Community Financial System, Inc. Q1 FY2020 Earnings Call

Welcome to Community Bank System First Quarter 2020 Earnings Conference Call. Please note that this presentation contains forward-looking statements within the provisions of the Private Securities Litigation Reform Act of 1995 that are based on current expectations, estimates and projections about the industry, markets and economic environment in which the company operates. Such statements involve risks and uncertainties that could cause actual results to differ materially from the results discussed in these statements. These risks are detailed in the company's Annual Report and Form 10-K filed with the Securities and Exchange Commission. Today's call presenters are Mark Tryniski, President and Chief Executive Officer; and Joseph Sutaris, Executive Vice President and Chief Financial Officer. Gentlemen, you may begin.

Thank you, Brent. Good morning everyone, and thank you all for joining our first quarter conference call. In addition to Joe and I, we also have Scott Kingsley and Joe Serbun joining us on the call this morning. This is clearly a different and challenging environment for all of us. Let me start with our COVID response. Early in March, we mobilized a pandemic response team consisting of senior leadership across the company. Our initial focus was and continues to be the health, safety, and wellbeing of our employees, customers, and communities. We began execution immediately by limiting travel and in-person meetings, instituting sick time policies, and enhancing cleaning protocols in all company facilities. Over the following week, we closed branch lobbies, instituted work-from-home, separated our critical operating functions into multiple teams in different facilities, utilized shift work, and instituted social distancing in all office spaces. We've been operating in this environment for over a month, and although less efficient, our operational cadence is quite stable. We're prepared to continue as we are until such time it is safe for our people to resume normalized operations. At this juncture in New York and Vermont, the governors' stay-at-home orders extend through May 15th, and in Pennsylvania that date was April 30th. We will remain abreast of circumstances and react accordingly in the best interest of our employees, customers, and communities. Also in March, we announced that Scott Kingsley, our Chief Operating Officer, would be retiring from Community Bank System, effective June 30, 2020. Scott served in that role for a year and a half after having served as Chief Financial Officer since 2004. Scott worked side by side with me for nearly 16 years, and the company will miss his energy, passion, and talent. At that same time, we further announced that Joe Serbun was appointed to the role of Chief Banking Officer. Joe began his career with Community Bank in 2008 and previously served as Executive Vice President and Chief Credit Officer. Joe will continue to bring exceptional leadership to our company for the benefit of all our stakeholders. I will make only a few comments on Q1. It was pretty good. Absent the COVID-related allowance build, earnings were very strong. I think a couple of pennies better than last year’s quarter. The total loan book was down much less than seasonally expected, and we had organic growth in commercial. Deposit performance was also good in the quarter. Our non-banking businesses had a good quarter. The pre-tax earnings of our wealth management business were up 12%, benefits were up 9%, and insurance was up 2%. With respect to the acquisition of Steuben Trust Company that we announced in October, it is proceeding, albeit at a somewhat slower pace. We continue to be hopeful that we can close in the second quarter, but have not yet received regulatory approval and there remains uncertainty around the ultimate impact of COVID-19. Joe will provide a little more color on the quarter, but there are obviously more significant matters to discuss. One of those I wanted to touch on was financial strength. In severe circumstances of economic and financial distress, there is no substitute for earnings, liquidity, capital, asset quality, core deposits, and revenue diversification. When I look at the fundamental financial strength of our company, I'm highly confident we are as well prepared for whatever the future may bring as we possibly can be. Joe?

Thank you, Mark. Good morning everyone. As Mark noted, the earnings results for the first quarter of 2020 were solid in spite of challenges related to the COVID-19 crisis. The company reported $0.76 in fully diluted GAAP earnings per share for the first quarter, excluding $0.01 per share for acquisition-related expenses. Fully diluted operating earnings per share were $0.77 for the quarter. These results were $0.04 per share lower than the first quarter of 2019 due largely to the COVID-19 crisis and its related impact on the company’s operations. In particular, the company reported $5.6 million in its provision for credit losses in the first quarter of 2020, reflective of expected credit losses due to rapidly deteriorating economic conditions. This amount exceeded the company's net charge-offs in the quarter by $1.6 million, or 9 basis points annualized, by $4 million. By comparison, in the first quarter of 2019, the company reported $2.4 million in the provision for loan losses and net charge-offs of $2.6 million, or 17 basis points annualized. In addition, the company's net interest income was negatively impacted in the quarter due to the significant and rapid decrease in short-term interest rates. During the first quarter of 2020, the company adopted the new CECL accounting standard. This resulted in a $1.4 million, or 2.7%, increase in the allowance for credit losses from $49.9 million prior to adoption to $51.3 million after adoption. Due largely to the expectation of increased credit losses from the COVID-19 adverse impact on economic and business operating conditions, the company's allowance for credit losses increased an additional $4.4 million, or 8.6%, at the end of the first quarter to $55.7 million. The total increase in the allowance for credit losses was $5.8 million, or 11.5%, between December 31, 2019 and March 31, 2020. The allowance for credit losses to total loans outstanding at March 31, 2020 was 0.81%, which represented over 8 times the company's trailing 12 months net charge-offs. As Mark noted, we believe the company's capital reserves, liquidity profile, diversified revenue streams, strong credit record, and experienced management team leave us well prepared to endure the impacts of the COVID-19 crisis. The company's net tangible equity to net tangible assets ratio was 10.8% at March 31, 2020. This was up from 10% at the end of 2019 and 9.8% from one year earlier. Similarly, the Tier 1 leverage ratio was 11.1% at the end of the first quarter, which is over two times the well-capitalized regulatory standard of 5%. During the first quarter of 2020, shareholders’ equity increased $121.4 million, or 6.5%. This included a $19.3 million increase in retained earnings and a $93.7 million increase in accumulated other comprehensive income, net of tax, due primarily to an increase in the value of the company’s available-for-sale investment securities. At December 31, 2019, the company’s Tier 1 risk-based capital ratio, total risk-based capital, and common equity Tier 1 capital ratios were 17.2%, 18.0%, and 16.1%, respectively, reflective of the company's lower risk asset base and high levels of regulatory capital. The company has an abundance of liquidity resources and is well-positioned to fund future balance sheet growth, including its current loan pipeline, potential advances on undrawn lines of credit, and pending Paycheck Protection Program (PPP) loans. The company's funding base is largely comprised of low-cost core deposits. At March 31, 2020, checking and savings accounts represented 68.5% of the company's total deposit base. The company's cash and cash equivalents net of float and reserves totaled $458.9 million at March 31. Total borrowing availability at the Federal Reserve Bank was $260.4 million and total borrowing capacity at the Federal Home Loan Bank was $1.83 billion. The available-for-sale investment securities portfolio was valued at $3.14 billion, $1.5 billion of which was available for pledging if needed. In total, these sources of immediate liquidity exceeded $4 billion. We closed the first quarter of 2020 with total assets of $11.81 billion. This was up $398.7 million, or 3.5%, from the end of the linked fourth quarter of 2019, and up $892.5 million, or 8.2%, from one year earlier. The increase in the quarter was largely due to a net inflow of deposits as seasonally anticipated; year-over-year increases in total assets were driven by both the third quarter 2019 acquisition of Kinderhook and organic growth. Average earning assets for the first quarter of 2020 were $10.04 billion, which is consistent with the fourth quarter of 2019 but up $664.1 million, or 7.1%, from one year prior due to both the Kinderhook transaction and organic growth. Average loan balances in the first quarter of 2020 were up $18.8 million, or 0.3%, when compared to the linked fourth quarter of 2019, and down $603 million, or 9.6%, when compared to the first quarter of 2019. On a linked quarter basis, the average outstanding balances in business lending, consumer mortgage and consumer indirect portfolios were up slightly or were offset in part by decreases in the consumer direct and home equity portfolios. The increase in average loans outstanding on an annual basis was driven by the Kinderhook acquisition, as well as organic loan growth. Ending total loans were down on a linked quarter comparative basis $24.5 million, or 0.4% as seasonally anticipated. Exclusive of loans acquired in the Kinderhook transaction, ending total loans outstanding increased by $174.1 million, or 2.8%, from a year prior. At March 31, 2020, the carried value of the company's investment securities portfolio was $3.19 billion. This includes net unrealized gains of $155.5 million, up from $33.1 million in net unrealized gains at the end of 2019 and $7.9 million in net unrealized gains a year earlier. The effective duration of the company's investment securities portfolio was 3.6 years at March 31, 2020. Average total deposits were up $651.5 million, or 7.8%, from the same quarter last year, but down $45 million, or 0.5%, on a linked quarter basis. The increase year-over-year was driven by the acquisition of $560.1 million of deposit liabilities in the third quarter due to the Kinderhook transaction. The company's average cost of deposits was 25 basis points in the first quarter of 2020, 1 basis point lower than the linked fourth quarter of 2019. The company reported total revenues of $148.7 million in the first quarter of 2020, an increase of $6.1 million, or 4.3%, over the prior year's first quarter. Net interest income increased $3.2 million, or 3.7%, to $90.1 million due to a $664.1 million, or 7.1%, increase in average earning assets between periods, offset in part by a 15 basis point decrease in net interest margin. Non-interest revenues increased $2.9 million, or 5.3%, between comparable quarters due to increases in banking and non-banking revenues, offset in part by a small loss on equity securities. Interest income and fees on loans increased $4.9 million, or 6.6%, over the comparable prior year quarter due to an increase in average total loans outstanding, offset by a 17 basis point decrease in average loan yield. As previously reported, the first quarter 2019 average loan yield was favorably impacted by 6 basis points due to $1 million in one-time loan fees. Interest income on investments decreased $0.5 million, or 2.9%, between the first quarter of 2019 and the first quarter of 2020. The tax-equivalent average yield on investments, including cash equivalents, decreased from 2.58% in the first quarter of 2019 to 2.45% in the first quarter of 2020, reflective of lower interest rates. Interest expense was $1.1 million higher than the previous year's first quarter, driven by a 4 basis point increase in the cost on interest-bearing liabilities and a $446.1 million, or 6.9%, increase in average interest-bearing liability balances. Employee benefit services revenues for the first quarter of 2020 were $25.4 million. This represents a $1.3 million, or 5.5%, increase over the first quarter 2019 revenues. The improvement in revenues was driven by increases in plan administrative and actuarial recordkeeping fees, as well as increases in employee benefit trust transfer agent fees. The company reported $8.1 million in insurance services revenues during the first quarter of 2020, a $0.2 million, or 2.5%, increase over the first quarter of 2019 results due primarily to an increase in group medical and property and casualty insurance revenues. Wealth management revenues for the first quarter of 2020 were $7.1 million, or $0.8 million, or 12.4% higher than the first quarter of 2019 due to both acquired and organic growth. Banking noninterest revenues increased $0.7 million due primarily to an increase in mortgage banking revenues. During the first quarter of 2020, the company increased its commitment to sell secondary market eligible residential mortgage loans, which drove an increase in mortgage banking revenues from $0.2 million in the first quarter of 2019 to $0.9 million in the first quarter of 2020. The company recorded $93.7 million in total operating expenses in the first quarter of 2020. This represents a $5 million, or 5.7%, increase in operating expenses over the first quarter of 2019. Salaries and employee benefits expense increased $4.9 million between comparable quarters and is reflective of the increased payroll costs associated with the third quarter of 2019 Kinderhook acquisition, merit-related pay increases, and an increase in employee benefits including significantly higher medical costs. Total data processing and communication expenses increased $1 million, or 10.8%, between comparable annual quarters, driven by higher payment processing and telecommunication costs. Occupancy and equipment expenses increased $0.5 million, or 4.4%, between the periods due largely to increased costs associated with the Kinderhook acquisition. These increases were partially offset by a $0.5 million, or 11.2%, decrease in intangible asset amortization, as well as a $0.7 million, or 6.4%, decrease in other expenses. The effective tax rate for the first quarter of 2020 was 18.8%, up from 18.5% in the first quarter of 2019. The company recorded lower amounts of stock-based compensation tax benefits in the first quarter of 2020 as compared to the first quarter of 2019. Exclusive of stock-based compensation tax benefits, the company's effective tax rate was 20.9% in the first quarter of 2020. From a credit risk and lending perspective, the company is taking actions to identify and assess its COVID-19 related credit exposures based on asset class and borrower type. No specific COVID-19 related credit impairments were identified within the company's investment securities portfolio during the first quarter of 2020. With respect to the company's lending activities, the company implemented a customer payment deferral program to assist both consumers and business borrowers that may be experiencing financial hardship due to COVID-19 related challenges. Through April 15, 2020, the company granted payment deferral requests for up to three months for 3,274 consumer borrowers and 1,018 business borrowers, representing $587.2 million of the company's loan balances. The company anticipates it will continue to receive COVID-19 financial hardship payment deferral requests throughout the second quarter of 2020. At March 31, 2020, nonperforming loans increased to 0.46% of total loans. This compares to 0.39% of total loans outstanding at the end of the first quarter of 2019 and 0.35% at the end of the linked fourth quarter of 2019. The increase in nonperforming loans is largely attributable to a single line of credit that matured on December 31, 2019 with a total outstanding balance of $9.9 million. Total delinquent loans, which includes nonperforming loans and loans 30 or more days delinquent, to total loans outstanding was 1.11% at the end of the first quarter of 2020. This compares to 0.88% at the end of the first quarter of 2019, and 0.94% at the end of the linked fourth quarter of 2019. The delinquency status for loans on payment deferral due to COVID-19 financial hardship was reported at March 31, 2020 based on the delinquency status at March 20, 2020. The Steuben acquisition is scheduled to close later this quarter. However, due to the COVID-19 crisis and pending regulatory approval, the ultimate closing date may need to be adjusted. As a reminder, Steuben Trust is a 14-branch franchise operating in a six-county region of Western New York with total assets of approximately $560 million. Community Bank currently serves four of the counties within Steuben’s current footprint and the other two are contiguous to our markets. The company expects this acquisition to be approximately $0.08 to $0.09 per share accretive to its first full year of GAAP earnings and $0.09 to $0.10 per share accretive to cash earnings, excluding one-time transaction costs. Operationally, we will continue to adapt to the changing market conditions. In the immediate near-term, the company will remain focused on assuring the timely intermediation of deposit responses, assisting borrowers that experience financial hardship with payment relief, closing and funding PPP loans and other loans, and maintaining service standards in our financial services businesses. Based on the current market conditions, we believe certain aspects of the company's operations will be more adversely affected in the second quarter of 2020 than they were in the first quarter of 2020. Auto sales have slowed considerably, which has reduced the demand for new consumer automobile financing and consumer installment loans. Although business lending activity increased in April due largely to the PPP program, the effective yield on the PPP loans is significantly lower than the company's first quarter average earning asset yields, which may negatively impact the company's net interest margin in future periods. Deposit and other banking fees, including part-related interchange revenues, are expected to decrease due to declining levels of commerce. Higher levels of unemployment will affect our borrowers’ ability to service debt, which may increase the level of expected loan losses, eventually resulting in the company reporting significant provisions for credit losses in future periods. The company's wealth revenues will likely decrease in the second quarter, consistent with a decrease in investment asset values. Insurance services may be negatively impacted by lower sales activities. And although the company's employee benefit service business is largely driven by participant headcount levels, employee benefit trust operations will likely be negatively impacted by a decrease in underlying plan valuations. The company's dividend capacity remained strong. Its full year 2019 dividend payout ratio was 47.5%. Accordingly, the company expects to continue to pay a quarterly dividend consistent with past practice. Undoubtedly the COVID-19 crisis has changed the near-term outlook for society in general, as well as expectations around economic conditions. First and foremost, we remain hopeful that effective treatment options are on the near-term horizon and that a vaccine becomes widely available later in 2020. The specific acuity of this crisis on our employees and their families, our customers, communities and shareholders is highly uncertain. With that said, we intend to support our stakeholders in a thoughtful, disciplined and passionate manner and believe the company is well prepared to endure its impacts. Thank you. Now, I will turn it back to Mark.

Thank you, Joe. Before we open it up for questions, as I said in the opening, I have Scott Kingsley joining us to say a few final words to the group that he has had the opportunity to work so closely with over the past 16 years. Scott?

Thank you, Mark. To our investors and analysts, I want to take this opportunity to say thank you for your continued confidence and support of the company and myself over the past 16 years. Your engagement and effective challenge have been critical in our strategic decision making and supportive of our continuous improvement objectives. As I prepare for my retirement from the company, I am both humbled by and proud of what our team has been able to accomplish, and pleased to have been a small part of that. I'm also supremely confident in our team's ability to continue to provide a differentiated level of customer and community service and superior shareholder returns. Again, thank you.

Thank you, Scott. And thank you for 16 years of tremendous service at Community Bank System. On behalf of our entire organization, we wish you and your family the very best for the future.

Thank you, Mark.

With that Brent, I would now ask you to turn the line over for questions.

We will now begin the question-and-answer session. Our first question comes from Joe Fenech with Hovde Group. Please go ahead.

We’ve seen some other companies report just large provisions here in the first quarter to build reserves. Some of them cited unemployment data and projections they used, I guess, from as late as April 12th. I know you guys generally are very conservative on credit, so I wouldn't necessarily expect the same reserve build. But just curious what metrics do you use specifically to determine the reserve build from the first quarter? And given your comment in the release and on the call here about the increase in payment deferral requests you expect in the second quarter, whether that means the big reserve build might be pushed out a quarter or two for you all, or if you feel like you mostly accounted for that with this 1Q provision allocation?

As you know, we adopted the CECL model in the first quarter. The model is composed of three significant components: the quantitative component, the non-economic qualitative adjustments, and the economic qualitative adjustments. In the economic qualitative adjustments, we used forecasts provided by Moody's. The most recent update we used was through March 27. So effectively that was the latest information we had from Moody's. We used that in our CECL model. The best way to describe the economic forecast is a Nike Swoosh shape — a significant drop-off in the second quarter with some recovery in the later quarters. That component of the model largely drove the adjustment in the first quarter. With respect to the second quarter, we'll re-evaluate the model, which will include the economic qualitative factor adjustments, but also we'll evaluate observed data with respect to delinquency and migration of risk ratings and those types of factors. So it is possible that in the second quarter there will be additional reserve build based on the factors we're observing in addition to changes in the economic outlook.

And then with respect to the decline in oil prices, you all in the 2015-2016 cycle, Mark, Scott I guess, I remember having some tangential exposure to lower oil prices. There wound up being really no discernible impact to you as far as I can tell. So maybe just update us on what you consider your exposure to be to these record low oil prices.

Yes, I think we have, Joe, pretty limited, at this juncture, exposure, mostly to the fracking industry in Northeast Pennsylvania. I'll let Joe Serbun comment further on that.

Yes, Joe, our exposure, as Mark said, is not only to fracking but we also have some pipeline contractors in the portfolio. Our overall exposure is probably somewhere in the $40 million range, and we monitor it quarterly. It's much smaller than it was back in 2016 and 2017. The clients in the portfolio by and large are very strong and very liquid. They are currently considered essential and are working, which is good. At the moment, we don't have any concerns with respect to that portfolio.

And then on the outlook for M&A, you all were already an acquirer of choice in the market. I'd have to think the resiliency of the stock here through this makes you even more of an acquirer of choice. If one of your preferred targets came to you during this period, would you feel confident enough to move forward and announce something during this time or is there just too much uncertainty at this point?

I haven't really given much thought to that hypothetical question. My initial answer would be that we would take a look at it. If you look at the financial strength of the company as I and Joe have discussed, I think we're pretty well positioned for whatever the future may bring in terms of economic distress and potential credit losses. Clearly, the valuation delta could give us an opportunity. With that said, I would think most who are thinking about being sellers are not going to be thinking about it for some time. So, I don't expect a whole lot of activity, but if we have an opportunity to have a dialogue with a high-value partner right now, we certainly would entertain that.

And on the Steuben pending deal, it just sounds like, correct me if I'm wrong, that it's more of a process issue given everything the regulators and others have on their plate right now?

Yes, everything is moving slower. The people at Steuben and our teams are working remotely, and that creates challenges when we're trying to work through integration and approvals. The regulators are also working remotely, so everything is slower. When we announced the transaction, we said we expected to close in the second quarter. As of right now, we still plan to do that, but the future direction of COVID and the environment will dictate the timing.

And then last one for me. Mark, your fee businesses and the contribution there give you diversity of the earnings stream that not many of your peers have, but the closures of the businesses really didn't occur until the middle of the month last month of the quarter, right in March. Can you walk us through your initial thoughts on how you expect your various fee businesses to fare with what's going on?

Sure. I'll start with wealth management. I think that business will probably be down in the second quarter. We estimate somewhere in the 15% range in revenues for the second quarter. The earnings probably won't be down quite that much because there will be some offsets in terms of commissions and the like, but you're probably talking a double-digit decline in the wealth business. I think the investments and insurance businesses will be slightly less impacted and we would expect a single-digit reduction in revenues and earnings in those businesses for the second quarter.

Our next question comes from Alex Twerdahl with Piper Sandler. Please go ahead.

First off, Scott, congratulations on a great career and good luck with your retirement — you'll certainly be missed. I wanted to start with the PPP program. In terms of the fees that are expected from the roughly $350 million of loans you originated, are those fees going to be recognized pretty much over the life of the loans or reflected in the second quarter? How should we think about how those balances will impact the balance sheet?

We intend to hold those PPP loans for investment and therefore recognize the fee over the life of the loan. Regarding any additional funding in the second round, we will continue to evaluate our strategy, but at this time we're planning to hold them for investment.

Is there a way to estimate the weighted-average fee? Should we assume these are tiered at 1%, 3%, and 5% levels, or how should we think about the fee income recognition?

It will come through margin as we account for it, and we intend to amortize the fees over the loan life. With respect to the weighted-average fee, we have not finalized that determination. There were fee tiers of 1%, 3%, and 5% for different size loans, so the weighted average will be somewhere between those amounts, but we don't have it pinned down yet. We will shortly.

And then in terms of margin, can you help us get a little better sense for the moving parts? Given the PPP loans, expected durations, and other balance sheet dynamics, how should we be thinking about margin in the second quarter?

On the loan portfolio side, we saw a decrease in short-term rates late in the first quarter and the full impact of that decrease was not necessarily reflected in the first quarter results. We'll feel more impact on the loan yield side in the second quarter. PPP loans will be at a lower effective yield, so overall loan yields are expected to decrease. On the funding side, our total cost of deposits was around 10 to 11 basis points back in late 2015; we're sitting at 25 basis points now. I could see the cost of deposits drifting down a few basis points in the second quarter. The investment securities yield I would expect to be maintained in the short term around the 240-245 basis point tax-equivalent level. Cash equivalents on the balance sheet will earn less, which will pressure yields. Considering all these factors, it's fair to assume a decrease in net interest margin of roughly 5 to 7 basis points in the coming quarter.

The only thing I would add is that while loan and deposit yields are reasonably straightforward to estimate, there is a potentially material variable related to PPP fee recognition. We amortize fees over the life of loans we hold. As PPP loans start getting forgiven, we will accelerate recognition of the associated fees tied to those forgiven loans. That acceleration could significantly affect reported margin in the near term, so that's an important variable to keep in mind.

And then just final for me, can you run through the hot-button loan segments and quantify exposure to lodging, restaurant, retail, CRE, etc.?

Sure. Retail trade represents about 4% of our total loans, roughly $260 million. Lodging represents about 3% of our exposure. Health care and social assistance represent about 2% of exposure. Construction is about 2%. Dairy farming represents about 1%. Food service is about 1%. Furniture stores — we have a couple of larger furniture store relationships — represent about 1%. Manufacturing is around 2%. Casino exposure is less than 1%. Transportation is also less than 1% of total loan outstandings. We continue to monitor these sectors closely.

Our next question comes from Erik Zwick with Boenning and Scattergood. Please go ahead.

I guess a couple of follow-ups on the PPP program. You noted that you approved just under 1,400 loans. How does that compare to the number of applications you received? And since the program ran out of funding last week, have you continued to accept additional applications? If the second round of funding comes through, will you be able to process and fund those additional applications?

It's been hectic and fast-paced. We do expect additional congressional appropriations for the program — hopefully upwards of $2.5 billion — and we have continued to process and validate applications and loan requests. We have continued to take in submissions and are working through them with the expectation that additional funding will be available. We have thrown a lot of people at it. We probably have roughly 1,200 applications that we were still working on that did not get processed in the first go-round. The expectation is we'll work through those applications if additional funding becomes available, and we'll take care of as many clients as we possibly can. It's been a team effort.

Switching to current credit — the 90 days plus delinquent and still accruing bucket increased from $5.4 million to $12.6 million. You mentioned delinquencies are reflected as of March 20th. Does that assume there was a pre-COVID increase in that bucket? If so, what was the driver?

Yes. We have one credit line of credit that matured just shy of $10 million as part of a larger relationship with a total relationship around $13 million to $14 million. We have two loans that are current and one line of credit that matured, and we're working with the client to come up with a solution. This was a pre-COVID situation and not a result of COVID-19. The business is in the retail sector, which is expected to be impacted post-COVID, and we're working to find a resolution.

Are you able to talk about what particular retail industry or what they serve?

No, just that it's in the retail sector.

Our next question comes from Russell Gunther from D. A. Davidson. Please go ahead.

I just wanted to follow up on the exposures you provided. Is that your effort to ring-fence or quantify the exposures that are most at risk in the near-term from COVID-19? Also, is there any material shift with Steuben or any of their exposures worth calling out?

Yes, those are the sectors we view as most likely to be impacted by COVID-19, expressed as a percentage of total loan outstandings. We will continue to evaluate and adjust that list as the situation unfolds. With respect to the Steuben portfolio, they have some exposure to certain industries similar to ours. At the time we underwrote the transaction, we estimated the impact and it would add a little bit of healthcare exposure, but it's not a very large portfolio and nothing that would move their needle significantly on a holistic basis.

Are there any portfolio characteristics you can share, such as weighted-average LTV or debt service coverage ratios, to help contextualize these exposures?

That's a bit of a challenge because many of our loans are seasoned and appraisals or collateral valuations are dated. LTVs reported today may not be meaningful. What I can share are our typical underwriting targets: loan-to-value targets are generally between 70% and 75%. We can go beyond that, but our target is around 70% to 75%. We typically underwrite with 10- or 20-year amortization assumptions where applicable, and we typically underwrite to debt service coverage ratios in the 1.1x to 1.2x range. We prefer personal guarantees and recourse where appropriate. We do have some loans above supervisory limits of 85%, but they are not a significant bucket.

Have you contemplated stress testing — similar to DFAST or a severely adverse scenario — to estimate potential loss rates from COVID-19?

We are below the size thresholds for DFAST testing. However, we did run internal capital stress testing when we participated in similar processes in the past. We maintain an internal stress testing routine; we completed one in December and include severely adverse scenarios that apply multiple factors to charge-off levels and other assumptions. In those stress tests, we were able to maintain regulatory capital levels well above required standards and our internal targets.

I appreciate that. Stepping back, could you give any sense of potential loss magnitude or order-of-magnitude thoughts on what COVID-19 losses could be?

We've spent a lot of time considering that. The real challenge is the breadth of impact — it's not localized to one sector like lodging. We have roughly $3 billion in business lending, $2.5 billion in residential mortgage, and about $1 billion in auto lending. If unemployment rose dramatically — say into double digits — the potential stress across these portfolios could be significant. If we had reason to believe losses would be substantially higher than the level embedded in our current allowance, we would provide additional coverage. At this stage it's extremely uncertain and difficult to determine an order of magnitude. We expect to get better clarity as the second quarter unfolds and as the economy begins to return toward normalization, at which point we will be better informed to make judgments.

Our next question comes from Collyn Gilbert with KBW. Please go ahead.

Mark, thanks for the color on unknowns. I want to dig in on the movement within the reserve this quarter. Joe, I think you said $1.4 million of the increase in the reserve from the fourth quarter was related to CECL. I thought previously you had said that increase might be closer to $5 million. Is that correct? Why was the CECL component lower than earlier estimates?

Collyn, we provided a range of estimates in prior discussions. Initially, we had a higher expectation, but as we refined our model through the fourth quarter and right before adoption, we lowered the estimate and came up with a post-adoption number that was only $1.4 million higher than the 12/31 allowance. Essentially, the incurred-loss model would have produced an allowance of $49.9 million, and the CECL model produced $51.3 million on adoption. The net increase due to conversion was $1.4 million. As the COVID crisis unfolded, we tested our model and relied primarily on the economic qualitative factor adjustments to reflect forward-looking expectations. Observed data such as delinquency and risk rating migrations had not developed meaningfully by quarter-end. We do expect to see more observable changes in the second quarter and will evaluate further reserve needs at that time.

On the payment deferrals — the $587 million you referenced — are you setting aside a specific reserve allocation for those deferrals with the expectation that they could become problem credits, or how are you thinking about reserves on both the $587 million of deferrals and the $1.5 billion of potential exposure you called out in the deck?

That's a fair question. We've discussed internally whether to make a separate reserve adjustment specifically for deferred payments. At this point we believe that would be speculative because we do not yet have clarity on how long deferrals will last or how quickly the economy will reopen. We did not make a separate adjustment in the first quarter specifically for deferrals. Remember that the deferral numbers were reported through April 15th, so the numbers changed after March 31st. If we experience a second round of deferrals or observe clear deterioration in observed data, we would evaluate and likely make additional reserve adjustments in the second quarter.

Was there any accelerated line draw activity during the quarter or post-quarter within these credits?

Interestingly, utilization of lines has been quite stable over the last nine quarters. The high point was about 53% to 54% and the low about 47% to 48%; we're sitting around 49% utilization now. So there's really not a lot of change. We did have one client with a sizable line who elected to draw down on it — a highly liquid firm that didn't necessarily need to draw on it but did — and that was an outlier. Overall, not much change.

On the expense side, you had previously provided a baseline for quarterly operating expenses in the $93 million to $94 million range. Is there anything COVID-related that will materially change your expense outlook from that baseline?

There will be some additional COVID-related expenses — for example, enhanced cleaning activities — but we are seeing reductions in travel and other items. Net-net, the COVID impact is not expected to materially change the trajectory of our operating expenses.

It looks like other operating expenses dropped a fair bit this quarter. Was there anything unusual that caused that, and what's the outlook for that line going forward?

I didn't see anything unusual in the quarter. There were some prior quarters where we had one-time items, and this quarter appears more normalized. I can follow up if you'd like more detailed line-item analysis.

One final quick question on PPP: what percent of the applicants were existing Community Bank customers versus non-customers?

The vast majority were existing customers.

Were you getting requests from non-customers as well?

We had a few limited requests from non-customers.

Our next question comes from Matthew Breese of Piper Jaffray. Please go ahead.

Curious on the $587 million of loans that were granted deferral: what was the asset class breakdown of those loans? And was there any overlap between that bucket and the PPP bucket?

A couple of points. Both the deferrals and the PPP loans were concentrated geographically — a majority came from New York State. About two-thirds of both the deferrals and PPP dollars were in New York versus Pennsylvania. In terms of sectors, retail trade, lodging, manufacturing, construction, and healthcare represented about 75% of the PPP activity. On the deferral side, much of the early consumer deferrals — mortgages and HELOCs — were 30-day deferrals for both principal and interest. As people watched the news, many asked for longer relief, and we extended some to 90 days in total. In the commercial world, early on we achieved principal-only deferrals with interest paid; later many borrowers requested both principal and interest deferrals, which we accommodated. We coordinated our approach with regulators, and they are aware of the approach we took with deferrals and durations. These deferrals did not require reclassification to troubled debt restructurings or special risk ratings at the time of the deferral because the actions were consistent with regulatory guidance.

On the counts — the 3,274 consumer borrowers and 1,018 business borrowers that received deferrals — what percentage of your total consumer accounts and total business accounts do those represent?

We have about 40,000 installment loans. I don't have a precise count on total residential mortgage accounts at hand, but from an account perspective we might have roughly 6,000 small business customers. I don't have the exact percentage breakdown by account type in front of me.

To give some directional context: deferrals represent roughly 4% of the outstanding balance of our consumer mortgage portfolio. On installment loans, it's probably around 2.7% of outstanding balances. So on a percentage-of-portfolio basis we're in the low single digits for these categories.

To add some granularity, our average installment loan — particularly indirect auto — has an average balance under $20,000. Our average residential mortgage outstanding is roughly $100,000. So the number of consumer accounts that requested deferrals is proportionally larger relative to the smaller-balance installment loans, which likely explains why consumer deferral requests were greater in volume.

Do you view your more rural footprint as an advantage or disadvantage in this environment?

Historically our footprint in non-metropolitan markets has been advantageous for our business model. From a COVID-19 perspective, the greatest health and economic impacts initially concentrated in urban areas. Our markets are generally less urban, and that could mean a faster reopening and potentially less economic impact, which may work to our advantage. Also, asset valuation appreciation in our markets has been more modest historically, so collateral values haven't experienced the same level of run-up seen in some metro markets. That may help stabilize LTVs relative to origination metrics.

Our next question comes from William Wallace with Raymond James. Please go ahead.

Scott, I'll echo the congratulations and wish you well. On PPP forgiveness: is there any reason to expect that the very large majority of these loans wouldn't be forgiven over the next 60 days?

No, we have no reason to believe they won't be forgiven if the borrowers met the program requirements. We expect that many will be forgiven, which is why I highlighted that forgiveness activity could materially accelerate recognition of associated fees. The main unknown is the SBA's capacity to process the volume of forgiveness applications in a timely fashion. That could create delays in the recognition and administrative processing of forgiveness requests.

You mentioned roughly 1,200 applications in process that were not closed in the first round. Can you give the dollar amount of those and what percentage of processed applications were approved?

I don't have the exact dollar amount off the top of my head, but the average loan size for those we did approve was about $70,000. You can use that to estimate the remaining exposure. In terms of approvals, the vast majority of applications we processed were approved — we had very few denials.

Last question: how should we think about loan growth, payoffs versus refinancings, especially since non-bank lending channels have been constrained?

Outside of PPP activity, loan originations have slowed. Indirect auto activity is quite slow because many dealer channels are shut down; that portfolio runs off at a steady pace and won't be replaced in the near term. The residential mortgage pipeline is hanging in — refinance activity is somewhat active given low rates, but purchase activity is weaker. On the commercial side, we're working through committed and in-progress construction loans, but new commercial originations are muted. Many of our people are focused on PPP and deferral work, which also limits origination capacity. So, excluding PPP, I would expect core portfolios to be softer in the near term.

What about credits coming due that would normally be sold or refinanced by others? Are you refinancing them, or are other investors able to pay them off?

For loans we want to retain, we're working to keep those relationships. For loans we'd normally sell or syndicate, market activity has slowed, so it has been harder to place loans with third parties. We'll work with clients where needed and adapt to market conditions, but there hasn't been a lot of third-party investor activity in the last 30 days.

Our next question comes from Collyn Gilbert. Please go ahead.

Just two quick follow-ups. Do you have the number in terms of loans that have been asked for forbearance post-March 31st?

Through April 15th, which is what we included in the deck, we had 4,292 loans deferred representing $587.2 million.

And finally, on reserves: historically your loss content has been low, and peak reserves after the prior crisis were much higher relative to the loan book. Given current book characteristics, could you carry a much lower reserve relative to historical peaks, or how are you thinking about that?

That's a difficult question to answer at this point. Looking forward with the impacts of COVID is tough to determine. Some larger banks maintain higher total reserves but also have higher loss content historically. We're currently at 81 basis points as a percent of loans. It's hard to say whether that will move closer to 1% or stay at current levels until we see more of how the crisis unfolds.

Joe, is it possible that purchase accounting has contributed to that differential over time?

It's possible — that gets into some technical accounting considerations — but there is a possibility it had some impact over time.

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

Great. Thank you, Brent. Lastly, we typically have many employees and directors who are listening to this call, and I just wanted to take the opportunity to thank every one of them for their understanding, effort, engagement and support. It's been a difficult and demanding time for all of us and we really have risen as a team to our standards as essential workers. Most important, I want to thank you for caring about our customers, our communities, our shareholders and each other. We are not in the banking business. We're in the people business. I could not be more proud of the work side-by-side every day of the 3,000 colleagues that make Community Bank System the organization that it is. Thank you all again for joining and be healthy, be safe. We will talk again next quarter. Thank you.

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