Popular, Inc. (BPOP) Earnings Call Transcript & Summary

All right. Thank you. Good morning, everyone. I'm [ Jerry Benson ]. And up next, we have Banco Popular. Banco Popular is a full-service financial services company with operations in Puerto Rico, the United States and the Virgin Islands. It's the largest banking institution by assets and deposits in Puerto Rico. And at September 30th, it had about $70 billion in assets. And presenting today for Banco Popular is Carlos Vazquez, the Executive Vice President and Chief Financial Officer. He joined Banco Popular in 1997, was President of Popular Bank in 2010 or named in 2010 and has been EVP and CFO since 2013. And prior to joining Banco Popular, he worked at JPMorgan for 15 years. So with that, please join me in welcoming Carlos.

Thank you. Good morning.

Good morning. And this is a fireside chat, but I'm going to pause everyone in a while. And if any audience members has a question, please raise your hand and feel free, don't be shy. I'd like to start off with recent events. We had Puerto Rico was hit by hurricane Fiona back in September. I guess, first off, how is everyone in terms of safety and how did they endure the storm?

Yes. Fiona is a very good place to start because I think everybody is suffering a little bit of PSTD from Hurricane Maria. So when Fiona hit the island and the next set of news hours later was that the island was without power. I think everybody rushed to a very quick assumption. This is the same as Maria. And hurricanes are always unfortunate, don't get me wrong, but it wasn't anywhere close to the same as Maria. We were a bit lucky this time. Fiona was much less storm of much less strength. Hurricane Maria hit the whole island, it went like through the middle of Puerto Rico. Hurricane Fiona, I just touched the southern part of the island, mostly the Southeast, and the Southeast is part of the island where there's less economic activity and actually less population. So I don't want to minimize the effect. A lot of people have suffered a lot as a result of the hurricane, but it was nowhere close to the magnitude of Maria. To give you one example, when Maria hit the condominium where I live, they didn't get power back for 3 months. In the case of Fiona, we had power back in 5 days. So it was very different. So we're very, very lucky that way. We're trying to help a lot of our clients who did suffer some damages. We're trying to keep them help. But the best example, I think, of the difference is that we set up a whole call center as with Maria to make sure we get the calls from our clients that are requesting assistance. And we ended up sending them home very quickly because the number of calls we got from clients requesting assistance was a fraction of what we had experienced in Maria. So that's probably the best example that the effect on the island -- on our clients was not material. We don't think it's going to be a big, big effect on credit either given the number of clients that have called us that they need assistance with their loans. And actually, it has a good chance of having a perverse benefit, the fact that were hit by Fiona. And the perverse benefit is that as a result of Fiona, we got a lot of visits from very high-ranking officials in Washington into Puerto Rico. And when they left, the biggest thing they took away from their business is that, wow, those funds from Maria are still very slow in showing up. So they went back to Washington with that in their mind, and that has already started to result in changes. To give you one example, a number of the different FEMA programs, the way they work is that they will fund the construction of a project, call it, $1 million, but the beneficiary of those funds has to put up 20% of the funds. So there were tens of projects in Puerto Rico that have been approved. They're ready to go -- funding was available but the beneficiary that could have been municipality or not-for-profit or some other institutions, they didn't have the 20% to get it going. So the project never started. So as a result of Fiona, among other things, FEMA started a program of what they call advances, where they've actually advanced the 20%, so the project would get going even if the beneficiary didn't have the funds to contribute to the program. So in a weird way, I think we may have benefited from Fiona in that there's a lot more attention in Washington to the speed at which the funds are showing up. The funds have already started showing up at a much better speed before Fiona, but we expect and we hope that will continue to go up as a result of the perception of high-ranking officials taking back to Washington. Another example is the Secretary of Energy, has been in Puerto Rico twice since September. I don't remember the last time Secretary of Energy went to Puerto Rico before that. So I think those are both positives. And they contribute to what I think is the thesis of many of our shareholders and our investors on the bank, and that thesis is anchored on a view that despite the fact that the Puerto Rico economy is linked to the U.S. economy and the macro things that are going to drive the results are similar, meaning interest rates, meaning overall economic -- national economic growth, that Puerto Rico has some idiosyncratic things that are affecting the performance of the economy. And this -- these hurricane assistance funds, mostly the ones coming from Maria from 4 or 5 years ago, are that difference. The expectation is that the funds that will continue to be deployed are probably in the mid billion, $4 billion or $5 billion a year, and that will roll on for 5 or 6 years. And for an economy, our size, an investment of $5 billion, $6 billion a year is actually quite material. So most people believe that while the Puerto Rico economy will continue to be linked to the Mainland, that it will be linked with that one important differentiator, which is that the Puerto Rico economy should do slightly better than whatever your outlook is from the mainland economy.

The 20% advance just curious, is that a loan? Or is that an actual grant?

I actually don't know how it works. That is a correct question, [ Jerry ]. I don't know the mechanics of how it works or they give them more time to fund it. I don't know how it works, but it will get some more programs going, which is the important part. The other thing to keep in mind on hurricane relief is I think there's -- everybody is looking for this one project of $2 billion that we can all go visit it and see it and put our hard hats on and walk around this project to be sort of the anchor of the reconstruction. At this point in time, there isn't such a project. I mean, there will be a lot of that, that happens when the reconstruction of the electric sector gets going, and that's taken some time, in part because of its complexity, we have to redesign the electric sector and then start the rebuilding. But what we have is not this on $2 billion project that everybody can go see. It's a 200 projects of $1 million or $5 million are going around and a lot of those are already happening. So we are seeing the benefit of the flow of funds, but it is harder to see it in action because it's a lot of smaller projects as opposed to one very large project that everybody can hang their hat on.

Okay. And this $65 billion still the aggregate amount that is earmarked for hurricane assistance?

Yes. It's in the mid-60s of which probably about 1/3 has actually been disbursed. So we have something in the ballpark of 2/3 of that still in process, and that is what would fund this $4 billion, $5 billion, $6 billion a year for many years to come.

Okay. And just one more question on it in the electric grid and the reconstruction is there -- I know there's a number of hurdles that need to be overcome first, but is there an estimated time line on... ballpark?

Most of the estimates I've heard are -- it's going to take a decade.

A decade.

Yes, to move it from where it is to where everybody wants it to be. They going to put that in the right context. We have a central generation integrated electric system in Puerto Rico. That was assigned 50, 60 years ago. It was implemented 50, 60 years ago, and it's probably very badly maintained for 30 of those 60 years. What most people seem to think is the answer now will be a system that is distributed where there isn't a central points of generation that we have now that are rather old, inefficient and dirty and most of the electricity in Puerto Rico is still comes out of oil and not particularly clean oil at that. So the idea will be that you have the distributed generation -- distributed system, such that we do not have a situation we have now that when generation goes, the whole island goes dark. And then you'll have more solar, you'll have more wind. We'll have cleaner inputs into the system that should be very helpful as well. By the way, the answer is going to be a mix of all these things. People like to think that green is the answer for everything. We can tell you for a fact that during very bad hurricanes, the solar panels are blown away and the windmills are destroyed, okay? So the answer has to be a mix of all these things. It's probably going to be first a shift from oil to natural gas and more renewable, more solar, more wind. As battery technology gets better, that will improve the usage of things like solar. So it will take a while. It will be a mix of all these solutions that will lead us to a better place.

I'd like to shift gears now. Year-to-date, Banco Popular has generated some real solid loan growth. How would you characterize the current environment for loan demand in Puerto Rico?

We're quite pleased actually. At the beginning of the year, when we were asked the question on loan growth, we expressed a view that we expect the net loan growth to start happening in the second half of this year. And a lot of that was driven by additional loan demand, but it wasn't big enough to overtake the repayment of PPP loans that were still happening and the normal runoff of things like our mortgage -- resi-mortgage portfolio, which is very large. Luckily, we're wrong on that one. And we started seeing net loan growth in the first quarter of this year. That continued in the second. The third quarter was particularly strong. We grew loans by about $1.2 billion. Now the third quarter was the beneficiary of some unique situations. First of all, we did close a very large commercial loan that contributes to the balance in a single transaction. And we also had lower repayments and prepayments than we expected, so ended up being a particularly good quarter. So I always make sure I tell people not to pencil in another $1.2 billion for the fourth quarter. But we are seeing very strong demand from our clients and a lot of economic activity. So we still think the pipeline is quite strong. So we are hoping for continued loan growth. One of the things that is good about the loan growth is that it's very broad-based. While it's driven by commercial loan growth, both in the Bank of Puerto Rico and the bank in the Mainland, all our loan categories grew last quarter. And I don't remember the last time we could say that. Our personal loans grew, our credit card balances grew. Our auto loan portfolio grew, our mortgage loan grew as well. Commercial grew both in the States and in Puerto Rico. So it was very broad-based, which is a good signal that the economy is still strong. The consumer is still strong. There's a lot of commercial activity, and that bodes well for the future. In the mortgage, the structure the same as in the U.S. It's exactly the same. In the case of Puerto Rico, if anything, it's even more tilted to 30-year fixed-rate mortgages. We don't really have many of the variable rate mortgages that are -- you can get in the states. So it's the same structure. The main reason our mortgage book grew for the first time this quarter in many, many quarters is that we were selling a lot of what we originated, FHA loans, and we decided to retain in portfolio now those loans. So we're not selling them into the market anyway. The net interest income from those actually is better than the fees we could get from selling them. So that made the mortgage portfolio grow. But we're seeing the same pressure in Puerto Rico as in the rest of the country. The housing market is slowing down quite a bit. And it's not the same -- 7% mortgage is not the same than a 3.5% mortgage, so it is affecting a lot of the buyers. Now in Puerto Rico, we still have a shortage of housing because we didn't build housing for a whole decade. And there's still a lot of household formation. So there is some fundamental demand for new housing, especially in the lower levels of values that will continue for a while.

Okay. And how does -- household loan growth demand or loan demand differ in Puerto Rico versus the U.S., if it does at all?

Yes. In Puerto Rico it is a lot more broad-based because we have more of a consumer business in Puerto Rico. Our U.S. bank is very much of a commercial bank. So in our U.S. Bank, the demand continues to grow mostly in our niche businesses. We have a condo association lending business that we lent to the associations that run the condos, not the condo owners themselves, that is very successful. And that continues to see a lot of demand. In fact, the very unfortunate accident in Florida when the condo came down has, if anything, accelerated demand in that sector because a lot of associations that may have been putting off some maintenance and things like that, can't do that anymore. So there's some additional demand there. We have in the mainland, what we call health care, which it really is in health care. It is mostly nursing homes and managed care facilities. And that business continues to grow very nicely as well. And then the normal commercial that you will see in the bank in New York and Florida, which is some CRE, and that's where our stuff that also has good demand. Again, in Puerto Rico is much more broad-based. So we'll see commercials in most segments, and then we'll also see growth in the consumer, auto, credit card, person loans as well.

Yes. I'm going to go back and just ask one follow-on question. Interesting comment about the household formation being strong. Has the whole work-from-home phenomenon in the pandemic create upward pressure on that. I'm just curious, people come from the Mainland over to Puerto Rico?

Puerto Rico has a very unique history on population. As you know, for more than a decade, our economy was contracting. Joblessness was very high and people were leaving the island. So for more than a decade, we lost about 2% of the population every year on average. That departure of people sold down a couple of years ago, and best we can tell, the information on population is not perfect. But best we can tell, population seems to be pretty steady now for the last couple of years, and that's a positive. Some people are coming back and working for Puerto Rico. It's very common with young people. A lot of the friends of my children are doing that. They were working in New York and sharing an apartment with the other people and saving no money. And then they figured out, you know what, I can work from home, it's free because my parents pay for everything, there's food in the fridge, laundry gets done. And I can have the same job and the same income I had in New York. So some of the young people came back. That's a nice story, but actually, the more interesting one is a lot of young people starting new businesses. So they either have quit their job and started the business, they always wanted to try out or start the business on the side. And if it takes off, then they eventually quit. And it's very encouraging to see a lot of the new businesses started by younger people because I think bodes well for the future.

Absolutely. Okay. Great. Let me let's shift gears a little bit and just the outlook for interest rates, obviously, has evolved in a dramatic fashion, actually.

More so this morning.

And I'm just curious how you characterize Banco Popular's asset sensitivity at this juncture. We were -- we tend not to try to place big bets on rates. And if you look historically no, -- and the reason behind that is that we're not sure we're any wiser than any of you or anybody else in this room as far as where rates are going to go?

So it's hard for us to add value [indiscernible] that. So historically, if the strong consensus rates are going up, we'll be slightly asset sensitive, and there's no consensus will be closer to neutral. And that's more the way we handled it. That is not the position we had at the beginning of this year. Beginning of this year, we had a huge cash account, mostly because we were expecting the restructuring of the government to happen in the first quarter of this year, which it did. And we were expecting when that happened, an outflow of about $10 billion of deposits from the government of Puerto Rico. That also happened -- so we had a very large cash position that means we ended up being significantly more asset sensitive than normal at the beginning of the year. We continue to be more asset-sensitive as rates went up. We sort of brought forward asset sensitivity into our earnings in the middle of the year roughly. And now we're probably pretty neutral as far as asset sensitivity.

Okay. And so -- and I think it was on the Q3 earnings call, you talked about actually seeing some NIM compression near term, but then getting a gradual expansion thereafter, what's behind that dynamic?

Yes. We have to separate -- we sort of have 4 buckets of deposits in the bank. Our U.S. bank and the deposit, dynamics of the U.S. banks are very much very similar to any other $10 billion bank in the East Coast. So they are a bit more sensitive to interest rates and the deposit costs. And in Puerto Rico, then we have sort of 3 buckets. Our retail deposits, our commercial deposits and our public deposits. Our retail deposits historically have had a lower beta than in the States, and that has continued to be true in this cycle. Our commercial deposits have also historically have a lower beta than the states. And so far, that wholesale has continued to be true in this cycle. And then we have the government deposits that have 100% beta with a lag of about 3 months. The big difference in our overall cost of deposits in this cycle versus the last cycle is that this time, government deposits represent about 27% of our total deposit. And the last cycle was probably 5%. But the dynamics of the other 2 buckets continue to behave pretty much the same, and it's part of the strength of the franchise. There is a very valid question that a lot of people are discussing of what's going to happen to betas ultimately when we see this whole cycle and the possibility that betas will end up accelerating as a result of the high rate of the high speed at which rates are going up this time. I think there's some merit to that thesis. I'm not sure it's going to play out, but it wouldn't surprise me as it does. If you look at our average medium commercial clients or retail clients, if rates move 25 basis points every quarter, it's probably not enough for them to take action and do anything. If they move 150 basis points 1 quarter and another 125 the following quarters, they're probably going to try to do something. So I think the speed does matter. Again, we haven't seen it very much yet in our betas to date, but we probably will see it. So at this moment, we continue to see the retail deposits with a reasonably low beta closer to history. The commercial operating deposit is probably the same. There's a bucket of commercial deposits that excess cash over the pandemic. A lot of our commercial plans accumulated excess costs much above their historical levels of balances. Those are probably at higher risk. Our good clients that have good treasurers are already calling trying to get better yields on that. And we'll treat those goals on a case-to-case basis. I know a very good client that has a very broad relationship with us, will respond to that request one way. Before you have deposits with us, we might be a bit more reluctant to respond to a request of higher rates. And then we have the government deposits that, as I mentioned, they move with the market, but with about a quarter lag. And those will have a big effect. And those -- that is one of the drivers of the margin coming down in the fourth quarter. We expect our roughly $15 billion balance of government deposits, the cost of that to go up 150 basis points in the fourth quarter.

Okay. In that we think about the government deposits that $15 billion, it has whatever, 100% beta on a lag. But how is the stability of it?

Well, we've gotten that balance very perfectly wrong for about 4 years. And I want to explain the reason why it's so hard to get a forward view of those balances that is accurate. We talk about public deposits in Puerto Rico and that bucket is roughly a bit less than $15 billion now. And we tend to think it's 1 client. And really, the central government is the biggest part of that. But that bucket of $15 billion is made up of something like 240 different clients. So we go to every one of them. We try to get their outlook on how -- what the deposit is going to look like, how they're going to use their funds, when they're going to use their funds. We add up all those 240 different outlooks and come up with our best guess of what's going to happen with those deposits. We have to be cognizant of the fact that some of those government entities themselves don't have very good visibility on what their cash flows are going to be. I mean the central government -- the economy is doing better than expected. So the central government is collecting more taxes than they thought, to give you one example. The Inflation Act is passed and all of a sudden, money shows up that they didn't expect to show up. So it's a bit hard for them to come up with exact forecast. Our best view right now is that, that summation of accounts will probably end up between 13 and $15 billion by year-end. The normal cycle of public deposits since it's driven by the central government is that they tend to gravitate up to the second quarter where tax collections come in, and then they tend to gravitate down from there the rest of the year as the funds are deployed by the government. That was the normal behavior in the past. That's been distorted quite a bit by pandemic funds by legislative funds, but all the special things that happen as a result of the pandemic, so it hasn't quite behaved that way. We'll see next year, we tend to go back to more normal patterns.

Okay. One last question on the balance sheet. I mean in this interest rate environment and likelihood of further rate increases, like how are you investing incremental liquidity? And what do you see as the optimal liquidity for Banco Popular?

Our investment portfolio is -- first of all, is very large. The influx of government deposits means that we've ended up with an investment portfolio that about half of our assets, that is not normal for us. Historically, we've been -- we looked like most of the banks in the high 20% or 30% of assets. So we have a much larger investment portfolio now than we historically have as a response of the growth in deposits. By the way, the growth of deposits have been in all segments. This is retail, commercial, but the government has been the biggest nominal numbers. We have a very conservative investment portfolio that's -- I'm not sure it's common in banks. Basically, we have 0 credit risk in our investment portfolio. It is almost all treasuries and agency-backed MBS'. And one of the reasons we would like to tell you that we're just conservative, but there's also other reasons why we came to this outcome, and one of the reasons is that for the bank in Puerto Rico, our investments in treasuries are tax exempt. So the risk-adjusted return is significantly better, obviously. So we have no municipals in our investment portfolio. We have no corporates. We don't have sub debt from other banks. We don't have any of those things. The investor portfolio is basically treasuries and mortgage-backed MBS'. Now it's still bigger than normal, and it's not our intent to look like a trust bank. But we so ended up there. Over time, as we now finally are seeing more robust loan growth. We hope to continue to deploy some of that into loan growth, which should be will be terrific. But we do have a much bigger investment portfolio than usual and more -- much bigger than most banks are size. So what that has also meant going back to your question on interest rates, is that we are a bit more exposed to the AOCI adjustment and maybe other banks may be on the tangible capital side. So it's something that we are paying attention to make sure we're comfortable with it.

Why don't we skip to that and why don't we talk a little bit about -- I mean, you just highlighted the reasons for having a little bit of an outsized securities book, but you also don't have to take any credit risk. You don't have any credit risk at a claim book. It gets interest rate driven. And obviously, there's been some significant marks for the industry, but I believe Banco Popular, it's been about 300 bps contraction year-over-year. I mean as investors, how should -- how do you think we should think about that? And how do you guys look at it internally? Because obviously, your CET1 is fairly robust.

Yes. Well, you've hit a very important point, [ Jerry ]. And there's different audiences that we have to think about when we look at this. Let's start with the most obvious one, which is regulators. Regulators are singularly focused on regulatory ratios, which doesn't include tangible capital, okay? And so far, our main regulator, by the way, is the Fed in New York. It is our main regulator. Historically, they have been singularly focused on regulatory capital ratios. And so far, we have not seen, or perceived any difference on how they are looking at the bank and the kind of things that they pay attention to. Now in our history with regulators, we have learned a lot of things as we went along. And we try to make sure we don't forget those learnings. And one of those learnings is that regulators don't care about things until they do. And when they do, then a lot of things can change. So the fact that they have not deviated from the historical focus on regulatory capital ratios doesn't mean that we shouldn't be attentive to tangible capital ratios because if at some point in time for whatever reason that we cannot predict right now or tell you the reasons right now, they decide to start paying attention, then it becomes an important thing to take a look at.

I mean, I guess we would have to compare it to maybe the GFC. Like what is it really credit that's going to be that inflection in the way they view TCE? Or what do you think it is? Or is there a certain level?

Right. Keep in mind also, regulators are different. So I can only comment in our relationship or discussions than we normally have with the Fed. We are not an OCC bank. I do not know how they react to things. We speak less regularly with the FDA, I'm not sure how they are reacting to things. But again, so far, we haven't seen any change on their focus. I'm not sure what will trigger it, to be frank with you. It could be a combination of credit, combination of other things. If they were to change their attention to other metrics. It's difficult to tell. I think maybe we'll be liquidity, maybe that will be more important to them. So -- but I'm not smart enough to talk for regulators. So I just try to understand where they are. And over time, we have learned also other lessons in our dealing with regulators. When the regulators are happy, life is good. So we try to do our best to make sure our communication is very broad. They know what we're doing. They know why we're doing it, so that there are no surprises. And when -- again, when they are happy, life tends to be good.

Right. Well, it's probably a good time to talk about I know the market was anticipating a capital management plan. And you guys delayed it out to, I think, midyear -- so maybe you could kind of share with the audience kind of your thoughts behind that.

Yes. I mean, we started our capital return program, I think, 6 or 7 years ago, something in that ballpark. And we've been managing it very actively since then and very successful, I think. What we've done this year, we announced -- let me put it in context, I'm sorry. We have a practice of trying to announce our capital actions for any given year in our webcast in January for the fourth quarter of the prior year. Now that is something that we just have chosen to do it that way. It's sort of worked reasonably well for us. So we've done most of the time that way. Nobody is requiring that we do it that way, but that's the way we manage this. So during this year, we announced our original capital plan for this year, which in January, which included a dividend increase and included a buyback of $500 million. And we started executing on that. We did an ASR for the first $400 million early in the year. And then in the second half of the year, when we did a transaction where we purchased a lot of assets from Evertec, our technology assets. We also sold our equity stake in Evertec that ended up resulting in a capital gain. We added another $130 million to our buyback this year. So we did $230 million in the second half of the year, and that's still ongoing by the way. So we are still buying back stock. What we have communicated to the market is that we will not be announcing our next steps or capital actions in January, that we will take a pause until the summer, so probably into the second half of next year. Until we have better clarity on how things are evolving. We think it makes sense to pause and take a look, and there's a lot of uncertainty and risk attached to interest rates right now, but we would like to have a better reading on that before defining the capital plan for next year. There is a lot of question marks that exactly what's going to happen with the economy. Again, we have a little bit of a breather in that whatever your answer to that question is, Puerto Rico should be a little bit better than that, we think. And most economies concur with that. But we still would like a bit more clarity. At this point in time, we have no reasons to have credit concerns at all. Our clients continue to perform very, very well in credit. Our delinquencies in almost all levels are continually at historical lows. Our average retail client still today has about 40% higher balances in their accounts than they did in February 2022. So the individual clients seems to be in pretty good shape. We have commercial clients in Puerto Rico that are also -- they have also proven to be very resilient. Most people lose sight of the fact that the commercial clients we have in Puerto Rico today, most of them also survived a 10-year recession. So they are pretty hard in businesses. They know how to respond to economic challenges. So, so far, credit is -- continues to look pretty good. But we thought it made sense to take a pause and retake what our capital return should be for '23 in the second half of the year. Now the change from announcing every January to announce in a bit later, obviously, it's a change and people react to change. Again, it is a little bit of a change that we have created ourselves by having the practice of announcing every January. And there's no reason why we have to do that. And most banks will just tell you what they intend to do next quarter. In fact, a lot of banks stop buybacks in the third and fourth quarter of this year, and we haven't. We have continued to buy back stock. As I mentioned, we ended up buying back more stock than the market expected in the second half of this year. But -- so we're just trying to be cautious until we have more and better information to make those decisions.

Yes, that makes sense. Before I go on, actually, we're getting close to the end of this. So I wanted to open up to the audience, Julien?

[indiscernible] What were your direct losses, either credit losses or expenses from rebuilding branches and stuff?

We had expenses in the multiple millions of dollars that resulted from that. We had as well with Fiona, by the way. So we had either expenses or forgiven fees of probably a couple of million dollars in the case of Fiona. In the case of Maria was a multiple of that. The credit behavior was surprisingly good. There was a lot of loss in Puerto Rico because of Maria, most of the loss was uninsured and therefore, unfinanced real estate. So there was real loss, but it doesn't mean it was in the bank's books. Most of the loss that happen in industrial space and things that sort were insured as well. So we -- when Maria happened, we added about $120 million to our allowance right after Maria happened in expectation of some losses to result from Maria. And over the next 2 years, we ended up reversing almost all of it. So there was a significant loss in the economy, so I don't get me wrong. But that loss to a large extent, did not reside in bank financing books.

So it sounds like your kind of direct exposure to hurricanes is actually pretty minimal. Maybe there's indirect exposure via kind of economic impact. Did it change the way you lend at all? I mean, it sounds probably not if the direct impact was pretty minimal.

I think we probably updated some underwriting criteria to make sure we're capturing any peripheral risks that we may have missed and learned from the experience. But the normal lending we do, especially real estate-related, will require insurance.

Can you share [indiscernible].

Absolutely. Absolutely. So again, hurricanes did result a significant loss. A lot of that loss ended up residing in insurance books, not in banking finance books. Again, hurricanes are a bit weird in that -- at the end of the day, if you look back now 5 years after Maria. Maria is probably a net positive to the economy. Now it's a very painful way to get to that net positive. Don't get me wrong. I'd rather not go through that. But when you add up the effect of all the assistance that results from the crisis, it probably ends up being net positive. Actually, if you look even today in New Orleans, that is more than 10 years since Hurricane in New Orleans. New Orleans today is still getting some federal funds from -- assistance funds from that event. So the events, again, may have a little bit of a perverse way to actually incentivize economic activity afterwards.

Well, that's actually all the time we have. Carlos, thank you so much.