M&T Bank Corporation (MTB) Q4 2020 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by, and welcome to the M&T Bank's Fourth Quarter 2020 Earnings Conference Call. At this time, all participants lines have been placed in a listen-only mode. And later the floor will be open for your question. [Operator Instructions] It is now my pleasure to turn the call over to Don MacLeod, Director of Investor Relations. Please go ahead.

Thank you, Maria, and good morning, everyone. I'd like to thank you for participating in M&T's fourth quarter and full year 2020 earnings conference call, both by telephone and through the webcast. If you have not read the earnings release we issued this morning, you may access it along with the financial tables and schedules from our website, www.mtb.com by clicking on the Investor Relations link and then on the Events and Presentations link. Also, before we start, I'd like to mention that comments made during this call might contain forward-looking statements relating to the banking industry and to M&T Bank Corporation. M&T encourages participants to refer to our SEC filings, including those found on Forms 8-K, 10-K and 10-Q for a complete discussion of forward-looking statements. Now I'd like to introduce our Chief Financial Officer, Darren King.

Thanks, Don, and good morning, everyone, and Happy New Year. Before we get into the details, I'll touch on just a few highlights in the recent quarter's results. PPP loan forgiveness ramped up in the fourth quarter. Average PPP loans declined by $351 million compared with the third quarter and were down $1.1 billion on an end-of-period basis. This resulted in the accelerated recognition of $29 million of PPP loan fees into net interest income during the quarter. In addition to the impact of accelerated PPP loan fees, net interest income increased as a result of improved deposit pricing across all customer segments. Notwithstanding the PPP forgiveness, average loans were up during the quarter, including growth in dealer floor plan loans, and mortgage loans purchased from servicing pools. Fee revenues held up well, particularly trust income due to continued strong capital markets and service charges due to the improved economic activity. Expenses were impacted by costs relating to the migration of our retail brokerage platform to LPL Financial and were otherwise in line with our expectations. As to credit, we saw an increase in nonaccrual loans this quarter that is consistent with the higher expected credit losses that we provided for earlier in the year. While CRE loans came off COVID forbearance and net charge-offs rose to a level just above our long-term average. Capital levels remained strong with our CET1 ratio growing to 10% at year-end. We'll review the numbers for the full year in a moment, but first let's turn to the results of the fourth quarter. Diluted GAAP earnings per common share were $3.52 in the fourth quarter of 2020, compared with $2.75 in the third quarter of 2020, and $3.60 in the fourth quarter of 2019. Net income for the quarter was $471 million, compared with $372 million in the linked quarter and $493 million in the year ago quarter. On a GAAP basis, M&T's fourth quarter results produced an annualized rate of return on average assets of 1.3% and an annualized return on average common equity of 12.07%. This compares with rates of 1.06% and 9.53%, respectively in the previous quarter. Included in GAAP results in the recent quarter were after-tax expenses from the amortization of intangible assets amounting to $2 million or $0.02 per common share, little change from the prior quarter. Consistent with our long-term practice, M&T provides supplemental reporting of its results on a net operating or tangible basis, from which we have only ever excluded the after-tax effect of amortization of intangible assets as well as any gains or expenses associated with mergers and acquisitions when they occur. M&T's net operating income for the fourth quarter, which excludes intangible amortization, was $473 million compared with $375 million in the linked quarter, and $496 million in last year's fourth quarter. Diluted net operating earnings per common share were $3.54 for the recent quarter compared with $2.77 in 2020's third quarter and $3.62 in the fourth quarter of 2019. Net operating income yielded annualized rates of return on average tangible assets and average tangible common shareholders' equity of 1.35% and 17.53% for the recent quarter. The comparable returns were 1.1% and 13.94% in the third quarter of 2020. In accordance with the SEC's guidelines, this morning's press release contains a tabular reconciliation of GAAP and non-GAAP results including tangible assets and equity. Included in the recent quarter's results was a $30 million distribution from Bayview Lending Group. This amounted to $23 million after-tax effect and $0.18 per common share. We expect that this distribution occurred in place of the usual distribution we have received from Bayview Lending Group in the first quarter of the past few years. Turning to the balance sheet and the income statement. Taxable equivalent net interest income was $993 million in the fourth quarter of 2020, marking an increase of $46 million or 5% from the linked quarter. The primary driver of that increase was the accelerated recognition of $29 million of fees on PPP loans following forgiveness of those loans by the small business administration. The net interest margin increased by 5 basis points to 3% compared with 2.95% in the linked quarter. The accelerated recognition of PPP fees added an estimated 9 basis points to the margin. A 5 basis point decline in the cost of interest-bearing deposits, repayment of debt outstanding and slightly higher income from our hedge portfolio, boosted the margin by an estimated 4 basis points. Continued inflows of excess liquidity, including DDA and interest checking resulted in a $4.5 billion increase in cash on deposit with the Federal Reserve. While this had a negligible impact on net interest income, it contributed to about 10 basis points of pressure on the net interest margin. All other factors, including lower premium amortization on acquired mortgage loans and mortgage-backed securities provided an approximate 2 basis point benefit to the margin. Average total loans increased by $456 million or about 0.5% compared to the previous quarter. Looking at loans by category, on an average basis compared with the linked quarter, commercial and industrial loans declined by $620 million or about 2%. Contributing to that decline was a $351 million decline in PPP loans, primarily reflecting loan forgiveness. Partially offsetting that, was a $231 million increase in floor plan loans as dealers seed to rebuild inventories following a very robust sales year. All other C&I loans declined by $500 million, largely from lower line utilization. We'd note that on an end-of-period basis, dealer loans were up $800 million and all other C&I loans were roughly flat, excluding the PPP forgiveness. Commercial real estate loans grew just over 1% compared with the third quarter, primarily as the result of further draws on pre-existing loans. New originations in the CRE space remained subdued. Residential real estate loans increased by $204 million or 1%, reflecting loans purchased from Ginnie Mae servicing pools, pending resolution, partially offset by repayments. Consumer loans were up 3%, reflecting higher indirect auto and recreation finance loans, partially offset by lower home equity lines of credit. Average core customer deposits, which exclude deposits received at M&T's Cayman Islands office as well as CDs over $250,000, grew by $4.5 billion or 4% compared with the third quarter reflecting higher interest and noninterest checking as well as money market deposit accounts. Turning to noninterest income. Noninterest income totaled $551 million in the fourth quarter compared with $521 million in the prior quarter. That increase reflects the $30 million distribution from Bayview Lending Group that I previously mentioned. The recent quarter also included $2 million of valuation gains on equity securities, largely on our remaining holdings of GSE preferred stock, while the third quarter included $3 million of such gains. Mortgage banking revenues were $140 million in the recent quarter compared with $153 million in the linked quarter. Residential mortgage loans originated for sale were $1.2 billion in the quarter, unchanged from the third quarter. Total residential mortgage banking revenues, including origination and servicing activities, were $95 million in the fourth quarter compared with $119 million in the prior quarter. The decrease reflects a lower gain on sale margin and residential servicing revenues declined very slightly. Commercial mortgage banking revenues totaled $45 million, encompassing both originations and servicing. The improvement from the third quarter was mainly a result of higher origination volumes. Trust income was $151 million in the recent quarter, up slightly from $150 million in the previous quarter. Business remains solid with slightly higher money market fund fee waivers, more than offset by higher levels of assets managed and continued good capital markets activity. Service charges on deposit accounts were $96 million, improved from $91 million in the third quarter. The improvement is largely the result of increased economic activity that resulted in growth in payments-related income. Excluding the BLG distribution, the improvement in other revenues from operations compared with the linked quarter also reflected an uptick in credit card-related activity. Turning to expenses. Operating expenses for the fourth quarter, which exclude the amortization of intangible assets, were $842 million compared with $823 million in the third quarter. Salaries and benefits declined by $3 million from the prior quarter. In accordance with the previously announced contract with LPL Financial, M&T took its first steps to transition its retail brokerage and advisory business to the LPL platform. In doing so, M&T incurred $14 million of transition expenses, including severance payments included in salaries and benefits, and a contract termination payment that is included in other costs of operations. Also included in other cost of operations is a $3 million addition to the valuation allowance for our mortgage servicing asset. This follows additions of $10 million to the allowance in each of the first and second quarters of 2020. The efficiency ratio, which excludes intangible amortization from the numerator and securities gains or losses from the denominator, was 54.6% in the recent quarter compared with 56.2% in the third quarter and 53.2% in the fourth quarter of 2019. Next, let's turn to credit. Under CECL, and as a result of the pandemic driven economic slowdown, M&T added $800 million to its allowance for credit losses over the course of 2020, while delinquencies, nonaccrual loans and net charge-offs have until recently remained relatively benign. In the CECL environment, statistical models helped predict expected loss content, which must be reserved for well in advance of default. The old loss reserving process didn't permit establishment of an allowance until loss content became incurred. Thus, loss reserves and charge-offs generally rose as delinquencies and nonaccruals increased. We're beginning to see the expected rise in nonaccrual loans and charge-offs that have already been reserved for under the CECL methodology. Net charge-offs for the recent quarter amounted to $97 million. Annualized net charge-offs as a percentage of total loans were 39 basis points for the fourth quarter compared with 12 basis points in the third quarter. It's interesting to note that the charge-off rate in the fourth quarter approximated our long-term average. During the quarter, we restructured substantially all of our limited exposure to the operators of regional malls. These had been under stress prior to the COVID-19 pandemic, which resulted in further deterioration and pushed them into default. The provision for loan losses in the fourth quarter amounted to $75 million, which was $22 million less than net charge-offs. The allowance for credit losses declined slightly to $1.7 billion or 1.76% of loans. That ratio was 1.79% of loans at the end of September. As has been the case since the beginning of 2020, the allowance at the end of the fourth quarter reflects an updated macroeconomic scenario. The scenario is different and less severe than those used at the end of the first and second quarters and modestly less severe than that used at the end of the third quarter, each of which model the uncertainty of the COVID-19 driven damage to the economy. In addition to losses that may be expected from newly originated loans, the allowance and the related provision for the recent quarter continued to reflect the ongoing impacts of the COVID-19 pandemic on economic activity. The uncertainty over additional economic stimulus and the ultimate collectibility of commercial real estate loans, most notably in the hospitality sector and retail sectors outside of the regional mall exposure. Our macroeconomic forecast uses a number of economic variables, with the largest drivers being the unemployment rate and GDP. Our forecast assumes the national unemployment rate continues to be at elevated levels on average 6.9% through 2021, followed by a gradual return to long-term historical averages by the end of 2022. The forecast assumes that GDP grows at a 4.1% annual rate during 2021, resulting in GDP returning to pre-recession levels by the end of 2021. Our forecast considers government stimulus, but not any further fiscal or monetary actions. Nonaccrual loans as of December 31 rose to $1.9 billion, an increase of $653 million from the end of September. That increase came primarily from the transfer to nonaccrual status of a handful of hotel relationships totaling $530 million. At the end of the quarter, nonaccrual loans as a percentage of loans was 1.92%. It is important to keep in mind that some of the usual credit metrics have been affected by the PPP loans on the balance sheet, which are 0 risk-weighted and carry little or no credit risk. Excluding the impact of PPP loans, the ratio of the allowance for credit losses to loans would be 1.86%. Similarly, the ratio of nonaccrual loans to total loans would be 2.03% and annualized net charge-offs as a percentage of total loans would be 42 basis points. Loans 90 days past due, on which we continue to accrue interest, were $859 million at the end of the recent quarter. Of these loans, $798 million or 93% were government-guaranteed by government-related entities. Government-guaranteed loans under COVID forbearance and which we have purchased from servicing tools are generally not reflected in these figures. As we noted on the October conference call, in the third -- and in the third quarter 10-Q, total loans under COVID-related modifications declined to $9.4 billion as of September 30. Those figures declined further to $5.3 billion at the end of the fourth quarter. Commercial and industrial loans with COVID-related modifications declined from $850 million to $433 million at the end of 2020. Of that figure, less than $100 million of those loans still have a form of payment deferral. COVID forbearances, other than payment deferrals, relate to things such as fee waivers and, in some cases, covenant waivers. Similarly, commercial real estate loans under COVID-related modifications declined from $5.1 billion at the end of the third quarter to $2 billion at December 31. Some $600 million of those loans have received a payment deferral. Mortgage-related loans under COVID-related modifications were $3.3 billion at the end of the third quarter, that figure declined to $2.7 billion as of the end of the fourth quarter. Remaining consumer loan modifications also declined to less than $100 million. Modification or forbearance status has not prevented us from updating loan grades within our commercial portfolio. The pace of downgrades into criticized slowed meaningfully in the fourth quarter, rising about 5% from the end of the prior quarter. Turning to capital. M&T's common equity Tier 1 ratio was an estimated 10% as of December 31 compared to 9.81% at the end of the third quarter. This reflects the impact of earnings in excess of dividends paid and slightly higher risk-weighted assets. Next, I'd like to take a moment to cover some of M&T's highlights of the past year. Overall, we believe the events of 2020 provided an illustration of the operational and financial resilience of M&T's franchise. Our colleagues performed like champions, dare I say like division champions, switching with barely a ripple to the work-from-home environment, helping clients through the extraordinary challenges presented by the lockdowns that all but brought the economy to a standstill, and helping customers navigate the PPP loan process that resulted in $7 billion of originations. The severe economic conditions brought about by the COVID-19 pandemic and the resultant 0 interest rate environment had a material impact on our financial results, including a 6% decline in pre-provision net revenue and a 30% decline in net income, partly the result of CECL accounting. Key financial highlights for the year were as follows: GAAP-based diluted earnings per common share were $9.94 compared with $13.75 in 2019. The net income was $1.35 billion compared with $1.93 billion in the prior year. These results produced returns on average assets and average common equity of 1% and 8.72%, respectively. Net operating income was $1.36 billion compared with $1.94 billion in the prior year. Net operating income for 2020, expressed as a rate of return on average tangible assets and average tangible common shareholders' equity, was 1.04% and 12.79%, respectively. Average diluted common shares declined by 4%, the result of repurchase activity in 2019 as well as the limited repurchases in the first quarter of 2020 prior to the pandemic. The total payout ratio for the year, including common stock dividends, was approximately 73%. Tangible book value per share grew to $80.52 at the end of 2020, up 7% from the end of 2019. And despite the challenges for the year, our CET1 ratio increased to 10% at the end of 2020 from 9.73% at the end of 2019. Now let's turn to the outlook. We're all pleased to see that the economy has continued to improve. The rollout of the vaccine holds the promise of a return to normalcy, but we continue to face pressures. Starting with the balance sheet, there are more moving parts than we would see in a typical year. PPP loans on our balance sheet amounted to $5.4 billion at the end of the year. We expect that substantially all of those loans will be repaid or forgiven in 2021, with the bulk of that occurring in the first half of the year. That process will be beneficial to net interest income and net interest margin in the quarters in which the Small Business Administration actually forgives the loans, similar to what we experienced in the fourth quarter. This week, we began accepting customer applications for PPP round 2. We expect these new loans to offset the decline in the original PPP balances to a certain extent. Another atypical element of the balance sheet as we enter 2021 is the level of cash and securities. It is our practice to maintain cash and securities such that we maintain sufficient liquidity and HQLA to meet our internal liquidity governance. The various stimulus programs enacted in 2020 have resulted in M&T, along with our peers, carrying higher levels of cash than normal. We believe that at year-end, we were holding close to $20 billion more in cash and securities than we would need under normal circumstances. The excess cash has little impact on net interest income, but significantly impacts the net interest margin. Every $1 billion in cash impacts the margin by 1 to 2 -- sorry, 2 to 3 basis points. One of the ways we've chosen to deploy the excess cash is through buyouts of Ginnie Mae mortgages from the pools of loans that we service or subservice. Those buyouts create net interest income in the short-term and fee income in the long-term when those loans return to performing status and may be sold to investors. As previously discussed, the active cash flow hedge position on our floating rate loan portfolio has increased to $17.4 billion in the fourth quarter and remains at that level until late this year. However, the fixed receive rate will decline as older swaps mature and newer forward starting swaps become active. Holding the unusual aspects of the balance sheet to the side, we expect total loans to be relatively flat in 2021. Commercial and industrial loans, excluding PPP, are expected to be flat to up slightly as increased economic activity leads to increased line utilization. CRE loans are expected to be flat to slightly down with a subdued outlook for new originations and slowing draws on pre-pandemic loans. Consumer loans are expected to grow at a mid-single digit pace. All in, we expect low to mid-single-digit year-over-year decline in net interest income, primarily the result of the challenging year-over-year rate environment. Turning to fees. We expect low single-digit year-over-year growth in noninterest revenues, similar to 2020. We believe the strong originations trends in mortgage banking will continue, but with continued pressures on gain on sale margins. That outlook also reflects our ability to resell the loans purchased from servicing pools, which could be delayed depending on State and Federal payment and foreclosure holidays. Trust income should be flattish with the full year impact of Money Fund fee waivers, offset by growth in other categories. This also assumes some growth in assets managed, combined with some stability in market values. We expect service charges to reflect the run rate in the fourth quarter and to improve on a full year-over-year basis. Receipt of the BayView Lending Group distribution this quarter pulled forward $30 million of revenue we previously expected to receive in 2021. Turning to expenses. Based on our expectations for growth in some fee revenue categories, primarily mortgage banking revenues and trust income, we expect expenses tied directly to those businesses will grow as well. Outside of those increases, we expect all other expense categories in the aggregate to be generally flat with the prior year. Overall, we expect expenses to be flat to up less than 1%. The trend in credit provisioning should show improvement in line with the macroeconomic outlook. Charge-offs, given the nature of the portfolio and the sectors most impacted by the pandemic, will be lumpy and will likely take longer to emerge. That said, we anticipate charge-offs will be higher in 2021 and likely higher than our long-term average. Given the uncertainty, we will focus our credit outlook on the near term. And for the first quarter of 2021, we don't currently foresee charge-offs higher than what we saw in the fourth quarter. Lastly, turning to capital. The Board has authorized us to repurchase up to $800 million of our common stock. We will continue to operate within the guidelines currently in effect by the Federal Reserve as well as taking into account the economic environment, our earnings outlook and capital position and any alternate capital deployment opportunities. Of course, as you are aware, our projections are subject to a number of uncertainties and various assumptions regarding national and regional economic growth, changes in interest rates, political events and other macroeconomic factors, which may differ materially from what actually unfolds in the future. Now let's open up the call to questions, before which Maria will briefly review the instructions.

[Operator Instructions] Our first question comes from the line of Ken Zerbe of Morgan Stanley.

All right. Great. I guess maybe starting off, you mentioned that you're certainly sitting on a huge amount of excess cash versus kind of where you normally have been there's obviously a lot of debate around whether it's prudent to invest the excess deposits in low rates or low rate -- low-yielding securities or keep it in cash. How are you guys balancing that debate? And where do you come out on it?

Yes. Ken, and thanks for the question. It's a discussion that we have every other week at our ALCO meetings, and our debate is always how long the cash will hang around, given the way it showed up on the balance sheet, it just nearly exploded in the second half of the year. And that impacts our decision and thought process about what kind of duration to take on. And so in the short term, as we work our way through that thought process, holding it in cash versus investing it in short-term treasuries, there really isn't much of a basis point gain from doing that. And so we're looking at alternative ways where we can get maybe a little bit better spread or yield on that cash without setting up a tremendous amount of duration risk. One of the things we mentioned in the prepared remarks was we've been using that cash for buyouts of Ginnie Mae securities. And we've found that to be an attractive use of cash in the short term, because it offsets an expense. The spread on those is better than what we would get on 1-year treasuries, and it creates the opportunity for some fee income. So we'll look for other opportunities like that. We'll watch and see how the PPP 2 goes and what the net change in loan balances is. And then we'll continue to watch the rate curve and the environment and we'll keep our powder drive. It's something we continue to look at on a regular basis to see where we can put some of that money to work because, obviously, we're not paid to hold cash. So that's always our objective, but we're trying to be prudent with how much duration risk we might be taking on.

All right. Great. Helpful. And then maybe just a follow-up question. Can you just talk a little bit more about the -- I think you mentioned $530 million worth of hotel credits that were transferred to nonaccrual. I'd love to learn more about the credit quality of those.

Yes. Sure. I'm happy to discuss that. When we look at those credits, I guess it's -- I don't need to take off my shoes and socks to count the number of them, which is a good thing. So we know exactly how many there are. We know exactly where they are, and we've had a long-standing relationship with many of these -- all of these clients. When I look at the loan to values of the top, let's say, we looked at the top 10, most are 60% or below. That 60%, obviously, is primarily based on at-origination, but we still haven't seen a material decrease in asset prices in the market with anything that's traded. We've seen the CMBS market come back a little bit as we got to the end of the year, which also should help sustain asset prices. And a couple of the downgrades are full relationships. And so there's a couple of larger ones, but it's not just 1 single property. It's multiple properties. And so there's good collateral behind these and good long-standing relationships. We can see some level of occupancy in the hotels. They tend to be in the larger cities. And so as the larger cities start to see more either business travel or tourism, they'll start to come back. But where we sit right now, we feel comfortable that we have our arms around these, and we have good visibility into them and are able to watch them closely. And so I would just view this as what would be the normal progression when you're in 1 of these economic cycles, right, that you start to see signs of delinquency, some of these were in the forbearance, and now they've gone to nonaccrual. And what's interesting about the new CECL environment is that you kind of take the provision and set up the reserve before you see stuff in nonaccrual. And so these are just kind of catching up to that provisioning. And generally, we called out hotels, because when you look outside of hotels, a lot of the CRE trends are actually pretty solid. Outside of that, there's really not a lot of concern today in multifamily. Retail portfolio has actually done reasonably well. And then we're actually seeing some parts of the portfolio, not necessarily CRE related, but seeing some upgrades like in the dealer book and that the dealers have done -- have just had a fantastic year, and in some cases, had record profits. And so that's really the sector that we're watching, and that's obviously why we did make that move with those loans and classifying them as nonaccrual.

Got it. Did you have to -- did you build any incremental reserve associated with those credits when they mentioned nonaccrual?

No, there wasn't anything material. Those that was in effect accounted for in the provisioning that we had done through the prior three quarters of the year.

Our next question comes from the line of John Pancari of Evercore ISI.

On the -- back to that $530 million, was the -- was that a result of more of a deeper dive into those credits this quarter that resulted them all moving to nonaccrual this quarter? Or was it just how it played out in terms of them coming off of forbearance?

Yes, John, it's really the latter. We've been able to identify in the hotel portfolio on a credit by credit basis, right from the start and being able to pay attention to each one of these relationships. Working with them, understanding what their NOI is and how it's moving and what kind of situation they're in. These would largely be folks that got to the end of that forbearance period. And we thought the appropriate thing to do, the most conservative thing to do, was to start to move them into nonaccrual and not continue down that forbearance path.

Got it. Okay. And then it looks like your 90-day past dues also increased about -- looks like more than 60% in the quarter. Was that also related to hotels? And then also maybe if you can comment on your office exposure and just how that's been holding up?

Yes. I guess I'll start with the office exposure. When we look at what's been happening in office, the trends in rent collection have been pretty solid. We haven't seen a big decrease in what the -- our customers have been able to receive from the tenants. And obviously, a bunch of that is with the leases that are signed, they tend to be longer-term and oftentimes with larger corporations. So it's been pretty steady. When I look at that space and I look at how many modifications there are in there, that are outstanding, it's like 1% of the portfolio. And so overall, the office space is doing very well. Again, I would say the multifamily space is also holding up quite well and retail, after our concerns early on, were -- is also doing well. When you look at the over 90-day to answer that question and what's going on. The bulk of that is driven by the residential mortgage loans and the things that we're buying out of the pools and they're largely government guaranteed. So it's kind of the way they're classified, but not something that we worry about from a credit perspective.

Our next question comes from the line of Bill Carcache of Wolfe Research.

Darren, following up on comments that you've made on credit, specifically the hotel credits that you moved to nonaccrual and are no longer applying forbearance. Can you just give a bigger picture view of what the trajectory is across the loan portfolio of downgrades? And then along those lines, is it reasonable to expect that we could see your reserve rate revert to the day 1 level of about 1.3% on the other side of the pandemic? And maybe you could just discuss how soon we'd get there in light of some of the longer tail concerns that you guys have cited in CRE in particular?

Yes, sure. So I guess, just watching the trends and what we've been seeing over the course of the last, I would describe it as 6 months. When we were in June in the thick of things, forbearance was quite widespread. There was -- it was across a number of industries and across quite a number of customers. And what's happened over the course of the last 6 months is is that we've seen stability, and we've seen improvement. I mean probably the most remarkable turnout was in the dealer book that was -- if my memory is correct, about $4.2 billion of forbearance. And all of those are off of forbearance. And in fact, all of those are current. They've not just -- they're not just off forbearance, but they've recovered what they had skipped. And when we look through the rest of the portfolio and what was in forbearance, I think I mentioned that -- first talking about criticized trends, so we've been -- even though forbearance has been on, we go through and we grade the book following our grading system and looking at our credit review process, looking at cash flows, looking at collateral, looking at ability to repay, intent to repay and grading the book. And so, what you're seeing is a slowdown in the migration to criticized, right; so we talked about that being up above 5% this quarter. And so the rate of increase in criticized is declining. And the nonaccruals is really just that progression of people going from criticized into nonaccrual. And what's really in the book is hotel. There's a little bit of retail, but retail has performed quite well. And a little bit of multifamily. And those are really the big three industries, but hotel is far and away the largest one. And what's interesting is when we look within the hotel portfolio, it's larger city hotels that are struggling. We have some hotels throughout our footprint that tend to be in less densely populated areas and perhaps are more like retreats or spas, and through this pandemic, they're actually seeing occupancy rates north of 70%. And so, there's a real range and skewness to occupancy rates within the hotel portfolio. And so the part that, I guess, we feel good about is we've been able to help a lot of customers stay in business and stay paying and work them through the process. And we've got a segment of the portfolio that still has some struggles, but we've got very clear visibility into who those are and what the issues are and are in a position to be able to work with them as much as possible to help protect the value of the assets and to try and keep them in business. The question about the portfolio and the long-term average, I think it's fair to say that going back to the pre -- the post-CECL allowance or reserve rate, subject to a similar mix of business is a fair assumption. It's important to keep in mind that some parts of the portfolio carry different loss rates under CECL than others. And so, all else equal, that's a good place to be or at least a good starting point as you think forward. Do we get there in 2021? Probably, it's a little hard to see that at this point. But right now, we think that, that's possible by the end of '22. I would think that's possible by the end of 2022.

That's super helpful color. If I could squeeze in another one. And just broadly, if you could discuss your thoughts around back book repricing dynamics for you guys? And really across the industry in the last cycle, loan yields continued to decline throughout the trip cycle until we got our first-rate hike in late 2015. And I was wondering if you could just discuss whether you expect to see a similar dynamic in the cycle?

Yes. I guess a couple of things on the back book. On the deposit side, we've seen a tremendous amount of repricing and the reactivity in the deposit book for us and for the industry, especially given the excess liquidity has been very rapid. And when we look at deposit pricing and yields, certainly for a lot of the interest-bearing categories, excluding time deposits, we're getting close to the lows that we saw over the last 10 years. And so, there's some room to go there, but it's single-digit -- low single-digit basis points. When you talk about loans and loan yields, we think about it as loan margin. And the yield is always of -- in reference to the benchmark rate because there's still variable rate driven product. But when you look at what the spread is on the new originations versus what's rolling off, we're actually seeing better spreads on new originations, and we've been seeing that for the last two quarters. And it's in the range of, call it, 40 to 60 basis points over where we had been pre-pandemic. And so what's nice about that is, obviously, over time, as the hedge benefit rolls off, we're starting to see -- we'll see a larger percentage of the book at the new pricing, which is a little bit better than the roll-off pricing. And so we actually were quite positive on that part of what's happening in the portfolio, especially in the consumer -- sorry, commercial real estate and C&I space.

Great. I mean how does that better spreads on new originations compared to the last trip cycle? Just to follow-up on that.

It's -- I guess, I would say, it's reasonably consistent with what happens in the cycle. And usually, what you see is you see leading into the cycles, you see margins drop because usually, there's lots of liquidity and lots of capital, lots of people looking for growth. And then you will see margins compress when you're in that environment. And then as you see a little challenge in the economy, then you tend to see -- and people protecting capital. You tend to see a little bit of an improvement in pricing because there's a little more pricing power. Obviously, with the liquidity, this time, probably not quite as severe as last time. And so there might not be the same level of increase in pricing, but we're certainly seeing it in the short term. And I would say that, that general trend in an economic environment like this and as you come out of it, is actually pretty consistent with what you see.

Our next question comes from the line of Steven Alexopoulos of JPMorgan.

So, the follow-up, not to beat a dead horse on the hotel nonaccruals. But I know you said there was no provision taken as you moved these into nonaccrual. But were there any charge-offs taken?

There were no charge-offs taken on that part of the portfolio. If you look at the charge-offs for the fourth quarter, there's really three large relationships that really drove that increase. Two of them were -- what we would describe as in closed malls and regional mall operators. And by taking those charge-offs, that pretty much eliminates our outstanding exposure to any closed malls. And then there was one company that was in the -- that's described as delivery service, highly related to the travel industry and with no travel going on, that necessitated the charge-off there. Outside of those, it was a variety of things, but generally relatively small compared to those three that I mentioned.

Okay. Darren, in terms of what you do now with these hotel loans on nonaccrual, I know you said they're long-term relationships. Are you planning to offer deferrals until maybe better days ahead? Do you plan to modify the principle, to move them back to current? Or do you plan on going into the hotel business and taking these collateral over?

Well, not the latter. The latter is always our last resort. We're bankers, not hotel operators. And so we'd rather let the experts do that. But it's generally the first two things that you talked about. So we'll work with the borrower and see whether we think that -- first question is, do they have any outside liquidity and can they bring something to the table to be in addition to the interest reserve. It could be some combination of that plus some extended payment relief. You could see some restructuring into something that looks more like an A note, B note kind of structure where you split to credit and might have a partial charge-off on those, but not a complete. And so there's a bunch of different options of ways that we can work with the clients to try and keep them in business and keep them operating as long as possible. Because, obviously, us being in that business is absolutely the last resort.

Okay. And if I could squeeze one more in. On dealer floor plan, I think you said it was up $800 million in the quarter. What was the balance at year-end? And can you talk about expectations for that business, right? It seems dealers have gone a lot more efficient with managing inventory levels during the pandemic. So can you talk about what you expect from the business?

Yes. I guess when you look at the number of cars on lots, we bottomed out. And I want to say, within the summer to early fall. And inventories have been building since then. There's a bunch of factors that go into the inventory that are sitting on the lots. I mean, not the least of which is what the car rental companies are doing. With the challenges in travel, there's been less demand for cars from that sector in the economy. In the early part of this year, as the manufacturers shut down, there was tremendous demand from the dealers to put used cars on their lots. And so they were buying up some of the inventory that was coming off of the rental agencies. And so what we expect is that we'll see an uptick in inventories as we go through 2021. We don't believe we'll go all the way back to what we saw pre-pandemic or in 2019, that the SAAR won't go all the way back into that $16.5 million, $17 million range, but we will see some pickup. And just, I guess, to give a little bit of sense of magnitude looking at balances. From where we are at the end of the year to where we were at the end of the year last year, there's roughly, call it, $800 million to $1 billion difference in what's outstanding at that point in time. So how quickly we get back there? I don't know that we'll get all the way back there in 2021, but we should be approaching that as we get to the end of the year, assuming the economy continues to operate the way it was. But we think there's still some room for that segment to show some growth.

Our next question comes from the line of Matt O'Connor of Deutsche Bank.

Did I miss any comments on the outlook for the tax rate in '21?

You did not miss that. Our expectation for the tax rate for next year is 24%, kind of plus or minus 0.5 point.

Okay. And are there opportunities to lower that? Like we're seeing some other banks heavy ESG kind of partnerships. I think some might be actually buying low-income housing credits, but some also seem to be doing some other things, structures, partnerships with customers or clients. And is that kind of another way to maybe deploy capital if there's not a lot of loans, can't really want grow securities and buying back stock after it's doubled is maybe less appealing, just theoretically. So are there opportunities there? Or are you trying to think differently of what you can do to your capital, given everything I just said?

Yes. No Litec is a part of the sector that we've always been in. We've actually kind of increased it in the last 18 to 24 months. It's part of being a community bank that being in the community, supporting those kinds of projects is critical. We found that over time to be effective in that space, you need to be not just on the loan, but in the equity side of those deals as well. And so we've been doing a little bit more of that. And so it's absolutely part of how we do it. It's also an important part of your CRE rating. And so for all of those reasons, that's absolutely a space that we have been in and will continue to be and as opportunities present themselves, we'll certainly be there for the communities and for the clients.

And then anything new specifically on some of these like ESG initiatives that other banks seem to be kind of leaning into pretty heavily that reduced tax rate?

Yes. We've done some of that. We're in the space. We see opportunity for there to be a little bit more. We haven't discussed it, because it's not been a huge part of our portfolio. And I guess one of the questions as we go forward is, even with the tax rate where it is, how much tax using capacity will there be with us making less money than we did a year ago. And -- but maybe there'll be more capacity to use taxes depending on if there's any changes with the new administration. So it's a space that we're familiar with, and we do, do some business in, but it's selective at the moment.

Our next question comes from the line of Ken Usdin of Jefferies.

Darren, just good to see that buyback announcement this morning. I was just wondering if you can just walk us through the December stress test results and the implied SCB that was brought forth in that document. Does it mean anything in terms of how you have to think about capital? And does it lead you to think about participating in this year's stress test process as a result?

Yes. So the CCAR stress test results in December, obviously, reflected a much more severe economic environment than anything we've seen before. And is meaningfully more severe than what the current environment looks like. And so the good news is we're not operating in that environment. That said, we learned a lot from that output, and we continue to work and talk with the Federal Reserve to understand a little bit more what is behind some of the outputs there. So we're using it to inform our thinking. We haven't -- as we go through that process and learn more from the Fed, that will help inform our decision about whether or not to participate in this year's CCAR. I guess if you look at the implications of that and where we are at the end of the year, the nice thing is when you look at where our CET1 ratio ended the year, we ended the year at 10%. And so there's a comfortable amount of space between where we sit and what might be implied by the outcome of that test. And with the restrictions that have been in place with capital sitting at 10%, we would feel, given also the reserving that we've done, that we're well protected and that we certainly don't need to see the CET1 ratio drop dramatically to 9% in the environment that we're in today with the restrictions. That wouldn't be possible. But we equally don't see a need or a concern that we would want to run the capital ratios up materially from here. And so as we think about those tests, we think about the feedback that came with them and we think about as we go forward, we'll be taking that -- those things into account as we determine when and if and how much shares -- stock to repurchase.

And ladies and gentlemen, we have time for one final question. Our final question will come from the line of Erika Najarian of Bank of America.

Just a follow-up to Ken's question. Is that -- should we think about your DFAS 2.0 results with the 5% SCB is binding relative to how you think about buybacks? I guess I was under impression that most banks were operating for the assumption that DFAS 1.0 results were sort of the binding results. And DFAS 2.0 wasn't binding, but I'm wondering, especially rolls have your $800 million buyback announcing your thoughts there?

Yes. I guess, I wouldn't consider that binding per se. It's obviously an input and an important one in our thought process, because the Federal Reserve told us something with that. And so we're paying attention to it. I think the -- by the letter of the law, the Feds indicated that they would not share with the institutions, if that was to become a new SCB until the end of March, if not sooner. And so we'll learn a little bit more about that over the course of the coming days. And so until that point, it's our understanding that the SCB that was calculated in the June results is the binding one. That said, when you look at the earnings and some of the restrictions on distributions relative to earnings in the forecast, moving the CET1 ratio down meaningfully would be pretty difficult. And so the announced buyback kind of takes into account our current capital position, our forecasted earnings our forecasted balance sheet growth and contemplate some of the restrictions that have been in place. And that was really what got us to that amount at this point. And as we mentioned before, we think that we're still not through the challenges of the pandemic. And so we wouldn't see -- it doesn't seem prudent to us to lower those ratios dramatically. But on the flip side, we don't think that we need to be higher, much higher than where we are. And so we'll kind of manage to that in the short term. And as we go through the year and see how the recovery unfolds, we'll continue to update our thinking and share that with you.

And if I could just squeeze in 1 more question on the NII guide. Darren, in the NII guide for down low to mid-single digits, what are you assuming, if any, in terms of cash deployment relative to $22.6 billion on average in the fourth quarter, and if you could remind us what the swap income realized in 2020 versus what's embedded in your guide for '21?

Yes. So within that projection, there is some expectation that the cash balances come down a little bit. And generally, it's the volatility in the cash balances as opposed to them going into higher-yielding securities or loan growth. And so I guess, embedded in that guide is still a relatively elevated cash position. So to the extent that we find ways to deploy that into higher-yielding assets, which we're always on the lookout for. There's some upside to that. When you look at the income from the hedge portfolio, what we earned this quarter was very similar to what we earned last quarter in terms of NII and what we should see in Q1 is also in line with that, and then it starts to decline, as we've talked about on prior calls over the course of 2021. There's still some benefit in 2020 -- 2021 from the hedges, but it's declining as we go through the year.

Okay. And could you care to quantify in dollars?

I've got it in front of me by quarter, so I'm just looking at it. It's -- it was around $300 million in 2020 and droid to about $275 million in 2021.

And that was our final question. I'd like to turn the floor back over to management for any additional or closing remarks.

Again, thank you all for participating today. And as always, a clarification of any of the items on the call or news release is necessary, please contact our Investor Relations department at (716) 842-5138. Thank you, and goodbye.

Thank you. Ladies and gentlemen, this does conclude today's conference call. You may now disconnect.