The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
MEMORANDUM FOR THE RECORD Event: Interview with Mark Winer Type of Event: Group Interview Date of Event: March 23, 2010, 9:30-12:30 p.m. Team Leader: Chris Seefer Location: FCIC, large conference room Participants - Non-FCIC: Mark Winer, Kenneth Winer (personal attorney), Jeff Kilduff (O’Melveny & Myers, LLP) and Evan Stoloff (Fannie Mae in-house counsel) Participants - FCIC: Chris Seefer, Tom Stanton, Mike Easterly, Clara Morain MFR Prepared by: Clara Morain Date of MFR: March 25, 2010 Summary of the Interview or Submission: Note: This is a summary of the interview, not a complete transcript. Direct quotations appear in side quote marks. Chris Seefer opened the meeting by summarizing the mandate of the Financial Crisis Inquiry Commission, noting specifically its statutory requirement to look into the role of GSEs in the financial crisis. He then asked Mr. Winer to explain his background with Fannie Mae. Mr. Winer’s Professional Backgroud Mr. Winer joined Fannie Mae in the spring of 2004 as a Vice President for market risk in the Capital Markets division, which he explained had overall responsibility for recommending measures of market risk and for recommended strategies to managers and to the trading floor to manage markets. He said that he dealt mostly with securities on balance sheet during that time. He held that position until approximately November 2005 when he became Deputy Chief Risk Officer (“CRO”) and Interim CRO. Mr. Winer said that at the time he assumed the Deputy CRO role “the whole existence of a central CRO was relatively new to the company.” He explained that previously, the company’s risk officers were embedded within business segments and reported to the business segment managers, with the exception of a brief time in 2005 when Adolfo Marzol required a centralized CRO. Mr. Winer said that upon assuming the CRO role, he designed a more centralized risk function for the company and began to develop and implement a set of company-wide risk activities that would give broad guidance about how to Page 1 of 13 think about risk at the company. He said that he also acted as point person between the risk policy and capital committees of the Board of Directors. He said that his fourth responsibility was to put in place management policies to manage risk. In the spring of 2006, he was appointed Senior Vice President for Business Analytics and Decisions (“BAD”), at the same time that Enrico Dallavecchia assumed the CRO position. Mr. Winer reported to Chief Business Officer Bob Levin from his appointment through Mr. Levin’s August 2008 resignation. 2005 Strategic Off-site Meeting Chris Seefer described Tom Lund’s 2005 PowerPoint presentation regarding Fannie Mae’s “Strategic Crossroads.” Mr. Winer said he regularly participated in the off-site strategy sessions, and that he recalled that the subject of whether or not to purchase higher-risk securities was frequently discussed during the timeframe of Mr. Lund’s presentation. Chris asked if a time came when the company made a conscious decision to purchase or guarantee higher risk securities. Mr. Winer said that the decision was made to purchase riskier loan products, but said, “don’t think of it as one decision. As I recall, the decision to expand into the Alt-A space was part of trying to provide a broader range of services to customers to maintain our importance to the customers we had. The decision to purchase subprime triple-A or higher [rated] securities was viewed as an attractive business opportunity to provide returns to company and liquidity to market.” He said that there were “fairly strict limits” on the types of products Fannie Mae could acquire and that there was “broad use of credit enhancements as part of our portfolio management when you got into higher risk activities.” Mr. Winer said that “Alt-A had always been a part of the portfolio products guaranteed by Fannie – since before I got there. They’d been around a long time. These discussions were both to expand our participation in that market and to take on a larger share of the risk of those loans. But they had always been a product of ours.” Chris asked if “taking a larger share of the risk of those loans” meant that Fannie Mae was buying less credit enhancements, and Mr. Winer said, “yes, that was part of the discussion. Fannie Mae continued to buy credit enhancement, but I don’t remember how to compare it.” In response to Chris Seefer’s questions about disclosure, Mr. Winer said that he did not know whether or not changes to the credit enhancements purchased on risky product were disclosed, but that he thinks there may be a relevant entry in the CRAM reports. Tom Stanton asked if the bank where Mr. Winer worked prior to Fannie Mae allocated capital across business lines so that managers knew how much capital supported their activities. Mr. Winer replied that the bank did allocate capital across business lines, but that he could not comment on its impact on managers. Tom asked to what extent managers at Fannie Mae understood the amount of capital backing each risky venture in increasing Alt-A. Mr. Winer replied that “the basic analytics used within Fannie Mae before I got there were pricing credit Page 2 of 13 risk – saying what do we think based on our economic capital-based model. So the basic analytics at the company was based on a calculation of a loan-specific economic capital measure… I can’t recall what language was used to construct limits on exposure, [however].” Chris asked if there were limits on exposures for certain asset classes, and Mr. Winer said that there were limits but that he does not recall specifically what they were. He said that the limits would be reflected in Single Family risk documents and capital markets risk policies and procedures. Returning to the discussion of Alt-A, Chris noted that the numbers reflect increases in purchases and guarantees of Alt-A loans from 2004 to 2007, and asked Mr. Winer if customers were asking Fannie Mae to purchase more non-traditional mortgages. Mr. Winer said that he did not know. Chris asked if Mr. Winer recalled whether or not in the 2003-2005 timeframe Fannie Mae was losing market share in terms of mortgage originations. Mr. Winer replied that, “I’ve seen the graphs showing the market share went down but I don’t know the numbers. Certainly, we all knew that market share went down.” Chris asked if that decline in market share influenced Fannie Mae’s decision to increase purchases of non-traditional mortgages and subprime PLS. Mr. Winer said “certainly some part of the decision was to ensure continued relevance in the market.” Chris asked if remaining relevant was the same as reversing market share declines. Mr. Winer said, “when I think about the issue, I remember thinking that we tried to make sure we had a security that had enough volume to ensure that acceptance and liquidity in the market was maintained. So relevance had to do both with volume and market share – we wanted securities to be liquid enough on the MBS side.” Housing Goals Chris referenced HUD’s November 2004 press release announcing new, higher HUD goals for the 2005-2008 time period and asked if Mr. Winer knew about the increase in affordable housing goals. Mr. Winer said, “when I was doing market risk portfolio I wasn’t paying much attention to housing goals and how they’d affect other businesses in the company. I do recall conversations about the HUD goals though and what they looked like.” Chris asked if increasing Fannie Mae’s purchases of non-traditional mortgage and subprime product was related to meeting affordable housing goals and Mr. Winer said “the Alt-A acquisitions were not supportive of housing goals and initially the subprime triple-A – it wasn’t clear how they’d fit in to goals. Eventually, they were determined to support housing goals and then that became a factor in increasing the purchases.” Chris clarified to make sure he correctly understood that at the outset of Fannie Mae’s increased participation in riskier market segments it was unclear if triple-A rated subprime PLS would help the company hit housing goals, and Mr. Winer said that was correct. Page 3 of 13 Fannie Mae’s Risk Management Function Chris asked if when he served as Interim CRO Mr. Winer was aware of the OFHEO Special Examination. He said that he was, and that his role in response to OFHEO’s exam was to build out a centralized risk management function, develop corporate-wide risk policies, and develop comprehensive risk reports for internal and board-level reporting. He explained that the reports would be at least quarterly reports that explained the chief risk characteristics of each business unit. He said that it was a challenge to determine how best to communicate the risks across business segments, but that most people within the business, especially within capital markets and single family business, were cooperative. Tom Stanton said that Fannie Mae had “a few powerful barons” in charge of businesses and asked what was needed to integrate the CRO into that “constellation.” Mr. Winer said that, “as designed by the CEO and board, the CRO was designed as an oversight function as opposed to having a direct risk manager in each activity… it was mainly a communication-escalation point than a business-directing position.” Tom Stanton asked if that was a common structure in Mr. Winer’s view, and Mr. Winer said that it varied from company to company. “It isn’t a role that has existed for decades so how it is structured is an evolving thing. What I did [when I started to design the structure] was go to five to ten other institutions to visit with the CROs to try to get a sense of their roles and responsibilities, and to see how they fit into their companies and to learn mainly how they think about the job,” he said. Tom asked if after his review of other institutions Mr. Winer was comfortable with the definition of CRO at Fannie Mae, and Mr. Winer replied that, “ours wasn’t an unusual definition. There was a variety out there and in some cases [the CRO] was very involved in all business decisions, and there were many like Fannie Mae with embedded risk officers and an oversight risk management function.” Chris asked if the risk management function had to be signed off on by OFHEO, and Mr. Winer said that he thought so but did not recall specifically. He said that OFHEO was kept apprised of the structure, however. Fannie Mae’s Increased Purchases and Guarantees of NTM Product Chris asked if as Fannie Mae increased its purchases of riskier products anyone within the company warned against the new strategy. Mr. Winer said, “at different times there were always people asking questions, and I know when I looked at the business analytics and in particular starting with the second half of 2006 and first half 2007, there were people on my team concerned about discrepancies with analytics and if we were properly compensating for risk, and if the compensation we were actually getting for the risks in the Alt-A space [was adequate].” Mr. Winer said that Bob Kazdin, a director in the risk department, was the most vocal about those discrepancies. He said that Mr. Kazdin was head of group that applied pricing and acted as a liaison between the builders of analytics and the businesspeople who used them. Chris asked Page 4 of 13 Mr. Winer to explain Mr. Kazdin’s concerns in more detail, and he replied that, “his concern was over pricing. He was concerned that we were under-pricing risk relative to what the analytics said.” Tom Stanton asked why the company would do that, and Mr. Winer said, “well there were numerous discussions within the company – the prices we were getting were consistent with what the market pricing was.” Chris asked if there was concern that the models were generating prices that were not sufficient for the levels of risk, and Mr. Winer said that the models were coming up with the pricing, and then the compensation Fannie Mae got was less than the model price. In response to Tom Stanton’s question about what kinds of models Mr. Winer’s group built, Mr. Winer said that everything in the guarantee book ran through his models, and that his group built all underlying models, and that other business groups took the models and built analytics. Chris asked if in the 2006-2007 timeframe Mr. Kazdin changed his opinion about under-pricing, and Mr. Winer responded that “Bob left. That was when issues were most pronounced and most visible. The issues continued through 2007.” Chris asked what was being done to address those concerns, and Mr. Winer said that, “models were always being changed, but this issue never went away.” Tom Stanton asked if that was a significant fact in his view. “It was viewed as a very significant issue by the executive management of the company because the difference was a big driver of some of our GAAP financial results. Day one losses on guarantee contracts were not entirely a result of change between model pricing and actual pricing, but it was probably the single biggest factor,” he said. Stanton asked if the problem was more systematic and persistent over time, and he asked what caused day one losses to be a significant concern. Winer said that although he is not an accountant, he understood day one losses to be the difference between the model fee for Alt-a and the charged fee, which was a significant issue in first half 2006. Chris asked if given that the model fee was greater than the fee Fannie Mae actually received, there were discussions within the company in the second half 2006 and into 2007 about not purchasing mortgages that were potentially under-priced, according to the models. “There was that discussion,” Mr. Winer said. He said that they compared the information from the company’s models with the available information about how others viewed the same risk – i.e. mortgage insurance companies and credit ratings agencies – and that “there was a question of who had it right.” Chris asked if they looked at the models underlying those views, and Mr. Winer said that they would see the outputs of the models, but couldn’t say what accounted for the differences. Chris asked if the prices generated from mortgage insurance companies and rating agency models were more in line with market prices and Mr. Winer said they were. Chris asked which companies Mr. Winer was specifically referencing, and Mr. Winer said Fitch, and mortgage insurers. Page 5 of 13 Tom asked if Mr. Winer was aware of a summer 2007 strategic planning meeting, and the strategic planning document that was presented during the meeting, which stated that Fannie Mae needed to start taking more risk. Mr. Winer said that he thinks that he recalled the meeting, and that if he was thinking of the right one, there was a discussion about Fannie Mae becoming a “broader participant in the market.” He also said that part of those discussions included talking about the risk transformation facility, or “RTF,” which could potentially enable the company could become a “broader participant” in the market it without actually taking more risk. Tom asked if the Chief Risk Officer was consulted about the RTF, and Winer said that by that time, he was no longer CRO, but that he recalled doing one small multi-family loan with the RTF. Tom then asked if Mr. Winer was familiar with Project Phineus, and he said “I believe so, yes. I remember seeing packages,” although he did not recall the specific recommendations of the report, and he did not recall if the CRO’s office was consulted on the project. He said that he remembered working with Citi and McKinsey regarding strategic choices for the company, although he didn’t recall specifically when that occurred. “As I recall, most involvement was around the commercial paper markets portfolio and how to think about how to add value to the capital markets portfolio,” he said. He recalled that the work with Citi and McKinsey involved thinking about managing manage market risk, liquidity risk, and about “how having a portfolio integrates with having a guarantee business.” He said that he was personally less involved in credit risk. Chris said that the investigation so far has shown that Fannie Mae increased its purchase and guarantee of non-traditional mortgage products – and Alt-A loans in particular - in the 20042005 timeframe. Chris asked Mr. Winer to comment on that, given his earlier remarks about concerns within the company in 2006 about the difference between the guarantee-fee the model found and the prices the market would actually bear. He also asked there a concern regarding the model versus market price prior to 2006. Mr. Winer said that as he recalled, the model price reflects both individual loan characteristics and the credit enhancements around those products, and that discrepancies between the model and the market prices really started to occur in 2006. Chris asked if there was any chance that his recollection has any connection with the time he left the CRO position for the Business Analysis and Development (“BAD”) group. Mr. Winer said that he saw the discrepancy more routinely in BAD, but that he would have seen it earlier as well. Mr. Winer explained that GAP reports – granular, 6-7 page reports - would show the difference between the market price versus the model price. He said there were also monthly reports on the fair value calculations of new guarantees, and the difference between the value of the credit obligation, compared to value of expected revenues from that credit obligation. Chris asked what else Fannie Mae did to address concerns about the guarantee fees. Mr. Winer said that credit enhancements were used to reduce the risk of some guarantees. Chris asked if he ever discussed his concerns with OFHEO, and he said “Yes, I did discuss with OFHEO,” although he said he was having a hard time remembering the people with whom he spoke. He did say that most of his discussions would have occurred with people overseeing the model risk. Page 6 of 13 Chris asked if the name Austin Kelley rang a bell, and Mr. Winer said that he knows Mr. Kelley now, but he wasn’t sure if he spoke to Mr. Kelley at the time. He said that when he was Chief Risk Officer, he met regularly with John Kerr, the Examiner in Charge (“EIC”), but that he had much less interaction with Mr. Kerr when he joined the BAD group. Mr. Winer explained that he would have at least monthly meetings with examiners at OFHEO to update them and answer their questions. Chris asked if Mr. Winer recalled any of the regulators’ responses to his concerns about outputs from models versus market pricing for guarantees. Mr. Winer said that he did not remember their reactions. Chris asked if he recalled any conversations about why the company continued to buy the product given those concerns, and Mr. Winer said that he didn’t. Tom asked if it was Mr. Winer’s regular practice to memorialize the meetings with regulators, and Mr. Winer said no. Chris asked what impact the fall 2007 mortgage market liquidity shocks had on pricing for various types of products. Mr. Winer replied that “my recollection is that there was a time lag between when markets would change and when our standards would change and when that would be reflected in our charge prices – 6 months or so.” He explained that the lag was due in part to contracts with customers, and also “because of a belief that when we had prices in place that those prices should cover loans originated during the times during which prices were in place.” He said that the issue was “less the contract than a belief that we should be dealing with a business where we would give them enough time to keep prices consistent for a certain loan so new charges wouldn’t be in effect for 6 months.” Chris asked if there were still concerns regarding pricing differences between market and model price even in the second half of 2007, and Mr. Winer said that there were, in part because of the lag time issue he just described. Chris asked where besides the GAP and Fair Value reports the gap between market and model price would be reported, and Mr. Winer said that those were the standard reports, and that there was also a lot of analysis performed by accounting regarding losses on guarantee contracts. He also said that his team would analyze the outputs of the company’s models with model outputs from Fitch and mortgage insurance companies. Chris asked if he continued to be concerned into 2008, and Mr. Winer said that “it diminished in 2008 but I think still a concern in early 2008.” He said that the concern diminished because Fannie Mae was able to increase the price it was able to charge. Chris asked if in the timeframe when GAP reports showed reason for concerns regarding pricing, there was a feeling within the company based on looking at other outputs that Fannie Mae’s model was wrong. Mr. Winer said that “particularly among those responsible for models,” “we believed our models were the best reflection of the risk.” He added a caveat that “we were also building models based on very limited data, especially for Alt-A, and I think the consensus in the company was that the models were wrong. Because there was a lot of uncertainty about what models could portray and because of what other people’s models had different pricing.” He added that “I didn’t think the models were wrong,” however. Page 7 of 13 Chris asked who did think Fannie Mae’s models were wrong, and Mr. Winer said that “ A number of people in the business, analytic people outside BAD were critical of models.” Chris asked if he recalled any specific individuals, and Mr. Winer said he didn’t really know, although “the most vocal was Rob Levin who felt that -- asked me can you show me why you think you’re right and everyone else is wrong, to which I said I think we’re doing the best job but I can’t prove it,” he also said that Eric Rosenblatt argued that models were wrong because Fannie Mae included national home price declines, which he thought were “unrealistic scenarios.” “We had a whole range – 2,000 paths that would allow home prices to go down 30-40% and he thought that was unrealistic,” Winer said of Mr. Rosenblatt. Tom asked if he was disappointed in the parameters Fannie Mae employed in its models. “The model assigned a low probability to the observed event, and I wish it’d been higher,” he said. Chris asked if Dan Mudd ever weighed in on the models, and Mr. Winer said that Mudd was in the room during discussions about models, but that he didn’t recall Mudd stating his positions. Chris asked the same question of Tom Lund, and Mr. Winer recalled that discussions with Mr. Lund were limited on the subject because the significant difference between the model output and Mr. Lund’s view of market pricing meant that model outputs “weren’t a useful tool to him in running his business.” Chris asked if he discussed the issue with Enrico Dallavecchia, and Mr. Winer said that he did not recall discussing the issue directly with him. Chris asked if Mr. Winer ever attended Board meetings, and if the topic ever arose during them if so. He said that he did attend some meetings, but that he would more commonly attend meetings of various board committees, including the risk policy and housing and community development committees. He said that “the discussions I recall were in particular on losses on guarantee contracts – the day one loss issue. I don’t recall that the general issue of model fee versus charge fee was discussed. Chris asked if when he was in charge of the BAD group, people in his department thought the models were inaccurate. “Well everyone thinks they’re inaccurate – they’re our best guess,” he said. “But no – nobody in my group thought they were overstating risk.” Chris asked if Mr. Winer would give him a sense of magnitude of these concerns. “I think it was a relatively big deal within the company,” he said. Chris asked if the issue was ever elevated to the Board of Directors. Mr. Winer said that that he believes the day one loss issue was elevated, but that he was not sure if the GAP numbers were. Chris asked if Mr. Winer ever expressed the view that the company shouldn’t have been buying the products given the model outputs. “The view I expressed wasn’t a do or don’t do, but a do within very prescribed and limited parameters,” he said. “It was a question of volume and extent, not do or don’t do.” He said that he didn’t set parameters. Chris asked if, grating that he wasn’t a decision maker on the issue, Mr. Winer agreed with level of Alt-A and non-traditional product Fannie Mae purchased and guaranteed? He replied that “I didn’t agree but I did recognize that I could well be wrong. So I thought it was not an irrational business decision.” Page 8 of 13 Chris asked why Mr. Kazdin left the company. The company offered an early retirement program and Mr. Kazdin decided to accept that program in late 2007 or early 2008, if he recalled correctly, Mr. Winer said. Chris asked if Mr. Winer recalled anyone within the company who was vocal about curbing the purchase and guarantee of riskier, non-traditional mortgage products. He said that Mr. Rosenblatt was later “equally vocal that we should not be in these activities” starting after the fall of 2007 “when the market started to turn.” He said that there were other people on the BAD team who were vocal internally but less externally vocal. He added that “eventually everybody did” change their minds about the activities after the market turned. Chris asked if there ever came a time when you he became comfortable with the decision to be in riskier, non-traditional mortgage products. He said that by the time Fannie Mae responded to market price increases, the company had changed underwriting guidelines, so the prices and nature of products had changed. He clarified that the changes to underwriting standards were announced at the end of 2007 and were implemented in the second quarter of 2008, if he remembered correctly. Chris said that we’ve seen in documents that Fannie Mae’s market share was going down, and then increased dramatically in 2007. Mr. Winer said that he thinks there was “an increased focus on using a model to adjust pricing to compensate for risk. So we worked on a number of pricing changes during that time.” Chris asked if he was involved in any discussions with OFHEO, Treasury and Federal Reserve representatives in the fall of 2007, and Mr. Winer said that he doesn’t recall being involved, but that he may have been. Chris asked if Mr. Winer was involved in any of the negotiations or discussions that led to the March 19, 2008 press release announcing the lifting of the portfolio caps and the relaxation of the capital surplus charge. Mr. Winer said that he recalled providing analytic support and document preparation support for those discussions, but that he really involved beyond that. Chris asked if he knew who was involved, and Mr. Winer said that he believed Mr. Levin was. Chris asked if Mr. Winer knew why Mr. Levin left the company, and Mr. Winer said that he didn’t know. Chris said that from the fall of 2007 through the spring of 2008, there liquidity was constrained and the Street was pulling back, and he asked Mr. Winer if people within Fannie Mae said that the company shouldn’t increase market share given what the models showed. He said that he recalled that the discussion was more along the lines of “we shouldn’t be earning excessive returns, but rather pricing in a reasonable way and supporting the market.” Chris asked if the company was getting a fair price for the risk based on what was going on in the market, and if there was added concern in the company about ratcheting up its purchase and guarantee of mortgages during that timeframe. Mr. Winer said that the “most heated debates were about whether we should continue to have the lag after we decided changes were put into place” on underwriting and pricing changes. Chris asked Mr. Winer to clarify if the lag was contractural or discretionary, and he said he believed they were discretionary. Page 9 of 13 Tom asked when Mr. Winer first became concerned that the amount of risk “was not just going to be a matter of pricing but could relate to more substantial losses?” Mr. Winer said that when they started seeing home prices go down nationally, he feared the credit losses Fannie Mae would experience. Tom asked what happened at that point, and Mr. Winer said that “people were already there. We were showing numbers projected in tens of billions. They had 40-50 billion in capital.” Tom asked what Mr. Winer’s team did with the information that loss forecasts were greater than capital, and Mr. Winer said that his team was asked to produce a number of different scenarios and to develop a capital management strategy for the company, to see what approaches Fannie Mae could take to try to maintain capital. In response to Chris’s question, Mr. Winer said he couldn’t recall when his group was asked to develop those reports. Mr. Winer clarified that his department was responsible for the loss forecast reports forecasting losses in tens of billions of dollars, and that the reports were “an input into [the] broader analysis” at Fannie Mae of the company’s financial situation. Chris asked if Mr. Winer’s group predicted insolvency (i.e. negative capital), and Mr. Winer said that he didn’t think so, although some forecasts showed capital falling below regulatory minimums. Chris asked if Mr. Winer would provide input regarding declining housing prices in the forecasts, and he said that several groups came up with home price scenarios, and the BAD group would run them through a loss projection model. Mr. Winer said that he did not recall the largest home price decline assumption that went into a scenario BAD analyzed. Chris asked if it could have been as large as a 30% home price decline, and Mr. Winer said that he could be wrong, but as he recalled, the most severe declines were in the 10% range. Federal Reserve & OCC Summer 2008 Review Chris asked Mr. Winer if he had any interaction with Fed and OCC examiners in the late summer of 2008 when they reviewed Fannie Mae’s financial condition, along with Morgan Stanley, at the Treasury Department’s behest. Mr. Winer said yes, and that he recalled that the representatives from the Fed and the OCC “were really focused on our allowance calculations” (as opposed to the cash flow models he had been discussing during the course of this interview), and that he and his team provided data and analysis to help the regulators in their review. Chris asked what they thought about the loan loss allowance, and Mr. Winer said that they should it should be higher. Mr. Winer said that he was not part of the team that calculated the loan loss allowance, although he was a member of the committee that advised the CFO on the allowance. He said that the methodology for loan loss allowance (which was calculated in the Finance Department run by Steven Swad) involved dividing the company’s portfolio into cohorts, looking at the average historical performance of the cohorts, and assuming that over time, that cohorts would perform like they had historically. He said that on that basis, they would assign losses to each cohort, then aggregate. Chris asked what Mr. Winer’s opinion was of that methodology. Mr. Winer said that the accountants told him that the allowance can’t incorporate views of future conditions, and since he is not an accountant, he assumed they were right. He said he found the methodology to Page 10 of 13 be “cumbersome and often hard to see through the detail,” but that it seemed consistent with what he was told was proper accounting. Chris asked if he thought the company was under-reserved. Mr. Winer said he did not think so, but he isn’t an accountant. Chris asked Mr. Winer if he thought that Fannie Mae was going to incur more losses in the future than it had in its loan loss reserves, and Mr. Winer said “Yes.” He said that to the best of his recollection, he first thought the reserves could be exhausted when “we started observing declines in home prices.” Chris asked if there was a difference in opinion between the Fed/OCC group and what Mr. Swad’s finance department projected. Mr. Winer said that “As I recall there was. [There were] two issues. One, they recommended a different analytic approach to computing the analysis of whether or not you take into account future economic conditions, and two, they thought you should include your view of future economic conditions.” Chris asked if Mr. Winer could elaborate on the first point, and he said that the Fed and OCC used a roll-rate analysis to determine the probability that delinquent loans would ultimately default. Mr. Winer said that he didn’t have an opinion on the validity of the roll-rate analysis, though he said that he did think that the OCC/Fed approach flawed from an analytical standpoint, but that he didn’t know if it would prove to be good if it was implemented. Accounting issues raised in March 2008 Barron’s article Chris asked if Mr. Winer recalled the March 2008 Barron’s article questioning whether or not Fannie Mae was insolvent as a result of questionable accounting. Mr. Winer said that he didn’t particularly remember it, although he did recall people calling Fannie’s accounting practices into question. Chris said that understanding that he didn’t recall the article specifically, one of the issues raised was the value of Alt-A and subprime private label securities on Fannie Mae’s books. Chris asked if he recalled any discussion in 2007 or later about whether or not the securities were overvalued. Mr. Winer said he didn’t. Chris asked Mr. Winer if he was involved in the preparation of Fannie Mae’s financial statements. He said that the BAD group provided some models used in the preparation. Chris asked if Fannie Mae valued its PLS by finding a price in the market or if it relied on pricing models. Mr. Winer said that the company used third party pricing to value the securities, although he didn’t know if they always did that. Chris asked if he recalled discussion during or after late 2007 about whether or not the company needed to take additional marks on the PLS, and Mr. Winer said he didn’t recall. Chris asked again if he recalled any discussion of being under-reserved in 2007, and Mr. Winer said that “part of what committee always did was [talk] about the loss reserves. I don’t remember a heightened level of discussion” during that period, however. Chris asked if he personally ever held the view that Fannie Mae was under-reserved. Mr. Winer said “No.” He said that he anticipated challenges with its reserves under the projected losses, but that the reserves were consistent with the company’s requirements, to the best of his Page 11 of 13 knowledge. Mr. Winer said that the BAD group did not have any role in determining the loan loss reserves. Chris asked if the BAD group had any involvement in looking at value of company’s deferred tax asset (“DTA”). He replied that they provided some analytic support, in particular on the part of the calculation that involved projecting future cashflows from low income housing tax credits. Chris asked if he thought that the DTA was overstated, and Mr. Winer said no. Chris said that he had heard that at some point Fannie Mae made unsecured loans to borrowers to bring them current, and asked if the BAD group had any involvement in valuing those loans when they came out of the trust. Mr. Winer said that again, the group would provide some analytic support, but did not have any further involvement. Chris asked if Mr. Winer knew that the company was “curing” delinquency in that way, and Mr. Winer said he doesn’t recall the details but the program did sound familiar. Chris asked if he knew when the program started, and Mr. Winer said that he didn’t know precisely, but he didn’t think it was a longstanding program that would have been initiated before 2007 or 2008. Chris asked if he discussed the program with the Fed or OCC examiners during the course of their summer 2008 review, and Mr. Winer said no, and that he didn’t hear anything from the regulators regarding that program. Chris asked if Mr. Winer dealt with Morgan Stanley in August 2008, and Mr. Winer said he didn’t recall. “I took part in responding to data requests which were for Morgan Stanley, and [I remember] having discussions with people to try to better define the data that they wanted,” but that was the extent, he said. He said that Morgan Stanley wanted information about the characteristics of the loans in the guarantee book, and that some of their worked involved producing loss forecasts, though he didn’t know what the forecasts were. Chris asked if the BAD group also produced loss forecasts, and Mr. Winer said they did, and those forecasts were incorporated into company reports, which were produced on a quarterly basis by mid-2007. He said that at some point it changed to monthly but that he didn’t recall precisely when that happened. Chris asked if the Fed and OCC representatives looked at those reports, and Mr. Winer said he didn’t recall. Chris asked if he recalled any criticisms of the reports. Mr. Winer said that he recalled that Herbert Allison had questions about them once he was appointed CEO after conservatorship was announced. “He questioned the reliability. He was concerned that they were believable – if they were too small. He seemed to have that concern the first day he was there,” Mr. Winer said about Mr. Allison. Chris asked if the loss forecasts were used to estimate the loan loss allowance, and Mr. Winer said they weren’t, that they were used to project future earnings and capital at the company. He said that he thinks they forecasts were provided to the forecasting committee under the CFO, who would take them and try to map them into an overall assessment, which he said that he assumed was shared with executive management. In response to follow up questions from Chris, Mr. Winer said that the BAD group really provided analytic support for all business, but there wasn’t a specific report it prepared regarding the PLS. He said that with regard to reports on capital, the group would look at prepayment Page 12 of 13 models, “back-testing reports” and would look at the reports on the relative value of different securities. He also said there was a home price monitor report that describes what the group sees happening to home prices across the country. Chris asked if the GAP reports discussed earlier show changes in the GAP over time, and Mr. Winer said he didn’t know one way or another. Chris asked where the decision would be made to go below the modeling price in setting a fee, and Mr. Winer said that there’s “not one answer.” He said it would be in the Single Family business “chain of command,” and that depending on the nature of the transaction, different people would sign off on the decision. He said that most of the Alt-A acquisitions were done through Fannie Mae’s investing channel. He said that model fees are just a guideline, that they are only one factor in determining the right price. Chris said that he did not have further questions, thanked Mr. Winer for his time, and concluded the interview. 4811-3784-2693, v. 3 Page 13 of 13
@FedFRASER