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

Good morning, everybody. This is Brian Meredith. I am the insurance analyst here at UBS. Apologies for all the technical difficulties we've been having here. But we are here for the AIG fireside chat. With us with us today, we've got Mark Lyons, who is the Chief Financial Officer of AIG; Sabra Purtill, the Deputy CFO; and Shelly Singh, who is in charge of Investor Relations for AIG. Mark's got a couple of quick comments he's going to give, and then I'm going to get right into the fireside chat and the questions. Mark, I'll hand it over to you.

Great, Brian. And thank you. And so rather than take up a lot of time on pre-comments, especially that it's been cut short a little bit, I'd rather just get into a lot of the questions. I think, first of all, thanks again, Brian, for inviting us for 2 days. And so we're very, very happy to be here. We think AIG is very well positioned right now, both from a capital liquidity, subsidiary capital and liquidity perspective as well as taking full advantage of the marketplace on the general insurance side. But -- and there's other aspects in the quarter you guys will probably ask about. So Brian, why don't I just hand it back to you?

Yes. Let's just go ahead. So the first kind of broad question, Mark. So as we look 12 months from now, as you look back on the impacts of COVID-19 on the P&C insurance industry and life industries, what do you think the long-term implications are? And how is AIG positioned to respond?

Well it's a -- you started off with a crystal ball question. So let's see how good my crystal ball is. So -- well COVID, I think it's clear, has been an accelerant in the -- certainly in the PC marketplace. The market was already, I think, hardening pretty materially. So I deem this an accelerant rather than a cause. AIG, so it's done -- it's lifted all boats, if you will. I think AIG has been taking maximum advantage of the marketplace and using our breadth and influence and pricing power and so forth to really push things really across the board. But I think the reason we are positioned is because of all those couple of years of hard, hard underwriting changes and redefining appetites, and having a protective reinsurance structure and so forth and so on, that you've heard us talk about before. I think longer term, there's going to be fewer businesses to insure, maybe that's medium term, Brian, more than anything else on that. On the L&R side, whether that affects long-term mortality or not, my -- I mean the actuarial review, assumption review for us will happen throughout the third quarter. But based on what I've seen to date, I would really believe it's going to affect things long term. It could be like a blip, but that blip would only come to pass to be elevated if there's feeling of secondary, longer-term issues associated with COVID. So deep-tissue organ damage, things of that nature that prolong on. But we have to wait and see on that. And I think there'll be some -- as a result of fewer businesses and other aspects, there's probably more credit hits coming in across various industries, affecting everyone differently. But I think the net-net is, for the medium term, smaller insurable exposures.

Great. So first question here for you, Mark. So COVID-19 charge you took in the quarter was based on claims activity to the first and second quarters. I guess the question is, does this mean that you're likely to have more COVID-19 development here in the next 12 months? And if so, where do you think it's likely to come from?

Well I think Sabra and I will ham and egg this a little bit. Let me just start off more conceptually, and then I think Sabra can pick it up on some of your more detailed question aspects. But -- so I think from an industry perspective, and then I'll give AIG's view on this, is I think we've been -- we've attempted to be clear on how we've done it, why we've done it and why we've taken some of the coverage positions that we have. But more broadly, I think there's confusion in the marketplace, whether it's because of commentary or misinterpretation of commentary or what have you. But the way I look at things, and the gray hair that I've amassed over these 40 years, is we have to watch reaching forward and try to reserve today for actions that have not yet occurred. Now I'm talking specifically from a PC perspective. So why don't do I bother to talk about that? Firstly, it's the responsibility of any property casualty company to reserve as of the balance sheet date for all accidents that have occurred on or prior to that date on an ultimate basis, management's best estimate on an ultimate basis. So at the end of the first quarter, it was for accidents that occurred 3/31 or prior. At the end of the second quarter, it's 6/30 or prior whenever you look to first quarter, so forth and so on. So if an accident hasn't occurred yet, you shouldn't reflect it. And the secondary question would be, why not? And the answer to that is that's the function of your unearned premium reserve. When you write $100 policy on day 1, all $100 are unearned and there's a loss provision in there. Let's say, $70 loss ratio. So $70 is associated with accidents that will occur as that exposure earns over time. So to have a loss reserve up for a future occurrence and to have a -- on a premium reserve up for a future occurrence is overlapped, and I don't think any accounting firm would actually ever allow that. So therefore, there's got to be -- it's got to be a misinterpretation. AIG has done what I said, which is we look at all the coverage triggers across all the policy forms that are out there and reflect what's occurred at that point in time. The reason I say coverage trigger is because, event cancellation is a good example of one where you noticed earlier, say, than the 6/30 date for a sporting event that's effective in October, let's say, and the decision is made prior to 6/30 as well to cancel that event. That's the trigger to reflect it into your statements as of 6/30, not waiting until the October sporting event. So in that way, it reflects and it's consistent with what I just said. But I wanted to clear that up from, I think, an industry confusion point of view. And with that, I'd ask Sabra to throw in her two cents on this.

Yes. Sure. So just as a reminder, we've put up about $730 million of COVID-related reserves year-to-date. About 70% of those are for the business interruption/property where we have affirmative coverage for infectious disease combined with what we put up for contingency and travel. Obviously, as we looked into the second quarter, we did see some reserves that we put up for some other lines, including trade credit, marine, some on casualty workers' comp, A&H financial lines and then some in our reinsurance business. As we look towards the second half of the year, I would expect that where we [ would see claims ] are going to be more on those, what I would call, lines that are impacted by the continuing economic crisis as opposed to exposure to the disease itself. So trade credit would be an example, for instance. But as Mark said, it will reflect the claims activity and events that occur in the third or the fourth quarter as we believe that we have reserves that are ultimate for what has been reported to us today for the -- through the second quarter.

Great. So second quarter commercialized pricing, AIG up 16%. I guess the question I have here is, one, do you think that continues to accelerate here and for how long? And then on that topic, look, I know Peter mentioned something about diversification of the Commercial Lines portfolio. What areas was he referring to?

Okay. Thanks for that. So I guess a couple of things. Peter highlights on the calls the North American Commercial segment, reportable segment because that's where most of the changes have occurred. That's where AIG has applied the most surgery. And the reason for that is that's where it's been a recurring problem for AIG over the years. So I think you lead with where the most correction was made. So the comment about 16% overall, I think he went further to say that it was 21% in Commercial North America and 10% internationally. And when you dissect that a little bit further, I think on a prior call, we had talked about the U.K. already being at 10%, and that's continued to accelerate. Which means, therefore, non-U.K. is less than 10%, but it is increasing at an increasing rate. So we see none of these areas abating. We see no evidence of it abating. And we still had seen things increasing at an increasing rate, which I think is good news for everyone. So the reason I brought up the focus on North America Commercial, Brian, is because what can get lost in the sauce is Personal Lines business. And when you look at the Japan book in total, for example, that's $5 billion of our $35 billion. So that's 15% of the book is basically getting 1 or 2 points, which weights and mutes the overall results down pretty materially in that regard. And similarly, in lines like Accident & Health and some of the others, which are very predictable year in and year out, but are really aren't a rate issue. So you get these flattening aspects to it. So I wanted to bring that up. In terms of diversification on Commercial Lines, I think it means more a function of where are the opportunities, and therefore where you're going to put time and attention and capital into those opportunities that are both improving the most rapidly on a relative basis and providing -- going from red to green or yellow to green, so that there, it hits profitability return standards. So you can accelerate, writing more business there and perhaps keeping more net there. So I think that it really constitutes a mix change that will be different a year from today than it is today.

Got you. Next question. So how is Commercial Lines pricing now relative to loss cost trends? You think that with this 16% average price increases, margin should be expanding pretty materially from here. And from the second quarter call, it kind of sounded like you're taking a cautious view on loss trends, particularly related to social inflation. Maybe you can kind of clarify that a little bit for us.

Sure, sure. Well I think some of what we can talk about is really derivative from your prior question actually. But I don't think I really said much different than I said in the past. I think I've talked about a range of 1 or 2 points on some lines up to upper, knocking on the door double digits in some other areas. I think I pointed out excess casualty and commercial auto in those as well. So I think it's still within that framework. I think I've talked about the social inflation in the past on prior calls as well as -- and there were conferences. So I think it's been fairly consistent with that. So it's really with regard to the expansion of it. And I think it's clear. I mean let's take North America Commercial. You're getting 21% rate increases. I think it's pretty clear you're going to have margin expansion. I don't think anyone on the call would think we're having 21% loss cost trends at this point, and therefore everything stays flat. So I think it's reasonable to expect that there's going to be margin expansion, but it's not going to be uniformly felt. Going back to our prior discussion, I think it's safer to say, in a rank order, there should be more of that coming through North American Commercial, probably followed by International Commercial, then followed by Personal Lines areas. So I think that's one kind of view of that. And then secondly, as I kind of alluded to, it comes down -- you can't just look at your rate changes, you got to look at where you are given those rate changes and making sure that's where you want to really invest capital for expansion. And the good part though is I think we showed the 6% and 7% increase in net written premium. Even with new business down, this shows the strength of the renewal pricing and the retention we have been able to maintain to get that kind of growth. So we're -- we've been so defensive, right? Now we can see some offense coming in on that, and there should be some expansion. But there is still that uncertainty on some aspects of loss cost trends. So I think we talked about this once before that we've been implementing a reserve philosophy that's a lot more, let's recognize, bad news quickly and good news more slowly. And I think you'll see that reflected on a go-forward basis as well.

Yes. And Brian, I would just add to that, that I really strongly encourage people, particularly because of the impact on personal insurance with the travel business as well as the syndicate changes, that they really need to look at AIG's combined ratios on a segment basis. So for North American Commercial Lines for the quarter, we improved 320 basis points. For International, it's 500. Year-to-date, North American Commercial's improved by 200 and International's 180. So clearly, we are building margin in the book on a Commercial Lines basis.

Got you. That make sense. And given where pricing is relative to trend right now, you think that trend should continue to be pretty good?

Definitely.

Let's see. Next question here. Generally, I'm just curious here. How much lower does your combined ratio target need to be today versus, let's call it, last year at this time in order to achieve an acceptable return on capital given the interest rate environment with the kind of decline we've seen?

Yes. That's a weighted average question. I mean I think we can go back to the same kind of commentary we've done in the last couple. And Sabra just said, that's a function of each segment, with each got its own characteristics. European trends -- America has 70% of the world's players, right? So -- which -- a catch phrase I always use which just highlights the fact that things are going to be different in different areas of the world. But we're going to get that margin expansion. When -- if you guys look at things on the componentry of an ROE and you look at leverage, I think it's pretty clear when you look at AIG, you're closer to a 1:1 premium to equity ratio on -- in the GI side, but you're closer to 3 3, 3 4, I think, on invested assets. So the leverage associated on ROE from a change plus or minus in the NII yield, because that's operating view, right, that's a total return that's unrealized in there at all, is more impactful. So to the extent that you want to stay even, I don't care what your target is, you're going to have to make up for it more because of the leverage difference on equity. So it's not proportional. So if you're going down 100 basis points in ROE because of reduced NII, you can't just go 100 basis points in the combined ratio, you're going to have to go more. And I think I'm informing you when you look at those relativities on leverage between invested assets and premium to surplus, that guides you to what it is you have to do.

I see. That's helpful.

Yes. And I would just add, kind of like back of the envelope. I think the way to think about it is you've got to look at the duration of the investment portfolio. Our portfolio in GI is about 3, 3.5 years on duration. So as you lose about -- if you were to lose like 10 basis points of yield on the total portfolio, that's maybe 25, 30 bps on the ROE as opposed to 50 basis points on the combined ratio gives you 30 to 40 bps. So there is obviously a balance. But as Mark's saying, it's -- the impact on the ROE comes much more from the combined ratio improvement than it would be detracted by the yields.

Got you. Makes sense.

The other thing -- I just want to add, Brian, on top of that very, very quick, on Sabra's point, is that you can't talk about a combined ratio without talking about duration. And whether we're talking of 1.5 year duration in Property Casualty business, let's say assuming [ property cas ], first, excess work comp, which is 16 to 17 years, and your decision to grow there is more a function of interest rates than pricing that's going on is -- has to be reflected in. So -- and Sabra's point of us being 3.5 years, let's say, on average duration, that's not a huge -- a monolithic block. There's radically different targets across the board as a function of that duration.

Great. Okay, let's pivot over to, why Syndicate 2019 been a high topic the last quarter. Maybe you can go through kind of the economics of the transaction, your fee income, CD commissions. Is this dilutive to earnings? And then another question, how much capital is actually freed up by the transaction? So maybe it's ROE-accretive, but earnings dilutive initially.

Well, I think Sabra and I will ham and egg this one as well. So we provided some guidance or info to you, which was -- on the last call, which was for the North American Personal Insurance segment, as we report it, which has more than PCG in it, but I know just trying to help you guys in totality, was $425 million of net written and $345 million of net earned is kind of the average by quarter, each quarter that we might expect for 3Q and 4Q. So there's that. I think secondly, you've got a couple of other consequences associated with that on like a weighted average impact on loss ratio and expense ratio. Sabra, I think, will go into that a little bit more, but it changes that mix a bit. But it has created some havoc on GOE, if you will, in that area because you have this bolt out of premium, so you have a depressed premium size. And when you get to some of the businesses, like travel and other things are in there, you're still carrying those GOE costs of underwriting and sales and call centers and whatever else is in there, especially if you expect travel business to not be a long-term sustained hurt industry, and therefore, travel insurance supporting the travel industry. And we don't view that as a long term, we view that as transient. So why would I get rid of all the underwriting -- the systems, everything else that we have associated with it. So that's a distortion that can come in for a while. And there could be other noise as well in the balance of the year dealing with like CAT costs and how that's shared amongst all the capital providers and everything else in there. So -- and that noise will oscillate its way out. So as I think I said on the call, going into 2021, we're going to have a much different view of us being back to mid- to lower 90s in that. And -- but to your point on how does that affect our recognition or our revenue model. Instead of being 100% underwriter now, we now have a distributed capital model. It was one of the last things Peter and his team really had to focus on. Because we had a lot of bites at the apple on that, you're under any side because of renewals over multiple years has discontinued to be a spike in the concentration risk in cat. As the old joke goes, rich people like to live together and they like to live on the coast. So you get -- you can't really avoid that concentration risk. So there's other ways to ameliorate and spread that. So we'll ultimately wind up, as you segue into 2021, with about 25% net on that in terms of enjoying the underwriting gain associated with it. And the rest morphs into more of an MGA model where we have fee income and possibly some reinsurance overrides associated with that reinsurance that will find its way into the AP ratio. And the fee income, which I don't have to put a loss reserve up against, right, that drops through. But hopefully, that answered your question.

Yes. And I would just add. I think the way people should think about it for the second half of the year is because we still have expenses in the travel business, although we're not really getting any earned premium from it right now, combined with just the, what I would call, kind of the impact of the syndicate, I would look at the Personal Lines business as being kind of breakeven from an underwriting basis. Obviously, depending on what happens with CATs for the portion that we retain, there might be some impact from that as well. But then as you look going forward for 2021, hopefully, we'll have the travel business back, but the premium itself will be much smaller going forward for the Personal Lines business. And hopefully, we'll cut off the tail on the underwriting volatility. Because really, I think one of the things that I keep trying to remind people of is, for Personal Lines, you can't really look at it as excluding CATs because that is one of the major loss drivers of a Personal Lines book, particularly a heavily dominant homeowners book like we have in PCG. And then as we get to year-end and we've rolled this portfolio or at least 75% of it fully into the syndicate, back of the envelope with respect to your question about capital, it's maybe roughly $500 million of capital that will free up. So you'll see that for our year-end RBC numbers.

Got you. Makes sense. I got a couple of questions coming in. And I just want to give a clarification on my last question, Mark, when we talked about interest rates and effect on kind of combined ratios and stuff. I guess the question is, as you think about the interest rate environment, have you actually lowered your ROE expectations on underwriting?

On an explicit basis, no. And I think what it really implies is that this is a great market right now. And this is a market where you attempt to get, within reasonable methods, as much as you can out of it and don't limit yourself to a line of business ex ROE which you can actually get more. So those areas are going to be pushed very hard, especially when you look at the elasticity between rate change and retention. So there's a lot of focus on that.

Okay. Great. Next question. Actually, I got another one coming in here from the audience. In terms of emerging markets strategy, how do you see your business in Latin America going forward? And are you contemplating any strategic alternatives there?

Well, it's kind of hard to comment on. It's not that large for us now. I think a lot of people have attempted to grow by going through the Brazilian reinsurance approach. AIG had gotten out a lot of Latin American areas in the past regime before the current regime got in, that's probably worthy of a revisit. I'll just leave it like that.

Got you. Okay. Another one actually just came in. With respect to your ceded reinsurance, obviously, you've done a massive rework of your ceded reinsurance program, really lower volatility. Should we, given the reinsurance markets have kind of firmed up here, too, should we expect that that's going to be a bigger cost here going forward? Or could we start to see you guys retain more risk as your book kind of improves here?

A couple of -- a great question. A couple of good views to that. First off, kind of on the -- dovetailing on one of Sabra's comments about capital. With the structure changes on the high net worth business, PCG business, that's on a forward basis should make the structure of our CAT program, maybe in [ the risk ] program, cheaper for AIG because you don't have that kind of concentration aspect in there. And it's in, but in to a much lesser extent. And so on a go-forward basis, everything else being even from that, you could enjoy a cheaper program going forward. Now on the second part of it, any company should be reevaluating its reinsurance structure. It's a function of the -- not just the marketplace, but what the marketplace is doing to your view of the gross profitability. So to the extent that you want to retain more because either the volatility feel is better and the odds of making that profit are better, you may wind up ceding less. So -- and if you have demonstrated a material change in the book of business that's audited by reinsurers and verified by reinsurers, you have a strong argument for actually a ceding commission, a positive change. So just look at it this way. In certain pockets, there is an opportunity to grow on the front end and possibly to grow because compoundedly because of less session out the door.

Got you. Makes sense. Here's another one that just came in, I think, is a good one here. So if we take a look at where your stock is trading right now, obviously, valuation is really low. And I think we could probably attribute part of that to the Life Insurance aspect of your business, right? Life Insurance are trading pretty low. But even on some of the parts, at least from my math, it still looks really, really cheap here. The question then becomes, what kind of thought have you kind of put into maybe breaking up the GI and the Life Insurance businesses here, would AIG be better off actually having 2 separate businesses? Being 2 separate companies publicly traded, whatever it is.

Yes. Well, I don't think that's not been in management and the Board's mind. I mean that kind of thing always is thought of, going back to years and years and years ago for when I joined the company. So I mean with the tax reform, of course, the DTA argument lessened, right, because of what you'd lose on foreign tax credits and NOLs, but it's still there. It's still a big number. Let alone the diversification loss and capital that you get mostly from the rating agencies. When I think of diversification loss, I don't think of an internal capital because that's inside baseball, what are your real binding constraints on capital, which is generally external, these rating agency forces and others. So it's the diversification change associated with their view of it. And the combination of those 2 things are still material, but I don't want you to think that it's not something that management would routinely strategically think about. Do you have anything you want to add?

Yes. Look, I would just add that those 2 reasons that are there are still obviously significant for shareholders and would have a significant impact on shareholder value creation. But I would just note that, strategically, there's no reason why our GI business needs to have an L&R business. They have separate sales force and separate operations and the rest.

Got you. Makes sense. Let's see. Next one. Given where we are with current yields, I guess it sounded like on the recent conference call that you're not expecting base spreads to kind of be greater than kind of -- greater contraction than the original kind of forecast was, which you kind of laid out. Maybe remind us kind of what the impact...

The 8 to 16.

Yes, the 8 to 16. Is that still what you think even though rates have come down even from those levels, I think, when you originally gave us the 8 to 16? And then maybe on that topic, what's kind of the outlook right now for sales and surrenders in the life and individual business?

Do you want to take that, Sabra?

Yes. I'll start with it. On the 8 to 16, that is up from where we were seeing the compression going back 9 months ago or so. And obviously, a 100 basis point drop in the 10-year treasury over that time. But definitely put some pressure on the reinvestments. But the reason why it's still in the 8 to 16 range, I would say, is that, first of all, credit spreads are a lot wider. So that has actually helped with some of that compression. Yes, we're largely an investment-grade corporate investor in the Life and Retirement. And the other aspect of it is that the portfolio duration of Life and Retirement is like 8 to 9 years, so there's only a certain amount of the portfolio that's rolling off. There's actually been -- and it will be interesting to see how this changes. Last year, we were seeing a lot more compression coming because there's a really active tenders and calls on corporate debt in the second quarter just with all the market disruption we had. We didn't -- we had some other yield enhancement income, but it wasn't to the same impact. So what I would kind of point people to is keep the eye on credit spreads as much as you keep it on the risk-free rate, and keep an eye on what's happening with bond call activity because that -- those are really the key drivers that would impact the spreads. And then, obviously, the other thing in Life and Retirement is really what happens with our alternatives. And that's been a tale of 2 cities so far. First quarter, we had a negative impact on the hedge funds. But the private equity, as you know, is reported on a 1-quarter lag, so we had decent private equity returns in the first quarter. Second quarter, we had big gains on hedge funds because of the market, but then we booked losses on the private equity. So as you look into the third quarter, at least for private equity, we know what the market was at the end of June. So knock wood, that should be a favorable margin on the hedge funds. You tell me what the market looks like at 9/30, but so far this quarter looks pretty decent. And that's -- like I said, that's another component of the net investment income for Life and Retirement.

Makes sense. Okay. I know we're just about out of time here. I know we've got other things -- other meetings to get to. Let me ask just one quick question here. Assumption reviews in third quarter, do you have an assumption [ review ] in the third quarter with your Life Insurance business? And any reads on potential charges there?

No. It's too early, Brian. They're still going through that. But the only thing I could offer at this point is my comment on mortality, where it wasn't severe enough of a change from the pricing assumptions that made us say, we got a -- let's front-load that into 2Q. So they're going through that now. So I don't have any antenna. I don't have any cilia vibrating.

Got you.

Yes. And Brian, I would just add to that because we have been asked by people about like a one-factor sensitivity for the change in the interest rate on the DAC. And I know some of our peers do that, but just remember that the assumption study looks at everything. And when you look at lower rates, you're also generally looking at lower surrenders on the product that you have that's fixed rate-oriented. So it's not just a one-factor sensitivity, we look at everything from mortality, surrender activity, policyholder behavior with respect to utilization of guarantees, which is one area that can surprise you. People don't always behave economically. And then obviously, expenses, market rate assumptions, interest rate assumptions and the rest. So it's really soup to nuts.

And just to highlight also, Brian, what Sabra talked about before on spreads. Everybody likes to focus on the 10-year bullet, right, and that's just one piece of the pie, right? And only one piece of the pie. And one of our competitors this quarter who does their assumption in the second quarter, I think went from a 3 75 to a 3 25, and we're already at 3 50. So on the bullet, if you will, that's just the risk-free, there's already not much of a difference.

Makes sense. Makes sense. Mark, Sabra, Shelley, I really want to thank you all for your time. I apologize about all the technical difficulties here, but we got, I think, some good content, good discussion. And I really, really appreciate your time. And everybody, just want you to stay safe, stay healthy, and thanks for joining us.

Yes. I appreciate the invite, Brian. Thank you.

Thank you.

Thanks, you all.

Bye, everybody.