FR EA
1045-00007 · Client 1045 · Witness Preparation
witness-prep-memo-hu.docx DOCUMENT Litigation
34 / 35 ← → ✕
Contract analysis Extract to table

ATTORNEY-CLIENT PRIVILEGED ATTORNEY WORK PRODUCT CONFIDENTIAL

This memorandum is a privileged and confidential attorney-client communication and attorney work product. It was prepared by counsel for Calderwood & Harkness LLP in connection with the firm's representation of Belmont Ridge Bank National Association in SEC Investigation File No. HO-14287. This memorandum is not intended for distribution to any person other than the named recipients and authorized members of the engagement team. Any unauthorized review, use, disclosure, or distribution is strictly prohibited.

February 12, 2024

INTERNAL MEMORANDUM

TO: Victoria Chambers, Partner Howard Quinn, Partner

FROM: Jasmine Thornton, Lead Associate Liam Donahue, Working Associate

RE: Witness Preparation — Raymond Hu, Chief Accounting Officer, Belmont Ridge Bank NA; Investigative Testimony Scheduled February 14, 2024

MATTER: SEC Investigation File No. HO-14287 / C&H Engagement No. SC-MAT-2023-4172

Right-click to update Table of Contents

I. Executive Summary

Raymond Hu, Chief Accounting Officer of Belmont Ridge Bank National Association ("BRB" or the "Bank"), is scheduled to provide investigative testimony before the Staff of the Division of Enforcement of the U.S. Securities and Exchange Commission ("SEC" or the "Commission") on February 14, 2024, at 9:30 a.m. EST. This testimony is being conducted pursuant to the Formal Order of Investigation No. EN-2023-0947, which authorizes the Division of Enforcement to examine, among other matters, BRB's allowance for credit losses ("ACL") methodology and associated financial reporting for the fiscal year ended December 31, 2022, and subsequent interim periods.

The testimony will take place at the SEC's principal offices, located at 100 F Street NE, Washington, D.C. 20549. Based on prior correspondence with the Staff and patterns observed in earlier testimonial sessions in this investigation, the SEC Staff members expected to be present include: Andrea Felton, Senior Staff Attorney; Christopher Yuen, Staff Attorney; and Diane Milosevic, Staff Accountant. In addition, there is a reasonable possibility that Assistant Director Gregory Bonham and/or Branch Chief Renata Moss may attend some or all of the session, particularly in light of the fact that this is the third and final scheduled testimonial appearance in the investigation. The potential presence of more senior Staff personnel is consistent with the investigation moving toward a recommendation stage and warrants heightened preparation.

Calderwood & Harkness LLP ("C&H") will represent BRB in connection with the testimony. Victoria Chambers, Partner, and Jasmine Thornton, Lead Associate, will attend as counsel for the Bank. It is important to note — and to reconfirm with Mr. Hu during the final preparation session — that C&H represents BRB the entity, and not Mr. Hu in his individual capacity. An Upjohn warning was previously administered to Mr. Hu during his initial interview on October 25, 2023, in accordance with the firm's standard protocol for corporate representation in government investigations. The Upjohn warning will be re-confirmed during the final preparation session scheduled for February 13, 2024, to ensure that Mr. Hu continues to understand the scope and limitations of the attorney-client relationship.

This memorandum follows two prior witness testimonies in this investigation. Patricia Navarrete, BRB's Chief Financial Officer, provided investigative testimony on January 18, 2024. Naveen Chakravarti, Senior Vice President and Head of the BridgeFin Division, provided investigative testimony on January 30, 2024. Post-testimony debriefs were conducted following each appearance, and the observations and lessons learned from those debriefs have been incorporated into this memorandum. In particular, the Staff's areas of focus during the Navarrete and Chakravarti sessions — which centered on the treatment of re-aged and modified BridgeFin loans and the vintage-cohort loss-rate methodology — provide a reliable preview of the questioning Mr. Hu should expect to face.

The key focus areas for Mr. Hu's testimony are anticipated to include: (i) the ACL estimation process under ASC 326 (Current Expected Credit Losses, or "CECL"); (ii) the quarterly true-up procedures for the ACL; (iii) the accounting treatment of the 12,847 re-aged and modified BridgeFin loans with an aggregate outstanding balance of approximately $236.4 million; (iv) the nature, magnitude, and documentation of qualitative overlay adjustments to the quantitative CECL model output; and (v) Mr. Hu's communications with BRB's independent auditor, Whitfield Keane LLP (engagement partner Lorraine Dupont, PCAOB Registration No. 2847), regarding the ACL methodology and related audit procedures.

This memorandum is organized to provide a comprehensive framework for the two preparation sessions scheduled for February 12 and 13, 2024. It addresses Mr. Hu's professional background, anticipated areas of Staff questioning, key documents for review, testimony guidelines, coordination considerations relative to prior testimony, logistical details, and an overall risk assessment with strategic notes.

II. Background on the Witness

A. Professional Background

Full Name: Raymond Hu

Current Title: Chief Accounting Officer, Belmont Ridge Bank National Association

Reporting Structure: Mr. Hu reports directly to Patricia "Trish" Navarrete, Chief Financial Officer. Ms. Navarrete, in turn, reports to BRB's Chief Executive Officer, Douglas Wellman.

Tenure at BRB: Mr. Hu joined Belmont Ridge Bank in March 2015 as Vice President, Technical Accounting and Policy. He was promoted to Senior Vice President and Deputy Controller in January 2018, and assumed the role of Chief Accounting Officer in September 2019. He has therefore been in his current position for approximately four and a half years, a period that encompasses BRB's adoption of the CECL standard and the launch and growth of the BridgeFin platform.

Mr. Hu holds a Bachelor of Science in Accounting from the University of Virginia and a Master of Accountancy from the University of Michigan. He is a licensed Certified Public Accountant in the states of Virginia and New York. Prior to joining BRB, Mr. Hu spent approximately twelve years at Severin Beck CPAs LLP, where he rose to the rank of Senior Manager in the Financial Services Audit practice. His experience at Muldoon Sabel CPAs LLP engagements with several large regional and community banks, providing him with substantial background in bank accounting, credit loss estimation, and regulatory reporting.

Role in CECL Adoption: Mr. Hu was a central figure in BRB's adoption of Accounting Standards Codification Topic 326, Financial Instruments — Credit Losses (commonly referred to as "CECL"), which became effective for BRB on January 1, 2020, as a public business entity that is an SEC filer. Beginning in 2018, Mr. Hu led the cross-functional CECL implementation team, which included representatives from the Controller's office, the Risk Management division (under Chief Risk Officer Annette Kowalski), the Credit Administration function, and the Information Technology group. He oversaw the selection and development of the CECL model framework, the establishment of the model governance structure, and the integration of the BridgeFin consumer personal-loan and micro-business loan portfolios into the bank-wide CECL model following BridgeFin's launch in the third quarter of 2020.

Mr. Hu's deep involvement in both the CECL implementation and the BridgeFin portfolio makes him a uniquely important witness for the Staff. He possesses detailed technical knowledge of the model's architecture, the data inputs that drive the loss-rate calculations, and the judgment-based qualitative overlay process. The Staff is likely aware of his centrality to these matters based on the document production and the prior testimony of Ms. Navarrete and Mr. Chakravarti, both of whom identified Mr. Hu as the primary accounting officer responsible for the ACL estimation.

B. Hu's Specific Responsibilities Relevant to the Investigation

Mr. Hu's responsibilities that are directly relevant to the SEC's areas of inquiry include the following:

1. Direct Oversight of ACL Estimation. Mr. Hu has direct oversight of the allowance for credit losses estimation for all BRB loan portfolios. This includes the BridgeFin consumer personal-loan sub-segment and the BridgeFin micro-business loan sub-segment, which together constitute the BridgeFin portfolio at issue in this investigation. Mr. Hu's team within the Controller's office — consisting of approximately eight analysts and senior analysts — performs the quantitative modeling work that produces the initial ACL estimate each quarter. Mr. Hu personally reviews the model outputs and the supporting workpapers before they are submitted to the Credit Loss Committee for review and approval.

2. Quarterly ACL True-Up Process. Mr. Hu is responsible for the quarterly ACL true-up process, which involves reviewing updated quantitative model outputs, refreshing macroeconomic forecast assumptions, analyzing current portfolio conditions (including delinquency trends and charge-off activity), and calibrating qualitative overlay adjustments. This quarterly process is the mechanism by which BRB updates its ACL to reflect the most current information available as of each reporting date.

3. Primary Contact for Whitfield Keane LLP on ACL Matters. Mr. Hu serves as BRB's primary point of contact with the Bank's independent auditor, Whitfield Keane LLP, on all matters related to the ACL audit procedures. He has had regular and ongoing communications with Lorraine Dupont, the audit engagement partner, as well as with various members of the Whitfield Keane engagement team, regarding the CECL methodology, model inputs and assumptions, the re-aging treatment, and qualitative overlay determinations. These communications span the period from the initial CECL adoption audit in early 2020 through the most recent interim review for Q2 2023.

4. Sign-Off on ACL Schedules. Mr. Hu is a signatory on the ACL schedules that support BRB's periodic financial reporting. Specifically, he signed off on the ACL schedules underlying the Forms 10-K for fiscal year 2022 (filed February 28, 2023), the Form 10-Q for Q1 2023 (filed May 10, 2023), and the Form 10-Q for Q2 2023 (filed August 9, 2023). His sign-off indicates his concurrence that the ACL estimate has been prepared in accordance with BRB's established methodology and governance procedures and that it represents a reasonable estimate of expected credit losses under ASC 326.

5. Credit Loss Committee Participation. Mr. Hu is a standing member of BRB's Credit Loss Committee, the internal governance body that meets quarterly to review, challenge, and approve the ACL estimates for all loan portfolios. Other members of the Credit Loss Committee include Ms. Navarrete (Chair), Annette Kowalski (Chief Risk Officer), Marcus Jennings (Chief Credit Officer), and representatives from the business line leadership, including Mr. Chakravarti for the BridgeFin portfolio. Mr. Hu's role on the Committee is to present the quantitative model outputs, explain the proposed qualitative overlays, and respond to questions and challenges from Committee members.

C. Prior Statements / Document Production Considerations

Mr. Hu has not previously testified in this investigation. The February 14, 2024 session will be his first appearance before the SEC Staff. He has not provided any prior sworn statements, declarations, or affidavits in connection with this matter.

It is noteworthy that no individual testimony subpoena was issued to Mr. Hu. His appearance is voluntary, made on behalf of BRB pursuant to the Bank's cooperation posture in this investigation. This is in contrast to the appearances of the two prior witnesses: Ms. Navarrete received an individual testimony subpoena on November 22, 2023, and Mr. Chakravarti received an individual testimony subpoena on December 5, 2023. Mr. Hu's appearance was negotiated directly between C&H (Victoria Chambers) and Senior Staff Attorney Andrea Felton, and was confirmed by letter dated January 12, 2024. The voluntary nature of the appearance is consistent with BRB's cooperative approach to the investigation, but it should not be mistaken for a lack of Staff interest in Mr. Hu's testimony. To the contrary, the Staff's willingness to schedule him without a subpoena likely reflects an expectation that his testimony will be forthcoming and substantive.

With respect to the document production, BRB has produced a total of 284,617 documents, comprising approximately 1.42 million pages, across six rolling productions from November 1, 2023, through the scheduled final production on March 1, 2024. Mr. Hu was designated as one of twenty-three custodians whose files were collected, processed, and reviewed for production. His individual document population totaled approximately 34,500 documents, including emails, memoranda, workpapers, spreadsheets, and other electronic files. These documents were produced across the various rolling productions as they were reviewed and designated for production.

The following categories of documents authored by or received by Mr. Hu are of particular relevance and may be used by Staff as exhibits during the testimony:

• ACL Calculation Workpapers. The detailed workpapers supporting the ACL estimates for Q4 2022, Q1 2023, and Q2 2023 (produced at Bates range BRB-SEC-078000 through BRB-SEC-082500). These workpapers include the vintage-cohort loss-rate schedules, the macroeconomic forecast assumptions, the qualitative overlay calculations, and the summary ACL roll-forward schedules.

• Credit Loss Committee Meeting Minutes. The minutes of the Credit Loss Committee meetings for Q4 2022 (held January 12, 2023), Q1 2023 (held April 6, 2023), and Q2 2023 (held July 13, 2023). These minutes document the Committee's review and approval of the ACL estimates, including any challenges raised by Committee members and the resolutions thereof.

• Emails Between Hu and Lorraine Dupont. Communications between Mr. Hu and the Whitfield Keane engagement partner regarding the vintage-cohort methodology, the treatment of re-aged loans, the qualitative overlay approach, and other aspects of the ACL audit procedures. These emails span the period from approximately October 2022 through March 2023 (for the FY 2022 audit) and April through August 2023 (for the interim reviews).

• CECL Model Governance Documentation. BRB's CECL Model Governance Framework document, which sets forth the policies and procedures governing model development, validation, use, and change management for the CECL model. Mr. Hu was the primary author of this framework document.

Staff may also present internal emails reflecting discussions among BRB personnel regarding the re-aging treatment, vintage-level performance deterioration, and adequacy of the ACL. During the preparation sessions, we will walk Mr. Hu through the key documents in each of these categories to ensure he is familiar with their contents and can testify consistently with the documentary record.

III. Anticipated Areas of SEC Staff Questioning

Based on our review of the Formal Order of Investigation, the Staff's document requests and subpoenas, the areas of focus during the Navarrete and Chakravarti testimonies, and the general scope of the investigation as communicated by Staff, we anticipate that the questioning of Mr. Hu will concentrate on the following areas. This section is organized to track the likely flow of the examination and to provide preparation points and recommended framing for each topic.

A. ACL Estimation Process Under ASC 326 (CECL)

The Staff will almost certainly begin with a detailed exploration of the mechanics of BRB's CECL model as applied to the BridgeFin portfolio. The objective of this line of questioning will be to establish a baseline understanding of how BRB estimated expected lifetime credit losses and to lay the foundation for subsequent questions about specific inputs, assumptions, and judgments.

Anticipated Questions:

The Staff will likely probe how expected lifetime credit losses are estimated for the BridgeFin portfolio, including the choice of loss-rate methodology. BRB employs a vintage-cohort analysis approach, and the Staff will want Mr. Hu to explain why this methodology was selected over alternative approaches, such as a probability-of-default / loss-given-default ("PD/LGD") framework or a discounted-cash-flow methodology. The Staff may also ask whether alternative methodologies were considered during the CECL implementation process and why they were not adopted.

Preparation Points:

Mr. Hu should be prepared to explain BRB's vintage-cohort approach clearly and concisely. Under this approach, loans are grouped by origination quarter (the "vintage"), and historical loss experience is tracked for each vintage cohort over time to project expected losses over the remaining contractual life of each cohort. The vintage-cohort approach was selected because it is well-suited to consumer loan portfolios with relatively homogeneous characteristics, such as the BridgeFin personal-loan and micro-business loan products, and because it provides a transparent and intuitive framework for tracking credit performance by origination period.

Mr. Hu should be prepared to describe the key inputs to the model:

(i) Historical loss data: Actual credit losses observed by vintage cohort, tracked on a monthly and quarterly basis from each cohort's origination date.

(ii) Current conditions: An assessment of current credit conditions, including prevailing delinquency rates, charge-off trends, and portfolio composition.

(iii) Reasonable and supportable forecasts: Macroeconomic forecast variables — including unemployment rate, GDP growth, housing price indices, and consumer credit indices — sourced from a third-party economic forecasting provider (Moody's Analytics baseline scenario) and supplemented by management judgment.

(iv) Qualitative adjustments: Overlay adjustments applied to the quantitative model output to reflect factors not fully captured in the model's historical data or forecast assumptions.

Central Disputed Item:

The central numerical dispute in this investigation is the adequacy of BRB's reported ACL for the BridgeFin portfolio as of December 31, 2022. BRB reported an ACL of $143.5 million, representing a coverage ratio of approximately 3.50% of BridgeFin outstanding loans of $4.1 billion. The SEC Staff's preliminary view, as communicated through the pattern of questioning in the Navarrete and Chakravarti sessions and through informal Staff comments, is that the ACL should have been approximately $180.8 million, representing a coverage ratio of 4.41%. This implies a potential understatement of $37.3 million, or approximately 26% of the reported ACL figure.

Mr. Hu should be prepared to articulate that the $143.5 million figure was the product of a rigorous, well-documented process involving quantitative model outputs supplemented by qualitative overlays. The estimate was reviewed and approved by the Credit Loss Committee, concurred in by the CFO, and audited by Whitfield Keane LLP without qualification. Mr. Hu should emphasize that ASC 326 recognizes that credit loss estimation inherently involves significant judgment and that reasonable professionals may arrive at different estimates based on different assumptions. The relevant standard is not whether BRB's estimate precisely matched any particular alternative calculation, but whether the estimate fell within a reasonable range of outcomes under the standard. Mr. Hu should express confidence that BRB's estimate met this standard.

B. Treatment of Re-Aged / Modified BridgeFin Loans

This is expected to be the single most critical area of questioning and the focal point of Staff Accountant Diane Milosevic's examination. Based on the patterns observed during the Navarrete and Chakravarti testimonies, this topic consumed approximately 40-50% of the available examination time, and the Staff's questioning was detailed, technically precise, and persistent.

Core Issue:

During the periods under investigation, 12,847 BridgeFin loans with an aggregate outstanding balance of $236.4 million were re-aged or modified. Under BRB's accounting treatment, these loans were classified as "current" following the modification, and the delinquency clocks were reset to zero. This treatment had the effect of removing these loans from the delinquent buckets in BRB's vintage-cohort analysis, which in turn reduced the historical loss rates applied to the relevant vintage cohorts.

SEC Staff's Position:

The Staff's position, as best we can reconstruct it from the questioning in the prior two testimonies, is that these loans should have remained in their original vintage delinquency buckets for loss-rate estimation purposes. In the Staff's view, resetting the delinquency clocks artificially lowered the 90-plus-day delinquency rate from 2.38% (the Staff's calculation, which treats re-aged loans as continuing to carry their pre-modification delinquency status) to the 1.72% reported by BRB (which reflects the delinquency clocks reset to current following modification). This difference in delinquency classification, when fed through the vintage-cohort loss-rate model, is the primary driver of the gap between BRB's $143.5 million ACL and the Staff's alternative $180.8 million calculation.

Preparation Points for Mr. Hu:

Mr. Hu should be thoroughly prepared to address the re-aging treatment. The following points should form the core of his responses:

(i) Consistency with OCC Guidance. BRB's treatment of re-aged loans was consistent with the Office of the Comptroller of the Currency's Comptroller's Handbook guidance on loan modifications and troubled debt restructurings. The OCC guidance permits national banks to re-age a loan — that is, to restore a delinquent loan to current status — when the borrower demonstrates a renewed willingness and ability to repay the loan. BRB's re-aging policy was developed in consultation with its OCC examination team and was subject to supervisory review.

(ii) Re-Aging Criteria. BRB's re-aging policy established specific criteria that a borrower was required to satisfy before a loan could be re-aged. These criteria included, among other requirements, that the borrower make a minimum of three consecutive on-time payments following the modification. This requirement was designed to ensure that re-aging was applied only in cases where the borrower had demonstrated a meaningful and sustained improvement in payment behavior, rather than as a blanket mechanism for cosmetically improving delinquency statistics.

(iii) Auditor Review and Acceptance. The re-aging treatment was reviewed and accepted by Whitfield Keane LLP during the FY 2022 annual audit. Whitfield Keane tested a sample of re-aged loans, reviewed the re-aging policy and criteria, and concluded that BRB's treatment was appropriate under the applicable accounting standards and regulatory guidance. Whitfield Keane issued an unqualified audit opinion on BRB's FY 2022 financial statements, without a critical audit matter or emphasis-of-matter paragraph relating to the re-aging treatment.

(iv) Disclosure. BRB's re-aging policy and its application to the BridgeFin portfolio were disclosed in the Management's Discussion and Analysis section of the FY 2022 Form 10-K. The disclosure described the nature and volume of modifications and re-agings within the BridgeFin portfolio and their impact on reported delinquency and credit quality metrics.

(v) Qualitative Overlay for Modified Loans. Recognizing that re-aged and modified loans carry an elevated risk profile relative to loans that have never been delinquent, BRB applied a specific qualitative overlay to account for this incremental risk. The overlay was designed to supplement the quantitative model output and ensure that the ACL estimate appropriately reflected the credit risk inherent in the modified loan population, even though these loans had been reclassified as current. This overlay is a key component of BRB's defense and demonstrates that the re-aging treatment was not a mechanism for avoiding loss recognition.

Critical Instruction:

Mr. Hu should avoid volunteering information about any internal debate regarding the re-aging methodology. We are aware from the document review that certain members of the Credit Loss Committee raised questions about the appropriateness of the re-aging treatment during the Q4 2022 meeting, and that there were internal emails reflecting differing views on this topic. If the Staff asks directly about dissenting views within the Credit Loss Committee or internal disagreements about the re-aging approach, Mr. Hu must answer truthfully. However, his answers should be concise and should frame any internal debate as a healthy exercise of the challenge function that the Credit Loss Committee was designed to perform — not as evidence of a recognized problem that was ignored. During the preparation sessions, we will explore the specific documents and communications that touch on internal debate to ensure Mr. Hu is prepared to address them accurately and with appropriate context.

C. Quarterly True-Up Procedures

The Staff will seek to understand how BRB updated its ACL estimates on a quarterly basis, with particular focus on the true-ups performed for Q1 2023 (as reflected in the Form 10-Q filed May 10, 2023) and Q2 2023 (as reflected in the Form 10-Q filed August 9, 2023). The Staff's interest in the quarterly true-ups appears to relate to whether BRB was sufficiently responsive to deteriorating credit conditions in the BridgeFin portfolio during the first half of 2023 and whether the ACL was increased in a timely manner as new loss data emerged.

Preparation Points:

Mr. Hu should be prepared to explain the quarterly true-up process in a step-by-step fashion:

(i) Updated Loss-Rate Data. Each quarter, the ACL team refreshed the vintage-cohort loss-rate data to incorporate the most recent quarter's actual loss experience. This updated the historical loss curves for each vintage cohort and provided a current view of credit performance trends.

(ii) Refreshed Macroeconomic Forecast Inputs. The macroeconomic forecast assumptions used in the CECL model were updated each quarter based on the most recent available economic projections. BRB utilized the Moody's Analytics baseline scenario as a starting point and applied management judgment to adjust the forecasts where appropriate.

(iii) Review of Current Portfolio Conditions. The ACL team analyzed current portfolio conditions, including delinquency trends by vintage, charge-off activity, loan growth, portfolio composition changes, and any other factors that might affect the credit risk profile of the BridgeFin portfolio.

(iv) Credit Loss Committee Review and Challenge Session. The updated ACL estimate, including the proposed qualitative overlays, was presented to the Credit Loss Committee for review, challenge, and approval. Committee members were expected to probe the assumptions underlying the estimate and to raise any concerns about the adequacy or appropriateness of the proposed ACL.

(v) Qualitative Overlay Calibration. Based on the quantitative model outputs and the Committee's discussion, the qualitative overlays were calibrated to reflect management's judgment regarding factors not fully captured in the model.

(vi) CFO and CAO Sign-Off. Following Credit Loss Committee approval, the final ACL estimate was reviewed and signed off by the CFO (Ms. Navarrete) and the CAO (Mr. Hu) before being incorporated into BRB's financial statements.

Net Charge-Off Rate Context:

The net charge-off rate is a related but distinct metric that the Staff has focused on in connection with the re-aging treatment. BRB reported a net charge-off rate of 2.14% for FY 2022 for the BridgeFin portfolio. The Staff's preliminary view suggests that the net charge-off rate should have been approximately 3.05% if re-aged loans are treated differently — specifically, if certain modified loans that subsequently re-defaulted are captured in the charge-off numerator at their pre-modification delinquency levels rather than only when they reach charge-off status following the re-aging. Mr. Hu should be prepared to walk through the charge-off recognition process — including the timing of charge-off, the treatment of partial charge-offs, and how re-aged loans that subsequently re-default factor into the net charge-off calculation — in detail.

Mr. Hu should also be prepared to explain that the quarterly true-ups for Q1 and Q2 2023 reflected evolving credit conditions and that BRB's ACL was adjusted incrementally as new information became available. The ACL increased modestly from $143.5 million at December 31, 2022, to $149.2 million at March 31, 2023, and to $157.8 million at June 30, 2023, reflecting the ACL team's ongoing assessment of credit deterioration in certain vintage cohorts. These increases demonstrate that BRB's process was responsive to changing conditions, even if the Staff may argue that the increases were insufficient in magnitude.

D. Qualitative Overlay Adjustments

The qualitative overlay adjustments applied to the quantitative CECL model output are expected to be a significant area of Staff inquiry. ASC 326-20-30-7 and related implementation guidance contemplate that entities will apply qualitative adjustments to reflect factors that are relevant to expected credit losses but are not fully captured in the quantitative model's historical data or economic forecast assumptions. The Staff will likely explore the nature, magnitude, basis, and documentation of these overlays in detail.

Preparation Points:

Mr. Hu should be prepared to explain that ASC 326 not only permits but affirmatively encourages the use of qualitative adjustments to ensure that the ACL estimate reflects all relevant information about expected credit losses. The qualitative adjustment process is an integral and expected component of CECL implementation, not an after-the-fact correction mechanism.

The specific qualitative overlays relevant to the BridgeFin portfolio for the FY 2022 ACL estimate include:

(i) Platform Youth / Limited Seasoning Data Overlay. The BridgeFin platform was launched in Q3 2020, meaning that at the time of the FY 2022 estimate, the oldest vintage cohorts had only approximately two and a half years of performance history. This limited seasoning data meant that the vintage-cohort model had not yet observed full-life loss curves for any cohort. A qualitative overlay was applied to account for the risk that actual lifetime losses could exceed the model's projections, which were based on extrapolating observed early-life loss patterns.

(ii) AI-Driven Underwriting Model Overlay. BridgeFin's underwriting relies on a proprietary artificial intelligence-based credit scoring model that was developed in-house and that differs materially from traditional underwriting approaches used for BRB's legacy loan portfolios. Given the limited performance track record of this model — which had been in production for approximately two years at the time of the FY 2022 estimate — a qualitative overlay was applied to account for the possibility that the model's predictions might not accurately forecast credit losses over the full life of the loans.

(iii) Modified / Re-Aged Loan Risk Overlay. As discussed in Section III.B above, a qualitative overlay was applied specifically to account for the elevated risk profile of loans that had been modified and re-aged. This overlay was designed to ensure that the ACL captured incremental expected losses attributable to the modified loan population, even though these loans were classified as current in the delinquency statistics fed into the quantitative model.

Potential Staff Challenge:

The Staff may ask whether the qualitative overlays were sufficient to capture the credit deterioration observed in post-Q2 2021 vintage cohorts. These later-vintage cohorts exhibited more rapid delinquency migration and higher early-life loss rates compared to the earlier cohorts originated in Q3 2020 through Q2 2021. The Staff may argue that the overlays should have been larger to account for this divergence in credit performance. Mr. Hu should be prepared to show that the overlays were determined through a documented and disciplined process, reviewed and challenged by the Credit Loss Committee, and discussed with Whitfield Keane as part of the audit process. The magnitude of each overlay was based on specific analytical support, including sensitivity analyses and scenario testing, and was not set arbitrarily.

Key Preparation Point:

During the preparation sessions, we will review with Mr. Hu the specific dollar amounts of each qualitative overlay in the FY 2022 ACL estimate, the method used to determine each overlay's magnitude (including the range of outcomes considered), and the documentation supporting each overlay determination. Mr. Hu should be comfortable discussing both the aggregate sum of qualitative overlays and the individual components, as the Staff is likely to probe each overlay separately.

E. Communications with Whitfield Keane LLP (Independent Auditor)

The Staff will probe the nature and content of Mr. Hu's communications with Lorraine Dupont, the Whitfield Keane engagement partner, regarding the ACL methodology and related audit procedures. This area of questioning serves multiple purposes for the Staff: it allows them to understand the auditor's level of scrutiny, to identify any points of disagreement between BRB and its auditor, and to assess whether the auditor was fully informed about the inputs and assumptions underlying the ACL estimate.

Preparation Points:

Mr. Hu should be prepared to discuss the following topics in connection with his communications with Whitfield Keane:

(i) Audit Planning Meetings. Mr. Hu participated in audit planning meetings with the Whitfield Keane engagement team at the outset of the FY 2022 audit cycle. During these meetings, the CECL methodology was reviewed, areas of increased audit focus were identified (including the BridgeFin portfolio and the re-aging treatment), and the auditor's planned testing approach was discussed.

(ii) Independent Testing of Model Inputs and Assumptions. Whitfield Keane independently tested BRB's ACL model inputs and assumptions, including the vintage-cohort loss-rate calculations, the macroeconomic forecast assumptions, and the data integrity of loan-level inputs. Mr. Hu's team provided requested data, workpapers, and access to model documentation to facilitate this testing.

(iii) Re-Aging Treatment Discussions. Mr. Hu and Ms. Dupont had specific discussions regarding the re-aging treatment of BridgeFin loans. These discussions addressed the re-aging policy and criteria, the volume and characteristics of re-aged loans, the impact of re-aging on delinquency statistics and loss-rate calculations, and the qualitative overlay applied to address the risk of the modified loan population. Mr. Hu should be prepared to describe these discussions factually and to note that Whitfield Keane reviewed and accepted BRB's approach.

(iv) Points of Challenge or Disagreement. The Staff may ask whether the auditor raised any challenges or objections to BRB's ACL methodology or to specific components of the ACL estimate. Mr. Hu should answer truthfully. If Whitfield Keane raised questions or requested additional analysis, Mr. Hu should describe the nature of the auditor's inquiry and how it was resolved — typically through the provision of additional documentation, further analysis, or refinement of assumptions. Any such exchanges are a normal part of the audit process and do not indicate a deficiency in BRB's methodology.

(v) Unqualified Audit Opinion. Mr. Hu should note that Whitfield Keane issued an unqualified audit opinion on BRB's FY 2022 consolidated financial statements. The auditor concluded that the financial statements were presented fairly, in all material respects, in conformity with U.S. generally accepted accounting principles. This conclusion necessarily encompassed the ACL estimate and the methodology used to derive it.

Privilege Considerations:

Communications between Mr. Hu and Whitfield Keane are generally not subject to the attorney-client privilege. They are auditor-client communications, and the presence of the auditor generally does not create a privileged communication. However, there is a limited category of communications that may be subject to protection:

• Any communications that were made at the direction of or through Calderwood & Harkness LLP or Caldwell Pratt & Stein LLP (BRB's securities reporting counsel, lead partner Martin Caldwell) in anticipation of the SEC investigation may be covered by the work-product doctrine. To the extent that C&H or CPS directed Mr. Hu to gather or prepare information from the auditor in connection with the investigation — as opposed to the ordinary audit process — those communications may be protectable.

• During the preparation sessions, Mr. Hu should be reminded that he must not disclose the substance of any discussions with C&H or CPS regarding audit-related topics. If a Staff question touches on legal advice that Mr. Hu received from C&H or CPS in connection with the auditor's review, counsel will instruct Mr. Hu not to answer and will assert the privilege on the record.

Hargrove & Millstein Coordination:

Hargrove & Millstein LLP (Samuel Hargrove, lead partner) represents Whitfield Keane LLP in connection with this SEC inquiry. Mr. Hu should be instructed not to comment on any communications with Hargrove & Millstein or to speculate about Whitfield Keane's legal position or defense strategy. If the Staff asks Mr. Hu about interactions with Hargrove & Millstein, he should state that he has not had any communications with that firm in connection with this matter (if that is the case) or defer to counsel.

F. BRB's Internal Controls Over Financial Reporting (ICFR) Related to ACL

The Staff may inquire about BRB's internal controls over financial reporting as they relate to the ACL estimation process. These questions may be directed at understanding whether BRB had adequate controls in place to ensure the accuracy and reliability of its ACL estimate, and at assessing whether the alleged ACL understatement reflects a material weakness or significant deficiency in internal controls. The Staff's interest in ICFR also relates to BRB's SOX Section 302 (CEO/CFO certifications) and SOX Section 404 (management's annual assessment of internal control effectiveness) obligations.

Preparation Points:

Mr. Hu should be prepared to describe BRB's ICFR control framework as it pertains to the ACL estimation process:

(i) Model Governance Controls. BRB's CECL model is subject to a comprehensive model governance framework that includes independent model validation by the Risk Management division, under the direction of Chief Risk Officer Annette Kowalski. The model validation function operates independently of the Controller's office and is responsible for assessing the conceptual soundness, data quality, and performance of the CECL model. Model validation reviews are performed at least annually and upon any material changes to the model.

(ii) Data Integrity Controls. BRB maintains controls over the integrity of loan-level data inputs that feed the CECL model, including reconciliation procedures between the loan accounting system and the CECL model data extracts, automated validation checks on key data fields, and manual review of exceptions.

(iii) Credit Loss Committee Approval Process. As described above, the ACL estimate is subject to review, challenge, and approval by the Credit Loss Committee, which serves as a key governance control. The Committee's deliberations are documented in meeting minutes, and the Committee's approval of the ACL estimate is a prerequisite for its incorporation into BRB's financial statements.

(iv) CFO and CEO Certification Process. The CFO and CEO review the ACL estimate and the related financial statement disclosures as part of the SOX certification process. This review encompasses an assessment of whether the ACL estimate is reasonable and whether the related disclosures are adequate.

Mr. Hu should affirm that BRB's management concluded that ICFR was effective as of December 31, 2022, and that this conclusion was concurred in by Whitfield Keane in its audit report on internal controls, which was issued without any material weakness or significant deficiency findings related to the ACL estimation process. The effectiveness of BRB's controls supports the reasonableness of the ACL estimate and undercuts any suggestion that the reported ACL was the product of an inadequate or deficient process.

G. Materiality Considerations

The Staff may attempt to elicit Mr. Hu's view on the materiality of the alleged $37.3 million ACL understatement relative to BRB's consolidated financial statements. This line of questioning is designed to establish that the magnitude of the alleged error is large enough to be "material" under applicable accounting and legal standards — a necessary predicate for any enforcement action based on reporting violations.

Context:

BRB's internal materiality threshold for the purposes of evaluating identified misstatements and adjustments is set at 5% of pre-tax income. For FY 2022, BRB reported pre-tax income of $612 million, yielding a materiality threshold of $30.6 million. The alleged $37.3 million ACL understatement exceeds this internal threshold by $6.7 million, or approximately 22%. This is a potentially adverse data point that the Staff may use to argue that even by BRB's own internal standards, the alleged understatement should be considered material.

Preparation Points:

Mr. Hu should be prepared to discuss materiality with care and nuance:

• Mr. Hu should frame materiality in the context of a range of reasonable estimates under CECL, rather than as a binary pass/fail assessment applied to a single-point figure. CECL inherently involves substantial estimation uncertainty, and the relevant inquiry under the accounting standard is whether BRB's estimate was within a range of reasonable outcomes — not whether it precisely matched any particular alternative calculation, including the Staff's.

• Mr. Hu should note that the concept of materiality involves both quantitative and qualitative considerations. Even if a quantitative threshold is exceeded, qualitative factors — such as the nature of the item, whether it affects a trend, whether it masks a change in earnings, and the overall mix of information available to investors — are relevant to the materiality determination.

• Mr. Hu should not concede that the $37.3 million figure is "material" without providing appropriate context. He should emphasize that BRB's reported ACL of $143.5 million was the product of a well-documented, well-governed process; that it was reviewed and accepted by the independent auditor; and that any alleged error of this magnitude in an ACL estimate must be evaluated in light of the inherent estimation range, not against a single alternative point estimate.

• If the Staff presents the $37.3 million figure and asks Mr. Hu whether it exceeds BRB's internal materiality threshold, he should acknowledge the mathematical relationship but immediately note that the internal materiality threshold is a tool used in the audit process and that its application requires judgment and context. He should not make any categorical statements about the materiality of the alleged understatement.

This is one of the more sensitive areas of the testimony, and Mr. Hu's responses must be carefully calibrated. Any unqualified concession on materiality could significantly narrow BRB's range of defensible positions going forward, particularly in the context of a potential Wells submission or contested proceeding.

IV. Key Documents for Preparation Session

A. Documents Hu Should Review

During the preparation sessions on February 12 and 13, 2024, Mr. Hu should review the following categories of documents. These documents represent the core documentary record that the Staff is likely to reference during the testimony, and Mr. Hu must be thoroughly familiar with their contents.

1. BRB FY 2022 Form 10-K (filed February 28, 2023). Mr. Hu should focus on the ACL-related footnotes to the consolidated financial statements (specifically, Note 5 — Loans and Allowance for Credit Losses), the Management's Discussion and Analysis section addressing credit quality, loan portfolio composition, and the ACL methodology, and the CECL methodology disclosures in Note 1 (Summary of Significant Accounting Policies). He should also review the MD&A discussion of re-aged and modified loans and the related credit quality disclosures.

2. BRB Q1 2023 Form 10-Q (filed May 10, 2023) and Q2 2023 Form 10-Q (filed August 9, 2023). Mr. Hu should review the same sections as identified above for the 10-K, focusing on any changes in the ACL estimate, credit quality trends, and updated disclosures regarding the BridgeFin portfolio.

3. ACL Calculation Workpapers for Q4 2022, Q1 2023, and Q2 2023 (Bates range BRB-SEC-078000 through BRB-SEC-082500). These workpapers are the most detailed documentary evidence of the ACL estimation process and include the vintage-cohort loss-rate schedules, macroeconomic forecast inputs, qualitative overlay calculations, and the summary ACL roll-forward. Mr. Hu should be prepared to walk through these workpapers in detail during the testimony.

4. Credit Loss Committee Meeting Minutes for Q4 2022, Q1 2023, and Q2 2023. These minutes document the Committee's review and approval of the ACL estimate for each period. Mr. Hu should be familiar with the questions and challenges raised by Committee members, the responses provided, and the basis for the Committee's approval.

5. BridgeFin Vintage-Cohort Loss-Rate Schedules. These schedules present the historical loss experience for each vintage cohort of BridgeFin loans and form the core quantitative input to the CECL model. Mr. Hu should be able to discuss trends in vintage-level performance, including the divergence in credit performance between earlier and later vintage cohorts.

6. Re-Aged/Modified Loan Policy Documentation. BRB's internal policy governing the re-aging and modification of BridgeFin loans, including the criteria for re-aging eligibility, the required number of consecutive on-time payments, and the reporting and monitoring requirements following re-aging.

7. Emails Between Hu and Lorraine Dupont (Whitfield Keane LLP). Communications between Mr. Hu and the audit engagement partner regarding the FY 2022 ACL audit procedures, the vintage-cohort methodology, the re-aging treatment, and the qualitative overlay approach. Mr. Hu should review these emails carefully to ensure that his testimony is fully consistent with the contemporaneous documentary record.

8. CECL Model Governance Framework Document. The comprehensive governance document that Mr. Hu authored, setting forth the policies and procedures governing the CECL model's development, validation, use, and change management.

9. Qualitative Overlay Documentation and Supporting Analyses. The documentation supporting each qualitative overlay applied to the FY 2022 ACL estimate, including the analytical basis for each overlay, the magnitude determinations, and any sensitivity analyses or scenario testing performed.

10. Brindoval Analytic Partners LLC Forensic Analysis Memorandum (Dr. Franklin Osei, dated December 22, 2023). This memorandum, prepared at C&H's direction, provides a forensic accounting analysis of BRB's ACL methodology and the Staff's alternative calculation. It is attorney work product and was prepared in anticipation of this investigation. Mr. Hu should review it for background understanding of the defense positions and to familiarize himself with the analytical framework supporting BRB's methodology. However, Mr. Hu must be clearly instructed that he may not reference the Brindoval memorandum, Dr. Osei, or any aspect of the forensic analysis during his testimony. If asked, he should not disclose that such an analysis exists.

B. Documents Staff Is Likely to Use as Exhibits

Based on the exhibit usage patterns during the Navarrete and Chakravarti testimonies and the general scope of the Staff's inquiry, we anticipate that the Staff may present the following categories of documents as exhibits during Mr. Hu's testimony:

1. ACL Summary Schedules. Documents showing the $143.5 million reported ACL and the 3.50% coverage ratio as of December 31, 2022. The Staff may present its own alternative calculation of $180.8 million / 4.41% and ask Mr. Hu to react to specific line items.

2. BridgeFin Delinquency and Charge-Off Trend Reports. Reports showing delinquency migration, roll rates, and charge-off trends for the BridgeFin portfolio, potentially highlighting the divergence between delinquency rates as reported by BRB (1.72% for 90-plus days) and the Staff's calculation (2.38% including re-aged loans).

3. Internal Emails Discussing Re-Aging Treatment. Emails among BRB personnel — potentially including Mr. Hu's own communications — that discuss the re-aging treatment of modified loans, its impact on delinquency statistics, or any internal views about whether the treatment was appropriate. The Staff may attempt to use these emails to suggest that BRB was aware of, or should have been aware of, concerns about the re-aging approach.

4. Credit Loss Committee Presentations. Slide decks and presentations from Credit Loss Committee meetings showing vintage-cohort performance deterioration for post-Q2 2021 originations. The Staff may use these materials to argue that the Committee was presented with evidence of worsening credit conditions but did not increase the ACL sufficiently.

5. Internal Communications Reflecting Concerns About ACL Adequacy. Any internal emails, memoranda, or other communications in which BRB personnel — including but not limited to Mr. Hu — expressed concerns about the sufficiency of the ACL estimate. These documents could be used to suggest that the ACL was known to be insufficient at the time it was reported.

During the preparation sessions, we will walk Mr. Hu through the most likely exhibit documents and practice his responses to questions about each one.

V. Testimony Guidelines and Instructions

A. General Conduct

The following guidelines govern Mr. Hu's conduct during the investigative testimony. These guidelines are standard for witness testimony before the SEC Division of Enforcement and will be reviewed with Mr. Hu during both preparation sessions.

1. Listen Carefully. Mr. Hu should listen carefully and attentively to each question before answering. He should ensure that he fully understands the question being asked. He should not anticipate or finish the questioner's sentences.

2. Answer Only the Question Asked. Mr. Hu should answer only the specific question that is posed. He should not volunteer information beyond what is necessary for a truthful and complete response to the question. The temptation to "explain" or to provide background context should be resisted unless the question specifically calls for such information. Short, direct answers are almost always preferable to lengthy narrative responses.

3. Ask for Clarification. If Mr. Hu does not understand a question — whether because it is ambiguous, compound, or uses unfamiliar terminology — he should ask the questioner to rephrase or clarify the question before attempting to answer. There is no penalty for asking for clarification, and it avoids the risk of answering a question that was not actually asked.

4. Do Not Speculate. If Mr. Hu does not recall a fact, a date, a number, or the substance of a conversation, he should say so truthfully rather than speculate or guess. Appropriate responses include "I don't recall," "I would need to review my records to confirm that," or "I don't have a specific recollection of that." Mr. Hu should not guess at numbers, dates, or other precise details unless he is confident in his recollection.

5. Do Not Guess at Numbers or Dates. The ACL calculation involves numerous specific dollar amounts, percentages, dates, and other precise figures. It is perfectly acceptable — and in fact preferable — for Mr. Hu to say "I would need to check my records to confirm the precise figure" rather than to offer an approximate number that could later be used to impeach him or to establish an unfavorable factual premise.

6. Truthful Testimony Is Mandatory. Mr. Hu must testify truthfully at all times. Any materially false statement made during SEC investigative testimony could constitute a violation of 18 U.S.C. § 1001 (false statements to a federal agency) and/or 18 U.S.C. § 1621 (perjury), both of which are federal criminal offenses carrying significant penalties. There is no exception to this requirement. If the truthful answer to a question is unfavorable, Mr. Hu should still answer truthfully; the preparation sessions will address how to frame truthful answers in a way that provides appropriate context.

7. Pace and Demeanor. Mr. Hu should maintain a calm, professional demeanor throughout the testimony. He should speak clearly and at a measured pace, both to ensure accuracy and to allow counsel time to evaluate each question and interpose objections or instructions where appropriate. He should not exhibit frustration, impatience, or combativeness, even if the questioning becomes repetitive or confrontational.

8. Breaks. Mr. Hu may request a break at any time. If he feels fatigued, confused, or uncertain about how to respond to a particular line of questioning, he should request a brief recess. He may also confer with counsel during breaks, although he may not confer with counsel to discuss a pending question that has already been asked.

B. Privilege Instructions

The following privilege instructions are critical and will be reinforced during the preparation sessions:

1. Attorney-Client Privilege. Mr. Hu must not disclose the substance of any communications with Calderwood & Harkness LLP attorneys — including Victoria Chambers, Jasmine Thornton, Howard Quinn, Liam Donahue, or any other C&H attorney — regarding legal advice provided in connection with this investigation or any related matters. Similarly, Mr. Hu must not disclose the substance of any communications with Caldwell Pratt & Stein LLP (Martin Caldwell) regarding legal advice related to BRB's securities reporting obligations.

2. Attorney Work Product. Mr. Hu must not disclose the substance of any analysis, memoranda, or findings prepared by Brindoval Analytic Partners LLC (Dr. Franklin Osei) or by any other consultant retained at the direction of C&H or CPS in connection with this investigation. These materials are protected attorney work product and are not subject to disclosure. If the Staff asks whether any outside analysis has been performed regarding the ACL methodology, Mr. Hu should defer to counsel.

3. Response to Privilege Questions. If the Staff asks Mr. Hu about the content of communications with counsel, he should respond: "I was advised by counsel not to disclose privileged communications." He should not elaborate or provide a partial description of the communication. Counsel will assert the privilege on the record as necessary.

4. Counsel Will Assert Privilege. Counsel for BRB will be present throughout the testimony and will assert privilege as necessary. If counsel instructs Mr. Hu not to answer a question on privilege grounds, he should comply with the instruction and should not attempt to answer the question over counsel's instruction.

5. Bank-Examination Privilege. Mr. Hu should not volunteer any information about OCC supervisory communications, examination findings, reports of examination, or matters requiring attention ("MRAs"). If the Staff asks about BRB's interactions with the OCC related to the BridgeFin portfolio or the ACL methodology, counsel will evaluate on a question-by-question basis whether to assert the bank-examination privilege and will instruct Mr. Hu accordingly. This area requires careful real-time judgment by counsel, as the scope and applicability of the bank-examination privilege in the SEC enforcement context involves unsettled legal questions.

C. Upjohn Reminder

During the final preparation session on February 13, 2024, counsel will reconfirm the following points with Mr. Hu, consistent with the Upjohn warning previously administered on October 25, 2023:

1. Calderwood & Harkness LLP represents Belmont Ridge Bank National Association, the corporate entity. C&H does not represent Raymond Hu in his individual capacity.

2. Mr. Hu's communications with C&H in the course of this representation are protected by the attorney-client privilege. However, the privilege belongs to BRB — not to Mr. Hu personally. BRB may, at its sole discretion, choose to waive the privilege and disclose the substance of communications with Mr. Hu to the SEC or to other parties. Mr. Hu cannot independently assert or waive the privilege.

3. There may be situations in which the interests of BRB as an entity diverge from Mr. Hu's individual interests. In such circumstances, C&H's duty of loyalty runs to BRB.

4. Mr. Hu has the right to retain his own individual counsel at his own expense at any time. He may also request that BRB advance legal fees for individual representation pursuant to the applicable provisions of BRB's certificate of incorporation, bylaws, and directors' and officers' liability insurance policy. The decision to advance fees is within BRB's discretion, subject to the terms of the governing documents.

5. If Mr. Hu has any questions about the scope of the representation or his right to individual counsel, he should raise them with C&H or with independent counsel of his choosing before the testimony takes place.

We recommend documenting the re-administration of the Upjohn warning in a contemporaneous memorandum to the file following the February 13 preparation session.

D. Handling Difficult Questions

Certain categories of questions are likely to be particularly challenging for Mr. Hu. The following guidance addresses anticipated difficult questions and provides a framework for Mr. Hu's responses. These scenarios will be practiced extensively during the mock examination on February 13, 2024.

1. "Did you ever believe the ACL was insufficient?"

If the Staff asks Mr. Hu whether he ever personally believed or harbored concerns that BRB's ACL for the BridgeFin portfolio was insufficient or understated, Mr. Hu must answer truthfully. During the preparation sessions, we must explore with Mr. Hu whether he in fact held any such concerns and, if so, the nature and extent of those concerns and the circumstances in which they arose.

If Mr. Hu did not harbor personal reservations about the adequacy of the ACL, he should say so directly and explain that he believed the estimate was reasonable and was the product of a well-governed process.

If Mr. Hu did harbor some degree of concern or uncertainty at any point, his response should be framed in the context of the collaborative estimation process and the exercise of reasonable professional judgment. He should explain that estimation uncertainty is inherent in the CECL framework, that reasonable professionals may hold different views about the appropriate level of the ACL, and that the Credit Loss Committee's function was precisely to surface, discuss, and resolve differing perspectives. Any personal reservations should be characterized as part of the normal