Logo of jester cap with thought bubble.

Image source: The Motley Fool.

M & T Bank (MTB 0.45%)
Q2 2020 Earnings Call
Jul 23, 2020, 11:00 a.m. ET

Contents:

  • Prepared Remarks
  • Questions and Answers
  • Call Participants

Prepared Remarks:


Operator

Ladies and gentlemen, thank you for standing by, and welcome to the M&T Bank second-quarter 2020 earnings conference call. [Operator instructions] I will now turn the call over to Don MacLeod, director of investor relations.

Don MacLeod -- Director of Investor Relations

Thank you, Laurie, and good morning. I'd like to thank everyone for participating in M&T's second-quarter 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 them along with the financial tables and schedules from our website, www.mtb.com and 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 on Forms 8-K, 10-K and 10-Q for a complete discussion of forward-looking statements and risk factors. Now I'd like to introduce our chief financial officer, Darren King.

10 stocks we like better than M & T Bank
When investing geniuses David and Tom Gardner have a stock tip, it can pay to listen. After all, the newsletter they have run for over a decade, Motley Fool Stock Advisor, has tripled the market.* 

David and Tom just revealed what they believe are the ten best stocks for investors to buy right now... and M & T Bank wasn't one of them! That's right -- they think these 10 stocks are even better buys.

See the 10 stocks

*Stock Advisor returns as of June 2, 2020

Darren King -- Chief Financial Officer

Thanks, Don, and good morning, everyone. As we noted in this morning's press release, M&T's results continue to be impacted by the economic slowdown brought on by the COVID-19 epidemic and the return to a zero-interest rate policy by the Federal Reserve. Our clients, both consumer and commercial, have adjusted to the new economic reality, which is reflected on our balance sheet by a slowdown in some loan categories and notably higher levels of deposits. In light of the challenging economic environment, our focus has shifted somewhat from capital distribution to capital strength.

As far as the impact on M&T goes, the low interest rate environment resulted in a decline in our net interest income. Payment-related fees suffered from the reduced level of economic activity due to the pandemic-related lockdowns. However, lower rates also led to a 13% improvement in mortgage banking revenue compared to the first quarter. Trust income remained solid, and operating expenses were well controlled.

The net result provided a solid foundation to support expected credit costs while also improving our capital ratios. In connection with the CECL loan loss accounting standard, which reflects our assessment of the future economic conditions as of the end of the quarter, we added $254 million to our allowance for credit losses. The common equity tier 1 ratio improved by 32 basis points to 9.51%, indicating that M&T is well-positioned to meet the needs of our customers and our communities. Now let's review the results for the quarter.

Diluted GAAP earnings per common share were $1.74 for the second quarter of 2020 compared with $1.93 in the first quarter of 2020 and $3.34 in the second quarter of 2019. Net income for the quarter was $241 million compared with $269 million in the linked quarter and $473 million in the year-ago quarter. On a GAAP basis, M&T's second-quarter results produced an annualized rate of return on average assets of 0.71% and an annualized return on average common equity of 6.13%. This compares with rates of 0.9% and 7%, 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 $3 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 second quarter, which excludes intangible amortization, was $244 million compared with $272 million in the linked quarter and $477 million in last year's second quarter. Diluted net operating earnings per common share were $1.76 for the recent quarter compared with $1.95 in 2020's first quarter and $3.37 in the second quarter of 2019.

Net operating income yielded annualized rates of return on average tangible assets and average tangible common shareholders' equity of 0.74% and 9.04% for the recent quarter. The comparable returns were 0.94% and 10.39% in the first 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. Turning to the balance sheet and the income statement.

Taxable equivalent net interest income was $961 million in the second quarter of 2020, down by $20 million from the linked quarter. This primarily reflects the lower interest rate environment following the Federal Reserve's emergency reduction to its Fed funds target late in March. These cuts led in turn to a 105-basis point decline in average one-month LIBOR compared to the first quarter. Overall, average interest-earning assets increased by $15 billion to $123 billion, while the net interest margin declined by 52 basis points to 3.13% compared with 3.65% in the linked quarter.

The government's fiscal and monetary policy actions were the primary drivers of our significant balance sheet growth in the quarter. First, loans made through the Paycheck Protection Program or PPP added $4.8 billion to average loans for the quarter. A significant portion of PPP funds currently sit in customer deposit accounts waiting to be deployed. Those PPP-derived deposits in combination with other stimulus programs led in turn to a $10 billion or 170% increase in our placement of cash at the Federal Reserve Bank of New York.

Those large balance sheet movements had a similarly large impact on the net interest margin. Cash held at the Federal Reserve reduced margin by an estimated 25 basis points, while having little effect on net interest income. The PPP loan portfolio was additive to net interest income during the quarter, but the combined impact of income and balances diluted the margin by about three basis points. The lower interest rate environment caused an estimated 22 basis points of pressure to the margin.

The net impact of lower rates was somewhat mitigated by a 38-basis point decrease in the cost of interest-bearing deposits. All other factors amounted to another two basis points of margin pressure. Average total loans increased by $6 billion or 7% compared with the previous quarter. Looking at loans by category, on an average basis, compared with the linked quarter.

Commercial and industrial loans increased by $5.4 billion or 22%, including the $4.8 billion of average PPP loans. C&I loans grew by nearly $600 million, excluding the PPP activity, largely the result of having line draws on the balance sheet for a full quarter, net of repayments. We saw a somewhat unusual decline in dealer floor plan balances as customers return to showrooms faster than manufacturer inventory could be shipped. Commercial real estate loans grew by 3% compared with the first quarter.

Residential real estate loans declined by just over 2% or $332 million, which primarily reflects the continuing measured rate of paydowns on acquired mortgages. Loans purchased from servicing pools, pending resolution, partially offset those paydowns. Consumer loans were up less than 0.5%, reflecting higher indirect recreation finance loans, partially offset by lower auto loans and home equity lines of credit. On an end-of-period basis, the PPP portfolio more than offset a net contraction of other commercial and industrial loans, reflecting paydowns of about half of the line draws that occurred late in the first quarter and a $1.3 billion decline in floor plan loans.

CRE and consumer loans each grew a little over 1%, while consumer real estate loans declined slightly. Average core customer deposits, which exclude deposits received at M&T's Cayman Islands office and CDs over $250,000, grew 17% or over $15 billion compared with the first quarter. That figure includes $10 billion of noninterest-bearing deposits. The factors driving the change included cash from the government stimulus programs held in both commercial and consumer accounts and higher levels of mortgage servicing escrow deposits.

These, in turn, led to the higher placement of cash at the Fed. On an end-of-period basis, core deposits were up $14 billion or 15%, reflecting those same factors. Foreign office deposits decreased 39% on an average basis and 29% on an end-of-period basis as on balance sheet sweep rates return to historic lows. Turning to noninterest income.

Noninterest income totaled $487 million in the second quarter compared with $529 million in the prior quarter. The recent quarter included $7 million of valuation gains on equity securities, largely on our remaining holdings of GSE-preferred stock, while the first quarter included $21 million of losses. Also, recall that during the first quarter of 2020, M&T received a cash distribution of $23 million from Bayview Lending Group. There was no such distribution in the second quarter.

Mortgage banking revenues were $45 million in the recent quarter compared with $128 million in the linked quarter. Residential mortgage loans originated for sale were $1.1 billion in the quarter, up 25% from $919 million in the first quarter. Total residential mortgage banking revenues, including origination and servicing activities, were $111 million in the second quarter, improved from $98 million in the prior quarter. The increase reflects the higher volume of loans originated for sale combined with stronger gain on sale margin, partially offset by lower servicing income.

Commercial mortgage banking revenues were $34 million in the second quarter compared with $30 million in the linked quarter. The comparable figure in the second quarter of 2019 was $36 million. Trust income was $152 million in the recent quarter, up slightly from $149 million in the previous quarter. The rebound in equity markets from first-quarter lows, good capital markets activity and $5 million of seasonal tax preparation fees, were all factors during the quarter, more than offsetting the emerging impact of money market fund and fee waivers in the zero-interest rate environment.

Service charges on deposit accounts were $77 million compared with $106 million in the first quarter. COVID-19-related waivers of many of the consumer service charge categories and a slowdown in overall payments activity were the primary factors contributing to the decline. The $46 million linked quarter decline in other revenues from operations reflects the Bayview Lending Group distribution I mentioned earlier, as well as lower payments revenues, that are not included in service charges, such as credit card interchange and merchant discount. Loan-related fees, including syndication fees also declined given the reduced pace of non-PPP commercial loan origination activity.

Operating expenses for the second quarter, which exclude the amortization of intangible assets, were $803 million, down some $100 million from $903 million in the first quarter. Recall that operating expenses for the first quarter included approximately $67 million of seasonally higher compensation and benefits costs, the largest of which related to accelerated recognition of equity compensation expense for certain retirement-eligible employees. As usual, those seasonal factors declined during the second quarter. In addition, we reduced our level of incentive accruals to reflect lower levels of new business activity following the pandemic-related lockdowns.

The impact of the pandemic also led to a noticeable decline in certain other expense categories. Advertising and marketing costs declined by $13 million compared with the prior quarter to under $10 million. Other costs of operations for the second quarter, included a $10 million addition to the valuation allowance on our capitalized mortgage servicing rights. A similar-sized addition was made during the first quarter.

In addition, travel and entertainment expense and professional services declined by an aggregate $12 million. The efficiency ratio, which excludes intangible amortization from the numerator and securities gains or losses from the denominator was 55.7% in the recent quarter compared with 58.9% in 2020's first quarter and 56% in the second quarter of 2019. Next, let's turn to credit. Net charge-offs for the recent quarter amounted to $71 million.

Annualized net charge-offs as a percentage of total loans were 29 basis points for the second quarter compared to 22 basis points in the first quarter. The increased charge-off activity largely relates to problem loans identified at the end of 2019 but whose deterioration was likely accelerated by the pandemic-induced economic slowdown. The provision for credit losses in the second quarter amounted to $325 million, exceeding net charge-offs by $254 million and increasing the allowance for credit losses to $1.6 billion or 1.68% of loans. The allowance currently reflects an updated series of assumptions, reflecting a somewhat more adverse economic scenario than either of the scenarios used at January 1 or March 31, 2020, as well as the impact of proactive risk rating changes within our portfolio to reflect the current economic environment.

Our macroeconomic forecast uses a number of variables with the largest drivers being the unemployment rate and GDP. Our forecast assumes the quarterly unemployment rate falls to 9% in the third quarter of this year from a peak at 13% in the second quarter, followed by a sustained high single-digit unemployment rate through 2022. The forecast assumes GDP contracts 6.7% during 2020 and recovers to prerecession levels by the second quarter of 2022. Nonaccrual loans as of June 30 amounted to $1.2 billion, an increase of $95 million from the end of March.

At the end of the quarter, nonaccrual loans as a percentage of loans was 1.18%. It's 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 zero risk weighted and carry little or no credit risk. Excluding the impact of PPP loans, the ratio of allowance for credit losses to loans would be 1.79%. Similarly, the ratio of nonaccrual loans to total loans would be 1.27%.

Annualized net charge-offs as a percentage of total loans would be 31 basis points. Loans 90 days past due, on which we continue to accrue interest, were $536 million at the end of the recent quarter. And of those loans, $454 million or 85% were guaranteed by government-related entities. Consistent with agency guidance, loans that have received some sort of relief, whether payment deferrals covenant modifications or other form of relief as a result of COVID-19-related stress are not reflected in our nonaccrual or delinquency numbers.

As the virus spread in mid-March to April, our customers reached out for relief actions and support from the bank. From that peak period, request for relief from both commercial and consumer customers are down by about 95%. M&T's relief actions in the commercial portfolios have been heavily influenced by auto and recreation finance dealers. Those dealer relationships, the vast majority of which are floor plan inventory account for $4.4 billion of relief requests, amounting to nearly 80% of total dealer balances.

High levels of forbearance for dealers has been seen industrywide. And given the strength in sales activity toward the end of the quarter, we expect further extensions of relief to be limited. Excluding dealer relationships, relief provided to commercial customers totaled $9.8 billion, comprising some 16% of balances. For the consumer portfolios, we provided assistance to approximately 30,000 accounts, representing $3 billion in balances of our combined mortgage, home equity line of credit in indirect recreation finance or auto portfolios, amounting to about 9% of total balances.

Of interest, approximately 30% of that population made a payment in the month of June. For mortgage loans that we don't own and that we service for others, relief was provided to approximately 70,000 accounts, totaling $13.2 billion. Turning to capital. M&T's common equity tier 1 ratio was an estimated 9.51% as of June 30 compared with 9.19% at the end of the first quarter.

This reflects the impact of earnings in excess of dividends paid and lower risk-weighted assets. M&T did not repurchase shares during the second quarter and will not be doing so in the third quarter. Turning to the outlook. As we sit here today, our outlook is somewhat clearer than it was 90 days ago.

However, there is still a fair amount of uncertainty. As far as the balance sheet goes, our liquidity assets, short-term investments, and deposits at the Fed rose somewhat beyond our expectations, with both a rate and volume-driven impact on the net interest margin. This was driven by inflows of deposits from the PPP loans and other government stimulus programs. While the pace is uncertain, we believe that recipients will use these funds and excess reserves will trend downward somewhat as we go forward.

Any additional government stimulus programs would, of course, have an impact on those assumptions. Excluding the impact from cash and PPP loans, the net interest margin experienced a 22-basis point rate-driven decline, following a 105-basis point decline in LIBOR. We expect average LIBOR in the third quarter to fall a little further as well deposit rates. Given all factors, we expect the printed margin will improve somewhat in the coming quarter.

We expect average PPP loans will increase from the $4.8 billion average in the quarter toward the $6.5 billion outstanding at June 30. Beyond that, the rate of prepayment and forgiveness will significantly impact the balance retained. As you know, forgiveness under the program is not currently automatic and is subject to review by the SBA. While we expect significant numbers of forgiveness request before the end of the year, it's difficult to handicap how much will occur in the third quarter versus the fourth quarter.

Commercial loan growth, which is to say excluding PPP loans, has slowed, and we expect those balances to remain flat to slightly down over the remainder of 2020 compared to where we ended the quarter. In a normal environment, we'd expect to see a seasonal slowdown in inventories and a corresponding decline in our floor plan loan balances during the third quarter. Recent vehicle sales volumes might necessitate dealers adding inventory against a backdrop of constrained production at the manufacturers and the upcoming model year changes. Residential real estate loans should continue to experience a measured pace of runoff as the vast majority of our loans are originated for sale.

However, as I touched on earlier, there are circumstances under which we can pursue buying delinquent loans or loans under forbearance out of the MBS pools we service. As a result, we have stepped up buyouts from the Ginnie Mae pools, which will lead to temporary growth in our residential mortgage loan portfolio. These are government guaranteed loans, so our credit risk is extremely limited. We'd expect to see some improvement in the growth of consumer loans compared with the recent quarter as recent indirect originations are on the balance sheet for a full quarter.

Our outlook for net interest income is also somewhat dependent on the eventual resolution of the PPP loans. While we expect net interest income to improve in the third quarter from the second, the rate of improvement is also heavily dependent on the pace of forgiveness or prepayments on the PPP loans. Turning to fees. Residential mortgage applications continue to be very strong with rates as low as they are, and purchase activity has held up well.

Waivers of money market mutual fund management fees will continue to impact trust income, while the zero-interest rate environment persists. We'll see a larger impact in the coming quarter than we did in the second quarter. Service charge income was impacted by lower levels of customer activity, higher balances and state-mandated waivers of certain consumer fees. Payments activity recovered by the end of the quarter.

However, certain categories of retail fees continue to be waived. In addition, higher commercial deposit balances have enabled those customers to offset the hard dollar fees arising from their treasury management obligations through the use of earnings credit. As we noted, the seasonal increase in salaries and benefits we experienced in the first quarter largely normalized during the second quarter. We have curtailed hiring and have been redeploying team members around the bank to address shifting business needs.

We're in a similar situation to last year where we expect expenses in the second half of the year to be roughly the same as the first half, excluding the first-quarter seasonal salaries and benefits figure. The third quarter is often higher than the fourth quarter from an expense perspective. In the CECL loan loss accounting environment, our allowance for credit losses at the end of the quarter reflects the macroeconomic variables I referenced earlier, the impact of the government stimulus, and the characteristics specific to our portfolio. As GAAP requires, we will reassess the allowance at the end of the third quarter based upon updated macroeconomic scenarios and M&T's specific credit data.

Finally, regarding capital, as noted at the beginning of the call, we're focused on capital strength. Consistent with the CCAR results, we don't expect to repurchase any shares during the third quarter, and we continue to meet the Federal Reserve earnings threshold for dividend distributions. Of course, as you're 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 Laurie will briefly review the instructions.

Questions & Answers:


Operator

Thank you. [Operator instructions] Your first question comes from the line of Gerard Cassidy of RBC.

Gerard Cassidy -- RBC Capital Markets -- Analyst

Good morning, Darren.

Darren King -- Chief Financial Officer

Good morning, Gerard.

Gerard Cassidy -- RBC Capital Markets -- Analyst

Question for you on the net interest margin. Obviously, you've cited the reasons why the margin came down so much this quarter, and the PPP loans and the excess deposits certainly were the contributing factors, as you pointed out. When you look at your asset yields, the loan yields, in particular, they came down quite steeply. Your cost of funding, not as much.

As we go forward, do you sense that the asset yields in the loan portfolio, the decline will start to diminish, but you still see continued reductions in the cost of funding?

Darren King -- Chief Financial Officer

Yes. Sure. When we look through the composition of the balance sheet, I think you saw that the yield on the C&I loans, in particular, came down with the CRE loans. That tends to be where the hedging impact goes, and we'll see those come down through time.

We also did see a nice decrease in the deposit cost. A significant portion of our deposit costs now are linked to the index, and so they came down proportionately. And so when we looked forward, if you hold the cash and the PPP aside and look at what we would consider to be the core margin, we think we wouldn't expect to see a drop like the 22 that we saw this past quarter that it would be more tied to movements in LIBOR, which given how close we are to zero, we would expect to be pretty minimal. And so we would see relatively small movements in the yield on loans and a similarly small yield in the deposit costs, with probably minimal exceptions from the managed costs where time deposits are still repricing, and there will be some of that go on through the second half of the year, as well as some of the other managed deposit costs like some of the money markets and the like.

So probably, all else equal, we follow LIBOR and maybe trickle down a couple of basis points in the core margin. But as we kind of pointed out, the printed margin will bounce around depending on how the cash balances move in the quarter, as well as the PPP funding.

Gerard Cassidy -- RBC Capital Markets -- Analyst

Very good. And then as the follow-up question, you obviously built up your reserves this quarter. M&T has distinguished itself as one of the best banks through a full cycle on credit. If the economic outlook does not deteriorate from the period that you just came through, in which you made your assumptions for the reserve buildup, and granted you'll have some reserves for any loan growth, should we anticipate provisions for loan losses coming down in third or fourth quarters, assuming this economic outlook doesn't deteriorate from this quarter under CECL accounting?

Darren King -- Chief Financial Officer

Well, I guess the math would say if the economic assumptions are identical, then the provision wouldn't really change with the exception of reflecting net loan growth and perhaps by category. So if you had a meaningful mix change, you would see an increase. But all else equal, as you point out, you probably see the provision not moving much and that the action would be predominantly actual charge-offs.

Gerard Cassidy -- RBC Capital Markets -- Analyst

But would the provisions still be at this level, though? I mean, could you build up reserves so much, assuming life of loan losses?

Darren King -- Chief Financial Officer

Yes. The provision would only be impacted by net loan growth if the macroeconomic assumptions were unchanged.

Gerard Cassidy -- RBC Capital Markets -- Analyst

OK. Thank you.

Operator

Your next question comes from the line of Dave Rochester of Compass Point.

Dave Rochester -- Compass Point Research -- Analyst

Good morning, guys.

Darren King -- Chief Financial Officer

Good morning.

Dave Rochester -- Compass Point Research -- Analyst

I appreciated the detail on the deferrals in some of those loan buckets. I was just wondering what the total amount was at the end of the quarter, if you happen to have that or the current total in terms of the total amount of loans in deferral. And then if you happen to have the criticized or classified asset trends for the quarter, that would be good to hear as well.

Darren King -- Chief Financial Officer

So we'll get you the sum of the loans in forbearance or with some kind of payment relief. When you look at the nonaccrual loans, the nonaccrual loans bumped up by $95 million for the quarter. More action was in the criticized space, and a lot of that was related to some of the forbearance activity. And so as we went through the portfolios and we looked at loans that were in forbearance, we took them down a greater to, depending on where they started.

And that was really the primary driver of an increase in criticized, but most of them are just over the threshold that we would start to consider criticized. Back to the balance figure of what's in the forbearance in aggregate for loans that we own, it's about $17 billion, about $14 billion commercial and about $3 billion for consumer, which would include the mortgage portfolio that we own. And then if you add in what we service for others, there's another $13-odd billion, but that's not our credit risk. And I guess it's important to point out that as we come to the end of the quarter, we're getting close to the end of some of the 90-day forbearance period.

And within that commercial portfolio, there's a large amount, $4.4 billion or around there of commercial or dealer floor plan balances. And we are not seeing a lot of activity in extension requests, and so we would expect that number to start to come down as we go into the third quarter.

Dave Rochester -- Compass Point Research -- Analyst

OK. And then for the total amount of criticized loans you have at the end of the quarter, what were those?

Darren King -- Chief Financial Officer

We typically don't disclose that until the 10-Q comes out.

Dave Rochester -- Compass Point Research -- Analyst

OK. Got you. But it sounds like those were up -- I guess I'm not sure with some of the other deterioration that you guys disclose. OK.

Got you.

Darren King -- Chief Financial Officer

Yes. It's primarily because of the down rates that we talked about.

Dave Rochester -- Compass Point Research -- Analyst

Yes. Got it. And then just switching to the NIM quickly. Just so I understand, you're saying that the core NIM moves a little bit lower, but the reported NIM with all the impacts on cash and whatnot could actually be higher if cash does actually decline in the next quarter or two.

Is that what you're saying?

Darren King -- Chief Financial Officer

That's right. Yes.

Dave Rochester -- Compass Point Research -- Analyst

OK. And I was just curious, just looking at the balance sheet. I mean, you have a decent amount of borrowings there. Your cash grew more than your entire borrowing balance this quarter.

Are there any opportunities where you can whittle away at that balance over time? I realize you may have some built-in prepayment penalties that you have to run into, but any opportunities on that side?

Darren King -- Chief Financial Officer

Yes. I guess if you'll look around -- given where rates are, we wouldn't be taking on any duration risk at this point with converting into mortgage-backed securities, but would prefer to stay short. We'll have some long-term debt that will mature and roll over, and we won't replace that in the short term. We can take advantage of those balances for funding.

And then we'll probably use some of it, as we mentioned in balances in the mortgage servicing business, where we would buy out some of the loans that are being modified and take them out of the pools. And we'll use some of the excess cash to fund those in the short term.

Dave Rochester -- Compass Point Research -- Analyst

And going back to the roll-off of the borrowings, what's the schedule for that look like over the next year? Do you have any lumpy roll-offs that could really help the NIM in any of these quarters?

Darren King -- Chief Financial Officer

Look, everything is swapped, so the savings is not materially large.

Dave Rochester -- Compass Point Research -- Analyst

OK. Got it. Great. Thanks, guys.

I appreciate it.

Operator

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

John Pancari -- Evercore ISI -- Analyst

Good morning. On the reserve, have you -- actually on your loss expectation, have you run your internal company run stress test yet for 2020 DFAS? Or have you disclosed it?

Darren King -- Chief Financial Officer

We don't run a company stress anymore as being a category for bank.

John Pancari -- Evercore ISI -- Analyst

OK. All right. So I guess I just want to see if you could help think about the reserve level now at the 1.68%, that's about half of the last time you did run the publicly disclosed company run stress tests in 2018. So how should we -- could you help us kind of triangulate that reserve coverage where it stands now versus the stress loss estimate? Is it really the macro assumptions behind it and dialing in of stimulus that you can attribute to that difference?

Darren King -- Chief Financial Officer

Well, I guess there's a number of factors, John, to think about when looking at that level of allowance. I guess the first place that I start is we think about it more as 1.8% as opposed to 1.7%, which is I know splitting hairs, but just because of the PPP loans and the fact that there's no credit risk there or very little. And then when you compare the CCAR process to the CECL process, there are a couple of key differences. One is the assumptions that you mentioned.

The stress test is specifically designed to really put the bank under stress. And so when you look at some of the assumptions, the depth of the decline in GDP and the length of time it stays at those depths, the drop in unemployment, while we saw a bigger drop in unemployment in the short term. When you look at where that trends out over the last four to six quarters of the projection, it's actually not as bad as what CCAR would have. And so there are some parts of the macroeconomic assumptions that are different.

A key difference would be HPI. And that we've seen HPI hold up really strong right now, and the decrease in HPI is a lot different in our economic assessment today than what it would be under stress. Another really key difference is within CCAR, you assume that all -- or we assume that all nonaccrual loans charge-off. And we know that's not our experience in the CECL allowance.

Our CECL process would reflect our experience at recovering or curing those customers that are in nonaccrual. And so those are some of the differences between the two. I guess the thing that we look back to is how the portfolio and the underwriting performed under the great financial crisis and what the actual losses were during that and what the peak losses were. And when we compare the portfolio to what we saw in that crisis, which was more financial-driven crisis, obviously, than as widespread as the pandemic.

We look at where our reserving ratios are relative to what happened then, so CECL more so reflects our underwriting experience than CCAR might. We're to use a little bit more of industry information and industry-level losses. And so those are a bunch of the things that go on underneath. And we're doing work.

I think we talked about it during the quarter, René talked about it, of the portfolios where we see the forbearance, the highest, really literally going bottoms-up and credit by credit to understand what we think their cash flow is and their ability to sustain revenue shortfalls for periods of time and looking at the values of the collateral and how they might hold up under stress. And all of those different things are factors in our thought process when we set the allowance. And so it's thorough, but it reflects both the economic environment and the view we have of our portfolio.

John Pancari -- Evercore ISI -- Analyst

That's very helpful. Thanks for all that color. And then separately, just wanted to see if I can get a little bit more granularity around the underlying commercial demand that you're seeing on the loan book. It sounds like understandably, given that you expect commercial balances to remain flat to slightly down.

It sounds like not a lot of underlying commercial demand. But just want to get your updated view on what you're seeing live in terms of borrower demand given where we stand right now in the pandemic.

Darren King -- Chief Financial Officer

Yes. We've definitely seen a slowdown in demand. I think PPP helps create some of that slowdown because the customers were able to take advantage of those programs. But really, what's encouraging to us that we've seen is a really rapid reaction by the customers to the new operating environment.

And we've seen them make changes to their business to manage the burn rate of expenses, and so when you look at how they're running their business, part of this you will see reflected in the deposit balances. The pace of turnover in their operating accounts isn't what it was pre-pandemic. And so when you factor that in, along with some of the PPP funds and their ability to manage, they're very liquid. And that's a great thing from a credit perspective because they can withstand a longer slowdown in economic activity.

But it doesn't necessarily lead to a lot of loan demand in the short term. And so that's really what's going on there. But from our perspective, when we look at how the customer base has actually responded, we're quite pleased with what we see. And the customers, I would say, are quite sanguine about what's going on.

I think we're in that period right now where for our part of the country, we got through the hard hit that happened in April and May. We're starting to see a rebound, and our customers are guardedly optimistic about how things are -- how they will unfold, but cautious, and all of that is what's leading to that forecast.

John Pancari -- Evercore ISI -- Analyst

Got it. Thank you.

Operator

Your next question comes from the line of Steven Alexopoulos of JP Morgan.

Steven Alexopoulos -- J.P. Morgan -- Analyst

Good morning, Darren. To start, just to follow-up on the reserve, how much of the reserve build this quarter was tied to the change in economic forecast as opposed to portfolio specific changes? And were qualitative overlays a material factor this quarter?

Darren King -- Chief Financial Officer

So if you look at the allowance, the biggest driver of the increase in -- or the provision was the economic forecast. When you look at the moves in the grades, in some cases, the grades moved by one or two, and those aren't generally the PD. The loss assumptions under those changes aren't big enough swings to cause a meaningful move in what you think your loss rates would be. It's really as you get up into the criticized and even beyond the criticized and the nonaccrual, where you really start to see an uptick in those rates.

I would describe it as the grating that happened this quarter kind of caught up a little bit to the macroeconomic assumptions that were in the model at the end of the first quarter. And so those are the drivers there. And when you look at the qualitative overlays, we really didn't change the overlays from an economic perspective in terms of the impact of the stimulus. That was pretty consistent from quarter over quarter.

I would say the place where there was a qualitative overlay was looking at some of the instances in the models where they haven't seen these types of inputs in terms of GDP changes and unemployment and in worse economic conditions tend to overpredict losses. And so for certain portfolios where we went through bottoms-up and looked at what we thought our loss was we kind of took that into account and made an overlay. But really, not much change quarter over quarter and not tremendously significant.

Steven Alexopoulos -- J.P. Morgan -- Analyst

OK. That's helpful. And then what are the areas that is a focal point for you guys? Is New York City commercial real estate exposure, as you know?

Darren King -- Chief Financial Officer

Yes.

Steven Alexopoulos -- J.P. Morgan -- Analyst

Could you give us what's the balance today? What are you seeing there? What are the deferral requests in that bucket? Any color would be helpful. Thanks.

Darren King -- Chief Financial Officer

Sure. So within New York City, I think when you look in the 10-K, including construction, the balances are right around $9 billion -- actually, sorry, a little bit high, $7 billion. And when we look at the requests for forbearance, they haven't increased since what we saw in the first quarter. And when you walk down the portfolio by category, if we look at the industrial type of space, it's really quite small, and that's the least impacted segment.

We're still seeing rent collection rates of 95% plus there and low levels of forbearance. Excuse me, the multifamily portfolio, which in total exposure, including construction, is about $2.5 billion. We're still seeing rent receipts above 90%. In fact, they went up slightly in July compared to what they were in April.

And we haven't seen any increases in requests for forbearance there. The office space continues to also see some slight upticks in rent being collected slightly over 90%, which is up from what it was in April, although not totally meaningfully. Hotel is probably one of the more challenged portfolios in aggregate, at least in terms of the economic impact, although the ability of the clients to handle things has been quite strong. And so the forbearance rates there are the highest, upwards of 70% plus.

But what's interesting is when we look at the forbearance rates across the whole hotel portfolio, they aren't materially higher in New York City than they are across all other geographies. And we have seen through our card activity that people are going back to hotels. In the retail world, which is another relatively large portfolio and then where there's probably the next highest level of forbearance activity after the hotel portfolio, is about $1.3 billion. And interesting, that's probably where we've seen the biggest increase in rents being paid since April and now is up over 50% from being slightly above 30% before.

And so the New York City portfolio is so far holding up well. We're seeing some trends that are positive in terms of payments, ultimately to the landlords. And then I guess the other thing that we keep an eye on is what the loan-to-values are in that portfolio versus any other. And the loan-to-values across all those categories, in New York City, tend to be five to 10 percentage points lower than what the LTVs might look like in a similar asset class in the rest of the footprint.

And so it's still early days. And so far too early to declare victory, but the trends and the change in the trends from early in the quarter to the end of the quarter haven't deteriorated and, arguably, are slightly more positive.

Steven Alexopoulos -- J.P. Morgan -- Analyst

All right. There was more color than I was hoping for. Thanks for taking my question.

Operator

Your next question comes from Matt O'Connor of Deutsche Bank.

Matt O'Connor -- Deutsche Bank -- Analyst

Hey, guys.

Darren King -- Chief Financial Officer

Hey, Matt.

Matt O'Connor -- Deutsche Bank -- Analyst

The PPP loans are a pretty big part of your loan portfolio, bigger than, I think, a lot of your peers. And if you just think about the origination fee that you'll get on that, it seems like a pretty decent chunk relative to earnings, capital reserves. Can you just give us an estimate on what you think the blended origination fee is on the $6.5 billion and how much you think will be forgiven or repaid the next few quarters in aggregate?

Darren King -- Chief Financial Officer

Yes. So the blended fee on the loans is probably somewhere between 270 and 280. And the pace of forgiveness is, as we said, a little bit tough to handicap. Although as we think about it, Matt, there's a couple of things.

The change to the program rules extended the time that the customers had to actually use the money. And so that pushes out the time period for the forgiveness. We thought a bunch of it originally, and we in the industry that it would be starting even as early as late in the second quarter. Now that could push all the way into the fourth quarter.

And so we expect forgiveness in the second half of the year, probably in the range of 60% to 70%. But how much is in the third quarter versus the fourth quarter. I guess conservatism would lead us to say, we think it's a little bit more back-end loaded in the fourth quarter than the third. But it remains to be seen.

And the other thing that's still just kind of hanging out there is the rules for forgiveness are still in consideration and will there be any automatic forgiveness or not for the small loans. I think those things will impact the timing that the longer it takes for the rules to be clear on forgiveness. I think that will have two impacts. It will slow down when that actually happens and when those fees get accelerated.

And then the second thing that will happen is, I think, customers are being cautious about when they use the money. And so that sits on the balance sheet while they want to make sure that they spend it appropriately so that they have a better chance of having their loans forgiven, and so it's a little bit fluid in there. But what's relatively certain is that that amount of origination fee or processing fee will come in over the next couple of years. The timing of when that shows up is a little bit uncertain.

Matt O'Connor -- Deutsche Bank -- Analyst

Got you. And then just separately, yeah, the expenses were very well managed this quarter. The guidance for second half to be similar to first half kind of implies 1Q expenses might have been bloated, 2Q expenses might have been unusually low. Is that a reasonable way to frame it? Because the cost came in a lot lower, I think, than anyone would have expected in 2Q.

And it doesn't seem like that's sustainable. So I just want to make sure I'm framing that right.

Darren King -- Chief Financial Officer

Yeah. I think you're in the right ballpark there. We think that it's closer to Q2, obviously, than it was to Q1. But this is M&T.

And when we see movements in revenue and headwinds, then we take action on the expense side. And that's kind of what happened in the second quarter and many of the things that we impacted in the second quarter should continue in the back half of the year. As we mentioned, there might be a little bit of lumpiness just because of some seasonal things that happen third quarter versus fourth. But if you average them out, you're in the ballpark, maybe slightly higher than Q2 on average for each of those last two quarters with a bit of a mix difference between third and fourth.

Matt O'Connor -- Deutsche Bank -- Analyst

Got you. That's helpful. Thank you.

Operator

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

Ken Usdin -- Jefferies -- Analyst

Thanks. Good morning. A question on the fee side of things. Darren, just wondering if you can help us understand some of the lines are better, some of the lines were worse.

And as you look out, can you just talk about -- can those service charges rebound? What do you see for mortgage? And just what were the fee waivers this quarter interim? And then what did they turn into? Thanks.

Darren King -- Chief Financial Officer

Sure. So just kind of going category by category, if you look at the mortgage business, second quarter was a bang-up quarter for us and for most of the industry. Whether we continue at that pace in the consumer space, it's difficult to say, but seems unlikely. But I don't think we go all the way down to where we were in the first quarter.

I would expect we're somewhere in between. We should continue to see some gain on sales just from where rates are and from some of the portfolios that we're servicing, where people will refi, which will be offset a little bit by slightly lower servicing fees. But net-net, I would expect us to be in the range between where we were in the first and the second. Trust income, you start to see a little bit of pressure on that line item just because of the fees that are associated with the money market mutual fund and the management fees.

And so well we might see a little bit of pressure there. And then the service charge line is really the one that was impacted this quarter. And it's interesting when you look at it that the payment-related fees, so interchange have come back to where they were and up year over year. The biggest impacts are on some of the other line items, some of the ATM-related fees where, from some of the government programs, state-related programs, we're not allowed to charge for those.

And so those will be off the table until that changes. And then one of the other big categories is NSF, and NSF is lower mainly for two reasons. There's less transaction activity and also higher average balances and customer accounts and then commercial account balances. So I think in a long-winded way of saying I think we start to see service charges come back up through time, but probably takes a couple of quarters to get back to the run rate we were at in the first quarter.

And then in some of the other income, again, you've got payment fees, that's where the credit card-related fees are just because they're not associated with deposit accounts. And so interchange and merchant should come back with business activity. And then that's where some of the loan origination fees are like syndications. And again, as those markets pick up, we'll see those fees come back.

Ken Usdin -- Jefferies -- Analyst

Got it. And then just a follow-up on the fee waivers. I think you talked last time that they could get to around $10 million. Do you have any different view there? And what were they this quarter? Thanks.

Darren King -- Chief Financial Officer

Well, they were slightly higher than that this quarter. And I would expect them to be in that range in the third quarter, and then we'll see how things unfold to the fourth. It's just because they're more policy-related. I'm hesitant to comment on what they might go beyond the third quarter.

Ken Usdin -- Jefferies -- Analyst

Oh, wait. I'm sorry, Darren, I meant the trust, the waivers with the money market fee waiver side.

Darren King -- Chief Financial Officer

Oh, OK. They were small in the second quarter. I mean, really not that significant. And we'll start to see them go up as we go forward, maybe in the neighborhood of $10 million a quarter, something like that.

Ken Usdin -- Jefferies -- Analyst

OK, got it. Thanks very much.

Operator

Your next question comes from the line of Saul Martinez of UBS. Sir, your line is open. Please state your question.

Saul Martinez -- UBS -- Analyst

Hello. Hi. Can you hear me? Sorry about that.

Darren King -- Chief Financial Officer

We got you now.

Saul Martinez -- UBS -- Analyst

So just to kind of make sure I understood the guidance on expenses and have my numbers straight. I mean, you mentioned that it would be similar to the second half versus the first half and stripping out some, I guess, some small amount of noncore items in the first quarter. By my calculation, that gets you a little under $1.7 billion or, let's just call it, $845 million, $850 million a quarter, which seems a little bit high versus what you kind of implied in your answer to Matt. So I just want to sharpen the pencil a little bit and make sure I kind of have a little bit tighter range in terms of the outlook going in the second half of the year.

Darren King -- Chief Financial Officer

Yes, sure. I think the difference might be the comment about first and second is it's the same, excluding the seasonal salary impact or similar, excluding the seasonal salary impact. And that, by itself, will be about $60 million that you'd exclude. And if you were including that in the first, equal second, then you'd get that higher run rate.

I think if you take the run rate that we saw in the second quarter and bump it up by $10 million or $15 million and think about that on average for the second half of the year, you're probably in a better run rate.

Saul Martinez -- UBS -- Analyst

OK. Got it. So basically, second half normalizing to the seasonality basically?

Darren King -- Chief Financial Officer

That's right.

Saul Martinez -- UBS -- Analyst

Is that what you're saying? OK. Got it. I just misunderstood that. OK.

And I guess on the NII, another bit more of a clarification question as well. I mean, you gave a ton of color on the moving parts around that interest margin, I probably got lost in all of it. But I think the punch line was you do expect some uptick in net interest income in the third quarter versus the second and -- first, I think I just wanted to make sure that was right. And is that outlook incorporating any of the PPP deferral gains that you talked about 60% to 70% in the second half, fully realizing it's completely uncertain and more fourth quarter tilted? But is that view that you will have some NII growth dependent on some realization of PPP forgiveness gains?

Darren King -- Chief Financial Officer

Yes. And if you go in the third quarter, we expect a slight uptick in NII, and that's because some small amount of forgiveness starts to come in. And then you'll have a full quarter of the PPP loans outstanding and the amortization of the fee and then really start to see more of that weighted to the fourth quarter.

Saul Martinez -- UBS -- Analyst

OK. But there is some stuff that view that there is some uptick is incorporating some forgiveness. And I guess the 60% to 70%, is that -- if you could just clarify on that, do you have a view of what percentage of the loans will be forgiven? And is the 60% to 70% based on the percentage of forgiveness, in other words, of those loans that are forgiven, you think 60% to 70% will in the second half without the right interpretation of what you said?

Darren King -- Chief Financial Officer

Yes. So at the highest level, when you look at the PPP loans, you're probably up to 75% over time. I guess I'm kind of hedging my bets a little bit and thinking that maybe it's more like 60% or 70%. What we know is just, again, because of the rules about when the funds are supposed to be dispersed, that it needs to happen by and large by the end of the year.

And then the question on timing is when the customers go and seek forgiveness and then how quickly the SBA processes it. It's likely to be concentrated in the next three quarters. And then there will be some -- the rest of it will hang on and show up kind of over the normal course of those two-year loans and the normal amortization. But that's kind of how we're thinking about it.

And as I said, a little bit of that is in the third quarter, which helps. And then obviously, whatever doesn't show up in the third quarter starts to show up in the fourth and the first of 2021.

Saul Martinez -- UBS -- Analyst

Got it. And I fully realize it's a very precise science. But I appreciate it. That's all.

Operator

Your next question comes from the line of Erika Najarian of Bank of America.

Erika Najarian -- Bank of America Merrill Lynch -- Analyst

Hi. I just had one follow-up question on NIM. Darren, you said earlier that the net interest margin was impacted 25 basis points by excess cash and three basis points by PPP. I think investors thought it was helpful when one of your peers said this morning what normalized NIM could be -- could stabilize as we look at your earnings power post all the PPP noise.

So if I exclude that, I would get to net interest margin in the 3.40% range. And as I think about your outlook for LIBOR decreasing but potentially offset by deposit costs, as you think about second quarter, third quarter of 2021, is three, four low or high three, three a good starting point for the margin?

Darren King -- Chief Financial Officer

Yeah. I think that's a good way to think about it that we talked about that, that we look at the core earnings power of the bank, and that 22 basis points gets you down to where we would be absent those large increases in cash balances at the Fed, as well as PPP. And you probably see that move down small amounts per quarter, just as the different hedge activity rolls on and rolls off, that the way the hedges were constructed through time that the received fixed rate on the hedges will slowly come down, and that will create some headwind. The flip side is the loan activity that is happening, we've seen some increases in the margin on that.

And so that will help offset. So I think you're thinking about it right, you're in the ballpark of how we would see it without those factors, really, one of the big questions in the printed margin. And I would just kind of hold this to the side. It's just where the cash settles out.

Erika Najarian -- Bank of America Merrill Lynch -- Analyst

Got it. Thank you.

Operator

I will now return the call to Don MacLeod for any additional or closing comments.

Don MacLeod -- Director of Investor Relations

Again, thank you all for participating today. And as always, with any clarification of any of the items on the call, the news release is necessary, please contact our investor relations department at (716) 842-5138.

Operator

[Operator signoff]

Duration: 65 minutes

Call participants:

Don MacLeod -- Director of Investor Relations

Darren King -- Chief Financial Officer

Gerard Cassidy -- RBC Capital Markets -- Analyst

Dave Rochester -- Compass Point Research -- Analyst

John Pancari -- Evercore ISI -- Analyst

Steven Alexopoulos -- J.P. Morgan -- Analyst

Matt O'Connor -- Deutsche Bank -- Analyst

Matt OConnor -- Deutsche Bank -- Analyst

Ken Usdin -- Jefferies -- Analyst

Saul Martinez -- UBS -- Analyst

Erika Najarian -- Bank of America Merrill Lynch -- Analyst

More MTB analysis

All earnings call transcripts