Earnings Call Transcript - FLG Q1 2020

Operator, Operator

Good day, and welcome to the Flagstar Bank First Quarter 2020 Earnings Call. Today’s conference is being recorded. At this time, I would like to turn the conference over to Ken Schellenberg, VP, Investor Relations. Please go ahead.

Kenneth Schellenberg, VP, Investor Relations

Thank you, Michelle, and good morning. Welcome to the Flagstar first quarter 2020 earnings call. Before we begin, I’d like to mention that our first quarter earnings release and presentation are available on our website at flagstar.com. I would also like to remind you that any forward-looking statements made during today’s call are subject to risks and uncertainty. Factors that could materially change our current forward-looking assumptions are described on Slide number 2 of today’s presentation in our press release and in our 2019 Form 10-K and subsequent reports on file with the SEC. We are also discussing GAAP and non-GAAP financial measures, which are described in our earnings release and in the presentation we made available for this earnings call. You should also refer to these documents as part of this call. With that, I’d like to now turn the call over to Sandro DiNello, our President and Chief Executive Officer.

Alessandro DiNello, President and Chief Executive Officer

Thank you, Ken. And good morning to everyone listening in. My apologies for the delay in getting this going today. Unfortunately, our conference call company is having some difficulties. I don’t think it has anything to do with all of us being remote. So hopefully, the rest of this call is much smoother than it has thus far. I hope that all of you and your loved ones have been able to stay safe and healthy. In the spirit of proper social distancing, I’m joined remotely this morning by Jim Ciroli, our Chief Financial Officer; Lee Smith, our Chief Operating Officer; Kristy Fercho, our President of Mortgage; and Steve Figliuolo, our Chief Risk Officer. I’m not going to spend much time today on our results, except for some commentary on our provision. I think it’s more important to tell you how we’ve adjusted our operations in the face of COVID-19 and how I think we’ll operate going forward given the uncertainty of the length of this pandemic. With respect to our results, it was a strong quarter with earnings of $0.80 per share, stronger still if you look at our pre-provision net revenue. Details will come from Jim and Lee in their remarks. Let me start with some comments about our provision, which totaled $14 million for the quarter. Admittedly, the provision for the quarter was difficult to estimate. We had no option but to use an immature model to estimate life-of-loan losses in perhaps the most uncertain economic time in our history. And when you add the unknown impact of government assistance programs to the mix, it was a daunting endeavor. Frankly, I don’t think that there is a model that exists that can accurately predict life-of-loan losses at this stage of the game. That said, based on our current data, we’re comfortable with our analysis. We have a strong commercial portfolio that, at the end of the quarter, had no nonperforming loans. Thus far, we’ve seen limited weakness. Our portfolio is diversified with no outsized exposure to any geography or industry, including little exposure to the auto industry. Our loan concentration policy is highly granular and conservatively structured, built to be resilient to a recession. Now we find ourselves in a recession, perhaps a deep one that may linger for the next 18 months. A hallmark of our lending philosophy has been to bank only proven operators. The eleventh commandment of Flagstar is: thou shall not lend to anyone that has walked away from a bank obligation. These principles, coupled with the fact that virtually all our borrowers are recession-hardened, will help mitigate credit concerns going forward. Finally, I’ll remind you that while we’re a $27 billion bank in terms of assets, our commercial credits, net of warehouse loans, total less than $5 billion. Thus far, every borrower is working with us. To date, they are all positive about their ability to weather the storm and remain going concerns. Thus far, 252 commercial borrowers representing $548 million in unpaid balances, just 11% of our commercial book, have requested payment deferrals and all were in good standing prior to their request. Business banking requests account for $71 million, with C&I and CRE representing $287 million and $190 million, respectively. It’s worth noting that we received no deferral requests from our homebuilder clients thus far. We are on top of every one of our credits, both monitoring and actively managing. We have assisted many of these customers with a PPP loan, and we will be assisting even more through the Main Street Lending program. We’re being careful in connection with new commercial loan requests outside of our warehouse business, and we are pulling back on commitments wherever it’s prudent and the opportunity presents itself. Even inside warehouse lending, we’ve taken a more conservative approach: essentially accepting no jumbo or non-QM loans as collateral, requiring a tighter credit box for government loans, and lower advance rates for loans that we believe are at higher risk. As to our consumer loan portfolio, it consists largely of low-LTV, high-FICO mortgage loans and high-FICO HELOCs. Thus far, we’ve seen no material weakness in those loans. I think it’s a given that the economic forecast at June 30 will be weaker than that at March 31. So I fully expect that our ACL model will suggest a significant allowance adjustment for Q2, but just how much is impossible to predict. My opinion is that GDP will contract perhaps as much as 10% in Q2, or as much as 40% on an annualized basis, followed by slow growth in the ensuing quarters. Unemployment might reach 20% in Q2, then gradually decline as GDP improves. So not quite what I would call a recovery. I realize this opinion is probably on the pessimistic side of the universe of thinking, but my view is that it’s most prudent to prepare for the worst and hope for the best, and that’s exactly what we’re doing. Next, I’d like to speak to how our company responded to the pandemic, and I’ll start by saying I could not be more proud of our team and how we’ve adjusted. We could not have gone much better. We prepared for a possible pandemic in our business continuity plan. As part of that plan three years ago, we moved a vast majority of our people to laptops. Additionally, we secured enough VPN licenses and bandwidth to move quickly. The transition went really well. Today, over 80% of our employees are working remotely. Interestingly, productivity has not dropped. In the last two weeks of the month, we processed over $2 billion of mortgage loans, increased warehouse balances by $1.5 billion, worked out arrangements with every commercial customer seeking a deferral, accepted thousands of requests for mortgage forbearance, and built a new technology platform, literally overnight, to launch the Paycheck Protection Program and got it up and running 36 hours after the guidelines were released. We took and processed over 3,000 applications for $400 million in loans. Remarkably, all these processes were performed and managed remotely. Additionally, on March 16, we eliminated in-branch service at select bank branches, and then a few days later, moved to drive-up service only with lobby access by appointment; not one customer complaint has reached my desk. In fact, deposit growth has been encouraging. We reduced rates aggressively following the Fed move and still saw deposit balances increase about $900 million during the quarter. We’ve seen a further increase in deposits in April of over $350 million from our banking customers. While I don’t know when normal will return or what it will look like, I can tell you we will be very conservative in moving people back to our facilities. That’s the right thing for our team. Given how smoothly we are operating, there’s no need to get ahead of a clear flattening of the curve. As always, we’ve been there for our communities. To date, we’ve contributed almost $600,000 across the country to support the COVID-19 cause, including $300,000 to a textile company that shifted to manufacturing masks that are being donated to hospitals. Only 17 of our 4,700 Flagstar employees and contractors have tested positive for COVID-19, and best of all, they all have either recovered or appear to be on the road to recovery. We believe that’s, in large part, because we started our social distancing early on March 13. Also, I’m very pleased to tell you that we surveyed our team and 98% felt that Flagstar responded effectively to both employee and customer-related concerns. As difficult as the last six weeks have been, I do think we retain the resilience of our business model in action. Here’s why. First, I think we can sustain net interest income on the back of our warehouse business, which currently has outstanding balances of approximately $3.8 billion, with a month-to-date average balance of over $4 billion. Warehouse lines are now almost entirely secured by conforming agency and government products. The portfolio provides strong returns with virtually no credit risk, and 91% of these loans had rate floors. We’ve been withholding new commercial loans outside of warehouse, and we expect to grow net interest income. Second, the mortgage business continues to be strong. Our gain-on-sale month-to-date is over $90 million and shows no signs of weakening with over 95% of the locks representing conforming agency loans. While I want to avoid opining on how long this performance will continue, I think it demonstrates the power of this business in the current environment. While we’re not guiding on net interest income, if industry projections of a 10% increase in originations for 2020 as compared to 2019 hold true, our warehouse and mortgage businesses will thrive, and we will be in a position to build a credit war chest without the need to raise capital should it be needed. To put it mildly, giving guidance right now is premature. However, I believe Flagstar is well positioned to weather the storm and come out stronger on the other side. Our profitability is strong, our capital is strong, our allowance is strong, our liquidity is strong, and our business model was working precisely as we designed it. Finally, I’d like to make it clear that we don’t expect to change our dividend posture in the near term, and we’re not contemplating stock buybacks at this time. Obviously, my comments today have been unlike anything before because we are in unprecedented times. My goal has been to give you my perspective on how I see Flagstar performing in the near future in the context of the SEC’s recent guidance and safe harbor protections. Drawing on more than 40 years of experience at Flagstar and a deep understanding of the company and the quality and caliber of its employees, we are committed to doing the right thing for our employees, customers and our communities. Protecting their health will secure Flagstar’s health. We will focus on the opportunities this environment gives us, and though we will continue to manage the risks we face conservatively, we will not dwell on them. If we do that, I firmly believe we will produce the best possible results for our shareholders. With that, I’ll turn it over to Jim.

James Ciroli, Chief Financial Officer

Thanks, Sandro. Turning to Slide 7. Our net income this quarter was $46 million or $0.80 per share. This performance compares to $58 million or $1.00 per share last quarter. The decline on a linked-quarter basis was largely due to the $14 million credit provision this quarter. We had adjusted net income of $37 million or $0.64 per share in the same quarter last year. Diving deeper into this quarter’s performance, our pretax pre-provision earnings were $70 million this quarter compared to $69 million last quarter. Net interest income was down $4 million or 3% over the prior quarter. Average earning assets grew $442 million. In line with our expectations, the net interest margin decreased by only 10 basis points despite 150 basis points of rate cuts in March that were unexpected. This performance was driven both by our strong core deposit franchise, consisting of granular retail customers, and by a well-positioned, well-diversified loan portfolio that has a large composition of fixed-rate assets and floors on many of its variable-rate loans. We’ll review these numbers further on the next slide. Mortgage revenues were $96 million, a decrease of $2 million or 2% compared to the prior quarter. During the quarter, we saw gain-on-sale margins increase significantly as primary-secondary spreads widened to historic levels and lenders worked to manage capacity. However, in the second half of March, extreme market dislocation and volatility occurred, and we experienced $45 million of hedge losses, which I’ll discuss further in a moment. Asset quality remained strong. Net charge-offs were 8 basis points, and nonperforming loans were relatively flat to year-end. Our allowance for credit losses, or ACL, which includes the reserve for unfunded loan commitments, was $152 million at quarter end, up from $110 million at year-end. At these levels, our coverage of loans held for investment is 1.5%. This was the result of our January 1 CECL adoption and the $14 million credit provision this quarter. We’ll provide more details when we get to the asset quality slide and take a deeper dive into CECL. Capital also remained solid. Despite our balance sheet ending the quarter at $27 billion, up from $23 billion at year-end, all capital ratios remained above the stress buffers that we’ve established based on our DFAST models. Total risk-based capital was 11.2% at March 31. Our CET1 ratio was 9.2% and our Tier 1 leverage ratio was 8.1%. We’ll go into more details on capital later, so let’s turn to Slide 8 and dive deeper into the income statement. Net interest income decreased $4 million to $148 million this quarter, which was 3% lower than last quarter. Nearly half of this decline resulted from having one less day in the quarter. The results reflect the 2% increase in average earning assets led by investment securities, loans held for sale and commercial real estate loans. Deposit costs came down 9 basis points, while average deposit balances were relatively flat quarter-over-quarter, excluding a decline in broker deposits. We’ll dive deeper into net interest income and our interest rate risk position on the next slide. Noninterest income decreased $5 million or 3% to $157 million due to lower mortgage revenues. Our gain-on-sale revenue was $90 million, representing a decrease of $11 million or 11% from the prior quarter. Fallout-adjusted locks increased 36% to $11.2 billion. The gain-on-sale margin decreased to 80 basis points. Channel margins were strong, especially in our delegated bulk channel. The Fed’s buying of mortgage-backed securities beginning on March 16 pushed prices on MBS much higher. This price change impacts the values of the TBA mortgage-backed securities that we sell short to hedge our pipeline and closed loans held for sale. As the pipeline and closed loans did not experience the same level of price appreciation, the short position declined in value more than the long position increased in value, creating $45 million of hedge losses. We’ve added Slide 40 in the appendix to provide more details on what we and the industry experienced. We were reacting quickly to this phenomenon, took actions to rebalance our hedge profile, and made pricing changes, which improved our position by month end. In April, the Fed modified their purchases and our actions took hold, which stabilized our gain-on-sale. Through last Friday, we recorded over $90 million of gain-on-sale revenue for April. While we won’t forecast what the second quarter gain-on-sale revenue will be, this partial month result indicates that the hedging issues have not persisted. We also had a $9 million improvement in the MSR return as a result of hedging gains this quarter. Loan administration income improved $4 million due to higher average loans serviced and a decline in the LIBOR-based credit that we provide to our sub-servicing customers for the custodial deposits that they control. Noninterest expense was $235 million, down $10 million from the prior quarter. The decrease was primarily related to $5 million of balance-sheet cleanup and discretionary expenses that took place in the fourth quarter and did not recur. Expenses were also down as mortgage loan closings declined 8%. Overall, mortgage expenses as a percentage of closings were consistent with the prior two quarters. Lee will provide more color on expenses later. We’ve added Slide 9 to provide more details on our interest rate risk position at the end of the quarter. While we won’t provide guidance on net interest income, we feel that we’re in a good position. Looking at our commercial loan portfolio, 95% of the loans are variable. As we parse that variable percent out, as Sandro mentioned, over 90% of our warehouse loans have floors in them. We also have floors in many of our C&I and CRE loans, either floors that don’t allow LIBOR to reset below zero or floors with LIBOR at a higher level. Consequently, we don’t see a lot of risk to spread compression coming from the asset side of the balance sheet. Looking at deposits, we have moved pricing down on our portfolio in reaction to the aggressive actions taken by the Fed to drive short-term interest rates. At quarter end, in the retail CD book, we had $1.8 billion of CDs that mature in the remaining nine months of this year and reset to lower rates. We also had $900 million of savings and money market accounts at quarter end. So we’re under some level of promotional rate that will come off; that promotional rate has an average cost of 1.9% and also repriced lower in the latter part of this year. Approximately 85% of our custodial deposits are controlled by our sub-servicing clients and naturally priced down as those rates are based on LIBOR. In summary, we believe that deposit rates could be a tailwind for the remainder of the year. Also on the liability side, we’ve executed interest rate swaps and entered into long-term FHLB advances to lock in lower-rate funding, laddering those out between three and seven years. We’ve now executed $1.5 billion of this strategy, securing long-term funding at an average cost of 61 basis points, well below our cost of funds in Q1. While it’s difficult to predict where rates might be for the remainder of this year, we feel that our interest rate risk position is in a good place, especially considering the liability-sensitive nature of the mortgage business. We feel that we can protect our net interest income and net interest margin in this environment, and believe that our net interest margin should be relatively flat to the first quarter. This perspective also contemplates the impact of the PPP loans that we have originated under deferral of the FDIC on those loans. To illustrate how it’s positioned us, I would share that while the first quarter net interest margin was 2.81%, March’s net interest margin was 2.84%, showing the effectiveness of the actions we’ve taken to protect net interest income. Let’s turn to Slide 10, which highlights our average balance sheet this quarter. Average loans held for investment declined $0.3 billion, driven by a decrease in average warehouse loans and residential first mortgages. Most of the decline in warehouse was seasonal and the decrease in average residential mortgages was the result of loan prepayments. At quarter end, we saw higher levels of warehouse loans and loans held for sale. The higher levels of warehouse loans resulted from our actions to continue to support customers in the mortgage finance business, even as we took a more conservative approach with non-QM and jumbo loans, tightened the credit box on government loans that we will support on our warehouse lines, and reduced advance rates on certain collateral. For loans held for sale, we saw elevated levels of newly created mortgage-backed securities that could not be sold at reasonable prices due to supply imbalances in the market. Ultimately, we delivered these bonds into our TBAs in April. While this inflated our balance sheet and compressed our spreads a bit, we were able to hold these assets without sacrificing execution. We expect that we’ll continue to have relatively higher balances in our warehouse and mortgage loans held for sale portfolios in the second quarter. Average deposits decreased $0.1 billion, driven by lower CD and wholesale deposit balances. Higher demand deposits partially offset this runoff. We continue to have a strong liquidity position, driven by the strength of our deposit base and access to multiple sources of liquidity, both on-balance-sheet with our high-quality securities portfolio and off-balance-sheet with our undrawn FHLB facilities. We also plan to fund our PPP loans using the Fed’s PPP facility and have ample access to borrow at the discount window. Finally, we continue to demonstrate significant capital generation capabilities with growth in our tangible book value per share of $29.52 at quarter end, up $0.95 from year-end. Let’s turn to asset quality on Slide 11. Credit quality in the loan portfolio remained strong. Delinquencies continue to be relatively low. Only 40 basis points of total loans were over 30 days delinquent, up 7 basis points from year-end. Our allowance for credit losses covered 1.1% of total HFI loans. This coverage now includes our reserve for unfunded loan commitments. Excluding warehouse loans from the denominator, given their relatively clean credit loss history and considering that substantially all of these are collateralized as agency and government-backed loans, our coverage ratio would stand at 1.5%, consistent with what we see other banks having under CECL. Considering the economic uncertainties related to COVID, we’re monitoring our entire loan portfolio, doing additional analyses within certain sectors and relationships, and staying in close communication with our significant borrowers. On Slide 12, we wanted to share with you our exposure to those industries that we believe are more likely to be most impacted. In total, we have just under $1 billion of outstanding loans in this category, representing only 7% of our loan portfolio. In the commercial loan portfolio, these balances totaled $328 million. Within this group, we’re focused on our exposures to automotive, hospitals and leisure and entertainment businesses, which includes restaurants. You can see that the exposure here is relatively low. We have no oil and gas exposure. In our commercial real estate portfolio, we have $647 million outstanding in the areas most likely to be impacted by COVID, including loans secured with hotels, retail properties and senior housing. Of the loans in this category, our average LTV is 55%, and our average debt service coverage was 1.6 times. We’ll be analyzing all of our borrower relationships this quarter. While we know this will be a challenging time for many people, we believe that our low level of exposure to these industries is the result of our well-diversified loan portfolio and the strong LTV and debt service coverage ratios, the result of our credit discipline. So let’s turn to Slide 13. We wanted to walk you through how we implemented CECL, how we got to the CECL balance at the end of the quarter and how we’ll be thinking about CECL in light of the uncertain economic outlook that we have right now. Overall, we ended the quarter with $152 million allowance for credit losses, consisting of $132 million of allowance for loan losses and $20 million in the reserve for unfunded loan commitments. The reserve for unfunded loan commitments is included in other liabilities on our balance sheet. Both are available to cover credit losses in the loan portfolio. In total, our allowance for credit losses at quarter end represents a 38% increase over what we reported at the end of 2019. In our adoption of CECL, we used three different forecasts of the next two years, which then reverted to a long-term average over a one-year period. These forecasts included an adverse projection that reflected severe economic distress. We weighted that model 30% in our day-one adoption. For our quarter-end estimates, we elected to use the March 27 Moody’s baseline, which reflects the economic distress caused by COVID and also contemplates some impact from the government actions taken to mitigate the stress. The forecast assumes that we are in a sudden, sharp and severe recession, only partially recovering later in the year. In this scenario, GDP contracted 18% in Q2 and HPI decreases 3% by the end of the year and unemployment spikes to 9% and moderates to 7% by the end of the year. We also judgmentally increased reserves in our CRE, homebuilder and C&I portfolios to provide additional coverage for industries and customers that we thought could be more exposed to the stressful conditions in our forecast. We’ve provided a portfolio-by-portfolio breakdown of the resulting ACL coverage ratios in our appendix. We also continue to maintain reserves for our loans with government guarantees and specifically measured loans whose amounts are not impacted by the CECL methodology change. Finally, as you’d expect, we’ve elected to defer the regulatory capital impact of adopting CECL until the end of 2021, after which it will be phased in at 25% per year. Turning to Slide 14. Despite tremendous balance sheet growth, our capital ratios remained solid and nicely above our stress buffers. Total risk-based capital was 11.2% at March 31, down only 31 basis points, while our CET1 ratio of 9.2% was relatively unchanged. Our Tier 1 leverage ratio of 8.1% actually increased 9 basis points on a pro forma basis with capital simplification as this ratio is based on average balances and the balance sheet growth in loans held for sale and warehouse loans happened toward the end of the quarter. I’d expect our Tier 1 leverage ratio to be lower next quarter. Between loans held for sale and our warehouse loan portfolio, we have approximately 540 basis points of total risk-based capital and over 400 basis points of Tier 1 leverage capital dedicated to these two asset categories that have very low-risk content. In warehouse lending, which has a 100% risk rate, we’ve had under $5 million of losses cumulative for the last 12 years, and we took a more conservative approach during the quarter for certain product categories. I’d remind you that we also hold the collateral for those loans while they are on our lines, and that collateral consisted almost entirely of agency and government-backed loans. Loans held for sale also had very little risk content, and over $2 billion of these balances at quarter end were Fannie, Freddie or Ginnie securities and were reflected in our trading portfolio. In summary, we believe that we’re operating at strong capital levels, given our low-risk balance sheet composition as more than half of our assets at quarter end were in categories that have very low-risk content: loans and trading securities held for sale, warehouse loans, investment securities, loans with government guarantees and cash. I’ll now turn to Lee for more insight in each of our businesses.

Lee Smith, Chief Operating Officer

Thanks, Jim, and good morning, everyone. We’re very pleased with our net income of $0.80 per diluted share for the first quarter, which increased tangible book value to $29.52, but more importantly, I couldn’t be more proud of how we’ve responded as an organization to the COVID-19 pandemic. Nothing is more critical than the health and safety of the Flagstar family. We were able to pivot quickly and ensure the vast majority of our employees were working from home by Monday, March 16. This didn’t just happen. We conducted a 2,500-employee remote work test on Friday, March 13 and expanded our network capacity in the days leading up to stay-at-home orders being put in place, which increased our remote connection capacity by 10,000 users. For those employees still needing to work from Flagstar facilities, we implemented distancing standards, an enhanced cleaning regimen and provided necessary protective equipment to ensure they felt as safe as possible while on site. It’s testament to the morale and spirit of the Flagstar team that during this transition, we didn't miss a beat and continued to serve our customers and partners with the same exceptional standards they’ve come to expect from us. As government relief programs have been rolled out, we’ve moved quickly and efficiently to ensure our customers can participate and benefit. We took over 3,000 applications for the Paycheck Protection Program and have worked with thousands of borrowers requesting forbearance relief on their mortgages. We’ve increased our call center capacity and have not seen material wait or hold times throughout this period. Furthermore, there have been no layoffs or furloughs at Flagstar. A lot has been thrown at us over the last seven weeks, but the team has stood tall. I believe the journey we’ve been on over the last few years, including multiple acquisitions, has helped prepare us for this moment. Now more than ever, I believe our business model will shine. We previously talked about how our different business lines act as a natural hedge. In this low interest rate environment, mortgage originations, because of increased refinance volume, have performed exceptionally well, as has our warehouse lending. These two businesses have been well supported by our sub-servicing operation. Fifty-two percent of our revenues in the first quarter were from non-interest or fee income businesses, and I believe this deliberate and well-balanced mix will ensure continued strong performance as we move forward. There were several other notable developments during the quarter, which included: average interest-earning assets increased $442 million or 2%, primarily a result of an increase in investment securities, while our net interest margin decreased 10 basis points to 2.81% as we effectively managed net interest margin compression through timely and thoughtful actions on the liability side of the balance sheet. We adopted CECL, increasing our credit reserves to $152 million at the end of the quarter, $18 million of which was the result of COVID-19 and its potential impact on our loan portfolio in the future. We maintain a diversified lending portfolio with quality credits and no significant exposure in any one industry. Mortgage banking revenues decreased only $2 million or 2% to $96 million in the first quarter versus $98 million in the fourth quarter as we continue to take advantage of the strong refinance market. Our sub-servicing business remained relatively flat in terms of loans serviced or sub-serviced. At the end of March, we were servicing or sub-servicing approximately 1.1 million loans, which generates consistent non-interest fee income for the bank. For the fifth consecutive year, we were awarded the Fannie Mae STAR Performer Award in general servicing, which is testament to the exceptional work and commitment of our servicing, call center and collections teams. Finally, our capital position remained solid, and we maintained strong liquidity, particularly given our broad deposit base and our access to Federal Home Loan Bank funding, which means we’re well positioned to not only weather this pandemic, but also support our customers and business partners who are not so fortunate. It’s undoubtedly been a challenging time, but we’ve rallied exceptionally well as an organization as a result of our fantastic employees, team-oriented approach and robust business model. I will now outline some of the key operating metrics from each of our major business segments during the first quarter. Please turn to Slide 16. Quarterly operating highlights for the Community Banking segment include average commercial and industrial and commercial real estate loans increased $127 million or 3%, with the growth being driven by the CRE lending group. We’ve been actively managing our commercial loan portfolio since the advent of COVID-19 and working closely with our customers given our relationship-based approach. Line of credit usage increased from 49% to 60% or $249 million between March 13 and April 17. We continue to be thoughtful in terms of new facilities and believe our strong credit policies and diversified portfolio will be a strength as the fallout from this pandemic becomes more apparent. Average consumer loans held for investment decreased $35 million or 1% as we ended the quarter with approximately $4.9 billion of consumer loans on our balance sheet, with 82% being residential first lien mortgages and HELOCs. Through April 24, we have received approximately 2,000 requests for deferrals on our non-mortgage consumer loan portfolio, which amounts to $119 million or 5.8% of all outstanding balances. Utilization levels on HELOCs have remained fairly flat throughout the pandemic period. I will talk about forbearance activity on first lien mortgages shortly. Average warehouse lending loans decreased $437 million or 16% to $2.3 billion in the quarter due to fewer days online resulting from the seasonal slowdown in mortgage activity, particularly in January and February. However, we made changes to our warehouse credit box at the end of the quarter as a result of the volatility in the mortgage market. We honored all loans that were on the line at the time of announcing we were going to stop financing non-QM and jumbo loans. Furthermore, we tightened our FICO limits around several other products in response to the market uncertainty and to protect our own position. I would add that during the month of April, we have been encouraged to see continued high outstanding average balances in warehouse lending. Given its positive correlation to the mortgage business, it acts as a nice hedge in a low interest rate environment when refinance activity is thriving. Overall, this means average loans held for investment decreased $345 million, which drove the $4 million or 3% decrease in net interest income quarter-over-quarter. In order to help our customers, we participated in administering the SBA’s Paycheck Protection Program just after the quarter ended and received approximately 3,000 applications, totaling approximately $400 million. This program was introduced very quickly, and the Flagstar team did an outstanding job of standing this up and ensuring our customers were able to participate and benefit. One hundred and fifty of our 160 bank branches remain open. The ten that are closed do not have a drive-up and are located in close proximity to another branch, so we chose to close them to further protect our employees. All of our ATMs remain operational. Average deposits, which include all interest-bearing and non-interest-bearing retail and custodial accounts, decreased approximately $100 million, but we did reduce the cost of interest-bearing deposits 13 basis points during the quarter as we moved quickly to compensate for the lower interest rate environment. We will continue to maintain our disciplined and relationship-based approach within community banking. We feel we have a loan portfolio with strong credit qualities, and we’ll continue to work closely with our customers throughout this pandemic. Please turn to Slide 17. Quarterly operating highlights for the mortgage origination business include fallout-adjusted lock volume increasing 36% to $11.2 billion quarter-over-quarter, while the net gain-on-loan-sale margin decreased 43 basis points to 80 basis points. As a result, gain-on-sale decreased $11 million to $90 million in the quarter. The increase in fallout-adjusted lock volume was driven by the robust refinance market due to low interest rates, particularly during the month of March. Refinance activity accounted for 64% of our lock volume during the quarter. Mortgage closings were $8.6 billion in the first quarter, an 8% decrease from the fourth quarter due to the anticipated seasonal slowdown in purchases. The big story happened toward the end of the quarter, where we incurred a substantial loss on our pipeline hedge for the reasons Jim has previously described. This reduced our margin for the quarter materially, but it was a short-term phenomenon and the hedge volatility settled right after quarter end and has been stable during the month of April. Similar to what we did on the warehouse side, right around quarter end we moved to stop originating higher-risk products and tightened the credit box in certain areas to protect our position and minimize any future write-downs or losses. Despite moving to a work-from-home environment, our mortgage operations team continues to excel and keep up with the additional volume. We have not seen any degradation in productivity levels, and I believe this is due in large part to our underwriting staff being 98% remote prior to COVID-19. This means the shift in work environment wasn’t unfamiliar. Furthermore, we have not been affected by offshore shutdowns as almost all of our operations are here in the U.S. We continue to run a compliant and efficient operation of scale. We’re pleased with how we reacted quickly at the end of the quarter to protect our mortgage earnings, both now and in the future, and are very encouraged by what we’ve seen during the month of April. We have already booked over $90 million of gain-on-sale revenue through April 23. Mortgage has always been a key component of our business model and strategy. It generates significant non-interest fee income for the bank and is a natural hedge to some of our other businesses in a declining rate environment. Moving to servicing. Quarterly operating highlights for the mortgage servicing segment on Slide 18 include that we ended the quarter servicing or sub-servicing approximately 1.1 million loans, of which over 917,000 or 85% are sub-serviced for other MSR owners. Of the 1.1 million loans we service or sub-service, 94% are backed by Fannie Mae, Freddie Mac or Ginnie Mae. The number of loans serviced to sub-serviced remained relatively flat in the quarter despite the high levels of refinance activity as we are able to replace runoff with new loans from our mortgage origination business. Today, we have the capacity to service or sub-service 2 million loans, as well as provide ancillary offerings such as recapture services and financing solutions to MSR owners. Those offerings are proving to be very beneficial in this current environment. If you look at Slide 39, you will see that we are achieving our $4 million to $6 million of operating profit before tax guidance for every 100,000 loans we add to the platform. The major issues for servicing right now are forbearance activity and liquidity. Through Thursday, April 23, 110,325 borrowers representing 10.7% of the portfolio that we either service or sub-service have requested forbearance relief because of COVID-19. Interestingly, 50% of those borrowers have made their April payment and not taken advantage of the forbearance option. This effectively means that right now, 5.3% of the loan book we service or sub-service is actually in forbearance. As part of the forbearance period, we’re also waiving certain fees, and there will be no negative reporting to the credit bureaus. As you may recall, we acquired a default servicing operation in Jacksonville in September 2019 in order to leverage our industry-leading oversight and monitoring and bring default servicing back in-house. Operationally, we couldn’t have been better prepared to handle the spike in core volume because of this pandemic. We have been monitoring call wait times closely and have been quick to react to the spikes to ensure we operate within our one-minute average speed-to-answer service level requirement. Furthermore, we’ve set up a hardship relief task force within our default servicing team to proactively reach out to borrowers in forbearance and work out an appropriate loss-mitigation solution after they exit forbearance. This will streamline the operation and allow us to get ahead of things rather than just letting these loans run through the normal process. During the quarter, we sold $6.6 billion of MSRs, $2.2 billion via bulk and $4.4 billion via flow sales and retained the sub-servicing on 85% of these sales. However, right now, there is no market for MSRs, given the recent market volatility and uncertainty surrounding liquidity on the advances. We do believe the market for MSRs will come back in the near future. In the meantime, we’re very comfortable retaining the MSRs we create through our origination business on our balance sheet. Our MSR to CET1 ratio is currently 13.6%, so we have plenty of room before we start to approach the 25% MSR to CET1 capital level and intend to use the runway we created through the end of the year, if necessary. Of the 1.1 million loans we service or sub-service, only 9% are MSRs Flagstar owns. It’s a small percentage, and the liquidity need on the advances is immaterial, given our overall strong liquidity position as a bank. Finally, custodial deposits averaged $4.8 billion in the first quarter, which was flat compared to the prior quarter. Again, this is just one additional benefit we get from our sub-servicing business as it provides liquidity that helps fund our balance sheet. Now more than ever, the strength of our sub-servicing business within a bank that is well capitalized and has plenty of liquidity stands alone in the industry, and I expect you will continue to see it thrive in the future. Moving on to expenses on Slide 20. Our total non-interest expenses decreased 4% or $10 million to $235 million quarter-over-quarter, and total revenues also decreased by $10 million to $305 million, generating positive operating leverage of 1.4% in the quarter. The main driver of the decrease in expenses was lower mortgage closings. If you look at Slide 21, our core non-mortgage non-interest expenses were $139 million in Q1. Within this amount is a one-time write-off of $2 million, which relates to a 2013 legacy sale, meaning the run rate is $137 million, a slight increase of $1 million from last quarter. Expenses tied directly to the mortgage origination business were $96 million, a decrease of $8 million versus Q4, given mortgage closings, which drive mortgage expenses, were down 8%. As a percentage of mortgage closings, mortgage expenses have been very consistent for the last three quarters, as you can see on Slide 21. Our efficiency ratio was 77% for the first quarter, which was an improvement of 1% from the prior quarter for the reasons I’ve just outlined. Given the uncertainty around the mortgage market and COVID-19-related expenses, we will not be providing Q2 non-interest expense guidance at this time. Ensuring the health and safety of the Flagstar family—our employees, customers, partners, communities and stakeholders—is our number one priority. The way the team has adapted to this new environment without breaking stride has been exceptional. We will continue to work relentlessly to help our customers through this difficult situation, and we will come out the other side stronger for it. I believe the strength of our business model and the logic of our business strategy will prevail in this uncertain environment. This concludes our prepared remarks, and we will now open the call to questions from our listeners.

Operator, Operator

Thank you. We will first hear from Scott Seifers with Piper Sandler. Please go ahead.

Scott Seifers, Analyst, Piper Sandler

Good morning. How are you? Thank you for taking my question. I guess the first question I wanted to ask is just on that increase in the trading securities of about $2.1 billion, and I guess the presumably related $2 billion increase in short-term and long-term FHLB advances. Maybe a little more color on what the strategy is there. And then as a follow-up, between the $2 billion in trading securities and then, of course, the $1 billion-plus jump in the warehouse, they really kind of dinged the tangible common equity ratio. So just curious what your thinking is on the balance between strong regulatory ratios but a comparatively thin tangible common equity ratio?

Alessandro DiNello, President and Chief Executive Officer

Let me address the $2 billion and the $1 billion, and then I’ll let Jim chime in on the capital piece of it. The $2 billion were securities that were kind of trapped in that period of time where there was Fed action going on, and we weren’t able to get them off the balance sheet at the end of the month. We did deliver them into the hedges; they were delivered into the TBAs in mid-April, so they’re off the balance sheet now. So that’s what that was. As for the warehouse, we’ve seen the ability to increase our warehouse commitments grow as other warehouse providers haven’t been in a position to take on more. We’ve kept a fairly low concentration limit on our warehouse business historically. Given the opportunity in the market, we narrowed the collateral that we accepted on our warehouse lines and reduced advance levels on certain collateral. So even though we’re at a higher level in our warehouse balances, the risk associated with that, we feel very comfortable with. If you look at the loss history in that business, it’s virtually nothing. Even during the financial crisis it was small. So we’re comfortable taking the level of warehouse exposure up given the spreads in that business and the low credit risk. It does impact assets a little bit, but I think the trade-off is helpful to us. I’ll let Jim opine on the capital piece of that.

James Ciroli, Chief Financial Officer

Yes. Thanks, Sandro. Just going back quickly, Scott. At the end of the quarter, there was a lot of selling of mortgage-backed securities going on, especially because there were margin calls and those entities needed liquidity. That supply imbalance helped dealers get more aggressive with their bids. When we saw the bids for our Fannie, Freddie and Ginnie securities — and that’s what comprises that trading securities line: newly created MBS that we weren’t able to sell off our balance sheet at prices that we expected to get — we had the option to deliver those securities into our TBAs. Rather than sacrifice execution at quarter end just to get those off the balance sheet, we let our capital levels go down a bit and our balance sheet balloon a bit. We had the liquidity and capital to do that and delivered those into the TBAs mid-month. Regarding capital, when you look at Flagstar’s balance sheet between warehouse loans, which Sandro elaborated on, and loans held for sale, those portfolios are low-risk. Also, 97% of our securities portfolio are Ginnie securities, so there’s no credit risk implicitly in that portfolio. When you look at the composition, more than half the balance sheet was in categories that have essentially no or very little credit risk. The 6.25% tangible common equity ratio at quarter end is a little low on an end-of-quarter spike basis because warehouse balances and trading securities spike at quarter end. If you look at the ratio on an average basis, it was roughly 7.25%.

Scott Seifers, Analyst, Piper Sandler

Okay. Perfect. That’s good color. Particularly on those $2 billion of securities because I had estimated that was just a loan at 60 basis points or so. And then separately, I was hoping to ask on the gain-on-sale revenue. I appreciate the color on the $90 million so far, and it certainly sounds like that transitory hedge stuff will not impact. So I can back into what sort of a core gain-on-sale margin would be. But given all the fluidity in the mortgage market right now, any sense for what ongoing gain-on-sale margins might look like and the big factors impacting them in your minds?

Alessandro DiNello, President and Chief Executive Officer

I’m not going to forecast margins, Scott. I don’t have a crystal ball. If industry expectations for stronger mortgage originations hold true, the warehouse and mortgage businesses will perform well, and we could have continued meaningful gain-on-sale revenue. I will say April is even stronger than what I cited earlier — the gain-on-sale through the partial month was over $95 million — but we won’t speculate on how margins evolve. Your guess is as good as mine.

Scott Seifers, Analyst, Piper Sandler

Fair enough. Thanks, guys.

Operator, Operator

Next, we’ll move to Bose George with KBW. Please go ahead.

Bose George, Analyst, KBW

Good morning. I want to ask about the loan sub-servicing business and how that gets impacted by the increase in delinquent loans. You said the 10.7% loan forbearance requires a more hands-on approach. How do your contracts work in terms of compensation for that? Does it drive additional fees to you?

Alessandro DiNello, President and Chief Executive Officer

Lee will address the details, but I’ll start by noting it’s different when you’re a sub-servicer versus a primary servicer. Lee?

Lee Smith, Chief Operating Officer

Let me pick up on Lee’s earlier point and provide more detail. First, of all the loans we service or sub-service, only 9% are owned by Flagstar, so our owned MSR exposure is small. From a liquidity point of view, that’s immaterial given our position as a bank with ample liquidity. Regarding sub-servicing contracts, there are a couple of things. As loans become more delinquent, the fees we receive do increase. Given the higher requests for forbearances and related activity, we have increased capacity on the collections side and in loss mitigation. So while there will be an increase in revenue because of higher servicing activity, there is also an increase in our cost base given higher activity levels. We are proactively reaching out to borrowers in forbearance and working to streamline loss mitigation after forbearance ends, which should help control long-term costs and outcomes.

Alessandro DiNello, President and Chief Executive Officer

If you look at our servicing portfolio, the sub-servicing number is large relative to owned MSRs, and that’s the different dynamic. Our owned servicing is comparable to a typical bank, but sub-servicing is where we have scale and a different set of economics and dynamics.

Bose George, Analyst, KBW

Thank you. Can you give a little more detail on your credit positioning within the CRE exposures, particularly hotels, retail and senior housing? What do you think would be reasonable provision levels for those exposures in a stress scenario?

James Ciroli, Chief Financial Officer

I can’t forecast a specific Q2 provision today. What I will say is the detail is in the slides; if you look at Slides 31 through 33, you’ll see the commercial lending breakdown and CRE detail. The exposures to the most impacted industries are relatively limited. The biggest exposure on the commercial side is homebuilders at about $900 million, and we haven’t had any deferral requests from our homebuilder clients thus far. This portfolio is small relative to total assets and is diversified. When you dig into the detail, you’ll see loan-to-value ratios are generally low and debt service coverage ratios are solid. We’ve always been conservative with our allowance and will remain so. We feel good about the allowance based on the economic scenarios at the end of March, and we’ll evaluate and act appropriately at June 30, but I don’t have a crystal ball to tell you the specific Q2 provision amount.

Operator, Operator

Next, we’ll move to Daniel Tamayo with Raymond James. Please go ahead.

Daniel Tamayo, Analyst, Raymond James

Hi, guys. How are you? Just on NIM and how the PPP program impacts that: I think you mentioned that NIM should be relatively flat in the second quarter, including the PPP benefit. How much benefit is assumed from PPP in that flattening?

Alessandro DiNello, President and Chief Executive Officer

I’ll let Jim answer, but I don’t think there’s a meaningful net interest margin benefit from PPP; if anything, it’s slightly dilutive and not material.

James Ciroli, Chief Financial Officer

That’s correct, Danny. PPP is dilutive to NIM, but the impact is not expected to be material.

Daniel Tamayo, Analyst, Raymond James

Okay. Thanks. On the floors in the warehouse business, did you benefit from those at all in the first quarter? How many are in the money at this point?

James Ciroli, Chief Financial Officer

If you look at Slide 9, a large portion of the warehouse portfolio has LIBOR floors. Roughly 71% of that portfolio have LIBOR floors above zero. So with LIBOR where it is today, many of those floors have been relevant. Approximately 1% of warehouse loans have a LIBOR floor at a level higher than zero that materially affects pricing today, but a broader portion has floors that keep LIBOR effectively above zero relative to the loan economics.

Daniel Tamayo, Analyst, Raymond James

Thanks. And on CRE, do you disclose how much of that is construction exposure?

Alessandro DiNello, President and Chief Executive Officer

If you go to Slide 32 you’ll see the CRE breakdown by category. The commitment levels are also shown. If you remove the homebuilder piece, total commitments would be lower. For precise construction detail, the March call report will break out those amounts as it does in the loan schedules.

Daniel Tamayo, Analyst, Raymond James

Perfect. Thanks for answering my questions.

Operator, Operator

Next, we’ll move to Henry Coffey with Wedbush. Please go ahead.

Henry Coffey, Analyst, Wedbush

Good morning. Given the reliance many banks have on Moody’s COVID forecast for CECL adjustments, is there a risk that that becomes countercyclical and starts affecting your lending decisions? Moody’s forecast is more negative in later iterations; does that affect lending behavior?

Alessandro DiNello, President and Chief Executive Officer

I don’t think Moody’s forecast directly affects our lending behavior. Indirectly, as the economy weakens, that affects our decisions in commercial lending. Right now, we have tightened new commercial approvals and require centralized approvals for new commercial loans. We’ve scaled back significantly. As conditions improve, we’ll make decisions to originate commercial loans where underwriting gives us confidence, but there is not a direct correlation between Moody’s inputs and day-to-day lending decisions.

Henry Coffey, Analyst, Wedbush

On the $90 million-plus gain-on-sale in April: do you have enough insight into the pipeline to get a feeling for May and June? Were there any aberrant outcomes in April that changed that, or is it just a big flow of business and improved margins?

Alessandro DiNello, President and Chief Executive Officer

Kristy can add color, but we did not see the hedge volatility in April that impacted late March. The business continued to be strong in April. The key question is whether margins will remain at April levels. Volumes look good for the near term, but margins are uncertain.

Kristy Fercho, President of Mortgage

One thing I’ll share is that our strategy over the last year has been to optimize returns through product, volume and channel mix and to bring value to our customers. We continued that strategy in April as the market was dislocated and took advantage of opportunities, resulting in the $90 million-plus benefit. There’s no reason to expect that this approach won’t continue in future months. We continue to optimize channel mix and pricing given market capacity and spreads.

Henry Coffey, Analyst, Wedbush

So watching primary-secondary spreads is a good guidepost. By the end of March those spreads reached historic highs and have tightened recently. Is that a fair conclusion?

Kristy Fercho, President of Mortgage

Yes. Primary-secondary spreads were about 233 basis points at the end of March and have tightened approximately 60 basis points over the last two weeks. That’s a good indicator to watch for market dynamics, pricing power, and capacity.

Henry Coffey, Analyst, Wedbush

On the servicing side: you mentioned owned MSRs are a small part of the book and liquidity needs are immaterial. What are your observations about sub-servicing clients — are they seeing increasing advance requirements? Are you working with them to provide financing lines, given FHFA and Ginnie responses?

Alessandro DiNello, President and Chief Executive Officer

Lee will provide detail, but the FHFA announcement limiting advances provided significant relief to sub-servicers and MSR owners. We’ll work with partners where needed.

Lee Smith, Chief Operating Officer

The FHFA announcement gave our sub-servicing clients more certainty by capping the amount of time they may need to make advances. That’s helpful for their liquidity planning. On the Ginnie Mae side, the PTAP program is helpful. We work with partners where we can and do have some financing lines available. If it makes sense and we can help our partners, we will. Our position as a well-capitalized bank with ample liquidity enables us to consider assisting where appropriate. The immediate liquidity uncertainty that existed a few weeks ago has eased with the FHFA announcement.

Henry Coffey, Analyst, Wedbush

Do you have the capacity from technology and people to pick up more market share if some correspondents pull back?

Lee Smith, Chief Operating Officer

We have capacity; we can service or sub-service up to 2 million loans today. We will evaluate opportunities thoughtfully and continue to grow where it makes sense. We won’t pursue aggressive, short-term opportunistic growth that doesn’t fit our strategy.

Alessandro DiNello, President and Chief Executive Officer

We’ll continue to be deliberate in growing servicing, taking advantage of appropriate opportunities as they arise.

Henry Coffey, Analyst, Wedbush

Thanks. This call was extremely helpful.

Operator, Operator

Next, we’ll move to Steve Moss with B. Riley. Please go ahead.

Steve Moss, Analyst, B. Riley

Good morning. Two questions. First, on non-interest-bearing deposit growth for the quarter, can you give some color on drivers?

Alessandro DiNello, President and Chief Executive Officer

Sorry, you cut out earlier, Steve. Could you repeat the question?

Steve Moss, Analyst, B. Riley

Yes — non-interest-bearing deposit growth for the quarter. Any color would be helpful.

Alessandro DiNello, President and Chief Executive Officer

The primary driver was custodial deposits.

Steve Moss, Analyst, B. Riley

Okay. Secondly, regarding the disclosures around leveraged lending and shared national credit portfolios, do you have any specific reserves for those portfolios?

James Ciroli, Chief Financial Officer

There are no specific reserves on either of those portfolios. They are included in the CECL model allocations by portfolio, but no separately identified specific reserves.

Steve Moss, Analyst, B. Riley

Okay. Helpful. One last question: the $35 million in loans that were special mention or substandard at quarter end — has that increased quarter-over-quarter or is that relatively flat? Any dynamics there?

Alessandro DiNello, President and Chief Executive Officer

It’s relatively flat and those had nothing to do with COVID.

Steve Moss, Analyst, B. Riley

Okay. Thank you very much; I appreciate that.

Operator, Operator

That concludes the question-and-answer session. At this time, I would like to turn the call back over to Sandro DiNello for any additional or closing remarks.

Alessandro DiNello, President and Chief Executive Officer

Thank you, Michelle. I’d like to close by talking about the Flagstar spirit. This company has come together in a way that I’ll never forget, not just in meeting daunting deadlines and the crushing workloads they shoulder, but in the way they have brought fun into daily life: a Flagstar COVID Facebook group, virtual happy hours, and a company-wide photo contest showing home workstations. While we’ve moved over 4,000 people to work at home, some folks still support branches or need to be in an office building. As Lee noted, we’ve done everything we can to put them in a safe environment and have rewarded them with over $1 million in bonuses. We changed the work week for our branch employees to four days but are paying them for five, and we continue to pay employees who are not working due to the virus. We also set up an employee assistance program through our foundation to help employees experiencing COVID-19-related financial hardships. In addition to the community support I noted earlier, we opened our Paycheck Protection Program to nonprofits that were not previous Flagstar customers and set up a special small-dollar loan program for people impacted by COVID-19 who live in low- to moderate-income areas. We know that we’re in a difficult fight, and we’ve expressed that sentiment with Flagstar versus COVID-19 T-shirts for every employee. We are tough, and we will prevail. Finally, thank you to the entire Flagstar family; I appreciate all you do more than my words can express. I urge everyone to stay home as much as you can, and I pray that you stay safe and healthy.

Operator, Operator

This will conclude today’s call. We thank you for your participation.