Please be advised that this conference call is being recorded. Good morning, and welcome to the BMO Financial Group Q2 2020 Earnings Release and Conference Call for May 27, 2020. Your host for today is Ms. Jill Homenuk, Head of Investor Relations. Ms. Homenuk, please go ahead.

Thank you. Good morning, and thanks for joining us today. Our agenda for today's investor presentation is as follows: We will begin the call with remarks from Darryl White, BMO's CEO, followed by presentations from Tom Flynn, the bank's Chief Financial Officer; and Pat Cronin, our Chief Risk Officer. We have with us today Ernie Johannson from Canadian P&C; and Dave Casper from U.S. P&C; Dan Barclay is here for BMO Capital Markets; and Joanna Rotenberg is here for BMO Wealth Management. After their presentations, we will have a question-and-answer period where we will take questions from prequalified analysts. [Operator Instructions]. On behalf of those speaking today, I note that forward-looking statements may be made during this call. Actual results could differ materially from forecasts, projections or conclusions in these statements. I would also remind listeners that the bank uses non-GAAP financial measures to arrive at adjusted results to assess and measure performance by business and the overall bank. Management assesses performance on a reported and adjusted basis and considers both to be useful in assessing underlying business performance. Darryl and Tom will be referring to adjusted results in their remarks unless otherwise noted as reported. Additional information on adjusting items, the bank's reported results and factors and assumptions related to forward-looking information can be found in our 2019 annual report and our second quarter 2020 report to shareholders. With that, I will hand things over to Darryl.

Thank you, Jill, and thank you for joining us this morning. We always value the opportunity to connect with you and particularly do during this extraordinary time. The global COVID-19 pandemic is having a profound and deeply personal impact on all of us. We want all of our stakeholders to know that as one of the largest banks in North America, we feel a clear responsibility to do our part to support the health of the economy. We're working together with policymakers, customers and other stakeholders to develop solutions to the complex challenges presented by the pandemic. The measures taken by governments and central banks to support the overall economy, individuals and businesses have been significant and unprecedented. For BMO, the strength and the clear momentum that we had going into the crisis means that we can provide the needed support to employees, to customers and communities and to ensure the operational and financial stability for the long-term benefit of our shareholders. Our top priority has been and continues to be the health and safety of all of our employees and customers, while maintaining critical banking services. We mobilized quickly to transition 90% of our non-branch employees to work remotely. We implemented strict safety procedures to protect those who need access to physical locations. We're acutely aware of the uncertainty and financial concerns our customers are facing, and we're committed to helping them navigate these challenges. We've provided personalized advice and access to a variety of flexible relief options, including payment deferrals for over 200,000 customers in Canada and in the United States. As of last week, we facilitated over $2 billion in funding under the Canada Emergency Benefit Assistance Program and over USD 5 billion in loans in the U.S. SBA Paycheck Protection Program, helping 75,000 businesses to keep…

All right. Thank you, Darryl, and good morning, everyone. My comments will start on Slide 9 with the highlights of our financial results for the quarter. Despite the challenges of the environment, our preprovision pretax earnings and our operational performance remained resilient. From a balance sheet perspective, we are also in a very good spot with strong capital and liquidity positions. Q2 reported EPS was $1, and net income was $689 million. Adjusted EPS was $1.04, and adjusted net income was $715 million, both down from last year, primarily due to higher provisions for credit losses, which drove 85% of the decline. Preprovision pretax earnings were approximately $2 billion, down 5% and comfortably absorbed the impact of higher PCLs. Adjusting items are similar in character to past quarters and are shown on Slide 29. Turning now to revenue. Second quarter net revenue was $5.5 billion, down 3% from last year. Higher revenue in P&C businesses from strong loan and deposit growth was offset by lower revenue in our market-sensitive businesses. Net interest income of $3.5 billion was up 12% or 10% in constant currency, driven by higher deposit and loan balances. Net noninterest revenue was $1.9 billion compared to $2.5 billion last year with the decline largely driven by lower trading, insurance and market-related fee revenues. Expenses decreased 2% from last year and reflects disciplined expense management. The provision for credit losses was $1.1 billion, and Pat will speak to this in his remarks. Moving to Slide 10 for capital. The capital position is strong and well above regulatory requirements. The common equity Tier 1 ratio was 11%, down from 11.4% in Q1. The change in the ratio reflects higher risk-weighted assets in the Clearpool acquisition, partially offset by the expected credit loss provisioning adjustment and other smaller net positive…

Thank you, Tom, and good morning, everyone. The current COVID-19 pandemic has had a meaningful impact on all of our risk types. With that said, we went into this crisis in a very strong risk position with a long track record of successfully managing risk through challenging times. As such, we expect to navigate the risks of the current crisis successfully and continue to serve our customers across all businesses. The most evident COVID-19 impact in the quarter is on our provisions for credit losses. As shown on Slide 17, our total provisions for credit losses were $1.1 billion or a provision rate of 94 basis points, significantly higher than what has been seen in recent quarters. The total provision was made up of a provision for impaired loans of $413 million or a provision rate of 35 basis points and a provision for performing loans of $705 million. Based on our estimates, we see approximately 1/3 of the impaired provisions being related to COVID-19 impacts in the quarter. As per Slide 17, this increase in provisions on impaired loans was due to increased provisions in Canadian P&C and in Capital Markets. In the Canadian P&C segment, the $212 million of losses were driven by elevated consumer losses, with the increase versus Q1, largely related to the impact of COVID-19, and elevated commercial losses, in part related to COVID-19 as well as one larger credit loss that occurred before the pandemic and that involved fraud. In Capital Markets, the PCL of $73 million, an increase of $20 million versus Q1, was driven by continued stress in the oil and gas markets, exacerbated this quarter by COVID-19-related demand declines for oil. In addition, Capital Markets had a larger PCL in the apparel retail sector, largely related, again, to the impact of…

[Operator Instructions]. We have a question from Ebrahim Poonawala from Bank of America.

I guess first question on just on credit, Pat. If you can talk to us around the reserve coverage? So when you look at Slide 18, one, like a comment you made about future performing PCLs, like why is 49 basis points reflective of this macro backdrop? And why is that enough? Because we've already seen some sort of concern from investors whether this stack ranks lower against your peers and whether this might be enough, given some of the COVID sector and exposures that you outlined? So I would appreciate some thought process around why 49 basis points is enough? And why we shouldn't expect a meaningful increase in that number going forward?

Sure. I think I caught most of your question. I would start with -- I don't know if you suggested that there was a meaningful increase coming in the provision. I wouldn't say that at all. The performing provision, we would expect to come down from this level in -- starting next quarter. And so we don't see an increase coming in the quarterly provision. With respect to the appropriateness of the coverage, obviously, we spend a lot of time thinking about that appropriateness. And I think, first of all, you can look at it on an overall coverage basis. If you look at the 2 -- roughly $2.4 billion of the total allowance, we think some -- a good way to look at it is versus trailing 4 quarters of specifics. And that gives us about 2x coverage of trailing 4 quarters, and that includes two quarters of, as you know, significantly elevated impaired PCL or you could even look at it based on the current quarter annualized. So a fairly stressed impaired quarter annualized for 4 more quarters. And that's still about 1.5x coverage. And so I think when you compare that to the peer set, you'll see that we're actually pretty much in line with most of the group. And then we also look at it by line of business. And I think for me, I look at the closing GA in terms of basis points and how that compares to the historical range of specific loss rates. And if you look across every business, U.S. commercial is a good example. We have 49 basis points of coverage now in U.S. commercial. The historic impaired rate of that business and for the last 6 years has ranged between 6 basis points and 27 basis points. So we think we're actually quite well covered in that sector. And then I think we look at the experience even this quarter in U.S. commercial, we actually saw the non-TF specifics go down this quarter, including incorporating some specific impacts from some of the sectors you talked about that are impacted by COVID. So I guess two comments. We're not really seeing significant stress yet in the U.S. commercial book. We think it's a really well put-together book that can withstand a lot of stress. And at 49 basis points compared to the historic loss rate of that segment, to us, that feels like adequate coverage. And I think if you look across all of the business segments and compare them to their historic loss rates or even some version of a stress loss rate like we're seeing now, you'll see that we're over covered in every single one of the business segments.

Got it. And just on...

Ebrahim, sorry, it's Jill. I'm just going to have to jump in with apologies, but I think we better keep it to 1 question just because we all know we have a hard stop at 8:15. Okay. Thank you.

The next question is from Scott Chan from Canaccord.

Pat, in the Capital Markets side, you talked about trading losses on separate -- several days on certain segments or sectors. Can you talk about those certain sectors? Is it -- the sectors that are more kind of commodity driven? Or is there something else kind of that was a bit unusual during the volatility?

Yes. Thanks for the question, Scott. I think what I'll do is maybe I'm going to ask Dan to make some preliminary comments on this. And if I have anything to add, I'll jump in after that.

Sure. Thanks for the question, Scott. I think as you'll appreciate, the quarter was a fascinating quarter where the early part of the quarter we had some adjustments as COVID came in. The two weeks in March were extraordinarily volatile. And then subsequent to that, in April and now going on into May, we've actually had, I would call it, exceptional results. In that period of high volatility, we had a couple of places where we had such extreme market dislocation around historical patterns that we had some exposure. The first is what we call our equity-linked note business. And then the second is how we're covering off the hedging of our exposure around our derivative book, which you will have known as ICA. So those are the two places that we work through some of that dislocation to the marketplace.

The next question is from Gabriel Dechaine from National Bank.

Just on the energy provisioning. I was a bit surprised that the -- is $54 million of specific provisions. And then thankfully, the -- you gave some additional disclosure on the provisions on performing. Similar question to Ebrahim there, given all the issues that we're seeing in that sector, why does that number makes sense? I'm just like devil's advocate here.

Yes. No, sure, that's a great question. And we think with that rate of coverage, we size it relative to a couple of things. I mean, first of all, that's well above the current loss rate we've seen on specifics over the last couple of quarters. And as you know, the sector is already quite stressed and that specific PCL rate has actually been pretty stable in the book for the last 3 quarters. And so we've had a pretty good picture of what stress looks like. The other thing you can look at is what the level of stress we would have seen in our book back in 2015, '16. '16 would have been the worst year during the last downturn. And there, we averaged about 150 basis points of loss in that year. As you know, we recovered a fair bit of it later, but that was the rate in that quarter. So that kind of gives you another bookmark. And then lastly, obviously, we do a ton of stress testing on the portfolio. And I would tell you that a $35 flat for 3 years stress test. And keep in mind, this is just a stress test. There assumptions baked into it. But our best guess of a $35 flat environment for 3 years running would be roughly about a 2% loss rate per year in that portfolio. And so that kind of gives us some sense that at 2.5x -- or 250 basis points of coverage or 3 plus, if you exclude pipelines, is a pretty reasonable number for a sector and actually would incorporate higher levels of stress than we've seen in some of the more stressed periods in the past for that sector and is quite consistent with our stress tests.

When you do take the specifics, are you seeing -- are you looking at when an actual account goes bankrupt or whatever? Are you doing any -- I think these guys are going bankruptcy, so I'll take a specific now?

No. We would take a specific provision on every oil and gas account that is currently classified as impaired, and that provision would fully reflect our complete expectation of what the loss would be. So everything in the oil and gas impaired portfolio right now has a specific provision that's been taken on it, if it needs one. I would remind you that a huge chunk of that oil and gas portfolio is reserve based. For example, we had a fair -- one very chunky addition into the impaired oil and gas portfolio this quarter. In fact, it accounted for almost 2/3 of the increase -- of the formations we had in the sector this quarter. And given the reserve base nature and the significant amount of junior debt that's below us, we actually don't expect to take any provision at all. And so that also gives us some comfort. And the other thing to keep in mind, too, is a lot of the producer clients that we deal with, particularly in the United States, are hedged, and a lot of them are hedged to the balance of the year. So they can weather quite well short-term storms. We'll see as prices play out. But in the short term, given that plus the reserve base nature of our lending, plus our stress testing and our own historic experience, we feel pretty well covered in the sector.

The next question is from Steve Theriault from Eight Capital.

Could we go back to Capital Markets for a moment? And just the equity-linked note marks. It didn't look like these products really rear their head to the equity line at the last time. I appreciate it wasn't as adverse, but when I go back to '08 and we had a big sell off. Maybe can you talk about the differences there? And would you expect that to come back linearly as the market comes back? Or it's come back in -- since quarter end somewhat? Or is it a situation -- I think maybe it was Tom earlier mentioning that you've taken some risk off the table in some spots. So is that -- does that come back with the market?

Yes. Thanks for the question, Steve. I think the piece I would give you is there was extreme volatility as we moved through that 2 weeks in March. And so as you think about some of that normalization, you're right, we will get some of that back over time with that normalization. But also as we went through there, we were doing dynamic hedging. And some of that dynamic hedging will then be permanent in terms of a realized loss. As we think about going forward, we've done a lot to take risk off the book, both in terms of market downside protection making sure that we have matched our term volatility profiles and then our exposure between different marketplaces, so I think Canada versus the U.S. and have worked very, very hard to take that book out and look at it today as being relatively balanced, strong and immunized. And again, I do think there'll be some recapture as we go forward. It's a strong business for us and has been and will continue to be so.

Is that a dynamic hedging program for that book new or just enhanced?

It's enhanced from where we were. And if you think we've taken a more conservative view of our risks. And so we're managing the business that way.

And if I look at the trading days in the appendix, the day that had the $180 million, I think I'm reading that right, loss date, is that primarily the -- this book? Or is that a combination of things?

In every one of those lost days is a combination of many things. If you think about the breadth of our positions. That day, in particular, was a combination really of the equity-linked notes as well as ex-VA. If you remember, that was the day of extreme market dislocation, just before the Fed came in and put in place its positions with dramatic asset sale prices moving and dislocating. And so you saw some of the volatility there to the negative. You see the volatile positive after that as markets started to normalize. But I think of it, in my mind, is about 2/3 for the equity-like notes and about 1/3 for the ex-VA.

The next question is from Meny Grauman from Cormark Securities.

As I'm thinking about sort of where PCL ratios peak and also going back to the discussion of the adequacy of coverage on the allowance, I'm wondering how appropriate is it to look back. You referenced sort of the past few quarters and years, but if you go back further to past deeper recessions in Canada, the early '80s, early '90s, how appropriate is it to look back to those historical examples to kind of gauge the magnitudes? And I guess government support is probably one factor, are there other factors that would make that comparison not the best in your view?

Yes. Thanks for the question, Meny. I think you answered your question a little bit right there. I certainly wouldn't discount the possibility that loss rates could end up being higher than what we factored into the performing provision. I mean when I think about impaired going forward, we're running this quarter at 35 basis points. That's clearly a rate lower than what you would have seen in the financial crisis or what you would have seen in prior recessions. But as you noted, there are some things that are quite a bit different this time around. Not the least of which, as you mentioned, there's some very, very substantial and quite targeted and direct stimulus provided by the government. The second thing I would suggest is loss rates tend to be very much a function of duration of the economic stress. And in prior recessions, you could think of those as running 2 to 3 years before we started to see recovery. In this one, just given the unique nature, most of the economic consensus forecast would see a recovery much, much faster than that. And think about stress on corporates or consumers, a lot of them, particularly when you get to higher-quality credit folks like ours can withstand short periods of stress, but things start to get nonlinear in terms of PCL as that duration goes longer. So I would say the risk to our forecast is if a sharper correction or a correction that is anticipated by the consensus turns out to be much longer than we expect, then you're going to see some upward pressure and likely to push up into some of those more stressed loss rates that you're talking about from prior recessions. The other thing I'd highlight, too, is mix has changed over the course of time as well. Our CRE portfolio is a great example. We had fairly elevated CRE losses during the last recession. That book looks dramatically different today than it did back then. And so when we run a stress test today, on that portfolio, we see very different loss rates than what we would have seen in '08. So I would say those 3 things: mix, duration and government stimulus would likely see loss rates lower. I'll caveat all that with we're in dramatically uncertain times. And so that forecast could be different. But I think about loss rates as from 35 basis points where we are today, they're likely to drift up higher, I think, likely into the 40s. The other end of that bookend would probably be at the 70 basis point range, which would be closer to what you saw during the GFC. So hopefully, that answers your question, Meny.

Yes.

The next question is from Doug Young from Desjardins Securities.

Questions for Pat. I think Pat, in your prepared remarks, you talked about just the build-out of performing loans. And I think you mentioned and I didn't catch it all, so I'm hoping you can flesh it out. The weaker outlook accounted for most of the build and change in scenario, weightings, models and whatnot that all netted out. So I'm hoping you can kind of just flesh that out a little bit more and talk a bit about -- more about the process? And maybe if you can give some weightings in terms of how much was driven by the scenario changes versus the forward-looking indicators and whatnot?

Sure. So as I said in my comments, the single largest factor by far was the macroeconomic variable change. So of the $705 million provision, that can -- that's the size of the macro impact was about -- was actually slightly larger than that. In addition, we had some balanced growth in there, which contributed about -- call it, about $40 million credit migration, with a little bit of credit migration for about the same amount. There is some FX in there as well that contributed about $50 million. So those were the things pushing upward on the provision. We had some model changes, which are pretty normal course in any quarter. That adjusted it down by about $60-ish million. And then we did change our scenario weights in the quarter because we're -- our base case is actually much, much closer to what our prior adverse case looked like and we're now clearly in a downturn. We felt that the probability of a benign scenario was about equally weighted to the adverse scenario now, just given that we're right in the middle of a downturn or entering into it. And so that scenario adjustment actually reduced the provision again by about $50 million. So those other things all netted out to about the impact of the macro change, and that's why I made those comments at the outset.

So the modeling reduced the performing loan PCL? And I guess the scenario weightings because you're not going to expect a downturn on top of a current downturn. And so that's the -- what you're thinking in terms of scenario. But when you think of the modeling, what did that relate to?

Yes. That's probably more detail than I think I want to get into on the call. There's a lot of technical things that we -- we are constantly evolving our models every quarter. We are always looking to improve the efficiency. And then sometimes we'll add sector-specific models where we don't have them and we are in. So these things tend to be quite small. This quarter was a little bit larger than you might normally see. But model changes are completely normal course. We have them fairly often, as I'm sure most other banks do as well.

The next question is from Mario Mendonca from TD Securities.

Can I just ask one quick clarification, first? Pat, when you talk about performing loan losses declining, but impaired increasing. Have you -- can you offer an outlook on just the total PCLs ratio looking forward? Could the impaired increase sufficiently such that we're looking at an impaired total PCLs ratio to a level similar to Q2 in subsequent quarters, Q3 specifically?

Yes. I would say, if you're thinking about totals, our current expectation would be that it would not be at the same level as what you would have seen this quarter. Beyond that, it gets pretty hard to predict. As you can imagine, the performing number, there's lots of moving parts in that, that can change from quarter-to-quarter. And so that one's harder to predict. But -- and based on where I sit today, I don't see a large amount of upward pressure on the impaired provision. And so that gives me some confidence that the net of those 2 things will be a lower total performing -- or a total provision number. But just given how much uncertainty there is in the market, I'm a bit low to try and pinpoint a specific number in terms of loss rate for you at this point.

Okay. The actual question I wanted to get to then was, Pat, you talked about the duration being very, very important to the trajectory of credit losses. When you think about setting your performing loan loss reserves, can you talk about when in your models, things are normal? What -- by that, I mean, unemployment returns to a pre-COVID level. GDP activity is essentially back to a pre-COVID level. Can you offer some commentary there?

Yes. I think first of all, I'll caveat with that by saying that those things are obviously important, but there are many, many other variables that go into the model, BBB spreads, levels of the VIX. So all of those things can move around, enhancing or offsetting some of the movement -- the move back to normal that you're talking about. And then beyond that, I would actually just simply point you, unfortunately, to the MD&A. We have some pretty good disclosure there about what we see. You'll see, for instance, in Canada, GDP declined this year of roughly about 6%. And then in 2021, rebound to 6%. Just the way the math works that doesn't get you back to flat by the end of 2021. But it gives you a sense of how we're predicting the recovery there. And then you'll see similar patterns for the unemployment rate as well.

The next question is from Nigel D'Souza from Veritas. And it is the last question.

So I actually wanted to follow up a bit on the macroeconomic variables for your performing loan loss modeling. And if I could turn to Page 32 of your shareholders' report, where you outlined your updated forward-looking indicators. [Technical Difficulty] have turned a little bit more positive, so you're expecting a higher home price depreciation, your base case scenario and in your adverse scenario, you're not expecting or forecasting a decline in home prices. So I was wondering if you could just provide more color on your thinking there and maybe also touch on what's the sensitivity of your performing loan allowances if in the future you did forecast a more material decline in the HPI?

Yes. And unfortunately, you cut out there. I don't know if it was on our end or yours, but I missed a good chunk of your question, actually. But maybe I'll try and answer it based on what I heard. I think you can -- we actually show the adverse scenario there on Page 32. So you can get a sense, and we also disclose if we had 100% weight in the adverse scenario. So that just gives you some perspective. It's about relative to the allowance today. It's about a $600 million increase in the allowance if we move to our 100% adverse case, and so that just gives you some perspective. And in there, we obviously don't -- or on Page 32, we don't disclose all the variables that it would be associated with our adverse case, but that just gives you a sense of the sensitivity. And hopefully, that answers the question part of which I did not hear.

So if we have time, just for clarification, just the assumption on your home price outlook, it looks to be more positive now. So I was hoping you could just provide more color on forecasting a decline in real estate prices even in the adverse scenario?

Sorry, your question is why are we generally bullish on housing prices?

Yes, according to your forward-looking indicator, it looks like you're expecting prices to be flat or higher in your scenario ratings here?

Sure. Maybe I'll ask Ernie, she's very close to this market segment, and so she's probably got some very good color.

Yes. Thank you for that question. What we're thinking on forecasting related to housing prices is basically what we're seeing now is some stabilization in certain sectors in certain geographies. And so I would say, right now, our position would be that we're going to be remaining quite flat on the housing prices. Obviously, in oil-affected regions, et cetera, we're going to see some nuances there. But for now, that's what our indicators are suggesting.

I would now like to turn back the meeting over to Mr. Darryl White.

Thank you, operator, and thank you to everybody on the call. I'll just wrap up with a couple of summary thoughts. In terms of bringing us back to our core messages, I would remind you that we are extremely proud. I'm extremely proud of the strong response our teams have had through the crisis. We're very comfortable with our capital and liquidity position, which drives the stability of the bank, the PPPT at $2 billion despite the softness in our market-sensitive businesses is very substantial relative to the prudent provisioning that we've got in our PCLs, and we're very confident in the outlook. So with that, operator, I will leave the call there, and I wish everyone continued health and safety. Thank you very much.

Thank you. The conference has now ended. Please disconnect your lines at this time. And we thank you for your participation.