OceanFirst Financial Corp. (OCFC) Q3 2020 Earnings Report, Transcript and Summary

Good day, and welcome to the OceanFirst Financial Corp. Earnings Conference Call. [Operator Instructions] Please note, this event is being recorded. I would now like to turn the conference over to Jill Hewitt, Investor Relations Officer. Please go ahead.

Great. Thank you. Good morning, and thank you all for joining us. I'm Jill Hewitt, Senior Vice President and Investor Relations Officer at OceanFirst Financial Corp. We will begin this morning's call with our forward-looking statement disclosure. Please remember that many of our remarks today contain forward-looking statements based on current expectations. Refer to our press release and other public filings, including the risk factors in our 10-K, where you will find factors that could cause actual results to differ materially from these forward-looking statements. Thank you, and now I will turn the call over to our host this morning, Chairman and Chief Executive Officer Christopher Maher. Chris?

Thank you, Jill, and good morning to all who've been able to join our third quarter 2020 earnings conference call today. This morning, I'm joined by our Chief Operating Officer, Joe Lebel; Chief Risk Officer, Grace Vallacchi; and Chief Financial Officer, Mike Fitzpatrick. As always, we appreciate your interest in our performance and are pleased to be able to discuss our operating results with you. This morning, we'll cover our financial and operating performance for the quarter, discuss the strategy behind the liquidation of certain loans and address our plans to manage the business through the next phase of the pandemic economy. As we did in the second quarter, please note that our earnings release was accompanied by a set of supplemental slides that are available on the company's website. We may refer to those slides during this call. After our discussion, we look forward to taking your questions. In terms of financial results for the third quarter, GAAP diluted earnings per share were a loss of $0.10. Quarterly loss was driven by a $35.7 million provision for credit losses taken during the quarter. The $35.7 million provision included a reserve build of $20.7 million, net charge-offs related to loans held for sale in the amount of $14.2 million and ordinary-course charge-offs of just $800,000. The $20.7 million reserve build was driven by commercial loan risk rating changes related to pandemic forbearance loans and other qualitative factors related to generally weak economic conditions. The outsized provision this quarter reflects a comprehensive risk rating review of the commercial forbearance portfolio and represents our best estimate for the credit risk related to the pandemic. Our reserve estimate is not reliant upon additional fiscal stimulus nor does it reflect an overly optimistic view of the resolution time for the pandemic. Grace will be walking through our approach to credit risk management and the allowance. Reported earnings were also impacted by merger-related expenses, branch consolidation expenses and the net unrealized loss on equity investments that totaled $5.8 million net of income tax. As a result, we pegged the core results for the quarter to be a $266,000 loss or less than $0.01 per share. Regarding capital management, the Board declared a quarterly cash dividend of $0.17 per common share and approximately $0.44 per depository share of preferred stock. The common share dividend represents the company's 95th consecutive quarterly cash dividend. The $0.17 common dividend reflects our view that earnings will rebound in the fourth quarter and into 2021. There are no plans to reduce or eliminate our common dividend at the present time. Capital levels remained strong with tangible equity to total assets of 8.4%. Please note that this ratio was negatively impacted by PPP loans, which decreased this ratio by 37 basis points. The TCE ratio, excluding PPP loans, would have equaled 8.8%. As noted in our earnings release, we moved a significant portion of our PPP loan portfolio to held for sale at quarter end. The sale of those PPP loans was completed this week at a net gain of approximately $5.3 million. The combination of loan sales and a forecasted improvement in profitability in Q4 will return us to a position of internally generating capital at a healthy rate. The company suspended common share repurchases on February 28. Since that time, we've been working with our clients to better understand how the pandemic has impacted their businesses. That understanding has allowed us to address our credit risk position as demonstrated by our sale of high-risk loans and reserve build. This data also informs our annual stress test process, which is currently underway. Following the completion of that stress test in early November, we will consider reactivating our share repurchase program. At this point, it appears that share repurchases could begin as early as the fourth quarter. The company has slightly more than 2 million shares remaining in the current share repurchase program. Should our level of surplus capital support a larger repurchase program, the Board will consider an expansion to the current authorization. Before we discuss the outlook for our business, I'd like to spend a minute reviewing market conditions in our area of operation. When we last spoke in July, our core market of Central New Jersey and the New Jersey Shore were beginning to reopen and there was a sense of optimism. That optimism continued throughout the summer and into the fall as a substantial number of people chose to leave the urban centers of New York and Philadelphia to spend the summer at the shore. In fact, many of these visitors have remained well past the traditional summer season. The influx of homebuyers for metropolitan areas is fueling a mini-boom in residential real estate with median single-family home prices in our core markets rising more than 10% compared to 2019. The population expansion and robust residential real estate market allowed many of our clients to salvage a solid summer season. The latest regional unemployment figures support the sharp snapback in New Jersey. Unemployment across New Jersey has dropped to just 6.7% from a high of 16.3% in April. New Jersey unemployment is well below the 13% and 11% results reported in the New York and Philadelphia metropolitan statistical areas, respectively. While COVID cases are surging nationally, they remain manageable throughout most of our markets, and there are no signs that a broad-based economic shutdown will be required in the near term. Although, we remain in uncertain times and circumstances could deteriorate very quickly. The recovery is uneven with many businesses and consumers continuing to feel the pain of the economic recession. Additional targeted stimulus is absolutely needed, but as I mentioned before, our forecasts do not rely on additional rounds of stimulus. Turning to the bank, our strategy is simple and conservative. We're using the same playbook as we would to address any economic crisis: First, we secured liquidity, increasing deposits by $1.4 billion this year. Second, we bolstered capital with $181 million in subordinated debt and perpetual stock issuances. Third, we ring-fenced the credit risk on the balance sheet and disposed of the highest-risk assets. The fourth and final effort will now be the rebuilding of margins and boosting operating margin -- profits. The decisions we made this quarter addressed downside risk to the balance sheet and allow us to focus on earnings and capital management strategies. The decision to accelerate the resolution of high-risk credits drove our financial results for the quarter. We're certainly disappointed to be announcing a GAAP loss, but OceanFirst has a history of acting quickly to corral risk in difficult times. In the first quarter of 2007, OceanFirst was among the first banks to acknowledge the developing recession and recorded a $0.47 per share loss during that quarter. Moving quickly in early 2007 allowed the bank to realize recoveries that were far higher than would have been the case later in 2007 and the years that followed. Our early action during that crisis allowed the bank to be firmly on the road to recovery more than a year before the Bear Stearns and Lehman collapse has been hit later in 2008. We certainly hope that the current challenges won't come close to reflecting the 2008 crisis, but we believe in the adage, "Sometimes, the first loss is the smallest." These actions also liberate the resources we need to focus on building our business. Our path to build margins and improve profitability is centered on executing a mix shift from cash into a combination of liquid securities and loans. During the third quarter, our average cash balances exceeded $800 million, and we experienced an additional $300 million of deposit growth during the quarter even after running off $80 million of certificates. Robust deposit growth has driven our loan-to-deposit ratio to under 90%, a ratio that will go even lower as we generate an additional $388 million in cash proceeds from the loan sales planned this quarter. While the deposits raised this year aren't immediately accretive, we are winning new relationships and will be patient as we deploy that cash over time. The mix shift will provide a significant opportunity to fuel earnings growth for quite some time. Excess deposits will be deployed in the same way we have built the business in the past: by recruiting commercial lending talent throughout our markets. Joe will walk you through our plans to accelerate recruiting and hiring to expand our lending capacity in 2021. Organic growth is our top priority heading into the fourth quarter, but it will take time to prudently deploy the massive amount of cash on the balance sheet. So we expect traditional measurements of profitability to be challenged until the mix shift is completed. An additional motivation to accelerate the resolution of pandemic-related credit risk is to allow us to entertain additional capital management strategies, including share repurchases and acquisitions. I've already covered share repurchases. We should also address acquisitions. Thus far in 2020, economic conditions and the need to focus on our core business have precluded consideration of acquisitions. However, as we wrap up 2020 and look forward into 2021, it appears that opportunities to acquire valuable franchises might begin to appear. Third quarter results in the banking sector have signaled that many institutions have moved past peak credit provisions. NIM pressure is now the focus and will probably continue to impact many banks over the coming quarters. Perhaps the best mitigating strategy for a protracted low-NIM environment will be improvements in operating leverage and relative expense reductions through increased scale. Shedding our highest-risk loans on an accelerated basis will allow us to consider strategic acquisition opportunities should they present themselves. At this point, let me hand the call off to Joe Lebel.

Thanks, Chris. I'll touch on deposit activity and loan originations and the effects of both on net interest income -- or net interest margin before some brief comments on expenses in the 2021 loan growth plans. We continue to grow relationship deposits with quarterly growth of $316 million and year-to-date growth of $1.4 billion, well ahead of our expectations. More importantly, this growth has included several high-profile, 8- and 9-figure relationship wins. And since we've seen most of the PPP loans utilized by borrowers, a very substantial portion of the excess liquidity is now available to be deployed into higher-yielding investments and loans. These new deposits are the result of the maturity of our corporate cash management business, which has increased by 33% year-over-year, as we've doubled staff to support the growth. You may recall that just a few quarters ago, we had discussed the loan-to-deposit ratio in the high 90s and strategies to raise deposits to provide the foundation for more loan growth. The current loan-to-deposit ratio was 86% and heading lower as we complete the loan sales mentioned earlier. We believe we've built a comprehensive competitive suite of treasury talent and products and can continue to accelerate deposit growth in 2021 and beyond. At the same time, we've continued the repricing of existing deposit accounts, reducing our cost of deposits from 57 basis points to 49 basis points quarter-over-quarter. Deposit costs continue to decrease as indicated by a quarter end weighted average deposit rate of 46 basis points. In the fourth quarter of 2020 and the first quarter of 2021, over 705 million of CDs with a weighted average rate of 157 basis points will mature, providing additional deposit cost reduction opportunities. Loan originations were $418 million for the quarter with continued strong residential closings and solid commercial activity despite the economic environment. Exclusive of PPP originations, loan volume of $1.3 billion year-to-date is 30% ahead of 2019. Loan growth is muted as we continue to exit some weaker acquired loans, as is our longtime strategy. We are continuing to sell most newly originated conforming residential loans with sales of $169 million through the 9 months of 2020. Loan originations from the commercial team were $188 million. And while the overall pipeline is stable quarter-over-quarter, there's no doubt that the pandemic has adversely affected what we expected to be a very strong year in commercial primarily from our Philadelphia, New York regions after they hit their stride earlier in the year. The residential business has helped in recent quarters with the combination of all-time low rates, moving back to suburban living and the scarcity of inventory all generating record activity in 2020 and a continued near-record pipeline. At these interest rates, we've chosen to continue to generate mortgage banking income while managing balance sheet exposures. Loan growth for the remainder of the year should be modest or flat as businesses and individual borrowers regain their footing depending on the state of the economy as it recovers from the pandemic. In the weeks to follow the election, we hope to hear more clarity from our customers. As you can imagine, the lack of clarity regarding future public policy and the need for more data regarding public health trends continues to restrain commercial investment and lending opportunities. Moving to the net interest margin, we saw a 27 basis point reduction. Core NIM declined by 19 basis points due to a few factors, notably the excess liquidity on the balance sheet, which we calculate at 13 basis points, and 6 basis points from the lower interest rate environment. Additionally, purchase accounting reduced the overall NIM by 6 basis points. New loans continue to be originated at lower market rates with average yields on interest-earning assets down 34 basis points quarter-over-quarter. Commercial originations contained swap loans with floating rate LIBOR spreads that continue to affect margin but provide floating rate flexibility for balance sheet protection. Moving to brief comments on expenses. Excluding merger-related and branch consolidation expenses, our operating expenses increased $802,000 primarily attributed to cover related expenses totaling $1.7 million, an increase of $600,000 over the previous quarter. We've also recorded an increase in the cost of employee benefits and higher professional fees. No branches were consolidated in the quarter. We expect some branch consolidation in 2021 and we'll provide more details next quarter. I'll finish up with some comments about the markets we serve and the expectations for loan growth. Even with the advent of COVID, we are very optimistic for loan and deposit prospects in 2021. Our newer market geographies in the New York City and Philadelphia metros have exceeded expectations. While I expect flat or modest growth for the remainder of the year, exclusive of underperforming loan sales, we are investing in the addition of commercial banking talent and will be prepared for strong loan growth to return in 2021. That said, the pandemic may restrain growth in Q1 and Q2. The interest rate environment is challenging and will require incremental investment in commercial lenders and teams to strengthen NIM and provide earnings momentum in 2021 and beyond. We have a track record of opening commercial loan production offices in new markets, attracting top-tier talent, and we will accelerate similar investments in 2021. I expect a few new LPOs during the 21 fiscal year as well as the deepening of our existing markets. At this point, I'll turn the call over to Grace.

Thank you, Joe. As Chris mentioned, OceanFirst has a long history of proactive identification and recognition of risk. We believe this philosophy enables a more effective management of risk and leads to the best outcome with regard to credit risk, which typically means the minimization of losses. It is simply good risk management and, I believe, is a key factor in our low loan loss history relative to our peers over a variety of economic cycles. This includes cumulative losses since 2007 that are 43% lower than our proxy peers and 70% lower than our UBPR peers. Specifically, we have been and remain, through the pandemic, quick to downgrade credit where appropriate and take active measures to work out problem loans. Chris mentioned our proactive approach to the developing recession in the first quarter of 2007. More recently, we actively derisked the Sun and Cape Bank portfolios shortly after acquisition, exiting several hundred million dollars of exposure that was inconsistent with our credit risk appetite and incurred zero loss through this process. We've maintained this philosophy of proactive risk identification during the pandemic. While the CARES Act provides exemption from nonaccrual and troubled debt restructuring status, it doesn't absolve banks from appropriately identifying the risk in their portfolios. We mentioned in the first and second quarter calls that we expected risk-weighted migration in the third quarter as forbearance periods began to end. Over the course of the third quarter, we continue to stay in close contact with our borrowers and monitor their financial condition and repayment capacity. This means we've talked with our borrowers, discussed their financial condition, verified their liquidity and have a reasonable basis for assessing their ability to continue or resume payments. As a result, we've updated risk rating for a significant portion of the commercial portfolio, including all of the commercial loans that received forbearance. During this process, we made no assumptions about additional stimulus funding, the timing of a potential vaccine or any other factors that may impact economic recovery. Given our close contact with our borrowers and these proactive efforts to measure and monitor credit risk, we are optimistic that we've identified and quantified all current material credit risk in the portfolio. Through this process, we identified $81 million in higher-risk commercial forbearance exposure and chose to accelerate resolution of these credits through loan sales. This includes $30 million in New York exposure and $51 million in New Jersey and Pennsylvania at recovery rates of 85% and 82%, respectively. These sales include $15 million in hotel exposure, $12 million in restaurant and food-related exposure and over $4 million in gym and fitness exposure. $18 million are substandard rate credits, and $32 million are special mention. Net of these loan sales but inclusive of all other risk rating migration this quarter, total classified balances, including residential and consumer loans, remain very manageable at just 2.2% of total held-for-investment loans. Both commercial- and residential-classified credits are well secured with low weighted average loan-to-value. Overall, the loan portfolio continues to perform well over 6 months after the pandemic-driven shutdown across our market areas. $4.1 billion or 77% of our $5.2 billion commercial loan portfolio never received forbearance. As of October 23, $1 billion or 91% of the $1.1 billion in commercial exposure that received forbearance has returned to payment. This leaves $88 million in commercial loans remaining on full forbearance as of October 23, all of which are expected to resume payments by December 31. In the residential portfolio, $207 million or 63% of the $329 million that received forbearance has returned to payment. This leaves $122 million in residential loans remaining on forbearance as of October 23. Inclusive of these forbearance loans that have returned to payment, total loan portfolio delinquencies are just 17 basis points, and nonperforming loans held for investment are 37 basis points as of September 30. Total TDRs should be low and essentially unchanged at $23 million. Net charge-offs, exclusive of the discount on loan sales, were just $800,000 in the third quarter, and OREO balances remain negligible at $106,000. Within the residential portfolio, we have just $9 million remaining in first 90-day forbearance period and $113 million remaining in second 90-day forbearance period exposure. The weighted average LTVs are just 61% and 68%, respectively, and weighted average FICO scores are 724 and 741, respectively. I'll point out that over 75% of our residential forbearance portfolio is located in Ocean, Cape May, Monmouth and Atlantic counties, where annual median home prices have increased substantially over the past year. It's quite reasonable to conclude that current loan to values and, thus, our risk of loss is even lower than these LTV figures indicate. The third quarter provision includes a qualitative credit reserve associated with residential forbearances. Depending on the final resolution of this portfolio, we may choose to sell a pool of residential loans in the fourth quarter. As with the commercial loan sale decision, residential collateral valuations remain strong, and it may be economical to accelerate the final disposition of any remaining loans that demonstrate high-risk characteristics. The existing reserve should support that approach if necessary. Inclusive of both commercial and residential loans and exclusive of loans held for sale, forbearance loans totaled just $210 million or 2.6% of total loans as of October 23. Our allowance for credit losses build of $20.7 million brings the funded loan loss reserve balance to $56.4 million or 70 basis points of total held-for-investment loans. This coverage increases to 1.10% with the addition of $31.6 million in unamortized credit marks. As Chris noted earlier, the allowance for credit loss increase was driven by commercial loan rating changes related to pandemic forbearance loans as well as qualitative factors related to generally weak economic conditions. The allowance for credit losses currently represents our best estimate of the credit risk related to the pandemic. We expected some third quarter risk rating migration as CARES Act forbearance periods ended and the impact of the economic shutdown and recession on individual borrowers became clear and the migration would drive an increase in quantitative reserves. Qualitative adjustments were made to account for the potential for further impact to our borrowers' repayment capacity, both within the commercial portfolio and as residential forbearance periods come to an end this quarter. To conclude, I'll reiterate that our credit risk position is modest and manageable. We're confident that we have a comprehensive understanding of our current credit risk profile given our thorough assessment of individual borrowers' capacity to repay, that we have identified the current risk of loss in our portfolio in light of today's economic conditions and that these losses are reflected in the ACL build and the loan sales. We've elected to settle a portion of higher-risk commercial credits to derisk the balance sheet and redirect our resources toward growth initiatives and reserve for risk associated with the end of residential forbearance periods during this fourth quarter. We're currently updating our stress test, which will include the Federal Reserve's latest adverse and severely adverse scenarios. Excess capital at the holding company further strengthens our position and provides ample growth capacity. I'll now turn it back to Chris for his concluding remarks.

Thank you, Grace. At this point, we will open the line up for questions.

[Operator Instructions] The first question today comes from Erik Zwick of Boenning and Scattergood.

First, just some questions on the loans you decided to liquidate. Just trying to think about -- so they've got about $67.5 million, and then I think the related charges that you recognized are at $14.2 million. So just curious, one, if I'm thinking about those 2, are those the right 2 numbers to compare? And then how did you estimate losses? And have you received any bids on the loans? Or just trying to kind of understand kind of the loss ratios there.

Sure. So those numbers are correct. And so the New York pool has already closed post quarter. So that has realized at the figures we expected. And we have what I would classify as firm bids from multiple bidders on the New Jersey and Pennsylvania pool. So we expect to close that in the next 2 weeks, but the marks that we took at the end of the quarter are reflective of those clearing prices. So we think we've got reasonable assurance. But New York is closed and Pennsylvania and New Jersey are pending closure.

And with respect to the bids, are these coming from other banks or nonbanks? Just curious what the buyer pool looks like at this point.

That's really interesting because they're very different markets. So in the New York metropolitan area, there's a long history of real estate investors being involved in note purchases for a variety of reasons, including the default rates are stronger or upheld more strongly by the courts in New York than elsewhere. So in the case of the New York credits, it was really a matter of matching a specific credit and asset class to buyers who specialize in those kinds of assets. And so over the years, we've done business with them and sold loans from time to time. Prices are a little weaker than you would typically expect in New York. You'd expect a little better price, but I think that's the pandemic discount. The buyers in New Jersey and Pennsylvania are a little different. That's much more of an institutional kind of credit fund buyer that is looking for distressed loan notes and prices them accordingly. So conversely, New York came in a little bit lower than the historical average. New Jersey actually is coming in a little better than the historical average. So -- but neither are very far off what we would expect to be pretty good recovery rates.

Okay. And then of that pool of loans that you decided to sell, any -- I'm sure it may be a mix, but any kind of breakdown between which loans were acquired through some of your recent acquisitions and some of where the relationships may have originated at OceanFirst?

Sure. There was not any overwhelming pattern of whether they came from a particular acquisition. So there was no clustering of concern around maybe a credit underwriting criteria or anything like that. There were a few that were long-term OceanFirst customers and a sprinkling probably of -- I'm not sure every acquisition, but maybe almost every acquisition may have had 1 or 2 credits. So it was a little bit of everything and that gave us comfort, too, because it indicates that there wasn't a pattern in the loan book that might be more problematic over time.

And then previously, you'd spoken favorably of the experience using forbearance following the Hurricane Sandy kind of impact. Just curious, what are you seeing this time that's different that leads you to believe that working with some of these borrowers longer would not lead to a more favorable outcome versus selling them today?

So that's a good question. So one of the things we look at is the chances that if we had held these loans, would we have recovered more than we recover today and the answer is almost certainly yes. So if we had decided to hold on to these loans, many were paying. In fact, a good section of these were past credits. However, what we had to weigh against that was ultimate recovery, which may take us several years and the ability to kind of put these behind us and focus on other things. So we certainly took a little bit of an extra liquidity discount. I guess that's probably a few million dollars. But we bought ourselves the time and attention by accelerating those. So I think that there was a cost to accelerate this. We don't think it was a giant cost. The other thing is that there's a finality that comes with the final disposition, and we were very conscious that had we merely kept these on the books, established a risk pool and taken, let's say, a reserve for the $14 million, that may have proven to earn us more money in the long run. But it may also have led to several quarters of discussions about valuations with a lot of stakeholders, with our regulators, with our investors, where now you'd be in a position of having to explain why you thought a fitness center or a hotel property was actually valued where you think it is. So this way, we get the final disposition. We know exactly what the answer is. We can be certain that we took those risks off the balance sheet.

And then last one for me, and then I'll step out of the -- step aside. With respect to the third quarter net interest margin, did it reflect any interest reversal related to any of the downgrades or the transfers to held for sale?

It did not. So there were no unusual entries in the net interest margin for the quarter.

The next question today comes from Russell Gunther of D.A. Davidson.

Just a follow-up on some of Erik's questions on the credits that you moved this quarter. Are you able to share how the loss rates translated from an asset class perspective, so discount on hotel, restaurant and gym and fitness?

So it was a little bit all over the place. And in any one category, we only had a couple of loans. But I'll -- Grace, you may have some extra data you could provide on that -- or Joe -- if you've got those figures handy.

Yes, Chris, I can give you a little bit. Russell, I think in hospitality, we're about $0.75 recovery. Again, it varied a bit, but I'll give you an example. In the -- and the food and beverage is very similar, about $0.75. So I think almost everybody was somewhere in that range. On an overall scope, it was interesting to watch how the bids came through and how things worked. I do want to add one other comment relative to the sale. 66% of the sale was criticized or classified. So while some were paying, we had rated them appropriately based on our expectations for long-term health.

Okay. And then were any of the -- let's see. $12 million in restaurant loans, any of that relate to the New York City pub portfolio?

Interestingly, that portfolio seems to be doing rather well. So those owner-operators have a significant amount of liquidity. And then they've got the liquidity in the bank in the form of bank deposits that we have and we can see. It is also -- for those families, it is their primary and only asset in many cases where they're operating. So a couple of things. They have been able to demonstrate more liquidity than you might expect from a real estate investor. They're also much more protective of their collateral. So as we talked to those folks, we said, look, we'd like to stand with you, but we need you to post the liquidity, in some cases, post payments in advance and those kinds of things. That was a particularly strong portfolio at the end of the day.

Got it. And then last one on this one. You mentioned that, Joe, 66% of this was out of the criticized, classified, which I think now is at 2.2% of held for investment. Ex those -- I know that these were particularly troubled assets in terms of hit -- hard hit from the pandemic, and the classified portfolio is not made up of all of these types. But can we -- is there any read-through from the clearing rate out of the criticized classified to what remains on criticized classified today?

So I'll make a couple of comments and I think Grace and Joe may chime in as well. We are an active commercial bank, which means we're not -- we don't look at things exactly the same way maybe a primary real estate lender would. We're used to managing relationships and credits. And we have a couple of tools: We have covenants. We have guarantors. They have significant liquidity. So moving a loan into special mention or substandard for us doesn't necessarily mean that we're significantly afraid of a loss. It means that the risk has increased and that we need to work on it and we need to work through it. And we've done this in the past when we acquired some of the banks. Cape and Sun in particular, we moved several hundred million dollars of loans into special mention, substandard. And then in the quarters that followed, we worked right through them. In some cases, you got paydowns or restructures or collateral -- additional collateral posted or you would get a sale of an asset or get refinanced out of it. So I wouldn't read through that we're overly concerned with these categories. We're just pretty aggressive about those designations because they give us the transparency to feel comfortable with the credit risk. Joe and Grace, you're intimately involved in that process, so you can chime in.

Before -- I'll turn it to Grace, but before I do that, I'll just add to Chris' comments that -- and we'll go back to Erik made a comment earlier about Sandy. A good example in that environment is that we had borrowers that were downgraded to criticized or classified that stayed in that bucket for a period of time and then returned to pass-rated credits. In this instance, I think we're going to see a lot of the same. And the credits that we sold, we felt, long term, would not have the same capability to return to the kind of performing asset that we needed. So is that fair, Grace?

Absolutely. And I guess what I would add is, in the context of what Chris was talking about -- I hope you guys can hear me better now than when I was speaking earlier. We tend to -- even though we have a preponderance of real estate collateral, we consider ourselves commercial lenders, and so we take that more proactive approach to risk rating and dealing with our customers in general. And so that's opposed to perhaps waiting for delinquency status to inform our risk ratings. So like both Joe and Chris said, this doesn't necessarily mean that these borrowers even have a payment issue at this point. And I can tell you that there is no real concentration in the nature of the downgrade. So for instance, they aren't all in one sector. They're kind of random in a sense.

And then just switching gears, if I could, quickly on the expense side of things this quarter. You mentioned some of the moving pieces. One of the kind of COVID related expense, I think, was $1.7 million. So I'm just wondering, trying to think of what the run rate would be. Perhaps that $1.7 million moves lower over time. But could you quantify what expenses are kind of below normalized based on perhaps less travel and expense, for example? Do those 2 things kind of net each other out? Or how should we think about that?

I think you should think about our expenses being elevated now predominantly because of COVID. And the 2 factors that Joe mentioned, obviously, direct COVID expenses, and we are investing a lot of time and energy into protecting our employees and our clients and our community. So we've got a significant amount of expenses with health care professionals. We are doing active testing within our employee base. We have outsourced health management where we've got health care professionals making decisions about people's work status and return to work. So we think all those things are responsible, but they are at a significant cost. Those expenses will decrease over time as we go into 2021. The health care expense, I just want to make a note about that. We're self-insured. So we bear a portion of the risk on health care. And our health care systems were closed in the second quarter. So even if you wanted to go do something as an employee, you couldn't find a place to go do it. So any procedures that people needed in the second quarter became procedures they got in the third quarter. So I don't think that, that was -- that should probably not prevail. But thinking about where are expenses today and where are they going, I think you're going to see 2 factors. You're going to see things like the pandemic expenses decrease. Joe mentioned we'll probably do a little bit of branch consolidation next year. There's not as much of that available as there was in prior years. But then at the same time, Joe also mentioned continuing to hire commercial bankers. So you're going to see a redirection of spend. It's hard to say exactly where that will net out. I don't think you're going to see a dramatic increase, and you're not going to see a dramatic decrease in expenses. It will bounce around kind of where we are now.

Okay. I guess you got to my follow-up question already. So the last piece of it would be, given that expense outlook, some of the other moving pieces on the top line you had discussed, do you think you're going to be able to generate positive operating leverage in 2021?

We do. So the positive operating leverage really needs to come from a fair amount of organic growth. And let me be clear about organic growth. We have about just under $1 billion to deploy before the balance sheet moves by $0.05. So we have a ton of cash. We average cash balances in the third quarter of over $800 million. So the first order of business is that our balance sheet says we're an $11.6 billion bank. We're really like a $10.5 billion bank. And that's why some of those margin numbers and return on asset numbers may be difficult to compare going forward. We have to grow into the cash we have. So when I say organic growth, I mean organic growth in the loan book. So that will occur over the next several quarters. As we do that, that will produce operating leverage because you'll see NIM stabilize and then start to increase. And then we will be working at operating leverage more on the revenue side than on the expense side.

Next question comes from Zack Westerlind of Stephens.

It's Zack Westerlind filling in for Matt Breese. So just on the deferral front, in the presentation, you mentioned that the remaining 200 should be worked out by year-end. I was just kind of curious about what factors give you confidence that, that's going to be -- those deferrals will be worked off the balance sheet by the end of the year.

So I think the main factor is that we have done a loan-by-loan review of everybody in forbearance and, actually, a significant amount of the commercial book that's not even in forbearance to make sure we understood -- not just a quick call of how you're doing. We had in-depth discussions about their access to liquidity. In some cases, we require liquidity to be brought to the bank and posted in advance of those payments. So we have done a fairly methodical review to make sure not just the people that have said, okay, I'll go back to making payments because our biggest concern was that we might have people come back. They hadn't made payments maybe in 6 months. They've got some cash put aside. They'll make 2 or 3 payments, and then in February, we've got an issue. So we wanted to get ahead of that and make sure we understood who really had the capacity not just to make a couple of payments or to hit a target or was contingent upon something happening that was external to the business, right? I'll make a couple of payments, but then I've got to have 100% capacity in my dining room before I can make payments after that. So it was a pretty rigorous review to draw out any areas we would have concern. And if you think about the pattern in those loans we sold, those were the loans where we couldn't find a way forward, where we looked at it and said, look, there's good collateral here, they're good people, but you just couldn't put together a string of circumstances with liquidity and operating characteristics where you could see a path out. So we don't think they're going to get better, and we don't think -- even if COVID eases more quickly than we think, it's not going to be a miracle for them.

Understood. And then just on the provisioning expense, there was a little bit of reserve build this quarter higher than the past couple of quarters. Are you comfortable with the level of reserves you have now? Or should we expect continued build going forward?

So I think we've had a lot of conversations over the last several quarters about the aggregate coverage of the loan portfolio. And if you think we've got this anomaly in unamortized credit mark, which kind of has a portion of the reserve that's not quite as apparent, if you add those together, we wind up now, as Grace said, at about 110 basis points, 111 basis points. For the credit risk profile of our institution, we think that's a pretty adequate reserve. As Grace noted, even with these charge-offs that we took this quarter, our long-term credit performance is quite favorable to our proxy peers as well as the UBPR peers. So our reserve should be a little bit lower than average. So we think we're in the ballpark of where we should be. In quarters 1 and 2, we were taking provisions not knowing exactly how the forbearances were going to fall out. We've gotten through that process now. The main concern would be what we think is a very small chance but a chance that economic conditions will deteriorate considerably from where they are today. So that would come in the form of maybe a regional very significant kind of stay-at-home order or those kinds of things. So hopefully, we avoid that. We don't see any signs of that today, but that would be something that we've not taken into account in our reserve.

Got you. That's helpful. And then just last question, moving over to the margin. We saw 27 basis points of compression this quarter. Moving into 2021, do you see that as a bottom? Or do you think that there's room to run lower there?

We're close to a bottom. I don't know. I'd be very careful about declaring things like an absolute bottom to things. But I think we're close. And what we have going on now is we'll have 2 factors in the fourth quarter, and then I'm comfortable we'll be through the bottom. The first is that, look, we're liquidating $388 million worth of loans. Now most of those were PPP loans, but they were still paying us interest last quarter. So we're not going to get that interest in. On the flip side, the $800 million we keep referring to, that was earning us 10 basis points. So even moving into mortgage-backed securities, you get a pickup. So we have some portion of the balance sheet that's at a negative carry. And Mike, maybe talk a little bit about just what that negative carry means in terms of margin to give you a sense as to what our stabilized margins will look like.

Yes. So when we look at the margin, if you look at short-term cash of $805 million in Q3, a year ago, it was $40 million. So with round numbers, it's $750 million excess cash earning 10 basis points. If we just drop the denominator of average earning assets by that amount, $750 million, that picks up 24 basis points of margin. Our margin would go from 2.97% to 3.21%. And if we just offset the 10 basis points that we're earning at the Fed with 10 basis points less in deposit cost and offset it, so just like Chris said, it's just an $11.6 billion balance sheet. Really, it's really a $10.5 billion balance sheet. And then the other thing is we've talked about broker deposits. We had $250 million in April when we were concerned about liquidity. We have a negative carry on that at 1%. So that's another 3 basis points in margin. Those start rolling off in October then January then April. So just there, you have 24 basis points of negative -- affecting the margins. So that takes your margin back to 3.24%. So it's not a margin problem. We have excess liquidity that we need to prudently invest. That's what we need to do.

Got you. That's very helpful. And then one quick last follow-up for me. The $350 million in securities that you mentioned in the deck that's coming on in Q4, could you give us a sense for what those securities are yielding?

There's a blend of securities there. So we'd be careful, obviously, not to put it all in one asset class, and it will range from things like garden-variety mortgage-backed securities. Those are not going to earn us a lot. Those may wind up being 90 basis points, maybe a little better than that. It's not sexy, but it's a lot better than 10 basis points. And then we've got some other investments we'll make. We have done some subordinated debt investing. We will continue to do dividend-paying equities that we believe in and a few other things like that. Municipal bonds will be another portion that we can -- we have the opportunity to increase our municipal bond book. But we need to be thoughtful about that because the credit risk in the municipal bond world is evolving, and we want to make sure we stay away from entities that may have higher-than-expected credit risk. I should mention, too, we talked a lot about asset quality in the loan book. We've done a review of our fixed income book as well, and that led to the sale of about $17 million worth of securities in CMBS pools that we think are at very high risk. We were able to -- we did that in the fourth quarter at a negligible gain. So it's not -- it's a rounding error. But we've been through not just our loan portfolio. We've been looking through every aspect of our balance sheet to make sure that -- in this case, these were CMBS portfolios that were invested in the hotel sector that had very high vacancy and very high nonpayment ratios, and we're able to get out of them clean. And I think we got out clean because we were early.

[Operator Instructions] The next question comes from Frank Schiraldi of Piper Sandler.

Just wanted to ask, Chris, about as you've gone out and marketed these loans. From a timing standpoint, I assume you had to go through the loan-by-loan review, so maybe you guys just weren't ready earlier to get something done. But was there any market earlier anyway? I'm just trying to get a sense, as you've marketed these things, if you've seen a significant pickup in interested parties and what seems like pretty palatable pricing.

No. We -- it was very interesting timing, Frank, because, first, we wanted to be on the early side but not first, and there were a couple of other banks that we watched closely that were actually liquidating portfolios. In some cases, our execution, I think, was a little better. But we wanted to see a few trades cross so we understood roughly what we were talking about in terms of value. But then we didn't want to wait until there was a rush at the door either because this is supply and demand. So we knew there was a robust buying pool. They're all rational buyers, so they're not going to go a little crazy. But it was a robust pool. But we wanted to make sure that our sales were conducted earlier in the fourth quarter just in case there is some rush to liquidated assets late in the fourth quarter, in particular, after the election. So we wanted to take as much of the election risk off the table as we could.

That makes sense. And then you mentioned the loan-by-loan review having been completed. Is that in, I guess, all the higher-risk categories? And does that imply that the sales that you -- that have taken place or that are in loans held for sale at this point is the bulk of what you're looking to move off the balance sheet?

The sales that we're talking about now are the bulk of what we expect, and we have been through both the forbearance pools in totality and a significant amount of the nonforbearance portfolio, and we've obviously focused on the higher-risk industries and geographies. I have to make a point here: the geography can be as important as the industry. So for example, we have cases where restaurants that were in resort locations near the Jersey Shore did pretty well. And they're in good shape going into next year. They're used to having kind of a tight winter season because they don't really have one. So the geography makes a big difference.

All right. And in terms of the -- I just want to understand -- make sure I understand when you talk about the moving back from forbearance by the end of the year. And if I look at, for example, $88 million in full forbearance that isn't making payment right now, are you saying that you feel pretty good that this pool is going to move back to just regular payments? Or could there be a portion that does need some additional attention through TDR or what have you?

So there will be a very modest pool of loans that will go to interest only, and in that pool, we are highly confident that we've seen the liquidity that's going to make those payments. And in many cases, we have asked for that liquidity to be moved into the bank here. So the majority of that is either going to go full payment or go to IO. And here's our outlook on how to talk about forbearances and how to be doing disclosures going forward. We think that this quarter, we will have wrapped up this kind of pandemic forbearance-related activity, especially so -- barring no -- return to a bigger issue in the pandemic. So we think the most important thing to do now is to just move to traditional credit metrics going forward. So as you see our year-end and going into next year, we're going to classify loans as performing or nonperforming, delinquencies, TDRs, the way we would normally, and we'll show you exactly what's going on in the balance sheet. But we don't expect to have a forbearance portfolio into next year, and we don't expect to be reporting on it that way. We'll show you delinquencies. We'll show you TDRs. We'll show you nonaccruals. We'll show you those kinds of metrics, and I think we'll be providing complete visibility into the loan book. So...

The next question comes from Christopher Marinac of Janney Montgomery Scott.

Chris, you may have alluded to this earlier in the call, but the criticized loans that -- and assets that we'll see in the 10-Q, those can include the held-for-sale loans. But there's a chunk of those loans that are already kind of selling in the quarter. So we kind of have to pro forma that the criticized number is going to be lower than what the upcoming Q is going to say. Do I have that kind of right in my mind?

Absolutely. So we have 2 events. First, the New York sale has closed, so that is done. I know it's not a criticized asset, but the PP loan -- PPP loan sales is also closed. And the remaining New Jersey and Pennsylvania sales will close. It's going to be tight probably around the Q. But if they close after the Q, I would imagine we will provide an 8-K just assuring people that those sales have been conducted and what the final metrics are.

Okay. So the level and the ratios will obviously change and go lower from here. So I guess my other question would be, is it -- are you comfortable that the migration of other things beyond this quarter should be limited just based on what you see right now?

There's 2 things. First, we have done an exhaustive review of loans. So there is not -- if we saw it, we have marked it or dealt with it. There is always the chance that something might pop up in the loan portfolio. What we've done to address that risk is taking a fairly significant set of qualitative adjustments into our reserve to account for the risk that there may be a credit here or there that pops up. The other thing that, I think, gives us great comfort is, when you look at our largest loan customers, we have -- they're all performing really well. So the top 20 customers will all be on some form of payment. There are only a couple of cases of IOs in that case. And we have a very granular loan book for a company our size. So even if you have an individual credit or relationship, it is unlikely to be one of scale that would come back and cause an issue. And once you get through the top 20 of our clients, you fall below probably a $25 million relationship exposure.

Okay. Great. That's helpful. And then just one last question that might be for Grace is just the -- on the reserve calculation within economic forecast. How different is that today compared to back in March and April? And do you see that changing a little or a lot as this next quarter or 2 come up?

So Chris, it's not as bad as it was in March and April. It's about, I think, an average of 8% unemployment over the next year or so. That's an average, obviously. That's higher now and to go down over time. And I think GDP levels out between 1% and 2% in the -- through to the end of the 2-year reasonable and supportable forecast. So not great. Pretty much -- very similar to last quarter in a sense. Slightly better.

[Operator Instructions] The next question comes from William Wallace of Raymond James.

I apologize if I missed this, Chris, but I'm trying to reconcile the difference between the -- it's $51 million in New Jersey and Pennsylvania, $30 million in loans in New York being sold to the text of the release that says $45.5 million of loans moved to held for sale, and I assume that's net of the $14.2 million in charge-offs. I'm just trying to figure out why those numbers don't add or match.

Okay. We'll let Mike reconcile. But you're right. We're showing the net amount in the earnings release. We're taking that -- in the supplemental presentation, we're showing you the principal balance.

Yes. And the $45 million...

So the -- go ahead.

The $45 million was forbearance loans that were transferred into held for sale. So they were loans that were...

Okay. So the difference were performing loans that were -- okay. Okay. That $80 million of principal value, what was the original or average loan-to-value of those loans? I assume these are all real estate loans. If I'm wrong...

They all had -- I think almost all of them had real estate collateral. Grace, do you have the weighted average? It's probably...

I don't. I can look for it while the questions continue.

Yes. So I will say, look, if you think about what's been going on in our markets, if you think about places like New York -- and we love New York, right? So we've got a good franchise there. We're going to continue lending there. We don't believe it's smart to bet against New York City. All that stuff said, it's under a great deal of stress right now. And you've got the highest unemployment rate in New York City that's been recorded, I went back and looked, since the '70s. You've got the issues pre pandemic about rent stabilization. Now you have the pandemic. We're watching vacancy rates. We're watching concessions, which inform the drop in actual realized rent in certain units. And if anyone has kind of walked around New York or, frankly, Center City Philadelphia, there are a lot of closed and empty buildings. And I know a lot of those rents are being paid. But just because the rent's being paid doesn't mean that it will continue to be paid. So it's a very hard market. And we went through and looked at each of our credits and came up with some that we thought it's better to be out, especially at [ these towers ]. We're very happy with the recovery rates.

Okay. And I don't know if you're still digging up the weighted average LTVs or not. But the other question I had...

There was nothing unusual -- I would say this. There's nothing unusual about those LTVs. So these were not like an 80% LTV pool or anything like that, and that was not the driving factor that we were using to liquidate. The driving factor was our visibility into the liquidity and cash flow to continue to pay these loans. That was the primary determiner.

Yes. No, I'm actually just kind of to your point about the first loss maybe usually being the best. I'm just kind of trying to gauge where the market might be on original value if we have banks continuing to need to sell assets.

I will tell you it's -- if you have a very small change in vacancy, so if you take like a multifamily building that's got -- Manhattan is historically like a 2% vacancy. Now it's just under 6%. The latest figures I saw were an 11% decrease in rent. So I did some math yesterday. If you -- if your vacancy goes from like 2% to 4%, your rents drop by about 10%, the actual rent on the unit, and your cap rate goes up by 1.5%. You could see in an average building of 30% to 40% decrease in the appraised value of that building because your NOI drops. And then as your NOI drops and vacancy rates go up, you would expect cap rates to go up on a slightly riskier asset.

Right. Yes. And then on the PPP loans, you may have given this, but do you have the premium that you'll make on that sale?

So the net would be -- $5.3 million is the net.

And does that include the fees that would need to be accelerated?

Correct. Yes. So that's the net. We actually sell them at a little discount. And then we take hold, and we accelerate the fees related to those loans. And that was -- it was only about half of the portfolio. We retained the other half. And frankly, those customers and the loans we retained were the more strategic customers for the bank. They were also the customers who were being very prompt about providing us the information we needed to file for forgiveness. So if they were helping us, it was easy to move them through. If we thought those were going to be long-term difficult loans to get through forgiveness, we decided to part ways.

Okay. And none of that was booked in the second quarter, correct?

That's correct. That will be booked in this fourth quarter.

So what was the net interest income contribution from the PPP loans in the second quarter?

I'd say we're at -- they were earning at 2.25% in the second quarter -- third quarter, I mean. Third quarter, 2.25%.

In the third quarter.

[Operator Instructions] The next question comes from Collyn Gilbert of KBW.

This is great color and coverage that you've offered. Just one -- a couple of things, too. First, on the PPP front. Just to make sure, is that gain going to be recognized through NII or fees on the sale?

I actually haven't thought about that yet. Mike, do you have the answer for that?

Yes, it's not fees. It's gain on sale of loan. It's not NII.

Okay. Okay. Got it. And then I know, Chris, you had indicated who the buyers were of the nonperforming loans that you're moving. But what about the buyer for the PPP slug? Who's the...

Yes, we found -- I won't give you a specific name, but we found -- there was just actually a robust market for that. So there's a little cottage industry going around of people buying PPP loans. And look, part of their calculation, I believe, is that they're pinning hopes that Congress will do a mass forgiveness, and that they will then not have to do any work and get the forgiveness, and that may happen. And if that happens, then we would've realized a little more had we held them. But the company that we sold to had bought from at least -- it was probably close -- between half a dozen and 10 banks prior to us. And there were few -- then there were a couple of bidders. So there are a couple of people putting together these pools of PPP loans in aggregate.

Okay. That's helpful. And then just on the comment that you guys made in the slide deck that you were assuming that a lot of the consumer deferrals are going to drop to nearly 0 by the end of the year. Just curious, what gives you that sense of comfort? I mean I get it on the commercial side. As you've said, you're in touch with the borrowers, you're having discussions, you know the businesses. But what gives you confidence and insight into these consumer forbearance loans that they'll become current?

You're right. It's a little more difficult to assess on the consumer side. We look at a few factors there. Probably the most important thing is we do have conversations with them as well. So we do have some information from our communication. I think in the slides, we indicated that $13.4 million of the customers who've come up to date have requested additional accommodation. Now it doesn't mean we're going to give it. We're kind of looking through those on a case-by-case basis. We've required they submit an application with more current information about their liquidity and all of that. So we had $13 million worth of requests. It may increase a little bit. We don't think it's going to be an unmanageable number. The other thing to point to, to Grace's comments, we expect some of that to appear in the fourth quarter. So in our reserve calculations for the third quarter, we anticipated that, and the provision helps cover that. So in the fourth quarter, if we have a pool, at this point, our best guess would be it might be $10 million, might be $20 million of loans that are higher risk on the consumer side. We have an adequate reserve to be able to sell those, too. So we've kind of covered that risk through the provision in Q3. And the underlying issue there is the #1 default characteristic around a residential loan is not actually income or FICO. The #1 default characteristic is LTV. And our LTVs were fine. But if you look at our lending area, and that's why we included the slides on the median home price values, most our loans are heavily concentrated in the shore communities that have had sharply increasing values, and those values continue to increase. So early in the summer, we saw some anomalies. So Cape May county was up. The median home price in Cape May county June to June from 2019 to 2020 went up 37%. So that's wonderful, but just stop and say is that sustainable, is it going to continue to happen. So we watched July and August and September. The 4 counties, and we include this in the supplemental, that constitute most of our loans and most of our forbearance, they all have double-digit median home price increases over the past 12 months. So it doesn't seem to be just a quick surge of a couple of panic buyers in April. It seems to be an enduring shift. And at the end of the day, if people can make money selling their home, they're not going to let it slide into foreclosure. So we are -- we're weighting that as well.

Okay. That's great. And then just to make sure, so tying what Grace's comments on selling -- potentially selling some resi mortgage or higher-risk consumer loans in the fourth quarter, is that then your -- when you just said, Chris, the $10 million to $20 million, that's all that you expect to do in that.

Correct.

Okay. And you think at this point, those credits are sufficiently reserved, so no additional provisioning would be needed.

That's correct.

This concludes our question-and-answer session. I would like to turn the conference back over to Christopher Maher for any closing remarks.

Right. Well, I thank everybody for taking the time to join us today. Enjoy the holidays. I hope everyone will stay safe, and we look forward to talking to you with our results in January. Take care.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.