BankUnited, Inc.

BankUnited, Inc.

$36.92
-0.05 (-0.14%)
NYSE
USD, US
Banks - Regional

BankUnited, Inc. (BKU) Q2 2020 Earnings Call Transcript

Published at 2020-07-29 22:51:07
Operator
Ladies and gentlemen, thank you for standing by, and welcome to the BankUnited, Inc. Second Quarter Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. [Operator Instructions] I would now like to hand the conference over to your speaker today, Susan Greenfield, Corporate Secretary. Please go ahead, ma'am.
Susan Greenfield
Thank you, Josh. Good morning and thank you for joining us today on our second quarter results conference call. On the call this morning are Raj Singh, our Chairman, President and CEO; Leslie Lunak, our Chief Financial Officer; and Tom Cornish, our Chief Operating Officer. Before we start, I'd like to remind everyone that this call may contain forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995 that reflects the company's current views with respect to, among other things, future events and financial performance. Any forward-looking statements made during this call are based on the historical performance of the company and its subsidiaries around the company's current plans, estimates and expectations. The inclusion of this forward-looking information should not be regarded as a representation by the company that the future plans, estimates or expectations contemplated by the company will be achieved. Such forward-looking statements are subject to various risks and uncertainties and assumptions, including, without limitation, those relating to the company's operations, financial results, financial condition, business prospects, growth strategy and liquidity, including as impacted by the COVID-19 pandemic. The company does not undertake any obligation to publicly update or review any forward-looking statement whether as a result of new information, future developments or otherwise. A number of important factors could cause actual results to differ materially from those indicated by the forward-looking statements. Information on these factors can be found in the company's annual report on Form 10-K for the year ended December 31, 2019, and any subsequent quarterly report on Form 10-Q or current report on Form 8-K, which are available at the SEC's website, www.sec.gov. With that, I'd like to turn the call over to Raj.
Raj Singh
Thank you, Susan. Welcome, everyone, to our earnings call. Thanks for giving us your time. Let me make a few comments about the environment before we get into the quarter. Three months make a big difference. This is not a traditional economic downturn. It's not caused by anything other than the virus. It is still a very serious situation, but we feel a lot better today than we did 90 days ago. There have been some encouraging signs in the economy over the last 3 months, whether it's employment data from May and June, or retail sales, or home sales or capital markets generally, but there is still a lot of uncertainty. Unemployment is still very high. There are mixed signals about certain sectors in CRE. The virus, obviously, is still not tamed, and there are a number of states that are reporting high levels, including Florida. The impact of all of this, then it gets further compounded by the fact that we are 100 days away from election. And the political scenario in the country will make it even harder to read when the economy is headed in the next 100 days and beyond. But overall, we are much more optimistic today, but I would still say we are cautiously optimistic than we were 3 months ago. When all this started, we started having Board meetings. Early on, it was actually on a weekly basis to inform our Board, but then eventually every 2 weeks. And in one of the early Board meetings, a question was brought up, which I want to share with the shareholders, which was about how are we going to deal with this crisis and, over the course of the next year or so? And what are the principles around which we will react to all this? And 3 principles already laid on the table, which we have used in doing everything that we're doing in the bank. First is intellectual honesty with yourself, which is to say, don't try and be overly optimistic and say, oh, this is okay, it will all be fine. So be intellectually honest with yourself. Second, be transparent with all stakeholders. That includes, of course, shareholders, but also includes our regulators, rating agencies, even our customers, our employees. Be transparent, provide more information than usual and don't try and hide anything. Third, be proactive. Or put differently, don't try to kick the can down the road because eventually you're going to have a deal with the issues. So be proactive and deal with them early, get in front of the issues rather behind them. So we've used these 3 principles in all the decisions that we've made over the course of the last 3, I guess, 4 months now. And with that, let me quickly turn into what the earnings were for the quarter. We reported $76.5 million of earnings this quarter, $0.80 per share. This compares to $0.81 per share last year this time. The annualized ROE and ROA was 11.6% and 90 basis points for ROA. We told you last time when we spoke to you that the PPNR would trend favorably. We told you that NIM will increase. The cost of funds will decline, will probably be the biggest decline in the history of the company. We told you operating expenses would trend downwards, and all of those things have happened. PPNR is up $37 million or 44% quarter-over-quarter. $10 million of that 44, $37 million came from net interest income, $15 million came from noninterest income and $12 million came from expenses. Leslie will get into the details of all that. But overall, across the board, all items went the right way. Unlike peers, mostly have reported declines in NIM. Our NIM actually improved from 2.35% to 2.39% as we had indicated 3 months ago, driven mostly because of cost of deposits. Our total cost of deposits declined 56 basis points from last quarter. So we went from 136 down to 80. And on a spot basis, at the end of the quarter on June 30, our API and our deposits was already down to 65 basis points. And into July, it continues to drop. So next quarter, you can expect another drop in cost of deposits. Maybe not 56 basis points, but it will be a nice drop again. And that trend, we expect that to continue into all of this year and into early next year. Provision also declined to $25.4 million from $125.4 million last quarter. Leslie will get into all the details around provision and reserve. But at a high level, I would say that our reserve build from the end of December, so for the full year, we have more than doubled our reserves, and they're now, I think this was 130% increase. We quickly yesterday looked at our peers and we looked at banks between [10 and 110 billion] to see just how much reserve bill people have actually done, and the average or the median was somewhere around 60%. So this is what I'm referring to in terms of getting in front of these issues and taking your medicine early, which we did in the first quarter, which is what resulted in the loss that we posted last quarter. Quickly switching to some balance sheet items. Noninterest DDA grew by $1.3 billion, 28% are, basically, not annualized. Just 28% quarter-over-quarter. Now DDA stands at 23% of total deposits. Last quarter, I think we were at 18%, so a very healthy trend. Average noninterest-bearing DDA was up also by $944 million compared to last quarter. Interest-earning assets were also up for the quarter. Loans and leases grew by $656 million. And securities portfolio grew by $819 million. And again, Leslie will get you a little more detail on that. We also saw a very substantial recovery in the unrealized loss on securities in the second quarter. So if you remember, in the first quarter, we had a $250 million mark, a negative mark, obviously, on the securities portfolio. And we had talked about how that was recovering nicely and that over the course of months, we expected to claw all of that back. We're happy to report that we clawed most of it back. We're down to only $2.6 million. So from negative $250 million to negative $2.6 million, that's almost 99% comeback in a 3-month period. Now there are still some unrealized losses in the CMBS and CLO asset class, and they're continuing to get better. And we always have been very comfortable with the portfolio. Book value increased by $2.59 this quarter. So overall, I couldn't be happier with the performance. Capital. Our capital position remains robust. Our CET1 ratio is 12.2% at the hold co and 13.4% at the bank. Oh, yes, we did issue $300 million in hold co sub debt this quarter at 5 and 08, and that helps our total capital ratio, which now stands at 14.3%. The dividend, we, in evaluating our dividend, we, of course, look at 2 things: we look at our capital adequacy and we look at our sort of medium-term or near-term and medium-term earnings. We feel good about both those things, and so we paid out a $0.23 dividend in the second quarter, and management at this time expects to recommend the same going forward. A few remarks about credit, and Tom and Leslie will get into it more deeply. But as we discussed with you, we've been very proactive in identifying the subsegments and the borrowers that will, in our estimation, be impacted by COVID-19 more than others. So that strategy has not changed. Our ratios, NPAs, NPLs are basically flat to prior quarter end, down marginally compared to December 31. At June 30, NPA ratio was 60 basis points. But again, if you exclude the guaranteed portion of SBA loans, it was 47 basis points. NPLs were at 86 basis points. But again, if you exclude the guaranteed portion of SBA loans, then it's 67 basis points. The efforts we've made to assist borrowers with PPP loans and all the deferrals are likely helping to mitigate and keep these numbers down. Annualized charge-off for the quarter was 20 basis points. The majority of this was linked to one loan in the franchise portfolio. This loan had been showing weakness before COVID, but it got resolved or worked out in the middle of the pandemic, they filed for bankruptcy just before, I think it was in February, so just before the pandemic kicked off, the worst time to work out a loan, which limited our workout solutions. So that's actually a large part of the charge offs, just that one loan. As you know, we've been very accommodating in granting 90-day deferrals. We have started to do that in the last week of March. Most of our deferrals came in over a 3-week period, from last week of March and the first 2 weeks of April, and then it really tapered off after that. We initially granted deferrals on $3.6 billion in loans, about 50% of our portfolio. However, far fewer people are asking now for a re-deferral. So the requests that we've received so far for re-deferrals is only $748 million. So think of it as early on, 50% of our portfolio or $3.6 billion ask for deferral. And now when it comes time, because we only did 90-day deferrals. We did not do anything more than that. So now that 90 days are expiring or have expired for a lot of these, the requests are coming in at a much lower clip, only $748 million or 3%. So that's a big drop and this is a very important number. I want to stress on this because these are hard numbers. This is indicative of customer behavior. These are not our estimates or a model telling us anything. This is actually what customers are doing. So I see that as a very positive number. Of course, it changes but where we are at the end of July, this is actually a very good place to be with the re-deferral rate. We did see an increase in special mention and substandard accruing loans. We did a deep dive in the commercial portfolio this quarter. And we reached out to individual borrowers, especially those who we thought were impacted. And we've increased monitoring of the portfolio to a totally different level, something we would have never done or never thought of doing until 3 months ago. But now we're monitoring this on a weekly basis, monthly basis. And we use all this information to re-risk the the portfolio. So this increase, think of it as lagging the CECL numbers because while CECL is essentially a modeling exercise which we did at the end of March, this is a very manual exercise. This is literally, you have to pick up a loan file, really analyze, make the judgment about whether the risk has gone up or not and then re-rate. So this happened over the course of the quarter and in some ways, it's catching up to the reserve that we put up at the end of the last quarter. So going-forward strategy quickly. This quarter, let's talk a little bit about this quarter, our safety and wellness of our employees is a primary, is the number 1 focus, and I'm happy to say everyone is fine. There have been some cases of COVID positives, but nobody is seriously ill. And the bank operationally is working just fine. There are no issues on liquidity. There was a lot more focus around the entire system last year, last quarter, but I think the Fed and everyone else has done a great job. Liquidity is not an issue at all anywhere. We directed much of our efforts this quarter towards PPP, and to say that it was a herculean task in the month of April would be an understatement. We did do 3,600 loans in a month. Just thinking about it, I even get surprised now. Even though we have achieved that, it was just an unbelievable task. The BU 2.0, which is, has been an ongoing initiative for the last 1.5 years, continues to move forward. As we told you last quarter, we are going to overshoot on the expense side. And on the revenue side, there will be some delays simply because launching some of these new efforts, new revenue efforts, does become a little hard when everybody is locked down and it's hard to get in front of new clients. But in terms of expenses, I think our target was $40 million. We're already at $47 million, and there's probably still some more that will come. And in terms of revenue, our numbers are still pretty good in terms of what we are shooting for, $20 million, but the timing is delayed. We are, just to talk about some of the higher, high level initiatives, we are launching the commercial credit card this quarter. The small business initiatives are also proceeding. We are going to launch automated underwriting platform later this year. That was a little bit delayed. And other initiatives include; we did sign a fee-generating agreement with Goldman Sachs this quarter, just a few days ago; and also a strategic shift in direction towards more treasury management and enhancing those products, it's all on track; we've also launched a new customer derivative program, which also will generate revenue on the commercial side. So overall, very happy with where we are. Let me turn it over to Tom, who can walk you through in a little more detailed loans, deposits, credit and so on. Tom? Tom Cornish : Great, Raj, thank you. So just to amplify some of the comments Raj made earlier. Excellent quarter overall for deposit growth. Total deposits were up $1.1 billion for the quarter. And as Raj mentioned, noninterest DDA actually grew by $1.3 billion. So looking into those numbers a little bit. While it's difficult to be exact about this, as we finished the quarter, we estimated that we had somewhere between $400 million to $600 million of noninterest DDA on the balance sheet that was related to the PPP loan proceeds that were still residing in the operating accounts of the clients. Remember, we funded a little bit over $800 million. So while that was a part of the growth, we still saw a strong overall growth in NI DDA across all geographies and business lines, and a significant portion of our deposit growth this quarter actually came from new client relationships. We had over 700 new business relationships generated over the quarter. And remember that this was a very different quarter for us in terms of generating business in a remote environment. We were learning how to do that. I think we did a really great job of it. We had close to $500 million in deposit growth that came out of those 700 business relationships. Again, each line of business contributed to that significantly. With this growth and our liquidity position, we were able to let some higher cost deposits run off during the quarter. Total time deposits actually declined by $811 million in the quarter as we reduced our CD rates. We do believe that some of that money moved into the money market product. As Raj mentioned, we continue to very systematically reduce our deposit cost. The spot rate on total deposits declined by 47 basis points for the quarter and 77 basis points year-to-date. And still, there are $2.3 billion in CDs that have not yet repriced since the last Fed rate cut with a weighted average rate of 1.91. So you can see from that, we still have significant opportunities as we renew CDs at substantially lower rates. Loan growth. Let's talk about loan growth just a bit. Loan growth primarily was concentrated in the PPP loans. In the mortgage warehousing area, we had $827 million of growth in PPP and $308 million in mortgage warehouse. Our C&I business was down by $317 million in the quarter mostly due to lower-than-normal production. And also, we have seen fairly substantial reductions in line utilization as we got into the second quarter, well below what we saw pre-COVID sort of line utilization numbers. And most other categories experienced either marginal increases or declines for the quarter. Those were really the 3 big change items for the quarter. From a PPP perspective, as Raj said, we did over 3,500 loans totaling $876 million. Almost all of these loans were made to existing clients of the bank, and we were able to accommodate all existing customers who came to us for a PPP loan. So that was a particular source of pride for us in the quarter: making sure that we took care of our clients. Our average loan size was $245,000. As of today, we have not yet started starting processes for giving this applications yet, but we expect to do that starting shortly. Just a few comments on growth and short-term growth opportunities from a strategy perspective. Obviously, the first half of Q2 was focused on PPP loans. As Raj mentioned, it was a herculean effort to essentially do 3,500 loans over a couple of week period of time. But we also saw good growth in the mortgage warehouse and some opportunities existing on the residential side. Pretty much everywhere else, we're taking a cautious stance. We're focused on the existing client base right now. And I would deem our overall current strategy to be strategically selective. We do have a number of clients that we're looking at increasing credit facilities where they have resilient business models and they're actually seeing growth in revenue. We're also starting to see some M&A opportunities within our client base at valuations that are pretty attractive to what we've seen them on a historic basis, so we do expect to see some opportunities there. And we're looking at new client businesses with companies that have strong financial conditions and operating performance in industry segments that are part of our longer-term strategic growth plans. So we've identified some areas, even prior to COVID, where we wanted to grow the portfolio, and we're still looking for new opportunities there. Beyond that, we're being cautious at this time until we can see some longer-term stability in the economy and the trajectory of where the health care crisis is going in the markets, particularly that we're in. It's likely, overall, excluding any forgiveness activity with PPP, which we think will be more of a Q4, Q1 2021-type event that we'll see the loan portfolio down just slightly for Q3. We continue to see opportunities to onboard new deposit relationships across all business lines, our pipelines from all of our teams. And our treasury management business is very strong. We're seeing movement of relationships to the bank in very significant numbers. We expect deposit growth in Q3 again, but probably not to the extent we saw this quarter. Our emphasis continues to be on creating full banking relationships that include noninterest DDA. So shifting that a little bit and talk about of loans on the deferral basis, Slide 21 in the supplemental deck provides more detail around this as well. Again, to emphasize what Raj said, our deferral rate has dropped significantly from the initial deferral rate that we saw. So we look at different parts of the portfolio. In the commercial portfolio, we granted 90-day deferral loans for the first time totaling $3 billion, which represented about 17% of the total commercial book. At this point, we received second deferral request on loans totaling only $696 million or 4% of the commercial book through July 17. So that's a very strong and promising reduction in the level of second deferrals that we've seen. Our initial deferral requests were mostly heavily concentrated in the CRE book, with 29% initial deferral rate and franchise finance with an initial 74% deferral rate. The re-deferral rate in those segments are now as of the 17th, down to 6% for the CRE portfolio and 25% for franchise. The CRE initial deferrals were mostly concentrated in the hotel and retail property type. So again, a substantial improvement in both of those numbers. I also want to emphasize that the re-deferral rate on the C&I portfolio was only 1% of the portfolio. So that was not large to begin with, and it's very, very minimal at this point. So, and each of these cases, also when we look at second level re-deferrals, the initial deferrals, we basically gave fairly readily. These, for the most part, we're looking at improving our collateral position, guarantees or other credit enhancements that we're looking at as we're going forward through the second level of deferrals. The substantial majority of the initial deferrals were processed in late March and April, so have recently reached or coming to the 90 days. While more second deferral request may still come, and there's obviously some changes going on in different parts of the economy, we think we've probably received most or all of them at this point. New first-time deferral requests trailed off materially after April and have really been pretty negligible since June 1st. Switching to the residential portfolio. We granted approximately $594 million, an initial 90-day deferrals, excluding the Ginnie Mae early buyout portfolio, that represented about 13% of the overall portfolio. Of these, 42% or $252 million have actually continued to make payments during the deferral period. $52 million or 1% of the portfolio have been deferred for another 90 days other, either by request or automatically by operations of the new regulations. Try to give you a little data on rent collections, particularly within the CRE portfolio in terms of what we're seeing, and this will take you through June. It's not 100% of the portfolio, but it's representative of a very large percentage that we're reaching out to and collecting this data. So if you go by product category, in the office segment, Florida's collection rates were in the 80% range, New York was in the 70s, multifamily in Florida was in the 90s, New York was in the 80s. Retail overall average was 62%. A little bit more dispersion of the mean in the retail area depending upon the product and where it is, but 62% was the average, generally higher in Florida. And in our warehouse and industrial segment, really minimal. Collection rates have been in the 90% range. I want to talk a little bit about hotel occupancy and the hotel market. Right now, substantially all hotel properties in both Florida and New York are open. There's a couple still closed, but the majority are open. Florida expected occupancy for June, for July, I mean, we're seeing ranges from 30% to close to 60%, trending up since April, where it averaged about 20%. Having seen a number of occupancy around holidays, weekends, where numbers have been higher than that. The summer season in Florida does tend to be a slower season, so we're tracking that carefully to see how occupancy moves up as the fall and winter season start to come into play. But this is an area that will be challenged, and this is one area where we expect to see some restructuring that's longer than the typical 90-day deferral. We don't have any occupancy data for New York yet because hotels have just really opened up recently. Our largest one opens up tomorrow, as a matter of fact. I want to spend a little time on the franchise side. Franchise restaurant. A fairly wide range of performance there with some good news overall. And it really is based kind of more on operating model to the extent you're open and have an established drive-thru and delivery model. The results have been pretty favorable. In many cases, we've actually seen year-to-year increases in same-store sales revenue for the April-May time frame. Full dine-in concepts, which is a much smaller portion of our portfolio, has struggled. That is a tougher recovery from an all-in dine-type concept. But in all cases where we have that, our May results all showed an improvement over April results. And while trends are still looking better, it's a bit early to see, but the second level of deferral requests have dropped dramatically in that portfolio as well. Fitness is another portion of what's in our franchise business. Stores are beginning to open up in many of the markets that we lend to, so it's still a little bit early to state that. But overall, the re-deferral rate in franchise is 7, is 25% compared to the original 74% for the first level of deferrals, so we're seeing much better performance. And we could have even been hopeful at that point. 25%, I think, is really excellent. So we'll turn it over to Leslie now to get into a little more detail on CECL and the quarterly results.
Leslie Lunak
Morning, everybody. Before I dive in, I'm just going to, we're not really doing a walk-through of our deck today with you, but we did put a supplemental deck out there for you to update a lot of the information that we had given you at the end of the first quarter, so it's there for reference. And so now diving into CECL and the allowance for credit losses. There are a lot of moving parts to the reserve estimate for the quarter, and I'll try to walk you through the more significant ones. So for your later reading pleasure, Slide 15 in our supplemental deck walks through the changes to the reserve by major portfolio segment. You can look there as I'm talking if you want to. But let me speak first to the economic forecast assumptions. So the loss estimate is impacted by the economic, by economic factors in two different ways; one is it takes into account economic conditions as they existed at the balance sheet date, which were significantly worse at June 30th compared to the model inputs at the prior quarter end. The second way is the forward path of the economic forecast, which was, in some respects, worse and in others better compared to the March 31st forecast. And one thing to note here is that at June 30th, unlike at March 31st, the worst point of the forecasted trajectory of the economy was behind us, whereas at March 31st, it was in front of us. To give you a few high-level data points. The forecast for this quarter incorporated unemployment starting at 13.4%, declining to 9% by the end of 2020 and to 7% by the end of 2021. Annualized GDP growth started at a negative 27%, recovers pretty significantly in Q3 and ultimately returns to prerecession levels by 2023. The mix trailing average started this quarter at 32. It stays elevated through 2020 and moderates from there, although it remains pretty choppy throughout the forecast horizon. The S&P 500 started at about 2,900, declines to about 2,700 at the end of 2020 and ends 2021 over 3,000. Again, kind of choppy. This quarter was a tale of three cities, so to speak, with respect to the reserve. At a high level, we use 3 different models to estimate expected losses, and we do that at the loan level. We use one model for residential, one for CRE and yet a third one for C&I, and each of those models works a little differently and has greater sensitivity to different economic variables and factors. And I'm looking now at Slide 15, if you want to go there. If not, just listen, then you can look later. So let's start with the easy one, residential. There wasn't a material change in the level of residential reserves this quarter. The residential model is most sensitive to HPI, which actually improved in the June forecast compared to the March forecast. The model is also pretty sensitive unemployment, which was worse than the June forecast. So those factors moved in opposite directions. An update to expected prepayment speeds also reduced the residential reserve marginally this quarter. In the aggregate for the CRE portfolio, the reserve increased from 57 basis points at March 31st to 1.54% at June 30th. The CRE model is highly sensitive to unemployment as well as to commercial property forecasts by property type and geography. Both of those deteriorated from the prior quarter's forecast in terms of the starting point and the forecasted trajectory. So as a result, the economic forecast increased to the CRE reserve by $48 million this quarter. Additionally, we applied a qualitative factor of $24 million, increasing the reserve based on data collected from individual borrowers that reflected a deterioration in current revenue levels or financial condition as compared to the most recent financial statements that were housed in the model. We collected this information from a substantial representative sample of borrowers, and we extrapolated the results across the relevant population and coming up with the qualitative factor. In contrast, the reserve on the C&I portfolio declined from 2% to 1.38% this quarter. The C&I model is very sensitive to changes in the volatility index, into the S&P 500. So while it's also sensitive to unemployment, this model is less sensitive to unemployment than the 3 model. And while the starting point of all of these variables was worse at June 30 than at March 31, the forward path of the VIX and the S&P 500 actually improved in the June forecast compared to the March forecast, as did expectations for credit spreads. So that offsets the deterioration in unemployment. Similar to the CRE portfolio, we made qualitative adjustments, increasing the commercial reserve based on data collected from a representative sample of our borrowers. One other thing that affected the commercial reserve this quarter, at 3/31, we made an assumption, with respect to prepayments, effectively assuming 0 prepayments for a period of time. Sitting at March 31, our thought process was kind of no one is going to prepay a loan in the middle of a pandemic. So we set prepayments to 0. Under CECL, the reserve is very sensitive to prepayment expectations, irrespective of the state of the economy because it's a lifetime reserve estimate. That 0 prepayment assumption turned out to be very inconsistent with our actual experience for the second quarter. Prepayments for the quarter actually picked up from our historical averages. So we removed that 0 prepayment assumption for the quarter, and that did have the impact of bringing the reserve down some, particularly in the C&I segment. That affected all segments, but the C&I segment was much more sensitive to it or much more impacted by it. The franchise portfolio continues to carry the highest reserve level at 3.12%, followed by CRE and then C&I. The franchise reserve is a little bit down this quarter as a percentage of loans, but that's mainly because of the large charge-offs that Raj referred to that we took that reduced the reserve this quarter. These relative reserve levels, higher-end franchise increase and low risk in C&I are consistent with our deferral rates, our re-deferral rates and our other observations about relative levels of stress being exhibited in those different portfolio segments. I would also like to point out that at June 30, our reserve was sitting at 60% of 2020 DFAST severely adverse losses. And under stress, our capital ratios remained comfortably above well capitalized levels. As to expectations about the provision and the reserve going forward, there are a couple of things that could lead to additional reserve builds other than new production. One would be if the overall economic outlook deteriorates materially from what our current forecast would suggest. The main catalyst we can see to that happen is some sort of return to widespread shutdowns or loss of confidence related to spread of the virus. Our reserve levels at June 30 aren't predicated on an expectation of a widespread return to shut down to the economy. We could also have some more granular borrower-specific or subsegment-specific individual builds if response of individual borrowers to the pandemic differ from our current expectations or what our model is currently projecting. But outside of a significant deterioration in the economy, I wouldn't expect a material reserve build. Any of these things could also, in theory, move in the other direction. Next, I'd like to give you a little bit of color around risk rating migration. There are some slides in the deck on this. Let me start by restating the guiding principle that Raj called out. When it comes to risk rating, we call it the way it is, and we attempt to be both proactive and objective. We started the downgrade process early. We called out any weaknesses that we saw. We believe that calling it what it is and recognizing an increase in risk when we see it is the only way to do this. I think it's difficult to conclude that given the current state of things with the pandemic, that credit has not increased, and we believe our risk rating methodology reflects that. Comparing risk ratings from bank-to-bank can be pretty difficult, particularly in this environment because every bank takes a different approach to this. We know there are some banks that have decided to wait to re-risk rate their portfolio, for example, until all their deferral periods are over. But that is not the approach we decided to take. Another thing I'd like to say real quick about risk ratings in relation to the reserve, we believe that the risk rating distribution of our portfolio is reflective of where our reserve sat at 3/31. In effect, our risk rating process is kind of catching up this quarter with our reserves. Having said that, to get into some of the details. Total special mention loans increased significantly from $288 million to $1.3 billion during the quarter. The segments where we saw the most significant increase were those where we expected to see it: CRE, retail and hotel. Within C&I, we saw it in the cruise line credits, in retail trade, food services and also we saw it in franchise finance. CRE increase in special mention was $527 million. Most of that was in retail at 1 68 and hotel at 2 70. C&I, we saw an increase in special mention of 3 29. We saw that, as I said, mostly in the cruise lines, retail trade and food services franchise that went up $147 million, not unexpectedly. We consider special mention ratings to be a transitional rating for loans that have potential weaknesses that could result in deterioration of repayment prospects at some future date if not checked or corrected. Total substandard accruing loans increased from $239 million to $561 million this quarter. Again, the segments with the largest increases were CRE, primarily hotel and retail. And within C&I, the most significant increase related to a couple of borrowers who supply or operate airport shops or concessions, which, of course, 6 months ago, looked like a great business. Let me shift now and talk for a minute about the NIM. The NIM increased by 4 basis points this quarter, from 2.35% to 2.39%. To get into the components of that, the yield on interest-earning assets declined by 44 basis points, reflecting a decline of 47 basis points in the yield on loans and a 33 basis point decline in the yield on investment securities. Reset of the coupon on floaters was the most impactful contributor to the decline in yields on securities. And overall, the decline in asset yields resulted from declines in benchmark rates, including turnover of the portfolio at lower prevailing rates. The cost of interest-bearing liabilities declined by 60 basis points quarter-over-quarter, with the cost of interest-bearing deposits down by 65 basis points and the cost of borrowings down by 54 basis points. Our expectation currently is for the NIM to be relatively stable for the third quarter. We believe both deposit costs and the yield on interest-earning assets will decline further. But on balance, the NIM will remain stable. We don't expect to see much recognition of fees from PPP forgiveness in the third quarter. We don't really expect to see that start to happen before Q4. Some specific items that impacted noninterest income and noninterest expense. Securities gains were $6.8 million this quarter compared to a loss of $3.5 million last quarter. Last quarter, if you recall, we had an unrealized loss of $5 million on marketable equity securities that ran through the P&L. This quarter, that component was a gain of $1.1 million. Other securities gains arose from our, just our ongoing management and positioning of the portfolio. Deposit service charges declined a little bit compared to the preceding quarter and the comparable quarter of the prior year. There was some forgiveness of service charges this quarter related to COVID, and there was also just lower volume of activity in some fee areas this quarter, which we believe was related to the pandemic as well. Employee compensation and benefits decreased by $10 million compared to the immediately preceding quarter. Some of this is seasonal payroll taxes, 401(k) contributions and HSA seating are always elevated in the first quarter. And additionally, variable compensation costs were lower this quarter, reflective of lower levels of production and earnings resulting from the pandemic. Probably a more valid comparison because there's a lot of quarter-over-quarter noise is the 6 months ended June 30, 2020, compared to the prior year, where we saw comp and benefits decrease by $14.7 million. We're really starting to see here the impact of all of our BankUnited 2.0 expense initiatives. Noninterest expense for the quarter does include $1.5 million of costs related directly to the pandemic. That could be everything from equipment to facilitate people working from home, to some costs we incurred, standing up our PPP program, supplies, equipping branches, things like that. There will probably be a little bit more of this to come. Our expectation is that operating expenses will be relatively flat for Q3 compared to Q2. And with that, I'm going to turn it over to Raj for some closing comments.
Raj Singh
Thank you, Leslie. The problem with transparency and providing you a lot of information is that these calls take forever. So sorry for taking so much of your time, but we'll open it up for Q&A.
Operator
[Operator Instructions] Our first question comes from Stephen Scouten with Piper Sandler.
Stephen Scouten
I'm curious, maybe first, on the expense run rate. I mean, a really impressive quarter-over-quarter change there. I know your guidance kind of previously, which, obviously, no one could be too specific, but it was just for expenses to be down year-over-year. So wondering if you could dig deeper into the salaries, migrations and what really caused that, if it was FTE reductions or more just the things you noted in the release? And then just lastly, if there was a FAS 91 impact within salaries as well.
Leslie Lunak
Yes. So I think the salary reduction, the most significant component of that, Stephen, is, in fact, FTE reductions year-over-year. There's a little bit of an element of that, as I mentioned in my remarks, that's reduced variable compensation. As to the FAS 91. If you look at Q2 compared to Q1, yes, there were a little bit more FAS 91 costs deferred in Q2 because of PPP. However, year-over-year, that actually went the other way. There was significantly more FAS 91 deferrals in '19 than in '20 because loan growth was higher in '19. And I don't have, I'm sorry, I don't have those exact numbers in front of me.
Stephen Scouten
No, that's okay. That's, yes. No, that's really helpful. Okay. And maybe just the one other question really I have is, so you guys gave a ton of detail on the migrations and the movements in the loan loss reserve, which is extremely helpful. But I guess maybe generically, how would you respond to kind of maybe a feeling if somebody looked at your reserve on a stand-alone basis, call it, the 1 27 without the PPP and the mortgage warehouse, the debt on an absolute basis, may look below peers. Would you just kind of highlight, I guess, all that you already said and say that you believe the risk rating changes are the lagging indicator? Or how can you maybe frame that up for us?
Raj Singh
I think you have to take into account the portfolio mix. So when you compare bank a to bank b, there's often a very big difference in portfolio mix. We have a big resi portfolio. We have other large portfolios that are not really impacted by this. So the municipal portfolio is hardly reserve or against that. That's 1.5 billion. So if you normalize for those things, then you will, that explains it more than anything else. We've always said our portfolio was never really created for high yield. It doesn't have the high-yield components. In other words, it doesn't have the high-risk components, which is why you see that difference. If we had a couple of billion dollars of construction loans or some other leverage loans, a couple of billion of that, you would see a much higher reserve.
Leslie Lunak
Yes. I would echo that. I think it's largely reflective of what we believe to be the quality of the portfolio. And Stephen, yes, I do think the risk rating migration is a lagging indicator. I think the risk rating distribution at 6 30 is much more reflective of where the, I think the reserve at 3/31, a quarter ago, was more reflective of the current risk rating distribution than of the risk rating distribution at that date. It's kind of catching up.
Operator
Our next question comes from Brady Gailey with KBW.
Brady Gailey
So if you look at the BankUnited 2.0 plan, I mean, clearly, you're coming ahead of where you thought you'd be on the expense side. I think you said you're at, you're already at $47 million versus, I think, the initial estimate was around $40 million. How much better do you think you can do on the expense side? It feels like the world has changed, and there's probably some more branches. And maybe people that weren't planned back when BKU was initially focused on.
Raj Singh
Yes. Brady, we have not gone back and majorly redone the initiatives to say, okay, let's take a second look at this and a 3.0, for lack of a better word. We haven't done that because the last two, three months have been rather sort of moving focus tactically on everything PPP and everything else that we're doing in this environment. But we will take a look at it, as the business model changes, as customer behavior changes, if there is an opportunity to look for more stuff, branches is a simple example of that, we are going to look at that. But that's not going to happen in the next quarter or two. That those will be longer-term because any decisions you make that are meaningful do take a year or two to really not to analyze but to operationalize, right? If we decide there's one more bank we could close, if you decide today, you're not going to see the impact of that for at least a year, if not longer.
Brady Gailey
Okay. And then looking at the cost of funds or the cost of deposits, I mean, it's a nice reduction there. But the cost of deposits is still 80 basis points, which it feels like there's definitely room to move that lower. How low do you think you can get the cost of deposits in this kind of 0 interest rate environment?
Raj Singh
Well, we're kind of showing you a little bit of what it's already looking like by telling you at June 30, it was down to 65 already, right? It was 80 for the quarter, but at a point in time, in June 30, it was 65. And it's already down from there in the first 3 weeks of July. I think we bottomed out in the high 50 basis points. Last time Fed was at 0. We're pretty confident that we will go well past that. Now do we get, do we bottom out in the 40s or the 30s? It's really hard to say. What we are focused on is basically bringing in operating accounts. And that actually, more than anything else, more than any of the highlight numbers, that's really the driver of long-term value. And we're having a lot of success. PPP was a gift as painful as it wants to go through it. And, but looking back at it, it was such a big gift. And the reason it was a big gift is a lot of big banks really, really kicked off their clients. And that has given an opportunity for us to go in. Even though we didn't do their PPP loans, but if you were not treated well by a bank in April, and even if you got a PPP loan eventually, but you have to go through a month of not knowing whether you'll get the money or not and not getting phone calls returned, that creates an opportunity for banks like us. And we're already capturing that. All the numbers you see here, some of the growth that we've gotten is because of that. And the pipeline, as Tom said, is robust. It's because of that. So that will be a gift that will keep giving for the next few months into next year.
Operator
Our next question comes from Jared Shaw with Wells Fargo.
Jared Shaw
Just wanted to circle back on the allowance and credit. First, I guess, on the deferrals. What percentage of the portfolio has already gone through that first 90 days? So I see we were down to 4%, but is that 4% a good indicative level of the whole thing, or that's just where we are right now?
Raj Singh
Yes. Most of the first run deferrals happened in a 3-week period which was last week of March and the first 2 weeks of April. So here we are sitting at the end of July. So I don't have the exact numbers, but I'd say a pretty large majority has already hit the 90-day window. And remember, they start talking to us about asking for another deferral, not on the last day, those conversations generally start at sort of a 2-month to 2.5-month period. So we feel that we have fielded a lot or most of what has to come in terms of re-deferrals, but there may be still some that trickle in. But a majority of, and what we're showing you is actually not approved deferrals. What we're showing you with these readable rates are the requests because they are still, some of them are still in process. And not every one of them will be granted. Most of them will be granted, but some may not be granted. We're showing you the request that have come in. So that's actually a larger number than what will actually get processed.
Jared Shaw
Got it. Okay. And then just looking at the allowance and the provision level. I'm looking at Slide 14, where you have the reduction in allowance of 8.4% from the economic forecast. That's definitely not what we've seen with the other banks. And I haven't really heard other banks focus on the VIX and the S&P. And when I look at like the baseline, you will use GDP expectation for 2021 from March to June, it's actually gotten worse. So I guess, was there a change in methodology from first quarter to second quarter on CECL or...
Leslie Lunak
No, no.
Jared Shaw
Can you just give a little more color on that broader economic deal?
Leslie Lunak
Yes. I would actually just flip to Slide 15, which I think is going to answer your question better than Slide 14 because it really breaks it down. Like in my remarks, I kind of said it's a tale of 3 cities. And you can see that the impacts were very different in the CRE portfolio than in the C&I portfolio, where the economic forecast really was punitive to the CRE portfolio. I don't, I can't speak to what other banks are doing or what models they're using, but I can say that the model that we use for our C&I portfolio, which is one of the suite of Moody's models, is very sensitive to the VIX and the S&P. I know a lot of other banks aren't talking about that. Maybe they don't have a lot of those types of loans in their portfolio that would run through that model. I can't speak to that. But that's really what's driving it. So no, there's no change in methodology whatsoever. Those particular economic variables in the June forecast actually improved compared to the March forecast.
Operator
Our next question comes from Steven Alexopoulos with JPMorgan.
Steven Alexopoulos
Before I start, I just have to say your credit-related disclosures, both the slide deck and on this call are really top notch. I think the best I've seen in the industry, so thank you for that. To follow up on Jared's questioning, I tell you, I think we're all staring at the Slide 15 and looking at the reduction tied to the economic forecast in C&I and scratching our heads saying, how would the market volatility in the S&P 500 impact BankUnited's credit quality on C&I? It just doesn't seem to make sense to us.
Leslie Lunak
Well, so it's a correlation. I can tell you the Moody's thousand page white paper on the risk count model, but you probably don't have time to read that. But what Moody's has determined is that with middle-market C&I modeling, that those are the two factors that are most highly correlated to perform, to credit performance. I think there's probably a whole lot of complicated math behind that, that may be over my head in some respects. But those are the factors that Moody's has determined are most, is from their credit research database with millions upon millions upon millions of observations, and these are the factors that they've determined that losses for those types of loans are most highly correlated to. Mathematically, that's how I answer your question. But to take a step back from that, I will tell you that we feel really good about where the reserve is sitting on our particular C&I book right now. We think it's in the right place. We know what's in that book, we know the quality of those borrowers, we understand their businesses, we understand their performance and their financial condition. We've spent a lot of time understanding how they're being impacted by the pandemic and we really feel like the reserve on that particular portfolio segment is right where it needs to be. And I think that's the more important thing than trying to have, we can have a long conversation about the math and the Moody's models, but I think that's really the more important observation, Stephen. Tom Cornish : So I think the 1% deferral rate is a fact...
Leslie Lunak
Supports that. Tom Cornish : Support that.
Leslie Lunak
Yes.
Raj Singh
There's another mechanical factor, which also went in, which is we have made assumptions around prepayments at the end of March, which were a little too draconian. We have assumed that like it generally happens during recession that prepayments would come to a grinding hold, and there would be no prepayments for the rest of this year. Now we look back at the last 3 months, if anything, prepayments actually picked up. So it applies us. That's usually not what happens. But we have to then react to the actual data as it's coming out, and we're seeing prepayments across all asset classes, by the way, not just one, that are at least as much as they were before. In some cases, are actually higher. So these models are also sensitive to prepayment assumptions, which, like I said, we had to dial back to more realistic numbers than the draconian ones we yield in March.
Leslie Lunak
And I'll also remind you that we did spend a lot of time in the month of June reaching out to our individual borrowers to find out not as of December 31 or as of March 31, but today, how is your business performing? What do your revenues look like? What do you, and all of those observations at the loan level informed our reserve estimate. So we feel very good about the C&I portfolio and the current reserve levels.
Steven Alexopoulos
Okay. That's fair. If I could just follow-up, too, on BankUnited 2.0 and looking at the $47 million of expense reductions so far, which is obviously better than expected. Maybe one, where are you coming in better than expected? And two, are there, have you now realized all of the cost saves, or are there still more to come as part of that plan?
Leslie Lunak
I would say a lot of it is comp and, a lot of its comps, Steven, just because that's the biggest expense line on the P&L. I mean, if you're going to save money, you're going to have to save in comps because that's where the dollars are. So that's where most of it is. We've had some of it, obviously, and the real estate spend has come down. Some of our technology spend, some of the things we've done, some of the investments we've made in technology are actually reducing our operating cost levels. So it's all of that. As to future opportunities, I'll echo what Raj said. I think there are future opportunities, but you're going to see them not all next quarter because they would be predicated on decisions that we would make about the real estate footprint, some things that are still to come in the technology areas and things that I still think are to come in the vendor spend area. But all of those are going to take time to materialize and really have an impact on the bottom line.
Raj Singh
Yes. Steven, I do want to make a point here. Please do not think that this is coming from starting the franchise from investments that we also are making as part of 2.0. So the number of $40 million is a net number. There is stuff that we're investing in, which will pay dividends in the years out. It's not like we've decided, let's just not do those things. So those, because on one hand, we have to tactically fight the fire. But at the same time, we have to keep building the franchise, and you cannot just keep fighting the fire and then have no franchise 2 years down the road. So those investments are still going, moving forward because once you make a decision on making investments, it may take 2 years to do it. And you don't stop in the middle, but that does a lot of damage. Make you feel good for a quarter or 2, but you really is the wrong thing to do. So those investments are happening. Like I said, the automated underwriting on the small business front, that was a good investment. It took a year of technology spend and training and hiring people, all of that, we haven't made a single loan on that. It will go live. Unfortunately, it's about 3 or 4 months later than we had originally wanted it to be. But we are going to move forward. That's the direction we're going in. In fact, PPP was a nice little test for us. We have to do automated underwriting on a very high level for PPP. And we learned a lot from that, and we're using it to invest in that platform. And we will launch it. Like I said, maybe 3-, 4-month delay, but we're not pulling back. That's not where the dollars are coming from. Some of it is, to be very honest, when we tell you $40 million, we internally are shooting for a higher number, as you would expect us to do, right? And it just becomes easier to achieve that when everything is locked down and people are generally, become a lot more sensitive to expenses. So it's a little easier to achieve.
Steven Alexopoulos
Okay. And maybe just one final one for Tom. You talked about all these commercial relationships that you're moving over to the bank. I think Leslie made the comment that we didn't expect anybody to prepay a loan in the middle of a pandemic. I wouldn't expect many commercial customers to move their account in the middle of a pandemic either, but you're doing it. Maybe give us some color on how you're, I mean, are these moving over Zoom meetings? Like how are you moving all these relationships over? Tom Cornish : They are Zoom-meeting driven. As I said, when I made my comments, we're learning how to sell...
Leslie Lunak
And I do think Tom has been [indiscernible]. Tom Cornish : Maybe I'll admit to that. No. But as Raj said, the PPP process, it's not all of it, but it, I think how successfully we executed that. And also compared to where others failed, we have picked up new relationships that have been with other major banks for 30, 40, 50. We got one that was there for 70 years that are very substantial relationships. And I think that our, each of our business units that are focused on kind of the niche market segments that they are focused on, have good reputations in those markets. And I think our overall performance has led to this kind of a flow of good operating business that's coming to us. And we have done it in a Zoom, noncontact environment. We've done a lot of internal training around how to do this better and getting people to be very comfortable with the technology. And clients have become as comfortable with the technology as well. So it's worked very well.
Raj Singh
Steve, for example, PPP loans, there are 2 rounds of PPP, round 1 and round 2. It's important to actually understand how many of your clients are taking care of in round 1 and how many in round 2 because anyone who wants a push to round 2 was probably not happy because they had to wait, and they didn't know there would be around 2 or not. So if you look at a bank and figure out that 10% of customers were done in round 1 and 90% in round 2, you have a lot of unhappy customers. Even though they got, everybody got the loan they were looking for. But it speaks to that one month of actually being on pins and needles, trying to figure out if your bank can do for you what other banks are doing for their clients. Our numbers were, I think, 85% or so of our clients we're taken care of in the first round. I get about 15% or roughly that many who fell into round 2. And customer satisfaction with those is absolutely lower than in round 1. Now then when a bank which has the numbers split, you have a very unhappy customer base. Even if they eventually got a loan, they haven't forgotten the number of times they have to call you and the number of times that you basically said, I don't know when we can help you. And that creates the opportunity for banks like us to be proactive and gather that business. Tom Cornish : Yes. I think if I look back on it, we had a really nice blend of an automated process. As Raj mentioned, we ran the first automated process during this that worked well, but we also combined it with high-touch personal communication. And it took a lot of hours. We were all on conference calls until midnight every night doing this. But I think the communication that went back and forth to the client in terms of keeping them informed and making it feel like a high-touch automated combination really worked out well for us.
Operator
Our next question comes from Ebrahim Poonawala with Bank of America.
Ebrahim Poonawala
A follow-up, I guess. Most of my questions have been asked and answered. Around the margin outlook, Leslie, you mentioned expect a stable NIM looking into third quarter. I guess, beyond that, I think just listening to Tom around deposit growth feels like you expect to retain the deposits that you got this quarter and then grow some as we look out. Where do you think the margin goes? Like, can the margin actually expand in this environment with, if the yield curve stays where it is? Or is the best case outlook for the margin to remain steady state?
Leslie Lunak
So I think looking out beyond the third quarter, there are just a lot of moving parts, so I'm hesitant to give much more guidance than that. The PPP forgiveness will be a factor. Our level of success and continuing to push deposit costs down will be a factor. How much we're able to grow the noninterest-bearing DDA book over the course of the rest of the year will be a factor. And another factor will be what opportunities we're able to identify to put interest-earning assets on the balance sheet and at what spread. And so all of those things right now, I think, are just difficult to predict. So I hate to not answer your question, but I'm hesitant to put a forecast out there for much beyond the quarter ahead.
Ebrahim Poonawala
Got it. And now if you can remind us, Leslie, what's the PPP fees that you expect to accrete?
Leslie Lunak
I'm sorry, could you repeat that?
Ebrahim Poonawala
The PPP origination fees that you expect to accrue.
Leslie Lunak
Yes. I got that number here somewhere, let me just grab it. I want to say there is $21 million as of 6/30 remaining to be recognized. And like I said, we don't really expect much for business activity in Q3. I don't expect that to start hitting until Q4-Q1 time frame realistically, but there's a total of $21 million.
Ebrahim Poonawala
And the entire PPP loans are funded by the Fed funding. That's correct? That goes away when the loans go there?
Leslie Lunak
Yes. Most of it. Right now, we've got $651 million in PPPLs borrowings on the balance sheet. Where we plan to position ourselves so by 9/30 as any of the PPP loans that are left on our balance sheet, we'll pledge up there, and we'll have them fully match funded.
Operator
Our next question comes from Brock Vandervliet with UBS.
Brock Vandervliet
Just to follow-up on Ebrahim's question. It seems like if you line it out, you've got 3 catalysts on the funding side: you got better mix, you got CD repricing continuing and you've got this operating account growth focus. And on the asset side, your, how close are we really to the kind of the burnout on resetting on the asset yields? I feel like that's really the question because I think you've got it on the funding side. It's really how much incremental pressure you see on the asset side.
Leslie Lunak
Yes. Everything held constant. I would say this would be the last quarter of that repricing down. The wildcard there, though, is prepayments. And if we continue to experience prepayments, the higher-yielding assets in the portfolio, that could also put pressure on asset yields, and it's difficult to predict how long that goes on. So that's the wild card in that equation, I think.
Brock Vandervliet
Okay. And on the expense side, you've got the BKU 2.0 saves juxtaposed with, hopefully, more normalization in the economy and an upward bias in some of those expense figures. How does that, how do you see that shaking out? Or I guess, put another way in terms of a, in terms of the core pickup in expenses with higher activity, how much could you dimension that at all?
Leslie Lunak
We really haven't put any guidance out there beyond 2020, and I'm not quite prepared to do that yet. I think things are just in too much of a state of flux. For Q3 and Q4, I would expect the run rate sitting here today to be relatively flat. And as we get a little deeper into the year, I'll be in a better position to really answer that question intelligently. I just don't feel equipped to do that today.
Brock Vandervliet
Okay, okay. But basically a flat run rate...
Leslie Lunak
For the next couple of quarters. And then by then, we'll have a better idea of what the glide path in front of us is.
Operator
Our next question comes from Dave Bishop with D.A. Davidson.
Dave Bishop
A question for you. I know it's just announced, but the fee income initiative, the partnership with Goldman Sachs, how should we think about that? Any guidance you can give in terms of what the revenue opportunity is from that? Tom Cornish : Yes. I would probably say it's too early to think about that. But as we look at the opportunity overall, this is a space that we obviously we're not in previously and did not have a product offering. And we have a large client base in the corporate, commercial, not-for-profit areas. So the 401(k), 403(b) and retirement services planning area. There are clearly substantial opportunities in our portfolio. And I think that this partnership will allow us to kind of unlock those opportunities and this alliance that we announced with Goldman the other, just earlier this week. But it's, we're just starting to now work through the training processes and whatnot. It be a little bit too early to put out a target on it. It could be important.
Dave Bishop
Got it. And then, Leslie, I know in the preamble, you gave some good detail in terms of the granularity regarding the uptick in special mention loans. Curious if you could just walk through that again. It looked like there were some upticks in some of the more granular areas in terms of the...
Leslie Lunak
[indiscernible] backout. Okay, special mention, in the aggregate, went from $288 million to 1.3 billion. We saw increase, that was up $527 million. $168 million of that was in retail. $270 of that was in hotel, only $50 million in multifamily and the rest just kind of drips and drabs. In the C&I book, it went up $329 million. $60 million of that was cruise lines, $54 million was in the retail trade sector. Food services was $53 million, and the rest is onesies, twosies. The franchise book went up by $147 million. And there's some detail on that, too, in the deck, if you want to, towards the back, but those are the high-level numbers. [indiscernible] it was concentrated in those areas where you would think it would be, which was the CRE retail, the hotel, the franchise, a couple isolated credits in the C&I book that are connected in some way to those stressed industries. So exactly what you would have expected.
Dave Bishop
And if I recall in the deck, crews lines exposure, right, they have not asked for deferments yet?
Leslie Lunak
Correct. We actually think the cruise lines are pretty well positioned. They've all been very successful in raising cash in the capital market. We, back to Raj's point about intellectual honesty, we think it's intellectually dishonest to pretend that the risk profile of that industry has not increased. And so it just seems the right thing to do to move those credits to special mention, even though we believe they are in a position to continue to service the debt.
Dave Bishop
Got it. And then maybe just commentary. I know there's some energy exposure through the operating leases. What's the current outlook there currently?
Leslie Lunak
So I think similar to what we said last quarter, oil prices actually seem to have rebounded some from last quarter. I don't, they're awfully volatile though. Railcar loads are down. So what we really think is going to happen there is as these assets are re-leased, they're going to be re-leased at lower rates. So I think we'll see that lease rental income kind of trend downward for the foreseeable future. It's difficult to say longer term, what's going to happen, but these are very long lived assets and are, by definition, having in the past withstood multiple cycles. So we'll see how it plays out. And I think what you'll see, it's not really credit exposure, it's really this rental income that is probably going to trail down some for the foreseeable future.
Operator
Our next question comes from Steven Duong with RBC Capital Markets.
Steven Duong
Sorry, I apologize if this has been answered. Jumped on late. But what are you guys seeing in your overall business activity in your Florida and New York markets, say, from the reopening through mid-June versus the recent spike in cases in Florida?
Raj Singh
Tom?
Tom Cornish
Well, let's start with Florida. I would say that while the case loads are definitely up, it's created a heightened sense of issues in the marketplace. The commercial market continues to move forward. Companies are operating. For the most part, businesses are open. Companies that manufacture and distribute goods or manufacturing and distributing goods and the overall market continues to move forward. It's almost a parallel kind of issue. When you look at what's happening, you see certain health statistics that are coming out and numbers that are obviously concerning. But at the same time, businesses that are essential businesses that are shipping food and components and auto parts and any of the myriad of industries that were in health care, are continuing to operate. Food services continue to operate. Franchise businesses that have good drive-thru models and delivery models are continuing to operate. Occupancy remains strong in the CRE markets in Florida, in the office. In industrial sectors. The industrial sector in Florida is actually on fire. I mean industrial purchasing right now is very, very strong as there's much more demand for warehouse space and e-commerce-related space and things of that nature. So overall, it's not, and we see it in the flow of business that we're getting and the flow of opportunities that we're seeing. It's not a negative scenario at all. New York has opened up, obviously, slower because it had more of a significant impact early on in the crisis, but we're starting to see the construction industry come back. It was early, identified as an essential industry. We have a number of significant construction relationships in the New York market. That's, they're up. The businesses are running. They're, they have projects underway. We're seeing operating businesses in the food and beverage distribution area, just like in Florida, continuing to ship goods and do well. And in the CRE basis, we're seeing, again, office, particularly Class A office, our exposure is a bit more in the Class A office side than it would be B, collections being very high. Multifamily business, we're also seeing rent collections that, at acceptable levels right now. So I would, the economies are functioning. And there are opportunities, and we're seeing business revenue in both markets. Every business is different, obviously. And we're in a lot of different businesses with a lot of different companies, but the overall financial results, I think, are playing out fairly well, and that's what we see in the deferrals.
Raj Singh
The Florida economy...
Steven Duong
[indiscernible] has gone there. Deferrals are the real test to payments.
Raj Singh
Yes. And the Florida economy, despite all the negative news over the last 3 weeks or so, at least to us, it doesn't feel like it has slowed down. So the numbers are obviously bad. On the health care front, they are beginning to level off or at least keeping up in their trough over the last 10 days or so. We're seeing them level off at still pretty high levels, but it has not impacted, at least from what we can see, the economic activity in Florida. And New York has just been more careful and slower with opening up the economy. The health care numbers look great and the economy is slowly opening up. And I think the trajectory would just be, the slope will be much less than Florida. Florida was a very steep increase starting in May 1. It just went straight up and, but we don't see signs of slowing down. Maybe a few bars will be closed here and there, but we're not seeing any signs of some kind of let's go back to a shutdown or slow everything down. That is not the sentiment on the ground. Tom Cornish : Yes. When you look at the C&I business, particularly in either Florida or in New York, but even more so in Florida, what is happening in Florida individually is important. But if you're doing business with a company that's a $500 million-company shipping goods around the U.S. and Canada and Europe, what happens in Florida is not necessarily the only driver. While the businesses are located in Florida, they're national and international businesses.
Steven Duong
Got it. So let me ask it this way then. If the spike in cases never happened, would you say that business wouldn't be that much better? It would just be similar to where it is now for Florida? Tom Cornish : No. It's kind of hard to say. That's a tough one to tell.
Leslie Lunak
The hard thing to measure is the impact that, that has on confidence. Would more people be traveling to Florida and staying in hotels if it hasn't been for that? Maybe. But it's really difficult to know.
Steven Duong
Right. So have you guys been seeing anything? Tom Cornish : [indiscernible] line.
Leslie Lunak
Yes.
Steven Duong
Have you guys been seeing anything as far as consumer confidence goes?
Raj Singh
Our evidence is more anecdotal, so, and maybe it's a little too early. But from what we see from talking to our clients, we haven't yet seen, I think your question is a good one, that if this had not happened, would the slope of this recovery, looking forward, been better, I would think it would have been better. I think more people will go to Disney if there weren't 10,000 cases a day in Florida. And so I'm sure that has an impact on Disney and Universal in Orlando. So it does have an impact, but it doesn't have a level of impact where things would backtrack that we go back to numbers in May or April, no, we're not seeing that. So, but I'm sure, if Florida is not able to control this, you will have, on the margin, it will be a slower recovery. The health care has to, health care situation has to improve. And if it really gets out of hand, then it might even go back. But we're not, I think the numbers, we monitor these very, very carefully, as you can imagine. There is some optimism. The numbers aren't getting worse over the last 10 days or so. And, not just in Florida, but I think all the other, the Sun Belt states have got hit hard in the last month are beginning to curb their numbers somewhat.
Steven Duong
Right. Yes. I guess it's always hard to try to assess the economic impact when these cases rise. And so just trying to separate, understand the news versus what you guys are actually experiencing.
Leslie Lunak
What I can tell you that on the street, and again, to Raj's point, it's anecdotal. Tom gave you some really good data point. Things we're hearing from our customers, I will tell you that I belong to an organization called the Florida Institute of CFOs. And when we meet and we speak, and this is CFOs from businesses all across the state of Florida, all different types of industries. Yes, it's an optimistic group. None of these people are saying, oh, my gosh, I think my business may not survive this. You're not hearing that at all. People are out and about. People are doing things. The level of activity is strong. Now all of that is anecdotal, not statistical, but...
Raj Singh
Now all we have to do is get the Floridians to wear mask and everything will be fine.
Steven Duong
Well, just a final one on this one. So let's just say like we were just to make some risk assessment here. If the economic recovery stalls, flat lines through the end of next year, would you, would that make you reassess your reserves?
Leslie Lunak
Well, of course. I mean, I kind of outlined for you what the assumptions are that are built into our economic forecast. If unemployment doesn't go down between now and the end of next year and GDP doesn't improve, and if nothing improves, if there is no trajectory of recovery whatsoever, then yes, of course. Not only ours, but everybody will have to reassess their reserves. If there's no, I have not seen one bank because economic forecast thus far does not include some trajectory of recovery from here. So if that is erased, sure. I think everybody will have to reassess the level of their reserves.
Operator
Our next question comes from Chris Marinac with Janney Montgomery.
Chris Marinac
I can keep this brief since it's been a long call. I know there was a lot of focus on Slide 15 on earlier questions. I was wondering if we shouldn't just focus on Slide 10 in addition to 15 just because your capital is strong and...
Leslie Lunak
No. [indiscernible] real quick, Chris.
Chris Marinac
But Leslie, doesn't that play into the whole reserve and capital question? I mean, to me, together, they're very strong, and it really gives you a buffer because of the CECL. I think you gave a lot of discussion about that. Just kind of want to ask that question on how you think of capital here.
Leslie Lunak
Yes. No, absolutely. I mean, I think Slide 10 is very illustrative that our reserves sits at 60% of 2020 DFAST, severely adverse losses, which a lot of our specific peers don't disclose because we're no longer required to run that DFAST scenario and disclose the results publicly. But when I compare it to what's coming out for the regional banks, our reserve is sitting at a higher level of 2020 DFAST severely adverse and a lot of the regionals are sitting at. You can see clearly that even with that level of stress, you can see where our capital ratios land and it's comfortably above any well capitalized minimums. So I think there's plenty of capital cushion even if we have further stress from here.
Operator
I'm not showing any further questions at this time. I would now like to turn the call back over to Raj Singh for any further remarks.
Raj Singh
Thank you, everyone, for giving us time. Sorry it took so long. I wait for the day that we could go back to normal and don't have to provide a 50-page deck. But until things get normal, these calls will be a little bit longer, our disclosure will be a little more extensive. But appreciate you spending time with us, and we'll see you in three months. Thank you.
Operator
Thank you. Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.