American International Group, Inc. (AIG) Earnings Call Transcript & Summary

Hello, everyone. I'm Tracy Benguigui, insurance analyst for Barclays. I'm pleased to moderate a fireside chat with Mark Lyons, CFO of AIG, We have about 40 minutes for this session. [Operator Instructions] We have a lot to cover. So I'm just going to turn it over to Mark for some opening remarks.

Great. Thank you, Tracy. And first off, thanks for having the event. Thanks for inviting me to participate and thanks to Barclays, I think, more generally. So it's always a good event. So happy to be here, number one. Number two, I don't want to burn the clock too much. So I'll just give a couple of opening comments, then we can get right into the Q&A. So I think 2021 pretty much shows a good, strong pivot for AIG away from fixing to -- and remediating to relaunching and having growth, more importantly, profitable growth, certainly in the General Insurance and enterprise, and Life and Retirement continued to have very strong stable results as well. We've laid out much more of a medium-term capital management view for everyone. We have strong liquidity. We have strong capital, whether it's at parent or whether it's in the subsidiaries that actually bind the business. We have great relationships with brokers and channels and our reinsurance partners. We think all of the efforts and everything that went into the last 3 years of remediation, again, we're very happy with the outcomes of that remediation, feel good about it. A lot of aspects to it, as Peter, I think, has enumerated over time in various earnings calls, but it really allows us to be in a position to now pivot in this way. And so we feel very, very positive that the marketplace is still positive, perhaps not quite as positive, but still positive nevertheless. As a global enterprise, we're not as just U.S.-centric in how we look at things, so that may be a point of conversation. But we still feel that I think there's a lot of strong momentum at our backs for just throughout the balance of the year. We have been growing the book. The unearned premium reserve has grown. We feel good about the margins embedded in that unearned premium reserve, which we're going to earn in over the next 4 quarters. And we just think good, strong days are ahead of us, and momentum is a great thing, and it's great to be right in that way. So with that, I'll turn it back to you.

Great. We're going to touch upon a lot of things that you just mentioned. But sticking with the theme of some strategic priorities, let's talk about the path to fully separate L&R. I understood the reason why you had to do a separation in 2 parts was because of the FTC that you wanted to wait for that rollout originally through 2023, and then you could do the second leg of that separation. But now that FTC is being consumed in 2022, is that some of the thinking behind the comment that IPO could go larger than 9.9%? Like could you even envision a full separation in that comment?

Okay. Well, interestingly, first, your premise is correct. So the FTC, we did say that would go through 2023. And that was, I would say, a limiting factor but a key factor associated with our time frame and the fact that we will now be consuming it, let alone having an anchor investor now that we wouldn't have had originally when we made those comments, makes it certainly more likely that we would have a higher participation out in an IPO. To your -- the latter question of whether you could envision it being -- I'll paraphrase you and say I can't ever say never in this business. It's a function of market conditions at the time and not only IPOs or SPAC success that's going on now, but also within the industry grouping. So if it's there, we'd like to see if we could take advantage as much as possible, but we're not -- certainly not counting on it.

Got it. And you talked about your anchor investor, so maybe the shift to investment portfolio and your asset allocation. I mean, which asset classes do you think Blackstone could offer expertise that would help enable your TA optimization? And also conversely, for GI, actually, GI holds assets that look like it should be on the Life Insurance balance sheet. So any portfolio shifting unify GI upon full separation?

Well, there's a lot there. On the Blackstone side, they've really got some -- well, they've got a global preference, which is very helpful because they can now basically handle anything. And they're very, very strong in a lot of asset classes. I would put real estate right up there with anyone, which I would correlate that into perhaps a life insurance asset. It's also an inflation-hedging asset, irrespective of the time periods, more so for longer-term holds, of course. But -- so I'd say that's probably a good example of where we think they're going to really add a lot of value right away because of their ability to originate and the breadth on which they can originate. On your GI question, interestingly, there will be some re-tinkering and tooling that goes on the side of that. We've always had different asset strategies, of course. With basically a 3.5-year loss reserve duration on the GI side and double digits on the at the Life and Retirement side, your asset mix should differ. But there are occasionally assets which are really bought at parent, hence our consolidation and eliminations that we talk about it every quarter, at parent and basically strip-owned by GI or L&R entities. So that would -- some of those might become less appropriate as we move forward to only pure GI company and then L&R, a pure Life and Retirement.

Got it. Other thing that I think about on the future state full separation is that your PMLs to capital is going to look very different, just given a smaller denominator. At least the rating agency I came from looked at enterprise capital. Will you need to cut back property or increase the use of reinsurance as a stand-alone P&C company?

Interesting question. I think actually, you just stated, it was your prior employer that really looked at it that basis, and the other 3 -- well, 3 core competitors, really looked at it more on an individual basis, either a legal entity basis or legal group basis. So that's a unique perspective. But overall, I have to start answering that question by a function of how we're putting property risk on the books to begin with. We really have a lot of levers. You have the retail channel, which had to be depopulated enormously and then we're happy with it and drop those limits on the front end. This is kind of a risk management question you're asking me, so let me just kind of go through that. And then with the Lexington being stood up in the wholesale channel, the whole complexion of that book is so radically different. I mean, I think the numbers are about 70% or so or $10 million in capacity amount. That's it. And maybe 80%, 85% is at $25 million. I mean, it's a completely different ball game. So that then can effectuate interestingly, not just your P&L, but where the buckets are filled because they're more localized exposures. They're not big, huge retail accounts. And lastly, Validus on both their assumption and session capabilities allows us to really mix and match that. But the net is, no, we don't expect to make material changes because of what we're seeing on the front end that allows that.

Excellent. So I guess, on the topic of property and catastrophe, so far, this quarter is shaping up to be quite active. And I'm not just talking about the impact from Ida, which feels like a 2-part event. I mean, so far, there's over 20 PCS-designated events just this quarter, which just feels high on a frequency standpoint if you look at the 10-year average. So can you share some early insights of your catastrophe experience so far this quarter?

Well, I think for context, kind of as you alluded to, you've got 20 PCS events. You've got differing exposures across the board. But what are the industry views of this? And when you look at what's coming out from a lot of the modeling agencies -- and some, they have to update it. Some have taken Ida. They'll have an initial one, and they'll say, "Whoop, it went out more than we thought we got included for the Northeast," and things like that. But it's shaping up between that and the European storms of Germany and storms of Japan and wildfires. And we still got 2.5 weeks in the quarter, right, to go. So this looks to me, in the industry, this is a $40 billion-ish, probably going to be north of that event in insured losses likely. So it's really, really tougher to tell. AIG, of course, being in a more of a high severity, lower frequency lines of business, not exclusively, but more so, it takes a little bit longer to get underneath it, and that certainly will be the case for Ida. So my preference is we want to really lock that down before I'd say anything publicly. But I can just tell you, if the industry is going to be elevated, I think it just goes through market share and other kinds of things that, that could impact us as well. However, knock on wood, because compared to many of the peers that others pit us against, there really isn't a peer with regard to our Japan exposure. So that peer are really the big 3 over there. So depending upon what happens in Japan, that could also dial it up, dial it down on a relative-to-others basis.

That's helpful context. Let's just shift gears to pricing. I'm actually struggling to find what is the real catalyst for the pricing momentum. It's not a capital replenishment. I mean, there has to be a narrative. Basically, you could tell your brokers or insureds to compel a continuation of rate increases. So there's just a number of stuff going around. People are talking about years of underpricing, complacency. We just talked about catastrophes, or we could lead that into climate change, social inflation, general inflation, lower investment yields or just heightened risk aversion. I mean, what do you think is the largest pull on this pricing story and your conviction that, that it will continue?

Yes, all of those are factors, but you asked for the dominant one or ones. Sitting on my perch, it's a little different, right, than sitting in a P&L or a line of business P&L or geography P&L. So to me, when I take the view of the industry over the course of the cycle, a couple of things have to happen. One is in the hard market years, you can't be targeting your long-term average. You got to be going above that because you know in a softer market, things are going to go -- what's the old adage? You go -- and I think even our IR guys use this, you go up in an elevator, you come down on the stairs with regard to pricing. So we know that will ultimately happen, so you've got to maximize the opportunities even when they're in front of you. I think from -- so from that point of view, I would say it's been years and years of underpricing. That was the catalyst for what it was before, and it's still some of the tailwind for what's needed now. And because the complexion of the C-suite, I'd say, over the last 20 years, in particular, maybe not CEOs, but the contribution of analytics and financially oriented people that understand the business is increasingly at the table. So understanding how a low interest rate environment in the long duration lines kills your ROE, it creates more discipline as a function of that duration. So you've got to make the right mix decisions and the right marketplace decisions in order to maximize it. The days of just looking at combined ratio are long over. But I think that helps from the conviction part of your question that there's more people having a common view of the necessity of that as an anchor to what you do.

Got it. That was a good answer. And I guess, speaking of talent, I mean, Mark, we're all the product of our history. And I've heard you say before that Arch underwriters could clearly under -- articulate their underwriting risk appetite. And before you came on board, that was not necessarily how an AIG underwriter thought. So on the cultural side, has everybody at AIG bought into that yet?

It's interesting. I'll divide your question into 2 pieces because the one aspect of it is there were areas in AIG that really understood the risk appetite, the classes of business, the industries, the sub-industries where they wanted to play. They just bounded anyway, whether the risk/reward trade-off made any sense. So that's why I'm kind of breaking it in 2. But the message, I think, is pretty inculcated now through the organization. And Peter really drove a lot of that. McElroy has driven it. We've got Jon Hancock internationally, really driving it. And it's reinforced all the time in many venues, not just one-on-ones that you have through executives and down the line. But the business reviews -- the quarterly business reviews are a real rigor. And their business reviews are not revenue reviews, right? It's not just focusing on premium in the door, the booking of it because there could be some cash delay on loss-sensitive programs and things like that, right? It's what's the profitability? Why are we all gross on this? Or why are we keeping such a small net on this if we have these convictions? Prove it to me. I mean, it's a completely different mechanism than I think the company was used to. And every open item is followed up on the next meeting. So there's really no escape.

Got it. Is it fair to say that you have not really seen appreciable underlying exposure growth yet? Like your growth is really coming from new business rather than growth within existing business like economic activity of increases in payroll or sales or car units or things like that. When can we anticipate that lift off come for AIG? Or is the makeup of your bulk less linked to GDP type of growth?

Yes. Good question. I think it's fair to say that it really has not been underlying economic engine -- changes that manifest themselves in the earned exposures that we rate against, I think, or written exposures that we rate against. But I'm thinking this through as you're asking. Our book -- there's so many sub-portfolios, just trying to see if I can make this connect. So we have frequency books. I'd say our PCG book, to some extent, it's frequency, but it's got such a cat component to it. And our concentrations to it makes it a little different animal. We don't have a big commercial auto book, back to your earned car years type of question. The GL book is really not a primary. GL book is really not that large. And we have not been a large comp book -- guaranteed cost comp book. It's been loss-rated -- overwhelmingly loss-sensitive, I should say. And interestingly, when you go to excess layers, let's just say, even on casualty, even if the exposures are growing, it's been tradition that excess business is not auditable. So you go in to get a bullet payment, and it is what it is over the course of the year. If exposures drop, you benefit; exposures increase, you may have got hit a little bit. So I think we're going to benefit from it but I would say less so compared to some of our frequency business-based peers.

So basically, the growth is coming more from new business and some higher retention rates.

Yes. What we just talked about in the second quarter, for example, yes, that would have been a new business and exactly what you said and greater retention rate.

Got it. And I'd like to talk to you about your ability to change terms and conditions potentially more quickly when greater force, given a large chunk of your policies are on manuscript form. So could you just help us out, like what percentage of your book is occupied by manuscript policies maybe versus standard ISO forms? And how frequently do you update your terms and conditions, which actually could magnify your return profile and serve as a risk mitigant?

Well, I'll pick the U.S. On an admitted business, you have much less flexibility because you've got to -- it's not just rate but form that you got to go through the filing process. As you know, some are for approval, some are file and use and so forth, so it's a mixed bag. Where you really have the opportunities are either the consent to rate areas, the non-admitted channels and what I'll call the big boys. So you got global companies who are very, very large national companies that are well grown generally through the big 3 that have unique exposures that is really a mano a mano type of negotiation. So it's in those that you have it. So -- but when it comes to, let's say, making an auto filing, PCG, right, you've got to go through. You're in X states and you've got to go through the admitted process, and it's hard to pivot quickly in those. Non-admitted, which is created from rate and form, prior approvals, you have a lot of flexibility. And the Lexington, as you know, with Dave McElroy, he came in and stood that up. And we've got a great team in there now leading it. And that was one of the big sources of growth of the new business, certainly in North America in the prior quarter and year-to-date. And it's because you've got that freedom. So I still view that as a favorable tailwind as we go through the balance of the year.

So a clear example I could think of is the leadership in all your global commercial policies that either explicitly affirm or exclude cyber. And if we could segue into cyber for just a second, how do you feel about rate adequacy for this line? And is there any changes in your risk appetite?

Well, it's been really gravitating up. It actually hit its highest effective rate change the last quarter, as a matter of fact. That's a line that we go against the thesis of dropping off, for example, that you've heard from other parties. So -- but it's been double-digit increases probably for the last, I would say, approximately 5 to 6 quarters. If I got that wrong, I'm sure IR guy will correct me. But the fact that it was the thickest last time is great. I mean -- you know that the complexion of the claims has changed dramatically, right? Ransomware is the headline of the day, frequency, and every successful one seems to embolden for more. And the aspect of data breaches, if you will, the frequency of those has really fallen off. So compared to 18 months ago, the kind of claim coming in is, night and day, different. So not just affirmative and non-affirmative but AIG has gone to a pretty tough rigorous application process, where the whole purpose of the detection is the degree of controls around the subject insured because they themselves may have a lot of external relationships that expose them like any company. Every company probably has interfaces of some type, right, and outside-in or brokers or intermediaries having access. So all of those kind of things are really focused on to be a point of differentiation between for risk selection purposes. And then depending upon the feedback on that, you may have co-parts or may be coinsurance along the way. And it could be different deductibles, limits could be cut or could be sublimits applicable to various features that we think perhaps aren't as strong as they could be.

Got it. It is my impression that as the cycle turns, international rate increases did not seem to run as aggressively as North America. Is it fair to say there's a lag there and more to come? And also, what is driving that exactly? And I have a follow-up.

Okay. Well, internationally, you can't put into a block, right? You've got to -- just like if somebody in Europe talked about the United States as a monolithic block, it probably wouldn't make sense either. So you got U.K., Europe, more Australasia and Japan in our world and Far East and all different markets. So first off, your premise was right. It started later than in the U.S. I think the U.S., on average, was more underpriced probably in property and the core casualty businesses like lead umbrellas that helps spark it. So I think it started there first. And of course, we live in the land of lawyers. So I think we're more litigious culture by definition. And so I think all of those contributed to being the U.S. first. You've got the U.K. followed probably 6 to 9 months later in any materiality and then really grew. So U.K. books' effective rate change, I think, has been 20% or north of 20% for a few quarters now. And others have been strong. They've been lower double digits. There are a couple where luckily, we don't have a huge book, might be single digit, upper single digit. Japan's book is dominated on personal lines orientation, and you have the same issues there you have in the U.S. about pivot ability because of being really admitted products. So -- but I do think there's more to come. And it really varies by area. So if you have a product line written abroad, that still may encapsulate some U.S. exposures inside it, whether it's D&O, ADRs on the middle or excess layer or it's a global risk that it's like a global program or reverse part of the program. Then you still have the U.S. push up, helping to drive up the overall on that program. But I still see a lot of good opportunity in that.

Got it. You actually dovetailed to my second part of that question. So maybe moving on to capital management. When I think about your holding company inflows and outflows, is it fair to say that liquidity will be somewhere in the double-digit billions at the end of the year? Looking ahead in 2022, if that's the case, if you could just go over your menu of capital deployment measures, hypothetically, how you would rank each of them.

So yes. So if we go through the rough math of it, so at year-end, let's just round the numbers, a year -- sorry, at quarter end, 2Q quarter end, we were $7 billion of parent liquidity, give or take. We've already talked about Blackstone closing at -- for the 9.9% of $2.2 billion. We had affordable housing at $5.1 billion. That's the headline number, kind of like the sticker number when you go to buy a car. And of course, there's a million other flows, right? There's dividend flows, tax sharing payments. There's outflows of interest and outflows of dividends and everything else, right? So -- but the rough math of that is $5.1 billion, $2 billion plus $2 billion -- actually $14 billion, $14.5 billion. We talked about capital management actions of $2 billion of share repurchases, so subtract that, and $2.5 billion associated with debt, the liability management actions. So round numbers, that math takes you to around $10 billion. So there's probably a plus or minus mill on that -- or bill on that. Sorry, million would mean nothing. But -- so I think that answers your question.

Okay. So you confirmed my math there. But then I guess, to my second point, if you think about your menu of options, and you already laid out your ultimate buyback number, but only really provide an outlook for the rest of this year. How would you think about those priorities heading into next year?

Well, I put the priorities as the top 3. And my guess is Peter probably also mentioned this, which is the debt repayment and the leverage targets we still want to be maintaining, that's a key priority; returning capital to shareholders through the share repurchase vehicle; and then investing in this hard market business. This is -- I guess I prefaced, this is the time to do it. And so if you've got to stretch every capital dollar, to me, that's a good use of time because that's what we're here for. We're here to find opportunities for our core competency, which should be underwriting as opposed to investment income bailing out underwriting. And the market is favorable. Our attack of the market is positive, and we think that's a fabulous use of our share capital.

Okay. So when I think about your priorities, you're also doing this liability management exercise. And you've already spoke about $2.5 billion of actions at least this year. I understand that may just be the first cut. Can you walk us through your thinking about getting you near 25% leverage for AIG, Inc. and 25% to 30% at L&R?

Yes. It's, I guess, a couple of things. One on the way, because it's public knowledge, right, we've already been out -- we had a press release about it, about doing a make-whole on $1.5 billion of our second quarter 2022, so just around the corner, 6 months plus. And that was 4 7/8 coupon, I think, so that was $1.5 billion. So that's already in -- you can count on that, let's put it that way. And then the other balance you could think of is in process and thinking. So just one thing about -- thinking about it, the other one is not doing it. So that's one reason I wanted to bring that up. With respect to -- just refresh me. Sorry, I lost my train of thought for a second. The core part of the question was about?

If it's just the first cut here, I mean, it's just doing $2.5 billion doesn't get you to your ultimate leverage thresholds targets. So it has to be like the first cut. So I'm just wondering, what does the second cut look like?

Okay. I think what might be helpful is we finished the quarter on a GAAP basis. Just look at our printout, the 27% even, 27% even. But that's what AOCI reflected. That's a GAAP view, and that can go all over the place. So we have a couple of forces, right? And when we talked years ago even, we've stuck to it of being sub-25%. That was really pre-separation. That was AIG in going-concern basis as constituted at that time. We still have that goal for RemainCo, if you will, for SeparationCo. But given that we're at 27%, given that we're going to be share repurchases and the liability management actions, we're still feeling the net of that is going to be favorable towards the -- and we'll evidence that, that will be favorable by getting that 27% down towards 25%. So we still believe that's going to be happening, let alone printing positive net income doesn't hurt either, yes, to helping that. On L&R, we've talked about this a little bit, is that our mechanism to enable separation is going to involve going after some public markets. And ultimately, without getting into the numbers, the L&R holdco would wind up issuing and then, in one form or another, effectively issue a debt-driven dividend to parent. So that's how you can see further downscale of RemainCo on it, and how we would set up the structure for SeparationCo would already be in the wheelhouse of the competitive position that we talked about.

So essentially, with L&R doing that debt dividend, that could be self-funded or to consume any more holdco liquidity getting to those ultimate ratios?

Other than some -- I'll call it relatively minor. These are costs associated with some tenders or make-wholes that I don't view as enormous, but we're still going to look. There could be some -- so for example, on the $1.5 billion that we already talked about on that one issue, because it's expiring or maturing so soon, make-whole is not a big deal. As you reach into future things, make-whole would become more significant, right? So that could be a use of parent liquidity and parent capital in that respect. So to that aspect to your question, yes, there could be that type of thing coming in. But I think more broadly is SeparationCo is setting up so that they're already in a very good position or soon to be very good position, and RemainCo continues to have the sub-25%. And I think I gave you a path how to get there.

Awesome. And then before closing up on capital management, I just have to talk about M&A. In light of L&R separation for AIG, how are you thinking about white space now? Does that create any urge to fill up the Board? Or is it better to be a more focused and nimble P&C insurer?

Well, we're clearly focused. We went through all this time of correction, right? So now we know the book cold. And when you go out and get something, you really think you know it, but until you're out of the hood, you never really know as well as it is. But given the capital priorities we've already gone through between the debt, the share repurchases and investing in our core competency in this marketplace, that's the focus. It's really that white space at this point. It doesn't mean we won't be looking at it, but it's not high on the radar right now.

Got it. That's very helpful. [Operator Instructions] I have some follow-up questions. Let's touch upon reserve adequacy. At this point, how do you feel about reserve adequacy for accident years 2016 to '18? Basically, the subsequent years that are out of scope of ADC and compared to all the actions you have taken, I guess, I would characterize the years of weaker underwriting.

Yes, that's a fair characterization. First off, just to preface that, the ADC has -- which is 80% of $25 billion ex $25 billion for accident year '15 and prior, so just to clarify for others on the call. There's $6.7 billion unused on that, which represents the 80% sliver of the session. So there's still a lot of room back there, just for information's sake. And when you go into '16, '17 and '18, I think I may have commented about those before, it's an interesting dichotomy because clearly, the underwriting, this accident year '16 would have been half of policy of '15, right, which is not a glory year. And therefore, '16 and '17 are certainly not up to the standards that we would be employing today and even if there wasn't a hardening market to help supplant it. So here's what I mean by the dichotomy. So you've got a lot of specialty lines. There's a lot of casualty in there. There's a lot of financial lines in there. Of course, all the short-tailed stuff is pretty much run off. And there's still comp, right, associated with those years. So we -- I grant you that the underwriting quality is not strong in '16, '17 and '18 partially, right? Because Tom Bolt got here in the beginning of '18 but focused mostly on property and third-party coverages and got in the second half of the year probably into the casualty businesses. But the age of these years -- so if you go back to accident year '16, where a lot of subject businesses that are still open, again, high-severity, low-frequency businesses, you need more time to get some seasoning to have any confidence in it, and now that's there. So more traditional actuarial methods seem to make sense now, where in the past, you go this -- I don't know, this looks straight. Either looks too good or looks too bad, either the 2. So we're gaining more confidence in those projection periods, but clearly -- on those accident years. But clearly, the underwriting is subpar on that. Never going to say anything different.

Got it. Maybe a little bit back to the pricing story, I mean, to some extent, all supply/demand. And I've actually heard you mentioned that AIG's walked away from $650 billion of limits. Who's absorbed this risk? And I guess, alternatively, how much of it was self-insured, thereby creating capacity constraints that might have contributed to this hard-market cycle?

Well, you have to look at the distribution of that $650 billion. A lot of it was first party, but many, many billions in casualty businesses and in financial lines businesses. I know it's easy to forget that in the prior regime, the large strategy, there were monstrous limits being put up, that on a gross basis, you never want to expose the balance sheet. And then secondly, you kind of misuse the capacity that any reinsurer would give a [ ceding ], right? They're only going to expose across all lines of business X. And I'll say that was not optimized on the session. I'll be a little generous there. So I guess I view it as that I don't see a lot of that being self-insured. I mean, there could have been some captive deals, but there aren't captive deals for $2.5 billion policy limits, right? So most of this had to be absorbed by the marketplace. And when I tend to think about -- well, I'm a little biased, of course, because of the chair we're in, but AIG and others were some of the sparks, I think, that we didn't drive the market, but we helped move the market. It's not just in the risk appetite and in the pricing and in the discipline. You also had Lloyd's doing that at the same time. And I've joked in the past that there was no less [ chance to learn ] anymore. And a little tongue in cheek, but there's some strength to that analogy. But if you put 2/3 of $1 trillion into the marketplace, everybody is going to reprice it. And I think that's what happened. So people -- I think I can't tell you exactly who, but I know it got absorbed. And there could have wound up being some level of higher SIR that occurred because of the price increases from a budgetary perspective, from an insurance CFO, for example. But I still think overwhelming proportions were just right around the industry and absorbed.

Okay. So maybe just a follow-up and you maybe could remind me. I think you and FM Global were the only ones that really offering like $2.5 billion limit in property. So are you saying that others stepped up? Or I think it was just a little bit more of just the market gravitating to that $1 billion level?

Well, what I think happened is a real overnight change for the brokerage community, where they could have a big slug of capacity in more than 2 places and instead have to have 25 markets fill up the gap. They had to really -- I don't -- I'm being a little flippant, but say, earlier commission on that renewal by having to buy for 25 markets, maybe going to 35 markets rather than just having a, we knew it was expiring with a big slug of capacity. So they had to do the whole -- the legwork.

Yes. And I think that's important because you have to also understand that structural elements that may play into the market cycle. But yes, in the call, we really only have one more minute. I don't know if you had any closing remarks as we head into 2022, how you're thinking about next year?

Well, I would just probably do a bit of a continuation in that we think we're well positioned. We think it took 3 years of shovels to get us to the point that we can maximize. We think that the market is still very strong on an absolute basis, and you've got to pick your shots and timing. And you don't want to look back in 5 years and say, would've, should've, could've. It's -- this is the time to identify the sectors, and this is the time to really -- to push it to the extent that your capital permits you to push it.

Excellent. I really enjoyed the discussion, Mark. Thank you so much.

As did I.

This concludes our session. Take care.