American International Group Inc (AIG) 2018 Q3 法說會逐字稿

Good day, and welcome to the AIG's Third Quarter 2018 Financial Results Conference Call.

Today's conference is being recorded.

At this time, I would like to turn the conference over to Ms. Liz Werner, Head of Investor Relationships -- Relations.

Please go ahead, ma'am.

Thank you, April.

Before we get started this morning, I'd like to remind you that today's presentation may contain 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 first, second and third quarter 2018 Form 10-Q and our 2017 Form 10-K under Management's Discussion and Analysis of Financial Condition and Results of Operations and under Risk Factors.

AIG is not under any obligation and expressly disclaims any obligation to 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 the slides for today's presentation and our financial supplement, both of which are available on our website.

As a reminder, for this morning's call, our Q&A session will have one question with one follow-up.

(Operator Instructions) On this morning's call, you'll hear from our senior management team, including CEO, Brian Duperreault; CFO, Sid Sankaran; CEO of GI, Peter Zaffino; GI Chief Actuary, Mark Lyons; and CEO of L&R, Kevin Hogan.

At this time, I'd like to turn the call over to Brian.

Thank you, Liz, and good morning, everyone.

Our third quarter results reflected volatility due to 14 global catastrophes, particularly the Japan typhoons.

We continue to execute on our reinsurance strategy, one of our key initiatives, which I will cover in more detail, and which Peter will also review in his remarks.

Other actions we're taking to improve our underwriting capabilities and profitability are taking hold, and we started to see some of the resulting benefits in our third quarter results.

We continue to expect to deliver an underwriting profit, including AAL for General Insurance as we exit 2018.

Over the course of the last year, Peter and his team have made significant progress in executing on our reinsurance strategy, and this work will continue as we approach the January 1 renewal season.

So far this year, we've lowered our North American CAT cover attachment point from a per occurrence of $1.5 billion to an aggregate of $750 million and added an additional international cover.

As you will hear from Peter in his remarks, AIG's national market share in Japan is 6%, and in the area most impacted, it's 10% of the average.

While our Japan reinsurance program was renewed in January 2018, maintaining its historical structure, which included 2 separate towers for the Commercial and Personal Insurance business, we have been working diligently throughout this year to get a single structure in place for 2019 to reduce our net exposure on both a frequency and severity basis.

We are pleased with the contributions and balance that Validus brings to our business mix.

The disciplined underwriting and risk approach that Validus takes was most evident this quarter in the estimated net CAT loss of approximately $200 million, which was in line with peers.

Validus was neutral to our accident year results this quarter and remains on track to contribute approximately one point to a combined ratio improvement as we exit 2018.

Our recent announcement of the pending acquisition of Glatfelter will provide further balance to General Insurance by improving our position in the programs market with the addition of one of the most respected firms in this space to our portfolio of businesses.

The closing of this transaction is expected to occur next week.

Turning to reserves.

We welcome Mark Lyons to the team this summer, and he has hit the ground running in reviewing our actual processes and procedures.

Net reserve additions of $170 million in the third quarter reflect the work Mark and his team performed, relating to approximately 75% of our book.

Year-to-date, net reserves development was flat.

Peter will provide additional detail of General Insurance in his prepared remarks, and Mark is joining the call and will give you more color on work he has done on General Insurance reserves.

L&R delivered another solid quarter, notwithstanding challenging year-over-year comparisons that reflect the impact of annual assumption reviews.

Underlying ROE continued in double digits, and in particular, our investments in businesses over the last few years are beginning to bear fruit with strong growth in Individual Retirement and Life, in particular.

Sid will discuss the results of the third quarter actuarial review, and Kevin will elaborate on the performance of this well-positioned business, which serves some of the most -- the world's most important needs, the need for sources of savings, lifetime income and protection.

Our steps to reduce exposures on our Legacy book and to allocate capital more efficiently underscore our capital management discipline and focus on long-term shareholder value.

The sale of 19.9% of Fortitude Re, formerly known as DSA Re, to The Carlyle Group is imminent and will free up capital as well as provide a platform for potential growth.

Lastly, we repurchased shares and warrants totaling $1 billion through the third quarter of this year, including $350 million in the third quarter.

We have $1.3 billion remaining in our authorization.

Share repurchase remains a capital management tool that we will continue to deploy in addition to other uses of capital that will contribute to our goal of making AIG better than it has ever been.

In conclusion, we are making progress towards positioning AIG for the long term.

We continue to work deliberately and thoughtfully and with a sense of urgency to improve our core underwriting capabilities, reduce volatility, deliver an underwriting profit, including AAL for General Insurance, as we exit 2018 and position each of our businesses for long-term success.

Now I'll turn it over to Sid, who will provide more detail on our third quarter financial results.

Thank you, Brian, and good morning, everyone.

This morning, I'll comment on our net third quarter financial results, our capital and liquidity position and provide an update on Fortitude Re and Validus.

Turning to Slide 4. We reported an adjusted after-tax loss of $0.34 per share.

Book value per share, excluding AOCI, was $66.83 and adjusted book value per share, which excludes AOCI and DTA, was $55.58 at quarter-end.

Note that approximately $500 million of the reduction in book value is associated with an increase in the deferred gain on our adverse development cover, which will amortize back into earnings and book value over time.

Overall net investment income from our insurance operations, including the Legacy Insurance Portfolios, was $3.4 billion in the quarter or $9.9 billion year-to-date, which is in line with the $13 billion full year guidance that we provided at the beginning of the year.

We continue to make progress on our restructuring initiatives in the third quarter, most notably in General Insurance, where general operating expenses declined 5% from the second quarter on an ex Validus comparable basis.

We continue to expect to achieve the full $450 million of annual run rate expense savings I referenced last quarter as we enter 2019.

General Insurance, Life and Retirement and Legacy results were all impacted by various noteworthy items listed on Page 5. We reported $1.6 billion of pretax catastrophe losses driven by multiple events, most significantly typhoons Jebi and Trami in Japan.

We remain comfortable with our initial pretax loss estimates for Hurricane Michael, which are between $300 million and $500 million, and this estimate will be included in our fourth quarter 2018 operating results.

Given these events place us close to attaching in a North American catastrophe aggregate program and that our third quarter severe losses attached on our aggregate XOL severe loss cover, this reduces the risk of volatility into the fourth quarter.

Peter will speak further about our catastrophe losses and reinsurance program in his remarks.

Note that Legacy incurred approximately $57 million of CAT losses associated with runoff Property policies in Japan.

As a result, Legacy's ROE was below our long-term 3% to 5% expectation.

Life and Retirement had a solid quarter with an adjusted ROE of 11.2%, inclusive of a $98 million charge associated with the annual assumptions review.

This was largely driven by adjustments in Individual Retirement due to refinements in Variable Annuity withdrawal assumptions, additional reserves related to older universal Life policies and refinements to interest crediting to the Life Insurance segment.

Note these adjustments did not impact our assessment of profitability on new business written going forward or profitability of the core portfolio.

We completed our third quarter detailed valuation reviews, and year-to-date net prior year reserve development was roughly flat while the quarter was adverse by $170 million in adjusted pretax operating income.

Mark Lyons will comment further with additional insight into this quarter's reserve review and actions taken.

Our third quarter adjusted effective tax rate was roughly 28%, which reflects the beneficial impact of tax discrete items this quarter on the pretax loss.

We expect our full year 2018 adjusted effective tax rate to increase slightly to approximately 25%.

This increase is primarily related to the impact of the third quarter catastrophic events on our full year net income projections, combined with our current assumptions around the impact of U.S. tax reform on our operations.

Our balance sheet and free cash flow remain strong.

As shown on Slide 6, current liquidity at quarter-end was $4.5 billion.

We continue to view the target level of liquidity at the holding company to be between $3 billion and $4 billion.

Cash proceeds in the quarter included $1.6 billion of dividends from our insurance subsidiaries and tax-sharing payments of approximately $200 million.

Our base case for annual dividends and tax payments from our insurance subs remains approximately $6 billion, although we see potential upside from nonrecurring flows.

Our capital ratios for our Life and Retirement and Legacy companies are above target levels, and our GI companies are comfortably at target levels even after the third quarter catastrophes.

Our strong balance sheet should continue to provide us ongoing financial flexibility.

During the quarter, we deployed approximately $350 million towards the purchase of common shares at an average price of $53.05, leaving our remaining authorization at approximately $1.3 billion.

Slide 7 depicts our capital structure and ratings.

We redeemed Validus preferred shares totaling approximately $400 million in October.

While our financial leverage ratio increased this quarter to 29.2%, reflecting the inclusion of Validus' debt and the net loss for the period, we're redeeming Validus junior subordinated debt in the fourth quarter through excess capital of the legal entity, which will reduce the financial leverage ratio by 0.6 points.

We expect to complete the sale of 19.9% of Fortitude Re to The Carlyle Group in the fourth quarter, at which time we'll begin to report this minority share in noncontrolling interest on the income statement, which will reduce Legacy's contribution to AIG's earnings per share.

Finally, with respect to Validus, we completed the purchase accounting during the quarter.

Of the $3 billion of cash consideration paid in excess of tangible net assets, $2 billion was attributed to goodwill and related intangibles; $444 million was attributed to the value of the distribution network acquired, or VODA; $298 million was attributed to the value of business acquired, or VOBA, with $268 million of other investment line -- lived intangibles.

The VOBA which was established approximated the amount of Validus' deferred policy acquisition costs that were written off in purchase accounting.

VOBA is being amortized consistent with the existing DAC runoff pattern of approximately 2 years and is included in adjusted pretax operating income and the underwriting ratios of the General Insurance segment.

VODA will be amortized over 15 years or roughly $30 million per year and will be reported in our other operation segment as part of adjusted pretax operating income.

To sum up, we continue to make progress towards delivering long-term profitability, reducing volatility and maintaining a strong balance sheet and free cash flow profile.

Now I'd like to turn the call over to Peter.

Thank you, Sid, and good morning.

Today, I will provide an update on General Insurance's progress against our key 2018 priorities to improve underwriting performance and position the business for long-term success, the recently announced Glatfelter transaction, third quarter financial results and our observations of current market conditions.

During the quarter, we continue to execute on our initiatives to strengthen General Insurance's core performance and achieve underwriting profitability.

I am pleased with the progress we're making to implement fundamental changes to our underwriting strategy and guidelines.

Under Tom Bolt, we have built a CUO structure to support our underwriters across the globe.

We recently welcomed Mike Price as Deputy Chief Underwriting Officer and have expanded the role of Lixin Zeng, CEO of AlphaCat, who will work closely with Tom to bring Validus' best practice in model development to AIG.

As we have shared in the past, our underwriting strategy has prioritized the reduction of gross and net limits in Property and Casualty.

We've reduced Property's gross limits from $2.5 billion to $750 million and its net limits from $611 million to $143 million.

In a similar way, we've reduced Casualty's gross limits from $250 million to $100 million.

With respect to underwriting governance, we've reviewed, validate and reissued 100% of global underwriting authorities to align with our revised risk appetite and instituted a new underwriter scorecard that measures performance against profitability and other key metrics.

As we reposition our portfolio, we are using reinsurance to support sustained profitable growth while prudently managing gross and net exposures and protecting AIG's balance sheet.

As a clear example, we recently expanded our existing $75 million excess of $25 million international Casualty excess of loss treaty into a global program that now includes exposure across U.S. Primary and Excess Casualty lines, which aligns the net retention to our overall risk appetite.

We are currently in the marketplace to place the U.S. Casualty quota share and preparing to enhance some of our existing covers during the January 1 renewal season, which we expect will yield additional benefits in 2019.

Organizationally, we continue to hire some of the industry's best talent, particularly in leadership positions across the world, with the primary focus on strengthening our underwriting capabilities.

Since last quarter's call, David McElroy has joined us as CEO of Lexington, and we've named Peter Bilsby of Talbot as Head of our Global Specialty business.

Both David and Peter have already demonstrated the value they bring in the organization as we focus on improving our underwriting portfolio.

I'm pleased to share that our new business leaders and the changes they are implementing have produced very positive feedback from our brokers, clients and reinsurers who recognize AIG as a better and more agile partner.

We continue to work diligently to reduce general operating expenses and free up capacity to invest in underwriting, actuarial and claims.

General Insurance's third quarter general operating expenses declined 5% sequentially, excluding Validus, and we remain on track to achieve our plan for 2018.

Finally, rounding out our efforts to strategically reposition General Insurance, we look forward to closing on the pending acquisition of Glatfelter Insurance Group next week.

Glatfelter has an excellent reputation as a highly selected program manager with world-class underwriting capabilities and in-house technology, a track record of underwriting profitability as well as a talented leadership team.

The addition of Glatfelter to AIG will accelerate the repositioning of our existing U.S. Programs business, where we have or are currently in the process of non-renewing over 50% of our current programs, in line with our primary objective of improving General Insurance's core underwriting performance.

Turning to General Insurance's third quarter results.

Slide 9 details our overall performance and profitability metrics.

Third quarter net premiums written declined 2%, excluding both FX and Validus, compared to the prior year quarter, largely driven by the execution of revised underwriting strategies in North America Commercial, partially offset by growth in North America Personal Insurance.

Validus contributed $440 million in net premiums written to our North America and International top line results.

With respect to profitability, the third quarter combined ratio included $1.6 billion in net catastrophe losses.

The breakdown of these losses is provided on Slide 10.

Validus' net catastrophe losses were approximately $200 million, over half of which were related to Japan, a result that was within our modeled expectation and in line with peers.

Our core insurance businesses, excluding Validus and Legacy, experienced net losses attributable to third quarter events of $748 million in Japan, emanating from 5 events, and $439 million in North America, emanating from 8 events, with the majority of the remaining losses arising from prior quarter catastrophes.

To provide further context on our business in Japan, AIG's national market share is 6%.

And in the Kansai region, home to Fuji Fire and Marine, and the location principally impacted by third quarter catastrophe events, our Personal and Commercial Property market share is approximately 10% on average.

As a result, our losses reflect both the unique severity and frequency of one of the worst catastrophe seasons for Japan in 25 years and AIG's footprint as the largest foreign-based insurer in the country.

As Brian noted, AIG's Japan CAT reinsurance program was structured in 2 separate towers for our Commercial and Personal Insurance businesses.

In late 2017, we began working to improve the program and successfully reduced attachment points for both towers during the January 1, 2018, renewal.

For the upcoming 2019 renewal, we plan to consolidate the program into a single tower to improve its effectiveness and further reduce our net exposure on a frequency and severity basis.

In North America, Hurricane Florence was a sizable industry event, which represented $325 million of our third quarter catastrophe losses, excluding Validus.

North America Personal Insurance's accident year results include increased loss estimates for the California mudslides that occurred in the first quarter of 2018.

The impacted territory is recovering from both the 2017 wildfires and the 2018 mudslides.

AIG's global claims teams have done a terrific job working with our insurers, often before the storms have made landfall, and began inspecting and paying claims as soon as regional conditionals allow.

We have responded to clients representing over 84,000 Commercial and Personal insurance claims, the majority of which are on Japan, given the severe nature of the damages in that region.

In the U.S., our claims teams have completed physical inspections for 100% of the sites impacted by Hurricane Florence where access has been permitted.

Shifting to accident year profitability, the adjusted accident year combined ratio was 99.4% and largely in line with our expectations.

The adjusted accident year loss ratio of 63.6% reflects underlying portfolio management improvements of 240 basis points and a more moderate level of severe losses compared to the prior year quarter and the second quarter of 2018.

The third quarter expense ratio was in line with second quarter results and our shift towards lower loss ratio and higher commission business in the North America Personal Insurance.

The GOE ratio does not yet reflect the expense reduction actions I mentioned earlier, which will begin to be evident in the fourth quarter and as we enter 2019.

Moving to Validus.

While the third quarter was challenging in terms of catastrophe losses, the business performed largely in line with our expectations.

As a heavy writer of catastrophe-exposed business, it is important to assess Validus' results on a 4-quarter basis.

Since the closing of the acquisition, our teams have been working closely to drive value, and we remain confident in the strong strategic contribution Validus brings to AIG.

Slides 11 and 12 provide North America's and International's third quarter financial results.

North America Commercial showed improvement across most lines, driven by business mix, rate and risk selection.

North America Personal Insurance had a solid quarter with an adjusted accident year loss ratio of 53.2%, and its acquisition ratio reflects the shifts we're making to grow lower loss ratio business.

In International Commercial, lower severe losses were partially offset by attritional loss activity in the U.K. and European Specialty businesses.

We expect that the growing momentum of our underwriting strategy will address attritional losses, particularly as we look to the January 1 renewal season.

International Personal Insurance continues to demonstrate profitability with an adjusted accident year combined ratio of 97.2%.

Moving to current market conditions.

The third quarter rate changes have been relatively in line with our experience over the past few quarters.

North America Commercial's overall rates increased approximately 4%.

We experienced admitted property pricing improvement in the mid-single-digit range while E&S Property rates improved in the low double-digit range.

North America Casualty rates increased in the low to mid-single digits, on average, depending on the line of business, attachment point, loss history and other factors.

Looking at the reinsurance market and the upcoming January 2019 renewals, we expect Property CAT and retrocessional rates to be relatively flat on average for non-loss-affected accounts.

Rates on loss-affected accounts will be determined on a case-by-case basis, including in Japan where the majority of the renewals take place on April 1. In Casualty, we expect to continue to see terms and conditions tighten with rate change in line with loss cost trends.

In closing, we continue to expect that the implementation of our underwriting and reinsurance strategies to manage gross and net lines, actions to reduce expenses and strategic contributions from Validus and Glatfelter will enable us to improve our underwriting performance as we enter 2019.

Our deliberate and targeted actions are positioning General Insurance for long-term success as we transform into a high-quality underwriting organization that is committed to delivering on its commitments to all of our stakeholders, and I'm very proud of what we've accomplished for the first 9 months of the year.

With that, I'll turn the call over to Mark.

Thank you, Peter, and good morning all.

I'd like to make some comments this morning, firstly, about some general observations I had since joining AIG earlier this summer.

Secondly, I'd like to provide more focused comment on loss reserves for the quarter.

Thirdly, I'd like to provide some clarity around the reserving numbers themselves.

And lastly, I'll give my views about the underwriting portfolio changes that have been implemented so far to date.

Over the last 4 months, I've had the opportunity to meet with many AIG executives and P&L owners to understand their strategies, historical results, budgets, competitive positioning and portfolio composition.

I've always operated under the assumption that an understanding the underlying business strategy is critical towards evaluating reserves, profitability, portfolio mix and the like.

Actuaries, after all, are charged with analyzing mountains of data and associated information to properly accomplish their jobs.

It's best then to understand the market conditions that's spawned the data being analyzed.

I found here a talented group of professionals who are hard-working and care about the company.

They've employed a host of appropriate actuarial methods that are suitable for the task in hand.

This group has helped me review, analyze and conclude on 50% of the General Insurance's loss reserves during the third quarter and also helped me review, challenge and understand the analyses done earlier this year.

Overall then, I've got my fingerprints on approximately 75% of the reserves through third quarter.

As for the fourth quarter, when the remaining 25% of reserves will be reviewed, the areas of focus will be U.S. financial lines, workers' compensation buffer excess policies, International Casualty reserves other than the U.K. and Europe, which were reviewed this quarter, and Personal line exposures.

At this time, on a preliminary basis and before the detailed reviews are completed, I see no material red flags to these fourth quarter lines that give me undue concern.

Of course, the work still has to be done, but that's how I see it at this point.

Lastly, I've discussed the reviewed lines through the third quarter with our outside independent actuaries as well to gain their insights and their views.

Now moving on to the third quarter areas of focus.

The only area where I felt a material reserve strengthening is necessary was in the Excess Casualty portfolio.

More specifically, the strengthening was predominantly centered on construction defect exposures through practice project and wrap policies as well as through Excess Casualty division other than those that specifically target construction.

The last several years, AIG has experienced the increase frequency of CD claims as well as an associated increased severity.

Given the latent nature of these reporting patterns, it was necessary to understand the historical and recent exposure characteristics of AIG's book.

Beginning around 2009, AIG began to simultaneously reduce their construction and CD exposure as well as veer away from residential risks in favor of commercial risks.

The book also shrunk exposure by approximately 70% as measured by on-level premium between 2009 and 2017 as well as proportionately away from the heaviest CD states, such as California.

We project that most claims as yet unreported will emanate from commercial exposures rather than the residential exposures that make up most of our known reported claims experience.

We anticipate that commercial CD claim emergence will be fewer in number but be more complex in nature of a higher-average cost per claim, and that higher average will be exacerbated by commercial insurers, who, as a matter of course, purchased higher towers and coverage historically.

Additionally, since AIG competed with capacity, among other aspects, more limits per insured were exposed and was the case for residential policies.

Furthermore, an exhaustive internal analysis outside claims audit, in close collaboration with Anthony Vidovich, who runs General Insurance claims worldwide and his team, provided valuable insight into the claims history and patterns.

Taking all these factors into consideration resulted in a $1.26 billion U.S. Excess Casualty reserves strengthening charge this quarter.

This figure is gross of the adverse development cover, or ADC, with Berkshire Hathaway.

There were other reserve adjustment this quarter in other lines of business that resulted in some partially beneficial net favorable development to the ADC, most notably from the commercial auto line.

Overall, given these pluses and minuses, the ADC experienced ceded adverse development of $723 million this quarter whereas AIG experienced net adverse development of $170 million, as Brian has said -- had already referenced.

Both figures are highlighted on Page 3 of the earnings release and on Page 44 of the financial supplement.

Now I want to take a moment and make clear that these reserve charges result from underwriting policies that existed in the past.

I do not view these as a carryforward issue into policies being written today under the revised underwriting strategy.

For additional clarity, I will walk you through the progression of some of these reserve changes.

You could reference Slide 13 as I discuss this.

As stated, the reserve charge gross of the ADC for Excess Casualty was $1.26 billion adverse.

The total charge across all lines of business for General Insurance, also on a gross of ADC basis, was $950 million.

This can be seen on Page 44 of the financial supplement by adding the unfavorable developments in lines covered by the ADC of $904 million and the unfavorable development from lines not covered by the ADC of $46 million.

The net of ADC adverse development for AIG of $170 million has already been reduced by the quarterly ADC amortization of $57 million.

Adding back in this amortization results a net adverse development charge of $227 million, which implies the $723 million ADC adverse charge we just discussed.

So hopefully, that helps you instead of confusing you more.

Now the ADC as of 9/30/2018, has nearly $7.7 billion of limit remaining at 100% level and, given the 80% cession, have $6.1 billion of remaining limits still available to AIG.

Lastly, from the perspective of the operating units, $170 million of net adverse development this quarter saw $134 million emanate from North America, mostly from Personal lines development on California wildfires, another $38 million from International and a minor $2 million of favorable development from the Legacy Portfolio.

As -- shifting gears a bit.

As part of Peter's leadership team, I have been involved in the strategies of the various operating units, and I have a fairly detailed view of the underwriting changes that have already been implemented.

Peter has already commented on the massive constriction of gross and net limits on the Property, Energy and Casualty side of the business.

And I'll just editorially add, when I say massive, when there's $468 million reduction in the vertical net on Property and $150 million reduction in International Casualty and -- roughly $80 million, those qualify as massive reductions, in my view.

As respect to other underwriting changes that have been implemented, clear underwriting appetites have been established.

Clarity with distribution around these revised appetites is happening as we speak, and this will result in a targeted and higher quality submission flow.

Thinner limits and higher attachment points and/or deductibles are being bound.

A market proportional shift to smaller accounts has occurred and continues, especially in the Excess Casualty area.

A definitive strategy to increase the mid-excess proportion of the Excess portfolio has also been implemented.

Clear ventilation rules exist between AIG entities on the same risk the Chief Underwriting Officer governs.

Heightened underwriter accountability now dovetails clearly with revised authority.

Improved pricing tools and portfolio monitoring has been developed for, again, Excess Casualty, in particular.

Peter referred to the successful binding of the Global Casualty $75 million ex $25 million excess of loss treaty, which shows the broad support of the reinsurance market towards our significantly altered gross underwriting strategy.

Lastly, for those of you who know me, I'm generally considered a skeptic and a cynic when it comes to purported benefits of underwriting changes.

My 40 years of property casualty experience, however, tells me that these changes are substantive and will lead to improved loss ratios in any market.

Now with all of these going on, this is really an exciting time to be at AIG.

So now I'd like to turn the call over to Kevin.

Thank you, Mark, and good morning, everyone.

As you can see on Slide 15, Life and Retirement delivered solid results for the quarter.

Excluding the adjustments from the annual assumption update that Sid discussed, adjusted pretax income of $811 million was in line with our expectations, along with an adjusted ROE of approximately 13%.

We continue to deploy capital to attractive opportunities, leveraging our broad product portfolio and channel strategy.

As market conditions improve, we increased premiums and deposits across our annuity lines, Fixed, Index and Variable.

We grew Life Insurance sales, especially in our International Life business and increased premiums and deposits in Group Retirement.

Higher general operating expenses reflect this new business growth as well as our investments to strengthen our platforms and enhance our digital capabilities.

If market conditions continue to improve, we are well positioned to deploy more capital at/or above our targeted economic returns while recognizing we will incur additional expenses associated with new business growth.

Now I will briefly discuss results for each of our businesses.

Turning to Individual Retirement on Slide 16.

Premiums and deposits grew by over 40% with particularly strong growth in Fixed and Index Annuities.

With these strong sales levels, we achieved positive net flows for the quarter, excluding Retail Mutual Funds.

Retail Mutual Funds, which is a comparatively small base of our business, continue to face headwinds in the quarter, and net flows may continue to be challenged for a period of time.

Total assets under management and fee income for Individual Retirement remained strong.

We continued our practice of active spread management, but as expected, we saw continued compression from our Fixed Annuity portfolio due to the roll-off of higher-yielding assets that are being reinvested at rates below the overall portfolio yields.

Base net investment spread for Variable and Indexed Annuities benefited from higher accretion in other investment income.

After adjusting for these items, spread decreased for these products in line with our expectations.

Turning to Group Retirement on Slide 17.

We increased premiums and deposits with solid periodic deposits and growth in individual product sales.

Our confirmed new group acquisitions for this year are strong, although the timing of some of these planned conversions may shift to early next year.

As I have mentioned on previous earnings calls, we will continue to see attrition of some large groups due to planned sponsors reducing the number of providers in their plans, M&A activity in the health care market and other competitive factors.

Our surrenders were impacted this quarter of a loss of 2 large plans, and the natural attrition of plans may impact surrenders in future quarters.

We continue to believe that our differentiated model, which combines high touch and high tech service, positions us well as a leader in the growing not-for-profit defined contribution market.

Similar to Individual Retirement, assets under management remained strong and we continue to actively manage spreads.

Base net investment spread for Group Retirement benefited from accretion and other investment income.

Adjusting for these items, spread was in line with the prior year quarter.

While we are more optimistic than last quarter, our great conditions are still below our portfolio yields, and spreads remain under pressure.

Also, in a rising rate environment, we will need to maintain market competitiveness on the crediting rates through our in-force business.

As we look forward, across Individual and Group Retirement, absent significant changes in the overall rate environment, our current expectation is that our base net spreads will decline by approximately 0 to 2 basis points per quarter.

Let's now move to Life Insurance on Slide 18.

We continue to make progress in our Life business, and separating our operating model from Fortitude Re will allow us to further focus on our new business platform.

Total premiums and deposits increased, and we continue to grow sales in the U.S. and the U.K. with particularly strong new business growth in the U.K. Lastly, our overall mortality experience was favorable to pricing assumptions in the prior year quarter.

Turning to Institutional Markets on Slide 19.

We did not execute notable opportunistic transactions during the quarter.

The market pipeline for pension risk transfer transactions over the next 12 to 18 months continues to be robust.

Overall, our Institutional Markets business continues to be well positioned to capitalize on available growth while remaining focused on achieving targeted economic returns.

To close, our results for the quarter reflect our ongoing strategy to leverage our broad product expertise and our distribution strength to deploy capital to the most attractive opportunities, which we believe continues to position us well.

Now I would like to turn it back to Brian to open up to Q&A.

Thanks, Kevin.

Let's go to Q&A.

(Operator Instructions) And we'll take our first question from Yaron Kinar from Goldman Sachs.

My first question is around the CAT losses.

So I guess the one thing I'm still struggling with after the pre-announcement and in the quarters, just get a better feel of the AAL.

And AY CAT losses were, at least, year-to-date seemed to be tracking well above the AAL.

And I understand that Japan losses were very significant.

But at the end of the day, I would would've thought that these are items -- or maybe exposures were known or the market weighting in Japan was known well into this year.

So maybe any additional color you can offer on your thoughts on AAL here, if the 4.5% is still a reasonable number to think about.

Yes.

Sure, Yaron.

I think it is, but I think Peter can give you a little bit more color.

Yes, look -- thanks, Brian, and thank you for the question.

Let me just start with the CATs.

We had said $1.6 billion in the quarter.

If you take out legacy Validus and then the $150 million of -- in the year, adjustments for mudslides, you're down to about $1.2 billion.

And so the Japan loss, again, it's one of the worst CAT years in 25 years, is $750 million, Jebi being the worst.

And so on an occurrence basis in terms of the return periods, it's actually exactly where we had thought.

And so there's -- with AAL and reinsurance, it was at expectations.

Now that had more frequency in the quarter than we've seen in the past.

And so therefore, taking a look at the AAL, then I'll comment on that in a second, we also had about $440 million in North America.

So when you look at those 2 and you look at the frequency that we had within the quarter on an aggregate basis, it's within modeled expectations.

So Mark commented a little bit on what we're doing on Property and gross limits.

And the AALs, if you look over the last 10 years, have been around 100 basis points or thereabouts lighter than our actual CAT experience.

So it's within line.

We're looking to revise total insured values, looking at PMLs and looking at different reinsurance structures.

Brian said in his prepared comments that we'll be relooking at Japan, so we'll have less frequency and severity as we enter 2019.

And we're running the model to recalibrate and take a look at the AALs, but they are very much in line with our expectations over a longer period of time.

And we'll take our next question from Elyse Greenspan from Wells Fargo.

So my first question is just on the underwriting profitability target.

Brian reaffirmed the goal for the end of the year.

So I just -- as you guys think about inflation, can you just kind of talk through that and what you have embedded in getting to that sub-100 target?

And then just a clarification there, because I know that that's the year-end target, is your expectation that General Insurance will print sub-100 underlying plus AAL in the first quarter of 2019?

Well, let me do the last piece first, yes, I do.

That's what I meant by under rigging 2019.

So we expect the first quarter to be an underwriting profit.

So with respect to inflation -- well, we got Mark here.

I think Mark's probably the best to answer that question.

Mark?

Great, Brian.

Thank you.

As respect to -- as Peter mentioned, we have 4% on the weighted average pricing increase.

We view this that we're gaining, so we're having a margin expansion.

It varied by line of business, as you know.

Averages can be deceiving.

Weak -- our loss cost trends, which is a frequency-severity combination, range from about 0 in some lines to about north of 8 in other lines of business.

Auto, for example, we see that as being fairly high at this point, around the 8% area.

We're getting materially more than that in rates, and that's actually a margin expansion line.

I -- most of our other lines, Casualty, Primary, Excess, we're feeling those are margin expansion lines as well, given the rate changes we're getting.

We're at -- some of the challenged areas, where comp is a little more flat, might be losing a little bit there.

D&O overall is -- we think we're getting -- it's close to flat.

But some areas are improving, and some areas, like EPLI, we could be marginally losing on.

And we'll move on to our next question from Josh Shanker with Deutsche Bank.

Can we talk a little about $148 million of California wildfire losses a year later?

That's a lot of houses or a lot of businesses.

What's going on there?

Mark, you -- I think Mark would've answered this question.

This is great.

So thank you.

Well, you get some of the same effects, whether it's wildfires or the mudslides, which were really early in 2018.

You have a lot of high-value areas that are very hard to reach, some are even on the sides of mountains.

You have -- so you have -- and regulatory ability to get in and even inspect them.

So you have some of that delay.

You have a lot of demand for contractor services, not just the rates per hour, but the cost of the equipment, and everything else is moving up as well.

So I would say, overall, that accounts for a lot of it.

You also have -- because they are mostly high net worth individuals, there's additional living expenses that tend to accrue, and I would say some are using those liberally.

I mean I just would add one thing is that when we look at the mudslides before and the current accident year, there wasn't a lot of difference.

As Mark points out, there wasn't a huge amount of insureds who was -- one is just being able to get in, physically inspect, assessing all of the different elements of the loss and then looking at some of the demand surge, and then you add in all the other components of it.

It's just something that was not able to assess within the first quarter.

So we got a much better look as we progress throughout the year.

Yes.

I'd say that -- when I've looked at our abilities to understand what the CATs are and price them properly, we've got a pretty good track record.

I think this is a bit of an anomaly, and I think it's just a unique situation.

We've corrected this particular loss in the insureds.

We'll move on to our next question from Kai Pan from Morgan Stanley.

My first question is on capital management.

Given the stock's trading at 60% of book value, do you find it just more attractive to buying back your stock versus other deployment on your capital?

Well, I mentioned earlier we bought $350 million of stock in the previous quarter.

Price was around $53.

So I'd say this price is pretty compelling.

Let's put it that way.

And we'll move on to Tom Gallagher with Evercore.

Can you comment on 2016 and 2017 Excess Casualty reserves strengthening, how much that was?

And how much rate do you think you need on this line now?

And also, can you comment on -- I think 3Q had some current accident year true-up in it.

How much did that negatively affect the loss pick in 3Q?

Mark?

Yes.

Thanks for the questions, Tom.

Yes, there's a couple of factors on that.

One of them relates to the area of focus, which was on CD and wraps, because you got to look at that by policy year, and the losses that emerged in the work or the core work has really already been done.

So we anticipate some of those accidents occurring from those policy years in '16, in '17, in '18.

So that's causative for some of the drift up, more so in '16 than '17, more in '17 than '18, which you'd expect with a shrinking book on that.

So that's a piece of it.

And on the balance, which will also include '18, we've made some, I call them, market cycle adjustments associated with where we think we are in the underwriting cycle.

Excess Casualty is a fairly volatile line of business, and we're extreme on the underwriting cycle.

So we've made some adjustments in that respect.

And I would say lastly, and this is one of the key underwriting strategies of moving attachment points materially, there's some auto exposure coming in on a small set of lead umbrella policies that has some auto exposure, and we're reflecting that.

That would've been more '16 than '17 though.

Yes.

And I think -- Mark mentioned earlier that the -- we think we're actually getting margin increases in the Excess at this point, just in general.

And with respect to construction, Peter mentioned that -- maybe we mentioned we're reducing our appetite in that business to a very limited amount.

We move on to our next question from Brian Meredith from UBS.

Yes, just quick follow-up there.

Mark, was there any current year development in the underlying loss or the loss pick in the third quarter in the General Insurance operations?

And a follow-up to that.

A lot of the discussion has been on changing limits profiles and reinsurance and all such things that I assume are going to improve your loss ratios going forward.

When are we going to see that?

Because Brian, I think a lot of the discussion has been improving the expense ratio going to next year, not so much the loss ratio.

Okay.

Well, let me answer that one first, and I'll let Mark talk about the other piece, which is the loss ratios in the third quarter.

What do we see?

Well, I think we have seen a little bit already.

Mark's talking about margin improvements.

I said we're going to get into an underwriting profit next year.

A lot of that is expense.

No question about it.

That's more immediate.

And I've mentioned this before, I think -- and I think Mark would support the same position, and that is we may believe that these changes are taking place, but we want to see it start to play out in the actual numbers.

So we're not going to -- we're going to take a more cautious approach to adjusting loss ratios down as we see the actual results start to show themselves.

That's not to mean that we don't think there is some improvement taking place.

But we want to make sure that we're doing this the right way, long term, that we don't have to go back and adjust upward reserves because we were a little too optimistic that the improvement took place early.

So you just have to understand how that's going to play out.

But we do see it coming, and we expect that 2019, we're going to be in an underwriting profit position, not a great one, but a profit.

And we'll move from there to a great one.

Mark, do you want to talk about the third quarter?

Sure, sure.

Thank you.

Brian, I think a more direct answer to your question is in the accident year was 90 basis points of loss ratio impact for the GI level in total of -- by all the causative factors I just mentioned emanating from the one product line.

But you have to look at it in 2 pieces.

Those -- the current accident quarter, if you will, was 30 basis points.

Then there was 60 basis points of catch-up, the true-up, the first 2 accident quarters of the year to -- through third quarter.

So 60 plus 30 and so it's 90 in total.

We'll take our next question from Jay Cohen from Bank of America Merrill Lynch.

Yes.

Most of my questions were answered.

Just one last one.

Selling off part of the Legacy business, can you give us some sense of the earnings impact that, that will have?

Well, I think we've given you an idea that the ROE is around 3 to 5, and we're going to take about 20% of that off.

So Sid, do you want to do the more of the arithmetic on that?

But it's a fairly simple arithmetic question.

Yes.

I think you can see for the full year that we've had an ROE in that range, so think of it as about 20% off the income is obviously the adjustment for the noncontrolling interest.

You just have to look at the third quarter because we had a -- we have this CAT in there and everything taking that out.

I think it's a pretty simple arithmetic.

We'll move on to our next question from Erik Bass from Autonomous Research.

I had a question for Kevin.

You mentioned that the ROE for the Life and Retirement business was in line with your expectations.

But the level of normalized earnings this quarter was well below where they were in the second quarter, and the recent trends spiked benefits from a favorable equity market tailwind and good mortality.

So just wondering, were there any other moving pieces to consider in the third quarter?

And how should we think about the run rate earnings power for the business?

Yes.

Thanks, Erik.

Well, I mean, there were no other unusual moving pieces in the third quarter.

I think looking past the actuarial review impact, in the current conditions and especially at the current level of premiums and deposits, which we're enjoying, which are higher than we've seen for some time, I really consider the third quarter run rate to be within our range of expectations.

And the returns in the low to mid-double digits is pretty much within the range that we have previously suggested.

And we'll take our next question from Jay Gelb from Barclays.

I had 2 questions.

The first is what level of return on equity do you think AIG is capable of achieving once you start seeing the improving Property Casualty coming through, thinking about kind of 2020 and beyond?

And then the second question, more near term, is, Sid, based on a 25% overall effective tax rate for 2018, that would seem to imply a fourth quarter 2018 tax rate around 30%.

Am I right on that?

Let's get that one last.

No, you're not right, and I think Sid will get that to you.

So return on equity, well, I would say that the -- you got to have -- we got to recognize we had a Legacy return on equity that's in that 3 to 5 range, but the rest of it, the core, I think, would be 9-plus.

And so weighted average maybe gets into the 8 range.

Of course, it depends on where Legacy goes.

But assuming it's still part of the family, I think that's the number.

Sid, do you want to talk about this tax rate?

Yes.

Thank you, Brian.

No, I think the 25%, remember, is simply intended to be -- as we look at, we've previously suggested that, for a full year, we're looking at 21% to 22%, excluding tax discrete.

And so as there's lower net income forecasted for the year, given the catastrophes, which would be taxed at 21%, we expect a higher expectation for the fourth quarter.

And so we concluded on our catch-up this quarter.

So it would be at that 25% level is our approximate expectation.

And we'll take our last question from Paul Newsome from Sandler O'Neill.

Any thoughts on the most recent earnings -- or equity volatility and financial market volatility this month on the Life Insurance businesses, both pro and con?

Yes.

Kev?

Yes.

Thanks, Paul.

As we've reported previously, our hedge program has performed and continues to perform within expectations.

Sid, if you have anything else to...

No, I think, as you know, we fully hedge for equity markets, and so we don't -- haven't seen anything in the recent market expectations around our Variable Annuity hedging program.

Just a reminder that, of course, on the investment side, there are securities in terms of private equity hedge funds and fair value options that do have equity market sensitivity.

But obviously, we need to see how the full year plays out for the Life Insurance business and the investment side there.

Right.

Okay, listen, I think we're going to end the Q&A.

I really do appreciate all your attention.

And we had a lot of content, so we cut the Q&A a little shorter than normal.

Thank you for your indulgence on that.

Look, but before I end the call, I want to recognize the hard work and tireless support that our employees around the world have provided to our customers and neighbors affected by the numerous global catastrophes.

Our hearts go out to all of those who have been impacted.

So with that, thank you very much.

This concludes today's presentation.

We thank you for your participation.

You may now disconnect.