American International Group, Inc. (AIG) Q3 2012 Earnings Report, Transcript and Summary

Good day, and welcome to the American International Group's Third Quarter Financial Results Conference Call. Today's conference is being recorded. At this time, I'd like to turn the conference over to Liz Werner, Head of Investor Relations. Please go ahead, ma'am.

Good morning, and thank you, everyone, for joining us this morning and all your efforts last night in the face of some of the challenges, given the hurricane. I do want to just let you all know that management's dialing in from multiple locations, so we just ask you to be a little patient, particularly during the Q&A period. But on the line today we have our senior management team, including Bob Benmosche, President and CEO; David Herzog, Chief Financial Officer; Peter Hancock, CEO of AIG Property Casualty; and Jay Wintrob, CEO of AIG Life and Retirement. I'd like to remind you that today's presentation may contain certain forward-looking statements which are based on management's current expectations and are subject to uncertainty and changes in circumstances. Any forward-looking statements are not guarantees of future performance or events. Actual performance and events may differ, possibly materially from such forward-looking statements. Factors that could cause this include the factors described in our 2012 Form 10-Q, our 2011 Form 10-K and our Form 8-K filed on May 4, 2012, under Management's Discussion and Analysis and under Risk Factors. AIG is not under any obligation to expressly disclaim any -- or update any forward-looking statements, whether as a result of new information, future events or otherwise. Today's presentation may contain non-GAAP financial measures. The reconciliation of such measures to the most comparable GAAP figures is included in our financial supplement, which is available on AIG's website, www.aig.com. Now, I'd like to turn our call over to Bob Benmosche. Bob?

Thank you, Liz, and good morning, everybody. We've got a crisis that we're all facing, but it seems like a lifetime ago, but during the third quarter, we did, in fact, do another major repurchase of our shares. And the Treasury was able to sell down to less than 16% of AIG. So for the year, we've been able to buy back $13 billion of our shares, $8 billion just in this quarter alone. And we've also -- by the way, I'm on Page 3 for those of you following the presentation -- also made good progress on our bank facilities and other unsecured lending especially at ILFC. Big question was once we did the sell-down, what would happen about regulation and who would be our regulator. And the Federal Reserve has, in fact, begun its supervision of AIG. I'm going to also add that people wondered what would happen with our capital management program once that happens. While our focus has been on share buyback, up until now our focus going forward is on capital management, working closely with the Fed, in terms of what we're able to do. We're going to now focus on our coverage ratio and that's going to be looking at our debt and our ability to cover the debt with earnings. On the Property Casualty side, again, we see good progress in terms of our reshaping that business. Peter will give you more color in just a moment. We are seeing rate increases and we should focus on our accident year, because that's where we're making progress in terms of the changes we're making. You'll see gradual improvement, which we said would occur. And we're continuing to focus on making sure that our reserves stay strong and we're working through making sure any development is dealt with very, very conservatively. On the AIG Life and Retirement business, we had, again, a good quarter. The markets helped us a little bit. We did have some adverse impact, and Jay will take you through a couple of the items from the past that we dealt with in this quarter. Our variable annuities sales and we've talked before about how we designed that product, we think we have a very good risk-managed product just in this design alone, and so our sales continue to grow. And Jay will talk a little bit about our base yields and what's happening in this low interest rate environment. Last, but not least, on the Mortgage Guaranty. We continue to see great progress in that business as the market begins to turn and we're having huge success with the way we're now underwriting mortgages, so that will be an add for us. One thing that David will talk about is ILFC in terms of where we are, in terms of managing that business, but more importantly, we still are committed to an IPO of that business once the markets are receptive. So let me turn it over to David who will start giving you some more color on the things I've just talked about. David?

Thank you, Bob, and good morning, everyone. As evident from our results, it's been another busy quarter here at AIG as we continue to execute, both on the capital management and on the operating front. As we look towards the remainder of the year and 2013 and beyond, capital management will remain a focus and core competency at AIG. We remain committed to our goal of $25 billion to $30 billion in capital management by 2015 and we're over halfway there or roughly halfway there at this point. We've described capital management to include more than just share buyback and this is a good time to remind you that we will consider acquisitions, investment in organic growth, debt capital management and certainly, maintaining strong capital at our operating companies, all aimed at effectively utilizing our deployable capital, increasing enterprise value, while also improving our interest coverage ratio that Bob just mentioned. All of the above options would be considered in the event of a sale of our AIA shares. As Bob said, we do not know what will be required under Fed regulation and we view this as being -- will be prudent given our expectations of becoming a nonbank SIFI. With respect to AIG's capacity for capital management, we believe, again, if we were to sell -- hypothetically sell our AIA shares, that we'd have approximately $2.5 billion of deployable capital now, and subject, of course, to discussions with the Federal Reserve, as well as rating agencies in terms of the actual deployment. For potential actions in the fourth quarter, we have about $1.1 billion of hybrids that become callable in December and we may consider calling those bonds. Turning to Slide 4 for the financials. You can see that after-tax operating earnings per share grew to $1 a share from a $1.58 loss a year ago. Growth was driven by increased earnings from both the Property Casualty and Life and Retirement segments. Book value per share was $68.79. Excluding AOCI, book value per share was $61.41, up 10% sequentially. Share buyback contributed nearly $5 to this quarter's book value growth. The effective operating tax rate increased during the quarter to just over 35%, due in part to a shift in the investments to taxable securities in our Property Casualty group and a third quarter, what I'll call, catch-up, reflecting higher full year income projections as well as a, what I would refer to as a return of provision true-up in the normal course that added a couple of points to the tax rate in the quarter. Going forward, we would expect our tax rate to be in the 30% to 31% range over time. As you know, we will not be paying -- actually be paying any taxes for some time, given the availability of our NOLs. On Slide 5, it provides a breakdown of the segment operating earnings, which reflects the 87% growth in operating income from our insurance operations, and Peter and Jay will provide additional insights in a moment. This quarter, the gains in the Direct Investment book benefited from credit spread tightening and gains recognized from winding down positions. As we have said in the past, our strategy has been to maximize the value during the runoff of this book. We may decide from time to time to terminate trades opportunistically for a variety of reasons, including derisking. We expect over half of the Direct Investment Book liabilities to run off by 2017. To date, we have realized significant value from an orderly and opportunistic wind-down. ILFC's results for the quarter reflected $98 million in impairment charges on aircraft, largely from the completion of our annual detailed portfolio review. While this is significant decline from the prior 2 years, it's a good time to remind you that there were significant industry trends driving those impairments, including the introduction of more fuel-efficient aircraft like the NEO. Also, as a reminder with respect to ILFC, AeroTurbine, our part-out company, gives us another end-of-aircraft life opportunity to optimize value. Slides 6 and 7 highlight our capital structure and liquidity. We remain well capitalized, with debt-to-total capital ratio of 20%. While leverage ratio is appropriate, we as a management team, as Bob said, are focused on improving our coverage ratios. The rating agencies have given us indications of expected improvements for an incremental increase in our coverage ratios over the next 12 to 18 months, and we intend to achieve these improvements through a combination of liability management and improvements in operating earnings in our core businesses. Cash flows from our insurance companies remain strong. We've already received roughly $5.3 billion in dividends from our insurance operations through the end of October, exceeding our annual target of $4 billion to $5 billion in dividends. These dividends of $5.3 billion includes $1.3 billion we received post quarter end. So, at this time, I'd like to turn it over to Peter for a discussion of AIG Property Casualty. Peter?

Thank you, David. Good morning, everyone. First, I'd like to make a brief comment regarding Hurricane Sandy. Right now, our immediate goals are obviously to safeguard our employees and to assist our customers. But it is too early to conclude how severe the storm is going to be from an insurance standpoint. So we're reaching out to customers, making sure our claims reporting information is available to them and standing by to assist them. But turning to our results, AIG Property Casualty continued to make progress in the third quarter. While I'd prefer the pace of change to be quicker, we remain on track with our strategic initiatives. As Slide 8 of the earnings presentation indicates, we reported operating income of $786 million in the third quarter, compared to $492 million in the comparable prior year period. The increase was driven by lower catastrophe losses, current accident year underwriting improvements and higher net investment income, partially offset by increased expenses. We recorded $261 million in catastrophe losses in the quarter, largely from crop losses related to the droughts in the U.S. Midwest and Hurricane Isaac. We believe that losses related to Hurricane Isaac were less severe than our business would have experienced in prior years, due to our focused efforts to carefully manage exposures to U.S. catastrophes. Third quarter net prior year adverse development was $145 million or 0.2% of our total reserves. The result was primarily driven by adverse development in environmental and primary casualty and the impact of changes in New York State benefit reforms on our workers' compensation business. The accident year loss ratio, as adjusted, was 66.5%, nearly a 2-point improvement over the comparable prior year period, despite an unusual increase in non-catastrophe property severe losses. We view the improvement in accident year loss ratio over the past several quarters as a strong indicator that our strategies to optimize business mix, pricing and risk selection through enhanced underwriting tools are succeeding. We continue to balance growth, profitability and risk, measuring the success of our initiatives based on overall risk-adjusted profitability. However, we maintain the capital and resource flexibility to respond to changes in market conditions. Our organizational structure promotes underwriting excellence as we've consolidated our underwriters into global teams versus the silo structure that existed in the past. We've empowered our employees to implement state-of-the-art underwriting tools and hire new talent to supplement our bench of existing talent. Partially offsetting the loss ratio improvement was an increase in the expense ratio. As we've communicated in prior quarters, the expense increase was largely due to higher acquisition costs, as we focused on higher-value lines and from greater investment in direct marketing. General operating expenses also increased from investments in people and infrastructure. Expenses will continue to remain elevated over the next year as we invest in these priorities. We expect to start to realize efficiencies and to experience a gradual decline in net expenses beginning in 2014. Property Casualty's net premiums were up 2.4% compared to prior year after adjusting for foreign exchange. Turning to Slide 9. Commercial Insurance net premiums were down slightly excluding foreign exchange. We completed the restructuring of our loss-sensitive business in U.S. casualty, which reduced premiums written by almost 1% in the quarter, but improved capital efficiency. The remainder of the decline reflected risk selection and rate discipline strategies, particularly in U.S. casualty. U.S. Commercial Insurance rates increased 8.4% while Property and Workers' Compensation rates in the U.S. increased 12.1% and 8.9%, respectively. We're encouraged by the rate environment, overall, but some lines and regions remain under pressure and we'd not characterize this as a hard market. Consumer Insurance continued to experience growth across its major lines of business. As you can see from Slide 10, this attractive segment, which includes the direct marketing channel, represents over 40% of net premiums year-to-date. We also continue to expand in targeted growth economy nations, which contributed over $1 billion in premiums. Notably, we grew by approximately 14% in Latin America and 10% in Asia Pacific, excluding Japan, during the third quarter. Slide 11 demonstrates our investment portfolio mix. Third quarter net investment income increased 20% as a result of positive marks on recently acquired structured securities. AIG Property Casualty is tailoring its strategic asset allocation to optimize profitability in the current low interest rate environment, while remaining aligned with AIG's overall risk appetite and tax position. We continue to diversify our portfolio into higher-yielding securities and reducing our concentration in municipal bonds during the third quarter. Reflecting our emphasis on capital management, we've made $2.4 billion in dividend payments to the holding company year-to-date, which includes $75 million in cash paid during the third quarter and an additional $800 million in cash paid in October. The commitment to capital efficiency places 2012 on track to be our largest dividend-paying year in recent history. Our capital adequacy levels remain in line with rating agency requirements, and we maintain strong financial strength ratings that carry a stable outlook with the 4 major rating agencies. Our legal entity simplification efforts have enabled greater capital efficiency, particularly in Europe, where the majority of our operations will be executed through one Pan-European insurance company beginning in December, subject to final U.K. court approval. In closing, I'd like to note that we have had several consecutive quarters of positive trends in business performance and execution of our strategic plans. I'm pleased to continue to make progress while recognizing that we have much more work to do. So let's turn over to Jay.

Thanks a lot, Peter, and good morning, everyone. I'm going to start on Slide 12. AIG Life and Retirement delivered another solid quarter, with pretax operating income up 75% over the same quarter last year, reflecting the positive impact from equity markets this quarter, higher net investment income and increased spread income. There were 3 items that we highlighted in the press release that are worth mentioning when analyzing our results this quarter. One is the previously announced resolution of the multistate examination related to the handling of unclaimed property, and the use of the Social Security Administration's Death Master File. In the quarter, the total earnings impact was negative $66 million. We also took a $20 million restructuring charge this quarter relating to the upcoming merger of 6 of our life insurance legal entities into American General Life Insurance Company at the end of this year, as well as for a project we call One Life [ph] that provides for the consolidation of all of our life insurance platforms, operations and systems. We expect these actions to result in an improved service delivery model, and importantly, to generate significant annual expense savings beginning next year as we continue to focus on disciplined expense management. And finally, AIG Life and Retirement results were negatively impacted by a $110 million interest credited adjustment for a runoff block of guaranteed investment contracts. That adjustment represented the cumulative impact of increased interest credited dating back to the early 2000s, and therefore, the annual earnings impact of this item was not a material amount over the individual years. Excluding these 3 items, our pretax operating income was very strong, at slightly more than $1 billion in the quarter. Also, as David mentioned, AIG Life and Retirement has been a significant contributor to capital management, providing $2.9 billion in dividends and distributions to date. Our risk-based capital ratio, which we currently estimate to be in excess of 475%, remains well above our capital maintenance agreement threshold of 435%. And stat earnings remain strong, providing capacity for dividends going forward. In this quarter, sales results once again reflect the value of our diversified business model. Variable annuity and retail mutual fund sales were strong in the quarter, while fixed annuity sales declined as expected, given the unprecedented low interest rate environment. Total life insurance sales were up from the year ago quarter, although down modestly sequentially. Variable annuity sales exceeded $1 billion in every quarter this year. And while still a relatively small percentage of total industry sales, our 39% variable annuity sales growth year-to-date reflects our reestablished position with core distributors -- in core distribution channels and a competitive and appropriately designed and priced product for the current market conditions. Due to the steep decline in fixed annuity sales, net flows, which we define as sales less all surrenders, death benefits, withdrawal and all other benefit payments, were negative for the first time in 7 quarters. We continue to make progress on implementing our new distribution organization structure, which is designed to fully leverage our distribution relationships and increase sales of multiple products across all of our distribution channels. Moving on to Slides 13 and 14. We have provided additional information on our asset allocation and on our portfolio yields. In the quarter, our annualized base investment yield declined sequentially as a result of lower accretion income on structured securities, lower income on certain equity method investments and lower yields on new purchases in the quarter, due to lower interest rates, credit spread tightening and higher credit quality purchases. It's important to distinguish between total and base investment yields. And while base investment yield excludes income from alternative investments, call and tender activity and other enhancements, total investment yield is based on the total reported net investment income. And total investment yield did remain strong for the quarter at 5.86% and 6.03% year-to-date, an increase over both the prior year and the prior quarter. We do not expect any material changes in our asset allocation in the near future, and we will continue to actively manage interest and crediting rates. At the end of the quarter, approximately 61% of our account values were at minimum crediting rates. And consistent with declining market interest rates, we saw a further sequential design -- further sequential decline, excuse me, in our cost of funds this quarter for all of our retirement services businesses. Turning on to Slide 15. Here, we've updated our disclosure regarding the impact of a sustained lower interest rate environment. Assuming the current interest rate and credit spread environment remain static, we do expect pressure on operating earnings going forward. To date, redeployment of our cash, including opportunistic investments in structured securities and disciplined management of interest crediting rates, have mitigated the impact. At this time, we do not foresee a significant DAC or stat capital impact as a result of the sustained low interest rate environment. Our projections are for a steady decline in base portfolio yields and in new money yields, which we are currently assuming to be between 3.75% and 4.25%, consistent with our actual base new money yield in the third quarter. We will continue to monitor reinvestment activity, credit spreads and overall new money yields closely and assess future impacts as appropriate. In the quarter, we continued to execute on a program to utilize capital tax loss carryforwards. Year-to-date, we sold approximately $5.9 billion of investments at a book yield of 6.3% and that generated realized gains of about $1.2 billion. With these sales and subsequent reinvestment of proceeds at lower yields, that impacted our operating income, but we captured real after-tax economic benefits. The program is now substantially compete, and the impact on 2013 pretax operating income compared to this year will be a reduction of about $33 million from the program. In closing, we've positioned ourselves very well for the challenges of the low interest rate environment and our earnings capacity remains solid. Continued focus on product innovation, disciplined product pricing, expense management and fully leveraging our resources and distribution relationships, remain core to our strategy. So at this time, I want to turn it back over to David.

Thank you, Jay. And turning to Slide 16 for just a quick summary of United Guaranty. UGC continues to see growth in its new business volumes and it's continuing its leadership position in the mortgage insurance industry. And UGC is generating returns in excess of 20% on new business and we continue to experience declines in delinquency. At this point, I'd like to turn it back to Liz to tee up the Q&A. Thanks.

Operator, could we begin the Q&A now?

[Operator Instructions] We'll take our first question from Josh Stirling, Sanford Bernstein.

So listen, I think we all appreciate the substantial clarity, I'd say, on your near-term plans for capital management and how debt is going to fit in. I'm wondering, David, if you could give us a sense of, in the near term, what sort of coverage ratios you might be trying to target? How much more beyond the hybrids you've identified you would sort of see as tools to get there? And as we -- when you think about sort of the firm's total capital generation, obviously, I think people sort of have a handle on things like operating earnings and tax benefits monetizing, I'm wondering if you can give us some other color around what the impact of the things you've been doing, restructuring your life insurance companies, consolidating legal entities in Europe and then percent of, prospectively moving assets from the liabilities from the DIB or liabilities from FP into the regulated entities that we should be thinking about as a future capital impact?

David, why don't you go ahead? That was a mouthful.

I'll start on the capital management and then kick it over to Jay for commentary on the Life legal entity restructuring, and then maybe Peter can comment further on the Property Casualty legal entity. And then we'll circle back to Brian for maybe a word or 2 on the -- some color on the Direct Investment book. With respect to the coverage ratio improvement, I think the agencies have been pretty clear that while our leverage is in a very strong position, they'd like to see a couple of turns improvement over that period of time, the 12 to 18 months. So again, there are lots of ways to accomplish that. One is with a reduction in the interest expense, and again, targeting financial leverage or the securities that count towards financial leverage. That includes senior debt. It includes some of the hybrids. So I think, again, that's one way. And obviously, the continued improvement in the operating earnings of our core insurance operations is clearly the other lever and that's consistent with our plans and expectations with respect to, again, the continued improvement. So those are the general direction that we're going. We haven't specifically identified any specific transactions. Again, that is under consideration. So Jay, why don't you maybe comment on the Life and Retirement legal entity structure?

Sure, David, thanks. Let me put it in context. In terms of directly answering your question, in terms of increased capital efficiency, we see that as modest, probably somewhere between $150 million and $250 million of less -- lower capital with the same amount of liabilities, principally due to the covariance of risk. But more importantly, the legal entity consolidation is going to diversify our risks to a greater extent. And that's going to allow us, both on an economic basis and also under all of our stress testing, to withstand far greater stresses within our largest life entity, American General Life Insurance Company. Also, there will be some modest, but not unimportant, expense savings from the consolidation. And as importantly, from the standpoint of our distribution partners and others that we do business with, it will make doing business with us easier, as we'll be more of a single life company with whom agents and financial advisors can be licensed to access a much broader range of our products. So there's several different benefits and capital efficiency is one of many.

Peter?

A similar story in our Pan-European legal entity consolidation. We're talking about going from 4 legal entities operating in 26 countries into one starting 1st of December and that has benefits, both in terms of stat capital as well as operating cost efficiencies and better controls. The stat capital benefit right off the bat is about $300 million, but will be greater over time, but dependent on a number of factors, including how we organize our internal reinsurance pooling arrangements and the extent to which we use Europe versus North America as a hub for internal reinsurance, as well as how we optimize our holdings in the investment portfolio, because right now our European investment portfolio has very, very high-quality assets relative to our North American one. So there's room to optimize where we hold what types of securities to increase our capital efficiencies. So I think that taken as a whole, very substantial efficiency benefits both in terms of OpEx, as well as capital usage.

Brian, you want to close on the DIB?

Sure, Bob. Yes, as we've said in the past, we're -- we manage the DIB to, again, maximize its profitability over time while maintaining adequate capital and liquidity to cover any risks. The DIB over time has generated several billion dollars of income and gains. As David said, we expect it to reach its half life by that 2017. And again, the significant NAV in that business will free up over time. As you heard in prior quarters, we have been able to move excess capital out of the DIB, so we do have some flexibility there and we'll continue to operate it prudently. I don't know if there's a whole lot more to say on that at the moment.

Okay. Josh, just keep in mind that buying back debt would be just one of the options and capital management, David went through a whole list, and I think it's important that we focus on the operating earnings within the insurance companies and where we can do more of that, as well as deal with the actual interest expense. Next question?

Yes, that's very comprehensive. The one other big question related to capital, and then I'll pass the floor, is when you think about the Fed and the prospects for you guys to be sort of held to a CCAR type of process, the rules on either for the savings and loan holding companies, which you are, but where the rules are not yet established and then the fact that you're not yet a SIFI, both of those lead one to say that it doesn't feel as though you'd be required at the moment to participate in 2013 CCAR. Is that the right read or should we be expecting you would be in CCAR this year?

That's true, what you said. However, we are working with the Fed because we're in a hurry. We want to make sure that we know where we stand and so we're doing our best to understand the requirements, run those requirements through our risk models and see what we can do to determine just where we stand. So while there's no requirements, we're still working informally to see if we can do something to assess where we are as we go forward. So we just have to wait and see how the Federal Reserve works with us and how they want to proceed with various tests. We are open to it and welcoming it, because we think it makes sense to get clarity now rather than later.

And you don't think that the Fed would feel -- I mean, they would prefer to have clarity sooner rather than later as well?

I can't speak for the Fed. I can only tell you they're in. They're just starting to work with us. We're a big company and they're working through it. And as soon as they're able to make some assessment, I'm sure they'll work with us. I think everybody wants to get clarity. It's just a matter of time and working through a schedule that makes sense for them.

Now we'll take our next question from Josh Shanker from Deutsche Bank.

My question mostly relates, Peter, on the P&C side. I want to touch upon some detail about how much these one-off, large non-catastrophe losses impacted the various international and U.S. segments in the former Chartis? And two, talk about the persistency of prior year unfavorable reserve development. I realize a lot of your competitors also take rather semi-frequent charges for environmental losses. But given the size of your book, sometimes you should be adjusting up, sometimes you should be adjusting down. It seems like reserves are always being adjusted upward.

Let me -- before I turn it over to Peter, I think maybe Peter should respond to the first part of your question in a second. But I'd like, maybe Peter and Charlie, because you raised this very important point on environmental, and the way we're doing it and quite frankly we at AIG are working really hard to do a bottom's up analysis of all of our reserves, starting with the claims themselves. And so maybe, Peter, if you could pick this up and then pass it to Charlie and go through that. It's a little bit more detail than some of you may want, but I think you need to understand the thoroughness of what we're doing here and it's not just the pattern of actuarial assumptions. Peter?

Yes, well, let's start with the first part of the question, which is just that the steady improvement in accident year loss ratio has been somewhat slowed in this quarter by some unusual non-CAT, but large losses in the international property side and so we view those as an outlier. A number of them were named losses where we were just below the threshold that we used to designate it as a CAT, so definitely some, what we believe, to be one-off anomalies. But on the reserves, let me just talk about that...

Can you just put a percent on that, by the way? What percent do you think that impacted your loss ratio there?

Josh, it's John Doyle. It's about a little less than 2 points, about 1.8 points. So as Peter mentioned, we had 3 storms in Asia that were named events that -- but didn't meet our $20 million threshold to call it a CAT. Then we had a number of larger losses, about 2x kind of our average number of [indiscernible] in the quarter, big fire losses in the quarter that impacted it as well. And we had a failed satellite launch, as kind of another example. It's double-digit severe losses, again, about 2x kind of an average number. Obviously, there's going to be some lumpiness to 2 large losses in property, but it was a bit of an unusual quarter for us. And as I said, it had close to a 2-point impact on the current accident year loss ratio.

Two points on international commercial?

Yes, 1.8 points overall globally and most of it was on the international side. Jeff?

Yes, Jeff Hayman jumping in for Consumer. There's about $24 million in losses for Consumer, exactly as John described and Peter named the storms that were under our threshold, Typhoon Jelawat, some torrential rains in the Kinki region of Japan, and that's about 0.7 on our accident year loss ratio.

Can we go to Charlie now on the environmental?

So a few things on environmental. We've now concluded a very detailed study of the environmental impairment liability portfolio, and that's the portfolio made up of 5 distinct, fairly heterogeneous books that we describe in the 10-Q. And we've reviewed claim-by-claim, 2,150-odd files on the most complex claims, so that we focused on those with the highest policy limits. Many of these were written in the period prior to 2004. And we did not do this prompted by any actuarial indications, and in fact, the actuarial third-party reviews we had would have suggested the environmental portfolio was redundant. We did it more because we think that the characteristics of those claims are such that you really need other experts, engineering firms, toxicologists and litigation experts. And in this quarter, roughly $60 million of the $77 million that you see that we've posted as prior year development, came from a very detailed analysis that we did on mass tort claims and we felt that the characteristics of those claims were sufficiently different to warrant a more conservative stance on their severity in particular. And then, the other part I would say about persistency of prior year development, in the quarter, we had $114 million of large commercial losses. Of that, $70 million roughly related to loss and legal on a few construction defect claims. And we also had large losses in the health care division, this quarter and also last quarter, and that's shown as primary casualty in the all other component of the 10-Q. And I would say the health care ones are unusual. The health care division has actually posted very favorable developments over many years and the industry, as a whole, has witnessed that. But these 2 claims are very peculiar and they really are, have unique characteristics that we don't think are indicative of a trend.

Thanks, Charlie. And so I think, as you can see, even on environmental, it's been a year of going through this and so we're probably in the $500 million to $600 million addition to reserve. But we think it's completely behind us now and we're continuing to do these reviews, to continually build confidence in this reserve and make sure we have it right. So a lot of work has gone onto it. But I don't think the trend is something that's just a trend, but it's actually our cleaning up to make sure everything is right.

So Bob, the initiative is complete on toxic waste?

Well, we've gone through the whole reserve. This is one area that was very complex and one that we really had to put a lot more time on.

Moving on, we'll take our next question from Michael Nannizzi from Goldman Sachs.

Sure. Just to follow up on Josh's question on international commercial. So if it's 2 points of unusual large losses total, that would be, if my math is right, that would be like 10 points on international commercial. If it's 2 points on international, it'd be 5 points. Just trying to get an understanding of what -- when you kind of back out these sort of lumpy losses, how did that segment perform. And if there's anything outside of those large losses that were abnormal, can you talk about what those trends were?

I'm going to turn it to Peter and John, then, if you would.

Yes, Michael, it was close 2 points globally. It was 1.8 points globally. James, do you have the number for the international side? I'm not sure what the number is exactly on the international side.

So there was overall, a 12-point increase in the accident year loss ratio for international consumer -- commercial, sorry. And over half of that impact -- over half of the driver of that change was due to these severe losses flowing through the P&L in a -- to a larger extent than in normal quarters.

Okay. And the rest of it is just. . .

The rest of it is there's a variety of events that flowed through the P&L this quarter, which, I think, the -- if you look at the historic trend in the Q1 and Q2, that would be more indicative of our go-forward run rate within international commercial. So we just had a number of unusual things popping up this quarter, the largest driver of which was the severe losses within international.

Okay, great. And then maybe switching gears a little bit, at other operating expenses, it looks like you had a reversal of an accrual. If you include that, it looks like expenses at -- the other expenses were up a bit compared to where you were last quarter and over the past couple of quarters. Just trying to get an understanding, what is that and how should we think about that line relative to the $1 billion in expense reduction that you've been talking about?

Sure. Let me turn this over to David, who will take you through some of that, but we are feeling very confident in the direction of 2015, our aspirational goals, but you're going to see some ups and down in any given quarter. But David?

Yes, thanks, Bob. With respect to the guidepost that we had set out in terms of what the -- from quarter-to-quarter, the general range of $200 million to $250 million of operating expenses at the holding company that are not allocated out to the opcos, I think that's still a reasonable set of guideposts. This particular quarter, there were a number of, what I'll call, infrastructure and transformation activities going on across the parent, the corporate center, including things like our data center consolidation, et cetera. And it's investments like that, that help us deal effectively with catastrophes like the one we've just incurred. Over time, I would expect, Michael, for the corporate expense not to -- a glide path down towards the low end of that range and then certainly below that as the $1 billion expense target is realized over the period up to 2015.

Great. And then last one, just U.S. Commercial, clearly the underlying loss ratio has improved. I mean, if my model is right, I mean this is the lowest underlying loss ratio we've seen at U.S. Commercial since the back half of 2010. Can you kind of talk about how that is occurring, what actions you're taking and where do you expect that to get to in order to kind of reach your overall goals by 2015...

John, do you want to pick that up?

Sure. It's -- we've talked about the various levers, right, being mix of business tools, claim initiatives and expenses, as David just covered and Peter talked about in his prepared remarks. Expenses are on the come, if you will, but we've made a lot of progress in terms of the mix of business. You've seen pretty aggressive actions we've taken -- in fact, in the presentation, there's a breakdown of U.S. casualty and the aggressive actions we've taken there. We've obviously have been pressing on the rate side, and I would say that the rate environment continues to improve. That flows through more quickly, obviously, in our property results. And we've not been aggressive about changing current accident year loss picks. We're in the process of bridging exercise with Charlie as we look about our loss picks as we go forward, but -- and we expect continued current accident year improvement in various lines and really on all lines in the United States. On the underwriting tools side, we've, with our science team and our underwriting resources, we've made some really exciting progress on some of the risk selection models that we're excited about as well. And then lastly, I would say we're about halfway through a multi-year program on the claims side. And we've seen some improvement in the United States. And we've got actually some exciting results we've seen in Europe as Eric Martinez and his team are executing on our global claims initiatives. So we expect to see continued current accident year improvement in the U.S. Commercial loss ratios.

Moving on, we'll take our next question from Brian Meredith, UBS.

I was wondering, could you give us some perspective just -- I know it's quite early, on kind of potential exposure here to Hurricane Sandy? How should we think about it with respect to AIG's exposure? And then maybe as part of that, is it possible to get what your retention is on your domestic property CAT reinsurance program?

I'll let John answer the second part. And then, as far as color, Peter, you might want to add to it, not that there's much there. We're just going through this on a claim-by-claim basis. We're working with clients right now. And it's hard for me to give you a scope of it because some of the flooding and water damage that was done, especially in the New York area, in Manhattan. But I can only say that we don't see this as being any kind of a huge issue for us financially, other than just dealing with the issues at hand. So I don't know that I could give you any more color at this point, but let me turn it over to Peter.

Yes, I think that, usually, we get about 80% of our claims notices within about 90 days, so we're obviously at the very beginning of that whole process. And I think to comment beyond the fact that the property damage comes in quicker than the business interruption and that we're looking at a broad range of commercial properties across sectors in the affected areas, it's a big broad area and we're getting new information every day, so it's just way too early to comment intelligently about it, so we'd rather not. But on the retention policies, and more generally, our approach to deductibles and flood supplements, we've made changes over the last 3 years, which, we think, helped us with Isaac last year and will also help us here relative to the exposures we've had previously. And we've made some pretty major changes in our global reinsurance strategy in terms of risk appetite internationally. And so John, maybe you want to just elaborate on the retention?

Well, I don't think we've disclosed where our CAT reinsurance attach is. But non-CAT, we've -- we're in the middle, there's been some press about our non-CAT kind of per-risk, if you will, reinsurance program. And as Peter said, really a byproduct of our new global view. Previously, we had structured reinsurance around legal entities around the world in a regional view, and so we're taking a more consistent risk appetite around the world. So it's an opportunity for us, primarily on the property side, to, again, take more risk back net. Simultaneously, though, we have actually just commenced a deal on the excess casualty side in the U.S. to actually seed out some of our U.S. excess casualty business. So we've got a kind of a global restructuring, if you will, of our overall reinsurance program. Pretty significant changes, but again, to drive a more consistent risk appetite and shift the mix of business to a better balance, really around the world.

Great. And Bob, just quickly, you mentioned M&A as a possible use of capital here. I'm just curious kind of what areas would be intriguing to you?

Anything that enhances what we already have anywhere in the world. So it's hard to say, I could give you some names, but I'm not sure that, that would be a good idea.

Moving on, we'll take our next question from Lee Cooperman from Omega Advisors.

All of us own the stock for a bunch or reasons, one of which is we see a big book value and we expect that the company to achieve a reasonable return on that book over time. So my first question is, what do you guys think is a reasonable return given the regulatory environment that now exists post-2008 and what's a reasonable time period for that return to be achieved? And secondly, if you had to hazard a guess about 2013, is 2013 going to be a year with cash dividends or additional significant buyback? If you had to guess.

One is, I think, we're absolutely on target from everything we feel right now, to the 2015, getting to an ROE of north of 10%. So we feel that, that's achievable. We have a lot of wood to chop to get there and a lot of things we're building today and that's the investments of infrastructure and so on that we're building within AIG. So we feel that, that's still attainable. There are still questions about what would be the capital required over time. There's a bias in the regulatory environment to want to put more capital restrictions on financial services companies. But I think that, sooner or later, people are going to realize that you could just go too far with all of that. But from what we know and see, we still think it's realistic to shoot for north of 10%, or north. And as far as 2013, we are working very closely with the Federal Reserve. We feel we're in excellent shape. We run what we believe to be our understanding of a CCAR, and so we're looking at binding constraints. And if I were to think about 2013, we're focused on the coverage ratio, which is important, because our credit ratings are important, because we want to show not only stable, but positive or even an uptick is important because we make guarantees and our companies look forward to having strong guarantees behind AIG. They stuck with us during the crisis, but we want to give them better assurances going forward. And I would think that a dividend on the stock probably would be something that if in fact we're able to do that, I would see that as something we'd like to do in 2013 if our capital position is strong enough that we can do that. And we won't know that until the Fed has more time to review where we are.

We'll take our next question from Adam Klauber from William Blair.

You've done a lot of work in the last 2 years on the U.S. commercial P&C segment. As we go into next year, I guess, 2 questions. One, do you think that business still needs more rate. And two, do you think you'll start growing that business?

Let me turn it over to Peter and then you and John can work that through.

Adam, the second part of that question was, I'm sorry?

Do you think you'll start growing that business going forward?

Okay. So there are -- obviously, we have a broad range of products in the U.S. commercial space and so all are not created equal. We've highlighted how we've reduced our overall exposure to U.S. casualty. So broadly, I think that the greatest rate need, if you think of it on a major line basis, we think is in the casualty space. Our property book in the U.S., our specialty lines, including financial lines, are performing better on a risk-adjusted basis than our view of the U.S. casualty market. Obviously, the interest rate environment there creates more pressure on that segment of the book than other segments. So in terms of rate need, I would say it's there. That's where we're getting the largest rate improvement at the moment and it's where we have a lot of our science resources focused on improving our risk selection. I mentioned earlier that we do expect current accident year improvement in those lines of business. So in terms of growth, we have modest mid-single-digit growth for commercial overall planned during the -- now through 2015 performance period. We'll see more aggressive growth outside of the U.S., but we do expect some low single-digit growth in our U.S. portfolio over the next couple of years.

We'll take our final question in from Mark Finkelstein from Evercore Partners.

I wanted to go back to the coverage ratio calculations. I think you mentioned that the intent is to get to 2 turns of improvement over the next 12 to 18 months. I guess, that's a little bit faster than I would have expected and maybe a little bit more dramatic than I would have expected. I think that kind of, if my numbers are right, would probably put you at about 5x coverage, correct me if I'm wrong on that. But I guess, can you just clarify that, that's the correct interpretation of what the aim is? And then, I guess, when I look at the maturities, I mean, obviously you have some hybrids that are coming or callable. Maybe just elaborate a little bit more on the strategy of getting there, if that is the, in fact, objective?

David, do you want to pick that up?

Yes, thanks, Bob, I'll start and then I'll ask Brian to elaborate a little bit further. Again, a 1 to 2, closer to 2, but somewhere in the 1 to 2x coverage improvement is what we're aiming at. And again, the exact calculation that each of the agencies performs is not precisely the same. So there's not a simple formula to plug into. But it's -- I think you're generally -- you've got it about right. We think it's around -- the existing is around 4x or a little less than 4, but again, it just depends on the particular agency. And again, I think that it's a combination of the 2 levers, one is on interest expense, certainly. And again, the monetization of our non-core assets and further capital generation at the -- both at the insurance companies, which gives us the deployable capital up to the holding company. And Brian can talk a little bit more about that in terms of what our plans and expectations are along those lines. But I think those are the 2 levers, it really gets back to improving the operating earnings. And you've heard from Peter and Jay on their core businesses and the trajectory that they're on, and then a focus on selected and targeted debt reduction and interest reduction. Brian, do you want to jump in here?

Thanks, David. I think you've nailed all the salient points, it will be a combination of interest expense reduction, as well as improvement in operating earnings. And we -- as you noted, we do have some bonds that are callable that, I think, make a lot of sense for us to go after. And when you think about our flexibility on an ongoing basis to go after this, as we said in the past, we expect $4 billion to $5 billion of subsidiary dividends coming up to parent every year. This year, we're already ahead of $5 billion as of today and expect a bit more in the fourth quarter. You take out the parent expenses and interest and you're sort of in a situation where we're going to have around $2 billion a year, give or take a little, for things like liability management and dealing with, opportunistically reducing our outstanding expensive debt. And that excludes the additional excess capital that will be generated from sales of non-core assets, as well as capital free-up over time from the DIB.

Okay, maybe just one quick question on the DIB. I guess I'd been a little surprised at how strong the DIB earnings have been over the last few quarters. I think, Brian, you talked about this concept of pulling to intrinsic quite a bit on the last quarter call. I'm just curious if you have a view on kind of a comparison of where the assets are currently marked versus your own view of what that intrinsic value is?

What we've said in the past and where we still feel pretty strongly about is that there is probably close to $5 billion -- $4 billion to $5 billion, I think we said in the past, of pull to intrinsic over the life of the portfolio. And what you're seeing now in a quarter like this is, again, the combination of net investment income as well as the CVA or the impact of spread tightening on the assets slightly offset by our own spread tightening. Because most of the assets and liabilities in the DIB or a substantial portion of which have been swapped, really what you're going to see coming through the P&L, when we think about pull to intrinsic, is effectively CVA. And it will not be consistent over time. It's going to be -- it's all market-driven, but we still feel pretty strongly about what we've indicated in the past is what the potential of this portfolio is.

Okay, I thank you all. For, I think, any other issues, please let us know, talk to Liz, but I thank you all. And good luck for those of you in the New York area. We've got a lot of work to just get back to normal. So I thank you all very much this morning.

Thank you. That will conclude today's conference. We thank you for your participation.