We're going to get started now. I want to welcome Peter Zaffino, who is AIG's Chief Operating Officer, Chief Executive Officer of General Insurance. We also have a solid team from AIG here, including Sabra Purtill, who heads Investor Relations, Rating Agency, Communications, Deputy CFO. I'm sorry?
Rating Agency.
Not Communications. I'm sorry. Rating Agency Relations, Deputy CFO, and Head of Corporate Development. Rose Marie Glazer, General Counsel. A solid team. Peter's going to start off making a few introductory comments, then we're going to head into Q&A. I do obviously have a list of questions, but if there are questions from the room, obviously, the point of this is for everyone to get the information that they need. Please do not hesitate to signal to me if you've got a question. We'll get the mic to you then proceed there. With that, Peter, please.
Great. Good morning, everyone. I thought I would start with some prepared remarks. A lot happening at AIG, I thought I would baseline, then we'll spend some time with Meyer Shields. Meyer, thank you for hosting us at the KBW Insurance Conference. I'm going to start with about 10 minutes of prepared remarks, and we'll begin with a few comments about Hurricane Dorian, which continues to unfold as we speak. Dorian proved to be an interesting and challenging storm to track. It strengthened into a major hurricane on Friday, achieving Category 3 status, reached Category 4 status by Saturday morning, and by Sunday was a Category 5, as you all know. Hurricane winds were peaking at 180 miles per hour. There's only been three hurricanes since 1950 that have been stronger. It was a serious storm.
Forecasts were continuously updated with wide dispersion in the storm track, resulting in a high level of uncertainty in landfall and potential losses, and that is still fluid. While there was concern about landfall in the Bahamas, Florida, Georgia, and Carolinas, in the end, Hurricane Dorian stalled over the Bahamas and has downgraded to a Category 2 hurricane. We continue to keep a close eye on Hurricane Dorian and in particular on the potential for landfall in the Carolinas. I'm trying to draw an analogy and a comparison, Hurricane Matthew in 2016 is an interesting comparison to Hurricane Dorian, both in intensity and storm track. The industry then, and again with Hurricane Dorian, avoided much of what could be a significant loss.
When we looked at Hurricane Matthew, we'll have to do the same thing with Hurricane Dorian once the information is more accurate, is that Hurricane Matthew could have been 10 times its size with just 30 miles west. They're really sensitive to the track. I just think it's a real reminder of the continued need to be diligent in the management of aggregation of property risk and also how it's priced. Working with the team over the last few days really crystallized for me the significant progress that we've made in General Insurance over the last two years, and I'll spend more time talking about that. Our focused and disciplined approach to repositioning the portfolio has led to smarter decision-making, more predictable outcomes, and improved performance. We went from an insurance company that had an inconsistent approach in the marketplace.
We took outsized risks, created significant volatility, and deployed substantial gross and net limits. Now we feel we're becoming an insurance company that has a clearly defined risk appetite, disciplined execution of underwriting and reinsurance strategies, significantly less volatility, and demonstrating market-leading behaviors that continue to improve our overall portfolio. We've taken significant action to position our businesses to be leaders in their respective markets by making foundational changes to our underwriting approach and reinsurance strategies while focusing on building and retaining a best-in-class team. We're operationalizing these actions through General Insurance by embedding more disciplined end-to-end business process and prudently managing expenses, which will ensure consistency and sustainability for future growth and investments.
The significant progress we've made towards achieving full-year underwriting profitability is the result of a change in our business mix, reduced volatility stemming from our underwriting actions, a very comprehensive reinsurance program, along with the acquisitions of Validus and Glatfelter that contributed. It's worth revisiting the significant changes we made to gross and net limits. I think it's really compelling. On our year-end 2018 earnings call, we disclosed that in property, for example, gross limits deployed in the field were going to be reduced from $2.5 billion to $750 million. Net limits were reduced from $611 million at the end of 2017 to a range of $5 million-$50 million as we enter 2019 on a per-risk basis.
In primary P&C, gross limits were reduced by over $8 billion in the aggregate throughout 2018. We also materially changed gross and net limits in Lexington, which is our E&S entity within AIG, Financial Lines and Casualty. I'd like to remind you of the dramatic change in our reinsurance strategy, particularly in light of Hurricane Dorian. You may recall that AIG's 2017 North America cat cover consisted only of a per-occurrence protection attaching at $1.5 billion. We experienced material catastrophe losses with hurricanes Harvey, Irma, and Maria in 2017, none of which result in recoveries from the North America cat cover. Since then, we've enhanced the program significantly by reducing the per-occurrence attachment point in North America to $750 million and adding a global aggregate protection that attaches at $750 million with a $100 million per event deductible.
In addition, in the second quarter of 2019, we entered into separate discrete cat covers that further reduce our retentions in the AIG Private Client Group and other geographies that have catastrophe exposure. We also made significant changes to our reinsurance programs in Japan and other international jurisdictions. Allow me to briefly comment on rate trends. As Mark and I said in our second quarter earnings call, we continue to see meaningful rate increases across our portfolio. In the high single digits in the second quarter compared to a mid-single digit improvement in the first quarter of the year. Note that nearly 60% of the second quarter's gross premium was associated with double-digit rate increases, whereas less than 10% was associated with rate decreases. Our comprehensive and disciplined strategy to reposition our businesses as market leaders has improved the quality and rate adequacy of our overall book of business.
Given recent news reports, I also briefly want to touch on recent developments concerning the litigation related to opioids, which have been filed primarily by state and local governments against manufacturers, distributors, and retailers of opioids. While it's too early to predict or quantify the outcome of all the litigation or the insurance that may apply, we've been very closely tracking these developments so that we can address these complicated risks appropriately as they continue to evolve. Before turning to AIG 200, I'd like to comment on our talent. I've actually talked about this quite a bit over the past two years. In general insurance, we continue to strengthen our underwriting leadership team and new talent to our reinsurance organization. We are expanding our operational capabilities that link to the rest of the organization and overall continue to deepen our bench.
I'm very proud of the team we've built. Our improving financial performance is a reflection of their dedication, expertise, and commitment to the journey. Now I'd like to spend a few minutes on AIG 200, which we first mentioned externally on our second quarter earnings call. Meyer wrote a little bit about it. Brian asked me to lead this program in my capacity as AIG's Global Chief Operating Officer. AIG is focused on the long-term strategic positioning. We're identifying opportunities to achieve operational effectiveness, recalibrate our standard, and become a high-quality, top-performing company. The elements of this program have been under development for several months. Colleagues from around the globe have been involved, providing information and suggestions regarding AIG's current strengths, development needs, and obstacles standing in the way of sustained growth and success.
By the second quarter, this program had matured sufficiently to officially brand it internally and announce it externally. The name AIG 200 was selected because it represents a pivot away from AIG's past and focuses on investing in critical areas as we build for the future. While the primary purpose is not about cost-cutting, we do expect our work will ultimately lead to a reduced expense base over time and, more importantly, an improved experience for our distribution partners, clients, policyholders, and colleagues. There are four core objectives to AIG 200 that the work will undertake and will align to. One is achieving underwriting excellence. Two is modernizing our operating infrastructure, including using straight-through processing and replacing some unnecessary manual efforts with technology and digitized workflow. Three, enhancing user and customer experiences. Four, becoming a more unified company.
While efforts are still underway to determine the size of required investments and the expected financial impact, I thought it would be helpful to add some detail by giving a couple of examples of the work we're undertaking. First, in Japan. In early 2018, we completed a merger of our consumer and SME-focused businesses into one entity in our go-to-market consumer brand, AIG Sonpo. As part of AIG 200, we will improve the strategic positioning of this business by transforming it into a next-generation digital insurance company. The designs of the new digital-first operating model has been completed and once executed, will bring the kind of anywhere, anytime, any device experience to our customers and agents that they expect in the digital age. This new operating model will drive significant efficiencies in our business as we will be replacing paper-intensive manual processes with automation and digitization. Second, shared services.
We've identified this area as a significant opportunity and priority for AIG 200. Shared services has delivered value for AIG. However, that value has not been fully realized, and the teams in the areas are siloed. As part of AIG 200, we'll enhance and transform AIG's shared services from its current by-function, by-business model, which is largely focused on lower cost transactional efficiencies to a best-in-class, fully integrated, multifunctional digital end-to-end process operation, delivering value through scale and simplification. Our new shared services operation will be focused on customer and user experience and deliver top-performing benchmarks with respect to cost, quality of service, and speed of execution. We're in the process of consolidating our existing shared services into one team and adding resources with significant shared services and functional domain expertise to execute on this transformation.
More broadly, I want to emphasize that AIG 200 is a multi-year, enterprise-led program that will define who we are as a company, how we differentiate ourselves in the global insurance marketplace, and how we create value for shareholders, clients, policyholders, colleagues, and other key stakeholders. AIG 200 is designed to move our infrastructure meaningfully into the future as we work to sustainably improve financial performance over that time. Similar to General Insurance, we're filling critical roles with seasoned executives with proven track records of achieving excellent results during transformation. Significant resources are being deployed to AIG 200, and we expect to provide an update on our third quarter call. Meyer, I hope that was informative and at least provided some insight on some of the questions you may have. I think with that, we'll have a discussion.
Fantastic. No, thank you very much. From my perspective, I want to say that one of the most reassuring aspects of AIG 200, and again, we're obviously learning a lot about it, is that you've had the time and been able to address sort of the critical underwriting pressures within General Insurance. There's so much lead time between you putting the people into place and their efforts bearing fruit that this is a demonstration, I think, of sort of the internal stabilization of the underwriting platform that you've had the wherewithal to think about the next phase and how do we move AIG to that level of stability. A lot of the questions that we get, obviously, with regard to AIG are with regard to General Insurance turnaround.
I want to put the discussion in the context of what is obviously a naive supposition, and that is that if underwriting results were unacceptable, one approach would be to raise rates on every single account such that the expected returns were okay. Again, that's not really feasible, and it's also not the approach that AIG has taken. You've done a lot more work in terms of refining underwriting appetite, broker communication, risk selection, et cetera. I wanted to drill down into some of those individual issues to really get a sense and to communicate a sense of what AIG has done differently. Let me start really with risk selection. It's been a phrase that Brian has emphasized over and over again, and he was talking about it well before there were signs of rate hardening.
I have to imagine that it involves more than going to individual desk underwriters and saying, "Write better accounts." What's involved in communicating and actually obtaining improved risk selection?
There's multiple strategic imperatives that took place. I think first was trying to define what AIG was going to be in terms of risk selection and its risk appetite. When we came in, there was just two verticals, one consumer, one commercial. In a company our size, had enormous implications of taking very large risks on multiple geographies, multiple classes, and actually having multiple points of entry within AIG. What we decided was to make certain that we had businesses that had clear risk appetite. It was clearly defined. I had a unique, as did Brian, perch from our prior roles. When you are doing significant limits that are commoditized, you do not get paid in the marketplace. If you're taking massive limits, at some point, it goes into more of a commoditized market, and then price becomes a driver.
You're not going to get paid for it. You cannot increase rates in a marketplace that gets commoditized in certain layers. We decided that one is it was too much volatility. You were not going to get paid for the risk if you're taking those substantial limits. There was no reinsurance that was mitigating some of the large limit strategies. You're taking huge losses on property and in casualty and clearly defining what that risk appetite was going to be for the Lexington, for our commercial property, what we were going to be doing in having a pure E&S player. Not telling the underwriters and leaders, like, "Okay, please get to good risk selection.
Please get the appropriate price in terms of conditions, but I'm going to measure top line every month as to where you are." We have to make sure that we're reflecting upon our growth, and we think that there are some real green shoots for growth. At the same time, we want to create a culture of underwriting profit, underwriting excellence, and that's been the journey we're on.
Okay. Can you give us an update in terms of where you are and where you need to be in fostering and communicating that culture?
Well, we have some terrific leaders that have joined the company that have tremendous track records of underwriting profitability. Dave McElroy comes to the top of the list, Mark Lyons, Tom Bolt. We have strength in our chief underwriting office by taking Kean Driscoll, who was the chief underwriting officer, is now the chief underwriting officer in property, was the CEO of Validus Re, Lex Baugh. We've really strengthened the senior leadership teams to be able to drive the messaging and then have started to be very active in building the bench. As you go down in AIG, there's a lot of very good people, and they've calibrated very quickly to what we wanted to revise for our underwriting strategy and risk appetite.
It's constantly operationalizing what we say we're going to do. We monitor and meet with the businesses every quarter to make sure we're making the progress. Quite frankly, what we've been able to do over our calendar year, I think, is tremendous.
Yeah. Fantastic. Again, if there are questions in the room, please don't hesitate to let me know. I want to touch on another, I would say, critical component of the strategy, and you've referenced this a little bit, which is the reinsurance purchasing strategy. There's been a significant uptick in reinsurance purchasing, I would say, going back to the immediate aftermath of Hurricane Irma. Since then, at more typical reinsurance points. If the business is better priced, then where does the importance of buying more reinsurance interact with that?
There's several categories to think about with reinsurance. First is the AIG that Brian and I entered did not buy a lot of reinsurance. When you look at the massive activity in 2017, having no recoveries on North American cat, that's probably not structured right. There was no such thing as an international catastrophe program for property at AIG. Again, not best practices. Some of the things we had to do was just core to the organization of making sure that we did not have outsized volatility and we did not have unexpected results, and we preserved capital. That was kind of the baseline strategy. Even the property cat program in 2017 had no reinstatement limit. You don't want to be out buying reinsurance after a cat. It's not healthy, and it's not cheap.
Coming up with a core program and then looking at the company as one organization where it was bought in silos. You saw what happened to us in Japan. That was largely done because we bought it in two separate towers. It was an old structure, consumer and commercial, same capital base. Just getting a philosophy of how are we going to look at aggregation across the globe. The other element of it is just not comfortable with taking significant net limits. Now, when I say taking significant net limits, it's relative to what we were taking. I don't apologize when we go to talk to clients that I can only put out $750 million of property. That's a healthy limit. That is a lead limit. That is something that's going to help and drive leadership in the marketplace.
I think also pivoting to the right language, that we are still a very large insurance company that can be very prevalent in solving risk issues for clients and for brokers. The last one is where we just felt, again, that large limit strategy wasn't just for property, it was also in casualties. That was an area where we felt the book was underperforming. In a period of time where we're transitioning into better gross underwriting, having reinsurance partners there for us to do a variety of things. One is to help us stabilize volatility. Two is to work with us in partnership. On our core casualty treaty, we have 15 reinsurers, they're in with us partnering all the time in terms of how we're doing certain things, how are we actually positioning ourselves in the market?
Are we doing the things we said on risk selection in terms of conditions and insights? I think that's helped us accelerate the turnaround.
One area that it's not quantifiable, but I think it's instantly important. Again, I'm drawing on my own experience because when I was actually in the industry, I worked for a large multinational company that wasn't making any money. I like to think that wasn't all my fault, but it was certainly reality that we were defenseless. We didn't understand enough about the risks we were taking. To some extent, the brokers serviced their clients well and said, "Hey, you can get cheaper insurance at company A, company B," without them knowing that they're underwriting at a loss. Can you talk a little bit about the broker communication? Obviously, based on where Brian had spent time and where you spent time, there are some strong relationships there.
Can you take us through the process and the product of redefining to the brokers what it is that you want to do?
That's a great question, you'll have to ask them because I'm just going to tell you what they tell me, then you can validate it. I think the brokers very much want a very active and very thought leadership quality in AIG. They very much want a strong AIG. I thought coming in from running a global broker, that would get more resistance to not, again, doing a risk appetite change, repositioning the portfolio, new people, the receptivity has been overwhelmingly positive. Certainly out of the gate, the first couple of quarters, getting that risk defined. Brokers want, if they could pick one thing and they'd like probably 50, but if they could pick one, it's certainty. Tell me what you're going to do and make sure you deliver on that.
When we're going through the initial phases of the turnaround, it was hard to articulate what we wanted to do and execute on it. That happened very quickly for us. I think the relationships with the brokers have been fantastic. You use data points. My first RIMS, it was that dialogue of, what are you doing to change risk? Who are we going to be doing the underwriting with? The second RIMS, which just happened this April of 2019, the conversation had totally changed in one year, which is how we can be relevant to their program, how we can help them. I think we have industry-leading cyber expertise, thinking through the implications of that and how we look at our multinational platform and where there's services that can differentiate. I think the conversation changed.
Quite frankly, we outlined to them in very clear words what we were going to do. We were going to strengthen our gross limit strategy. We were going to not have multiple channels where brokers can find ways into risk. We were going to have the Lexington be a wholesale distribution channel and be an E&S player. Where we think we need more rates, we're driving more rates. Again, I think that the receptivity, it's not like they're overly ecstatic about it, but I think they're receptive to it. They've partnered with us, I think it's worked well. They've really been very helpful.
Right. It's an interesting thing because simplistically, if you're reducing your gross limit size, you're making more work for the brokers.
Yes.
For them to respond positively, they have to be optimistic enough in terms of the coming stability of what it is that you're trying to do so that there's not another plan in two years.
Correct. I don't see us changing. We will revise the plan. Where we see opportunities for growth, we will make sure that we pivot. Again, if we were sitting here a year ago, would we talk about being an industry leader in certain classes? Probably not. I feel like we're pivoting and accelerating to be able to do that. What next year looks like, I think it's going to be more areas. I think we'll be able to articulate pockets where we think we really want to grow and make sure that we further refine the capabilities that we have. Again, AIG 200 will be part of that as we bring the company together to have a much better user experience for all of our different stakeholders.
Okay. Again, if there are questions, is there a microphone in the room that we can bring?
Just following up on that, given all the changes in your growth and net appetite, but also the pricing environment now, could you maybe talk about your outlook for NPW for next year?
We're going through a budgeting process now. There's a bunch of different variables that will contribute in terms of what the growth will be or if we shrink. We're very focused on, again, underwriting profitability. There will be opportunities. We're still earning in some of the treaties that we purchased on some of the core casualties. I'm going to stick with the reinsurance for a second, is that as we improve on the overall growth portfolio, I think this is part of Meyer's question, there will be opportunities where we may not need as much cat reinsurance just because we don't have the same PMLs that we started with or where we are today. Will we bend the curve in 2020? We will be able to do that. How much and to what magnitude is yet to be seen.
Again, I don't think I was talking about growth at this point last year. I think there's some opportunities, particularly in A&H. We have a great A&H capability. We now have a single leader. Again, back to AIG was highly fragmented. We've brought in a team. We've been investing in digital capabilities all year, and very much want to grow that as one of the single largest opportunities for us in the global insurance market. You'll start to see opportunities of growth there. We'll take a look at other portions of the portfolio where we think there may be unique opportunities. Our messaging is delivering underwriting profit.
We focus and I think that we've exhibited that we are watching the net premium written and net premium earned very carefully because you haven't seen our expense ratio increase during the period of time where we've seen shrinkage. We've managed that very carefully, and we'll continue to do so. I do think there's going to be some opportunities for growth. The remediation isn't fully done, so we'll be worked on in 2020. That may require us to shrink a couple of lines too.
Okay. Ben?
Thank you very much. Can you talk a little bit about some of the, I guess, what could be emerging liability concerns that are in the market over the last six to 12 months, reopening of laws and others, and how your reinsurance program comes into play for some of those legacy liabilities?
That's why I put them in my prepared remarks because I knew I'd get the question. It's a very complicated issue. It's emerging. As I said, I think that the fact that state and local government is dealing with a lot of the issues, we've seen a lot more in terms of whether it's revised or statutes. There's a lot happening. We're very coordinated in AIG and focusing on it, and we'll provide relevant information that's fact-based as we can each and every quarter where there's relevance. It's something that we're watching very carefully, and there's a lot more, as I said, in the last couple of quarters that has focused us on making sure that all the different areas where we may have exposure that we're carefully thinking through.
Maybe at a high level, could you just touch on the years of coverage that the NICO cover?
Again, it's a very difficult question because I don't know what I'm covering. To look at what exactly emerges and what is covered under products or not in certain cases or what might be excluded. I have to wait to see how it emerges on the direct side before I would understand how it's going to apply to the reinsurance side.
Okay. Thank you.
Okay. We've got a question in the back.
You mentioned earlier that you're likely to give an update on the Q3 call for the AIG 200. Is it likely that we'll get some numbers at that time? Meaning, potential positive impact on expense ratio, likely investment charges, that sort of thing, or is it still too early for that?
We may be prepared, I think if I was to look at it today, I would say it's probably premature to give a lot of specificity. The reason is I gave you two examples. There are multiple programs in flight that we want to make sure that we're getting the right priorities, that even talking through the shared services, that's fully scaled. We understand the implications of the cost to achieve the benefit, and that we can commit to our stakeholders of what the run rate benefit is. The work that I said that we've been doing over the past few months has been bottom up.
There's been a lot of great ideas that have surfaced from colleague input. We want to make sure that we're not rushing to the answer and that we're looking at the implications and the intersections that exist in the organization to prioritize the biggest opportunities. Again, I highlighted two. There's many more. We want to be able to come out with a coherent strategy and not piecemeal it as to what's going to be the cost to achieve and what will be the run rate benefit. The third quarter may be premature. We're working towards getting much more specificity.
I have another question. I'm going to change direction a little bit. I'm actually very happy that this is a question that's coming at the end rather than the beginning of our session because it's about pricing. I think a key message from Brian the entire time has been that pricing is not the primary issue and that there are improvements that can manifest themselves without benefit of pricing. Now, we are seeing pricing move. There's probably some give and take in terms of how much of that is a cause of AIG's movement and how much of it is an effect of AIG's movement because we do have a lot of market clout. I was hoping you could talk a little bit about what you're seeing.
In general, I want to frame this in the context of pricing changes that AIG is now enacting versus market pricing. Are there areas where you're able to, because of what is, I think, still sustained market dominance, maybe some pricing issues that are a vestige of previous averages that actually allow for a bigger gap in pricing relative to loss trends?
There's several ways to answer that. One is, again, this is more of an anecdote, it will emerge over quarters, is that we may have liked the risk. The way it was structured, the terms and conditions, the limit management just was poor. What I observed over the past several quarters is that we have been on accounts that have had losses, but historically would have had 50, $100 million, we're now at 10. I think that you, again, you don't have the volatility, that risk selection and how you position the portfolio, it takes time, is improving. On the pricing, again, we're a lead.
What happens with the rest of the market, I always viewed, again, in my prior role that the large global insurance companies or the big domestics are really the ones that have to demonstrate leadership. We're very focused on what we want to do, in all aspects of our business. When I look at what I cited on the second quarter about sort of primary D&O that we were pushing rates around 30%. Our policy retention is north of 90%. We're shrinking our limits dramatically. When you look at the premium, you're like, well, they're not really driving as much premium, but we're getting, again, 30% rates with much less aggregate, that I cited, again, on the second quarter call.
There's a dynamic there of if you're pushing rates, and you're having significant retention issues, then maybe you're not really achieving what you want to achieve. If you're pushing rates, shrinking limits, you like the way your book is performing, and you're getting 30 points, and that's not where the loss cost trends are, that should position that portfolio well over time. The Lexington I talk about it a lot because it's just such a dramatic change. If we were changing our distribution strategy and changing risk appetite, we wouldn't be getting 50% increase in submissions. That submission flow has been dramatic. It's coming from the wholesale market. There's two dynamics there. One is demonstrating leadership position and making sure that our risk appetite is well outlined, taking less limits, all of that coming in, allowing us to underwrite better.
They're responding to that. The second piece of this market that's changing is more is going into the E&S market. Our timing of strategically repositioning that and being very disciplined in the risks that we want to underwrite and getting that risk appetite out and having a clear and consistent strategy to our stakeholders is really proving to be very rewarding. Again, it's early days, but all of that is very positive.
Yeah. Fantastic. We are at the end of time. Obviously, we could spend hours talking about what's going on at AIG. This has been very informative and very helpful, and I think very positive. Thank you very much.
Thanks a lot, Mark. Thank you.