulations, or determining the reliability of computer-processed data obtained from the FDIC’s computerized systems. Our assessment of internal management control was limited to a review of DSC’s applicable policies and procedures as presented in DSC’s Manual of Examination Policies, Regional Directors Memoranda, and Examination Documentation modules, and our review of the implementation of these policies and procedures in the course of selected safety and soundness examinations.

LIFE CYCLE OF A BANK FAILURE

History of the Eighties–Lessons for the Future, Volume I, An Examination of the Banking Crises of the 1980s and Early 1990s, published in December 1997 and prepared by the FDIC’s former Division of Research and Statistics details the life cycle of a bank failure, as presented on pages 487 to 488:

In interviews with bank and thrift regulators, rapid loan growth was identified again and again as a precursor to failure. Whether or not loan growth is the primary risk in which banks engage, one regulator’s description of a three-phase process by which rapid loan growth evolves into a major problem does a good job of laying out the long-term nature of the development of a bank’s financial distress.

In the first stage, there is rapid loan growth; loan concentrations emerge, and lending is aggressive (internal controls in the growth areas are weak, and underwriting standards are lenient). The increased lending may be, but is not always, funded by a volatile lending source. This growth could occur throughout the entire institution or within a specific asset type. If the growth is in a specific asset type, the increase could stem either from growth in concentration in a loan category or from a shift into a new activity, with subsequent growth. If the rapid growth draws the attention of the relevant regulator, management usually points to the excellent earnings and contribution to capital that the growth has provided. This stage of the development of the problem can take up to two years.

In the second stage, the institution has rising loan-quality problems. Associated expenses may far exceed industry averages. Nonrecurrent sources of income are used to maintain the same level of profits that existed during the growth phase. Eventually profits begin to decline, and inadequate reserve levels become apparent. At this point the bank may be "loaned up" (that is, have a high loans-to-assets ratio). Management may still believe that the problem is manageable. This stage may take an additional one to two years.

In the final stage, deteriorating asset quality is a serious problem. The institution is incurring large loan losses, and charge-offs have increased. If the institution is large, the capital markets have recognized that the institution has inadequate loan-loss reserves and are unwilling to provide fresh capital. At this point, major changes in the bank’s operations are necessary. Dividends may be cut, expenses (mostly personnel) are slashed, and assets are sold to cover charge-offs and operating expenses (especially in larger institutions). This crisis phase may last up to a year and results either in the failure of the bank or, if dramatic and fundamental changes are made, in its eventual recovery.

As this account of the life cycle of failure makes clear, only in the course of years do changed behavior and the acceptance of greater risk lead to financial distress or failure. After all, neither growth itself nor most other risk taking is necessarily bad for a financial institution. Banks earn their income by assuming risk; to increase risk through growth can therefore be a sound strategy. Such a strategy would ideally be accompanied by increases in capital as a buffer against higher losses, maintenance of high underwriting standards, and attention to proper risk management–in other words, by prudent management of the institution’s growth. Moreover, regardless of whether the increased lending is prudent, ill timed, or very risky, the growth will generate added revenue from increased loan fees and interest income. In addition, because these are all new loans, initially there are no delinquencies and no loss charge-offs, so that the growth is almost always accompanied by growth in income and capital (assuming retained earnings). Only over time do the effects of growth or other risk taking–whether these effects are good or bad–become apparent. This long lead time before problems appear makes it difficult to identify future problem banks accurately.

CORPORATION COMMENTS

Federal Deposit Insurance Corporation
550 17th St. NW Washington DC, 20429
Washington, DC 20429
Division of Supervision and Consumer Protection

December 5, 2002

TO: Stephen M. Beard, Deputy Assistant Inspector General for Audits, Office of the Inspector General

FROM: Michael J. Zamorski [Electronically produced version; original signed by Michael J. Zamorski], Director, Division of Supervision and Consumer Protection

CONCUR: John F. Bovenzi [Electronically produced version; original signed by John Bovenzi], Deputy to the Chairman and Chief Operating Officer

SUBJECT: Draft Report Entitled "Examiner Assessment of High Loan-Growth Institutions" (Assignment No. 2003-003)

This memorandum represents the Division of Supervision and Consumer Protection’s ("DSC") response to the Office of Inspector General’s ("OIG") draft report entitled "Examiner Assessment of High Loan-Growth Institutions" ("Draft Report"). This Draft Report is the OIG’s second phase of its review of DSC’s assessment of commercial real estate loans in the course of safety and soundness examinations.

DSC is committed to ensuring that examiners carefully and accurately assess loan quality in FDIC-supervised financial institutions, particularly financial institutions experiencing high loan-growth. Although we share many of the OIG’s underlying desires to identify and correct potentially harmful loan trends at early stages, we believe that existing practices, guidance, and procedures adequately address these issues.

The stated objective of the OIG’s audit was to determine whether the examiners’ strategies for assessing a significant level of commercial real estate loan growth were sufficient for identifying increased risk. The OIG auditors concluded that they had insufficient assurance that examiners are consistently performing a comprehensive review and analysis of newly originated loans in high loan-growth institutions. Further, the OIG concludes that, due to their perception that there is inconsistent review and analysis performed, DSC supervision appears reactive rather than proactive.

DSC has a number of significant concerns about the scope of the OIG audit, which causes us to question the Draft Report’s assessments and conclusions. More specifically, the OIG audit did not include discussions with examination staff, including the examiner in-charge and examination loan manager (the examiner who led the loan review process). Discussions with these individuals may have resulted in a greater understanding of the examination process. In addition, DSC participates in numerous outreach activities, periodic interagency meetings, and internal sessions where commercial real estate and high-growth programs are discussed. None of these activities or programs is noted in the OIG Draft Report.

Further, the Draft Report’s audit sample consists of only 15 examinations. These examinations were of financial institutions where loans increased by 40 percent or more over the prior year. However, it is important to understand the context of the source for much of this growth. For example, seven of the financial institutions (or 47 percent) in the sample were de novo banks (including one bank that recently completed its third year of operations). De novo banks usually experience such growth rates initially, and special supervisory programs exist for these newly chartered financial institutions. The seven de novo banks included in the sample ranged in asset size from $32.3 million to $61.7 million and their growth was consistent with regulatory-approved plans.

The special scrutiny of de novo banks is a key element in our supervisory program. Examiner assessments of these financial institutions encompass comparisons with business plans and strategic projections submitted with the deposit insurance application, as well as a rigorous examination and onsite program where de novo institutions are evaluated more frequently than other categories of financial institutions. Newly originated loans are assessed on regular examinations and visitations in a comprehensive manner. Deposit insurance applications consider business plans and must favorably resolve seven statutory factors (including the general character and fitness of management). These factors may be found in Section 6 of the FDI Act.

The audit sample also included three financial institutions (or 20 percent) in which the loan growth was primarily due to recent mergers that were approved by the FDIC. In such cases, the expansion of loans largely results from the combination of loans originated by other financial institutions that were subject to onsite examinations. During the FDIC’s assessment of a merger application, we consider the asset quality of the targeted institutions, the qualifications of the anticipated management team, the risk management systems, and the deposit insurance risks.

The Draft Report (page 15) states "… in over one-half of the examinations reviewed (9 of 15 examinations), examiners did not initially target new loans for sampling purposes." The number of examinations that did not capture new loans for sampling purposes is not provided, but instead the Draft Report focuses on the pre-examination planning memoranda description of loan sampling. The OIG appears critical of the examination loan sampling descriptions in the pre-examination planning memoranda rather than the actual examination activities performed. Examiners use information garnered from the review of loans to assess the effectiveness and use of underwriting policies and procedures, and if the policies and procedures result in high quality loans. Also, information obtained from the loan review process is used to assess a financial institution’s risk management practices, as described on the Risk Management Assessment pages of the Report of Examination.

We have reviewed the existing records of the banks in the sample to better understand the amount of newly originated loans that were reviewed by examiners during the safety and soundness examinations. While workpapers for some of these examinations are no longer available, our review of available information shows that examiners did assess newly originated loans as part of the loans sampled. FDIC examiners routinely review newly originated loans as a basic examination procedure. DSC’s Loan Examination Module states that significant loans extended since the past examination and new types of loans should be included in a loan sample by examiners. The Regional Director Memorandum, entitled Loan Review, dated September 12, 2001, reminds examiners of the importance of loan reviews and the loan sampling process, including the assessment of newly originated loans. Loan line sheets document the origination date of loans and the Automated Loan Examination Review Tool ("ALERT") provides useful parameters to target newly originated loans. We believe that the loan line sheets and ALERT records show that FDIC examiners target and sample newly originated loans.

The table below lists the fifteen banks included in the Draft Report sample. The table shows the percentage of total loans reviewed by examiners, and the percentages of those loans that were newly originated loans, based on available information.

Table: Banks Included in the Draft Report Sample

Examination workpapers on 10 of the 15 examinations reviewed in the Draft Report are available. These documents show that examiners included a sizeable amount of newly originated loans in their loan samples. In nine of the 10 cases, at least half of the loan sample (either by dollar or number) consisted of newly originated loans and in the other case a meaningful level of new loans was also reviewed.

The Draft Report discusses subsequent examinations (page 26 through page 32) and found that four banks’ CAMELS ratings were subsequently downgraded. The Draft Report asserts that loan quality weaknesses in these four banks could have been detected "…earlier if a more thorough review of the newly originated loans had been performed." Two of the four banks were de novo institutions, in which loan samples were exclusively newly originated loans. In the third case, the bank’s composite rating declined from "1" to "2," during the OIG audit period. However, over half of the loans reviewed at the examination consisted of newly originated loans. In the fourth case, the bank’s composite rating was downgraded from a "3" to a "4." That bank’s growth was planned under an approved branch expansion program. Based on the findings of the prior examination, the bank was placed under a Board Resolution. Four of the nine provisions dealt with asset quality, credit administration, and loan policy. During the following examination, examiners determined that underwriting and credit administration remained troublesome. The Board Resolution was replaced with a Memorandum of Understanding, which limited growth, prohibited lending to classified borrowers, and required strengthening of lending policies and procedures. Therefore, in the fourth bank case, lending deficiencies were identified at an early stage during the institution’s growth period, and it became subject to close and increasing supervision. In each of the four cases cited in the Draft Report as lacking review of newly originated loans, existing examination workpapers show that examiner assessment of newly originated loans was appropriate and thorough.

OIG’s Recommendations

The following six recommendations are proposed by the OIG in the Draft Report and DSC’s response is included below:

(1) – Revise policies and procedures to define the term significant loan growth and to require examiners to target and specifically sample new loans for examination when a financial institution has experienced significant loan growth since the last full-scope safety and soundness examination.

DSC does not concur with this recommendation. We disagree that polices and procedures need revision to define significant loan growth and to target the review of such loans. A strict definition of "Significant Loan Growth" may hinder the identification of risk in individual banks and reduce examiner discretion to risk focus. What is significant will be different for each bank. Also, there could be significant loan growth in one segment of the portfolio, which one might want to specifically address. With the high number of variables, examiner judgment is a critical component in the assessment of risk and development of an appropriate sample. Attempting to tightly define significant loan growth, then requiring certain actions could misdirect examiner resources and be detrimental to an effective examination.

New loans are a part of the scope of loan review any time a bank is originating or purchasing new loans, regardless of growth or constriction. Moreover, new loans are captured within loan scopes through other criteria (size, internal rating, and loan type). Whether to sample, or the size of the new loan sample, should not be based on an arbitrary percentage of growth but instead on examiner judgement. This judgement should be applied in the context of management, lending controls, and the types of new loans being originated or purchased. For example, if significant loan growth occurred as a result of the purchase of a consumer loan portfolio, the new loan sample size generally would be managed differently than for the purchase of a commercial real estate loan portfolio. Similarly, if controls over credit administration were strong at the previous examination, the sample size would be managed differently than if controls were unsatisfactory.

DSC’s policies already require examiners to sample loans not on a bank’s watch list and newly originated credits. The Loan Examination Module (ED), Question 38 A2, states that the examiner should "Sample loans NOT on the watch list [i.e., pass loans] to validate the internal loan review process, considering . . . 4. Significant credits originated since the previous examination. 5. New types of loans . . . "

(2) – Revise policies and procedures to require examiners to report on new loans sampled for review purposes when a financial institution has experienced significant loan growth since the last full-scope safety and soundness examination.

DSC does not concur with this recommendation. Examiners routinely review newly originated loans as part of their loan sample and additional documentation of such reviews is unproductive. While policies could be revised to require that the new loan sample size be included in the loan scope comment currently on the A-page of the Report of Examination, regardless of whether significant growth has occurred, we do not find it necessary at this time.

(3) – When loan growth is a high risk factor, clarify existing policies and procedures to specifically detail how examiners should assess and comment on (a) the loan quality of newly originated loans, (b) the loan review system, and (c) loan underwriting.

DSC Response:

DSC does not concur with this recommendation. We believe that our policies and procedures adequately describe how to assess a bank’s loan quality, its loan review systems, and shifts in underwriting, and that further specific guidance for high-growth institutions is not required. Examiners are well aware of the risks inherent in high-growth institutions and employ reasonable sampling techniques to monitor this growth while still employing a risk-focused approach to the supervision of these institutions. In addition, the examiners determination of the rating for the Asset Quality component rating requires consideration of existing practices in loan administration and loan underwriting and is not solely based on the current condition of the loan portfolio.

(4) – When loan growth is a high risk factor, clarify existing polices and procedures to detail how examiners could incorporate a loan migration study into the assessment of loan quality and underwriting.

DSC does not concur with this recommendation. The OIG recommended process, or portions of the recommendation, already exist as part of the Allowance for Loan and Lease Losses ("ALLL") analysis, loan underwriting review, and the Loan Underwriting Survey.

The core ED modules clearly direct the examiner to review the ALLL and consider the bank's loan loss history but provides appropriate leeway as to when migration analysis is required. Also, the expanded ED module for loan procedures, Question 34C states that the ALLL review should consider loan loss history, and Question 34C, item 3 specifically asks that loan migration be considered. In addition, the Interagency Policy Statement on the ALLL, 12-21-93 (page 2), states that the bank's methodologies can range from simple calculations "… to more complex techniques, such as migration analysis."

Implementing this recommendation unnecessarily increases the burden on small banks and adds little value to the examination process. It is not proven and it is not apparent that expanding focus to include gradations of "pass" loans will generate conclusions any more accurate, meaningful, or supportable than those presently derived. Nor is it likely the regulatory response or corrective measures implemented by management will be more effective than actions precipitated by review of "Watch List" and "Special Mention" loans.

(5) – Re-emphasize to examiners the need to assess and report on management’s processes for controlling risk when potential high risk indicators are present.

DSC does not concur with this recommendation. Examiners appropriately assess and report on bank management’s risk management policies and practices in the report of examination. Much of this recommendation is already covered by ED modules and other instructions for assessment and reporting of risk areas. In addition, DSC has numerous mechanisms to assist in identifying risk factors (Growth Monitoring System and Statistical CAMELS Offsite Rating System, for example) and individual Reports of Examination receive review from Field Office Supervisors, Case Managers, Assistant Regional Directors, and others which helps assure that risk areas are appropriately addressed.

(6) – Revise existing policies and procedures to require examiners in their review of high loan growth banks to perform a risk assessment of a bank’s internal loan risk rating process that is based on a methodology that incorporates a review of non-adversely classified loans.

DSC Response:

DSC does not concur with this recommendation. The proposed process is already performed during the loan review in which the vast majority of loans in the sample are non-adversely classified loans. Examiners compare the bank’s internal loan grades with the examiner-assigned grades on all credits reviewed. The ALLL analysis prepared by the examiners also usually includes a stratification of the loan portfolio by credit grades and an assessment of the actual reserve percentages allocated for each internal loan grade assigned by the bank. Internal loan ratings are used to identify higher-risk loans and assign risk weights used in an assessment of the ALLL. Examiners make adjustments to these percentages, as warranted by examination findings, in the preparation of the ALLL analysis.

The ALERT menu provides a means for selecting variable cut-off limits for all loans, and examiners typically use a lower cut off for the bank's Special Mention and Watch List loans (lower quality pass loans). Asking examiners to take additional time to decide if a loan fits the bank's definition of high quality or very high quality does not provide meaningful data.