Mark Kociancic: Well, first of all, I go back to my other remarks that I was making. So you’ve got payment patterns for 2016 to 2019, well developed. And so we’re showing ultimate loss ratios in the ’16 to ’19 period, which are markedly elevated from the initial loss picks. So that’s one part. When we switch over to 2020, 2023 time frame, out of an abundance of prudence, we started with elevated loss ratios to begin with. So over and above what we would have expected. Then you’re getting the rate in addition to that, and you’ve got the portfolio management, which is eliminating some of the root cause of the ’16 to ’19 development. So it’s not just applying a raw number, a social inflation factor. There are other things that go to mitigate the development that could happen from 2020 to 2023 and probably more importantly, how do we get comfortable with those years.
So GL for us from 2020 to 2023 still looks very good, good to us. There’s — we don’t have an issue there. From 2016 to 2019, it was the lion’s share of the problem that we’re solving with this reserve charge today in insurance.
Juan Andrade: Yes. And again, Josh, this is Juan. Just to add a little bit more color on that. Look, the bottom line is the factors you’re talking about have already been addressed. And again, I would reiterate some of the things that Mark said, I think it’s very important. They’ve been addressed, number one, by much higher loss picks that we started to put in place really at the end of 2019 for 2020. That’s number one. The fact that we raised our inflation assumptions essentially in our loss trend select, so we price to it. Number three, the fact that additional pricing in excess of trend was coming in through that period of time, and on the underwriting side, the fact that we did a lot of different things, for example, increasing our loss sensitive mix, lowering limits, racing deductibles, all of that basically helped to give us the confidence that we’re talking about today in addition to exiting social inflation prone industry classes.
Josh Shanker: Okay. I know it’s a complicated issue. I’ll take anything else have offline. Appreciate it.
Operator: And our next question today comes from Elyse Greenspan at Wells Fargo.
Elyse Greenspan: Following up on the reserves as well. I was hoping can you give us a sense on where you’re working your reserves overall and in each of the segments, insurance and reinsurance relative to the actuarial midpoint and where was it before?
Mark Kociancic: So we’re clearly looking to best estimate plus a margin overall. So we do feel confident with that. I can tell you two pieces to the margin point. So one is what’s embedded in the balance sheet. And I think we’ve still got a good chunk of seasoned or embedded margin in the reserves. And then we have the more green years, particularly in long-tail casualty, which we feel strong about as well. But that’s going to take some time to play out. So I would say there’s an embedded margin in that’s diversified in the current set of reserves. The second part is really the flow or the engine that’s producing it. And this is what we’re trying to emphasize with a lot of the remarks we made in the model lives. The steps that we’ve taken to ensure that we are producing margin accretive business, and we’re just taking our time to let it season and we’re not touching it right now.
So when you think about the reserves that we’ve released, for example, you’re seeing mortgage, for example. There was — most of that release was from 2013 to 2019. So well-seasoned, very prudently reserved. We’re letting that out. Similarly, on the property side, you’ve got 2020 to 2022 as the lion’s share of those releases. Again, short-tail, well-seasoned, well defined, we’re letting that out. And so we have this engine that is producing and embedding margins. And for longer tail lines, it takes more time to let it season for shorter tail, it can become available sooner. And so to get back to your point, best estimate plus a margin, feeling good about our positioning and the flow of margin in the future.
Juan Andrade: Elyse, this is Juan. And just to add to what Mark is saying. To think about what we have been saying and said in this call today about the quality of the underwriting really over the last four years, the pricing environment that we’ve been in ahead of loss trend the portfolio management actions that we have done and how we have crafted both books to be higher quality. What that basically means is that there’s more in the tank and there are lots of good news in the future that we just haven’t touched for the reasons that Mark just articulated. They’re just not well seasoned yet.
Elyse Greenspan: And then my follow-up question, you guys said that you made an adjustment in reinsurance, but I think, Mark, you said that it was marginal for some of those years, 2016 to 2019. Can you give us a sense of what the reinsurance charge was for those years? And I guess, why you didn’t think you had to embed some extra conservatism and move that a little bit further?
Mark Kociancic: Yes. It was marginal. You’re looking at a very small percentage of the total reserves, low single-digit for those exposed years. I think there’s a couple of things that you can look at, and I would start with the reserve charge we took in 2020 that was fairly meaningful at $400 million, and most of that was going into the casualty years from 2016 to ’19. So I think we took a good chunk of that apple. Over the last few years, both sides, Insurance and Reinsurance, there has been some minor adjustments for those years throughout the time. So we have been nibbling away at the data that we’ve been seeing. 2023, I think, was more pronounced in terms of industry loss data for those years. But we were in a — we were just in a much stronger position from a reinsurance point of view, on those years. I don’t see any problem going forward on either side for ’16 to ’19 in either segment.
Elyse Greenspan: When you say low single-digits, do you mean low single-digits millions or low single-digits as a percent of.
Mark Kociancic: Percentage, percentage, yes.
Operator: And our next question today comes from David Motemaden with Evercore ISI.
David Motemaden: Just following up on the reserves. So Mark and Juan, you guys spoke about specific programs that are driving the insurance charge mainly on GL, but I guess I would have thought that the trends impacting those lines are the same trends that are impacting the rest of your book. So I’m just wondering, as you guys take a look at it, how do you make the conclusion that it’s isolated to these books in these accident years and not other programs across GL, but other lines in general?
Mark Kociancic: I think the data for us is quite definitive when you look at it. So in particular, 1 program and 1 block of business is driving a very strong majority of the development that we’re seeing in GL. It doesn’t mean it’s 100%, but it’s clearly a strong majority. And just in terms of other lines, made clear, umbrella, for example, on our side was performed well during that 20 to 16 — to 2016 period. Professional liability, very minimal impact for us. So this is really general liability, and there were two main components for us, which we identified some years ago and began to act upon. But the 2016 to 2019 aspects of those two are what we’re dealing with today.
Juan Andrade: Yes, David. And what I would add is those two items that both Mark and I have mentioned, those are niche type businesses, not sort of standard type of GL. And essentially, we have closed that book for all intended purposes.
David Motemaden: And then. Yes, just on that point, just closing the book. I guess I’m sort of just wondering any other color you can give us that would help us feel comfortable that this is behind us because I kind of look at the charge that you guys took in 2020. And then after that, there have been more additions. It sounds like you guys did a little bit here in 2023. So what makes you think we don’t sort of see that same trend happen on the insurance side as well?
Mark Kociancic: I think there’s several points. First of all, you’ve got very well-developed reporting patterns for the ’16 and ’19 years. So again, anywhere from 70% to 90% development, completion of development for 2016 to ’19. So that gives us a lot of confidence that we’re dealing with actual data. We also have assumptions that I think are very prudent in terms of the social inflation impact. And so when you combine that altogether, we’re able to capture with a high level of confidence what we think the ultimate loss ratios are going to be for really ’17, ’18 and ’19, ’16 is pretty much I think, done and really didn’t move that much for us. The larger point is 2020 onwards, and I think this is what we’ve been trying to emphasize is we’re benefiting from several factors, you can see with the process that we’ve embedded not only on the reserving side, but how we act upon information inside the company.
