American International Group Inc (AIG) 2019 Q3 法說會逐字稿

Good day, everyone, and welcome to AIG's Third Quarter 2019 Financial Results Conference Call.

Today's conference is being recorded.

And at this time, I would like to turn the conference over to Ms. Sabra Purtill, Head of Investor Relations.

Please go ahead, ma'am.

Good morning, and thank you all for joining us.

Today's call will cover AIG's third quarter 2019 financial results announced earlier this morning.

The news release, financial results presentation and financial supplement were posted on our website at www.aig.com at 7 this morning, and the 10-Q for the quarter will be filed later today after the call.

Our speakers today include Brian Duperreault, President and CEO; Peter Zaffino, CEO, General Insurance and Global Chief Operating Officer; Kevin Hogan, CEO, Life and Retirement; and Mark Lyons, Chief Financial Officer.

Following their prepared remarks, we will have time for Q&A.

Before Brian begins, please note that our commentary and discussion may contain forward-looking statements relating to company performance, market conditions, business mix and opportunities and strategic priorities.

These statements are not guarantees of future performance or events and are based on management's current expectations.

Actual performance and events may differ materially.

Factors that could cause results to differ include the factors described in our first and second quarter 2019 reports on Form 10-Q, our 2018 annual report on Form 10-K and other recent filings made with the SEC.

AIG is not under any obligation and expressly disclaims any obligation to update any forward-looking statement, whether as a result of new information, future events or otherwise.

Additionally, some remarks may refer to non-GAAP financial measures.

The reconciliation of such measures to the most comparable GAAP figures are included in our earnings release, financial supplement and presentation.

I'll now turn the call over to Brian.

Good morning, and thank you for joining us this morning.

As I've said over the past few quarters, our leadership team has taken significant action on a number of fronts to lay the foundation for long-term, sustainable and profitable growth at AIG.

The execution of our strategy is reflected in our results this quarter, which were in line with our expectations.

Our efforts in General Insurance have been focused on fostering a culture of underwriting excellence, outlining a consistent risk framework and reducing risk and volatility in our portfolio.

Peter and the team at GI are executing with focused urgency that is impressive, and the marketplace has taken notice.

We're beginning to see these significant efforts pay off in our results with the third quarter performance in GI yielding remarkable improvement over prior years.

I am particularly pleased that the GI reinsurance strategy played out as designed, dramatically reducing volatility in CAT season and preserving capital.

We've been leading the market with our professional approach to judiciously deploying capacity, appropriately addressing loss cost inflation and continued underwriting discipline on our pricing models.

These actions are playing out against an industry backdrop of prolonged pressures on accident year loss ratios, loss cost inflation, significant catastrophes over multiple years, pressuring the property lines, a lower interest rate environment, a complex retro market and a fatigued alternative capital market.

This market dynamic is different from the past because we now see the industry as a whole is acting more rationally, and this combination of changed behavior and external forces reaffirms my belief that this market cycle is sustainable for the foreseeable future.

One example where AIG has taken a leadership position in the industry is the way we tackle the increasing industry-wide issue in cyber insurance.

Companies are more focused on the severity of losses, increased systemic risk, a more stringent regulatory environment and ransom demands that are dramatically escalating.

In September, we formally announced our affirmative cyber initiative and have provided clarity to the market on how cyber coverage should be addressed and have mitigated ambiguity that existed.

Turning to our third quarter financial results.

As you saw on our press release this morning, adjusted return on common equity for the third quarter was 4.1% and 8.6% year-to-date.

Adjusted after-tax income was $505 million or $0.56 per share for the third quarter compared to an adjusted after-tax loss of $301 million or $0.34 per share in the prior year quarter.

Year-to-date 2019 adjusted after-tax income was $3.57 per share, reflecting a $1.80 per share improvement over the first 9 months of 2018.

General Insurance delivered a third consecutive quarter of adjusted accident quarter underwriting profit.

The adjusted accident quarter combined ratio, excluding CATs, was 95.9%, and we remain on track to deliver an adjusted underwriting profit for the full year.

In addition, last week, GI announced a new syndicate at Lloyd's serving the $50 billion specialist U.S. high net worth market.

Lloyd's has undertaken a comprehensive plan to regain its preeminent position in the marketplace.

We've been actively seeking opportunities to be part of that effort, first with our acquisition of Validus, which included Talbot, a respected Lloyd's platform; and now with this further partnership.

Peter will give you more detail on the great progress we are making in General Insurance as we position that business for the future.

Turning to Life and Retirement.

We continue to produce solid results in the third quarter with adjusted pretax income of $646 million and adjusted return on equity of 10.1%, including the impact of the annual assumptions review.

Adjusted return on equity was 12.5%, if you exclude the impact from the actuarial review process.

Despite continued headwinds from sustained low interest rates, we remain optimistic about our ability to deliver return on common equity in the low to mid-teens for the year given the strength of our diversified product portfolio and broad distribution network.

Kevin will provide additional information on the work he and his team are undertaking in Life and Retirement.

Net investment income was $3.4 billion for the quarter and $11 billion for the first 9 months, which is ahead of our original expectations.

Our year-to-date performance was due largely to strong alternative returns.

Mark will provide more detail on this and our overall financials.

Finally, I wanted to touch on AIG 200, our multiyear, enterprise-wide program to improve our core processes and infrastructure.

We have 4 core objectives for AIG 200, which are achieving underwriting excellence, modernizing our operating infrastructure, enhancing user and customer experiences and becoming a more unified company.

This work is absolutely critical if AIG is to become a top-performing company, and we are carefully prioritizing areas for investment and transformation.

We will provide estimated levels on investment, charges and savings for 2020 on our fourth quarter earnings call.

Peter, as the executive leading this important initiative, will elaborate further on the progress we've made through the third quarter.

Our ability to tackle transformational projects is supported by our strong balance sheet.

We are focused on maintaining financial flexibility, while reducing our leverage and reinvesting in our businesses.

Looking to the remainder of the year and into 2020, we are confident in our ability to achieve our goals for 2019 and continue to believe that the momentum we have generated will enable us to deliver a double-digit ROCE by the end of 2021.

We still have a lot of work ahead of us, and I'm very pleased with the progress we are making across the organization, and remain confident that AIG is well on its way to being a leading global insurance company.

With that, I'll hand it over to Peter.

Thank you, Brian, and good morning, everyone.

Today, I will provide an update on the General Insurance third quarter financial performance, the turnaround that we are executing in key business units, along with our observations on the current rate environment.

I will discuss the recently announced new Lloyd's syndicate serving the U.S. high net worth market.

And finally, I will close with comments on AIG 200.

The General Insurance third quarter results reflect steady progress towards our goal of achieving sustained underwriting profitability.

We continued to execute on the bold strategic moves we identified as critical to reposition our global portfolio, which include disciplined underwriting and reinsurance strategies, a clear focus on operational excellence and investing in talent for the future.

We continue to significantly reduce volatility through improved risk selection and aggressive limit management and align our businesses with our redefined underwriting strategy.

We're achieving better rate adequacy and, in many cases, are leading the market.

As we expected, it will take time for these actions to fully earn through our reported results, but I'm very pleased with our improved financial performance.

Turning to our results.

The third quarter adjusted accident quarter combined ratio was 95.9%, a 350 basis point improvement year-over-year, including a 210 basis point improvement in the adjusted accident quarter loss ratio and a 140 basis point improvement in the expense ratio.

In North America, the adjusted accident quarter loss ratio was 69.5%, a 30 basis point improvement year-over-year.

Our financial performance in the third quarter reflected the favorable impact of changes to our business mix, which was achieved through reductions in higher loss ratio businesses, strong results from Glatfelter, improved rate adequacy across a number of lines, lower frequency of Property attritional losses across Retail, Wholesale and Western World and the benefit of treaty and facultative reinsurance.

These positive drivers were largely offset by our Crop and Specialty businesses.

Crop negatively impacted North America's adjusted accident quarter loss ratio by approximately 100 basis.

Significant spring flooding in the Midwest affected many of our insured farmers, which drove a higher volume of prevented planting claims.

As a result, we increased our 2019 loss ratios despite remaining uncertainty regarding the ultimate profitability of this crop season.

We experienced higher losses in Inland Marine, Aviation and Energy, which negatively impacted North America's adjusted accident quarter loss ratio by approximately 250 basis points.

We continue to take actions to improve performance in U.S. Specialty lines through enhanced risk selection, rate improvement strategies and reinsurance.

Despite these headwinds in North America, on a global basis, our Specialty business performed very well with international seeing lower expected loss activity.

North America Personal Insurance performed as expected with 110 basis point improvement in the adjusted accident quarter loss ratio and a 100 basis point improvement in the adjusted accident quarter combined ratio.

Overall, I'm very pleased with North America's core performance and believe the trends we are seeing demonstrate a higher quality, better price and more balanced portfolio that is improving each quarter.

Moving to International.

The adjusted accident quarter loss ratio was 53.9%, a 430 basis point improvement year-over-year.

In International commercial, we had fewer severe and attritional losses that were below trend, particularly in Specialty and Talbot, our Lloyd's syndicate.

Additionally, reunderwriting efforts in Property within the U.K. and Europe are beginning to earn into our reported financials.

The International Personal Insurance adjusted accident quarter loss ratio improved 290 basis points as we continued to see a favorable trend in Japan personal auto.

The third quarter expense ratio for General Insurance of 34.4% was in line with our expectations and reflects our sustained diligence around expense control.

The third quarter of 2019 was an active CAT quarter.

Total net CAT losses were $497 million compared to $1.6 billion in the third quarter of 2018 and $3 billion in the third quarter of 2017.

There were 9 events in the quarter, 6 of which occurred in North America and 3 occurred in Japan.

Our net CAT ratio in the third quarter was 7.5% with 2.9% attributable to events in North America, 3.5% attributable to Japan and 1.1% attributable to Validus Re.

Net CAT losses from events in North America are over $203 million.

Hurricane Dorian was the single largest driver of net losses at $135 million, of which $10 million was attributable to Validus Re.

The 5 remaining smaller events in North America range from $10 million to $14 million in net losses per event.

Net CAT losses from the Japan events were $294 million, of which $65 million was attributable to Validus Re.

Typhoon Faxai, the strongest typhoon to hit the Kanto region since 2004, was the main driver of these losses.

I want to provide an overview of our business in Japan, given our significant market presence and the actions we have taken to address our CAT exposure.

Japan is AIG's largest market outside the United States.

It represents approximately $5.1 billion in gross premiums written with 82% in personal lines and 18% in commercial lines.

AIG is the largest foreign-owned General Insurance company in Japan with an average market share of 6% in the regions most impacted by recent CAT events.

As I've said in the past, as part of our strategy to contain volatility and manage CAT within our risk appetite, we enhanced our reinsurance coverage in Japan to reduce net retention and protect against tailwinds.

When we restructured our worldwide CAT program for 2019, this included moving to a single-occurrence tower for Japan with a modeled expected attaching point of the 1 in 7-year event.

This dedicated occurrence protection exhaust at the modeled 1 in 69-year event, and we have additional protection from our global CAT cover that provides protection in excess of the 1 in 500-year event.

Also, additionally, we also purchased international annual aggregate protection, which attaches at 1 in 10-year event for losses outside of North America, with approximately 80% of the model expected losses coming from Japan.

Overall, Japan has been a profitable business with an average historical adjusted accident year loss ratio of approximately 50%, and its prospects for improved profitability are very strong.

With respect to the recent CAT events in Japan, it is still early days from Typhoon Faxai, and as a result, industry loss estimates from the modeling firm span a broad range of between $3 billion and $9 billion.

Typhoon Hagibis, a fourth quarter event, is also expected to be sizable with AIR's early industry loss estimate currently at $8 billion to $16 billion.

To put the recent CAT activity in Japan in perspective, there have been 12 designated catastrophic tropical storms since 1980, 4 of which occurred in the last 2 years.

2018 and 2019 will be the 2 largest years for insured losses from tropical storm event in the last 40 years.

However, irrespective of the eventual size of these industry losses, AIG's expected maximum retained net loss is limited due to the benefit of both occurrence reinsurance and the international CAT annual aggregate protection.

Based on our current estimation of year-to-date CAT losses in international and excluding Validus Re, our net loss before reinstatement premium arising from this event is estimated at no more than approximately $75 million.

We have significant additional covers in place to protect against further fourth quarter CAT activity.

In our International business, including Japan and taking Hagibis into account, our maximum retention at any one loss is now $20 million before any additional reinstatement premiums for the remainder of the year.

With respect to Validus Re, we have an aggregate retro in place.

Given the wide range of modeled loss estimates, it's too early to provide a net CAT loss estimate from Typhoon Hagibis.

However, we expect the retro aggregate to attach with another $155 million in CAT losses.

As I've done in past calls, I'd like to highlight a few businesses that exemplify the work we're doing to reduce risk, reposition our portfolio and establish AIG as a market leader.

I recognize that I've mentioned Lexington often, but the magnitude of the turnaround in this business over the past year is powerful and worthy of another update.

In Casualty, submission volume increased 62% in the third quarter, and we were successful in further diversifying in the middle market while reducing total limits by 58% and improving rates by 31%.

In Property, submission volume increased 47%, while we reduced total limits by 74% and increased our overall proportion of excess policies from 11% at the end of 2018 to 25% on a 2019 year-to-date basis.

In addition, we increased deductibles by 18%, continued to improve terms and conditions and achieve third quarter renewal rate increases of 15%, which we expect will continue to increase, particularly due to the recent global CAT activity.

Turning to North America Retail Property, we continued to reduce aggregates and exposure to certain classes, increase deductibles and achieve meaningful rate increases.

We reduced total gross limits by over $20 billion or 37% in the third quarter and over $80 billion or 49% year-to-date.

Average deductibles increased by 27% in the quarter and 30% year-to-date.

These significant underwriting actions are reducing volatility, improving attritional loss exposure and significantly reducing CAT exposure.

Multiyear agreements represented approximately 30% of the Retail Property portfolio at the end of the third quarter.

These agreements will be approximately 20% of the portfolio by the end of 2019 and will trend lower during 2020, which will drive additional momentum in our reunderwriting efforts.

In the third quarter, we saw North America Retail Property rate increases accelerate as there was broader pullback in market capacity and disciplined underwriting was more prevalent.

The portfolio yielded high-teen rate increases and achieved mid-20s rate increases when you exclude the impact of long-term agreements.

We expect to continue to see rate increases through the remainder of the year and into 2020.

In North America Financial Lines, we accelerated remediation of challenged segments, and in the quarter, we achieved rate increases exceeding 30% across Commercial D&O led by increases exceeding 35% in public D&O.

At the same time, we reduced Primary Commercial D&O aggregate limits by over 40% compared to a 30% reduction in the second quarter, and we reduced Primary Commercial D&O policies with limits greater than $10 million in lead layers by over 40%.

While we've done significant work to reunderwrite our global portfolio, we're also focused on investing in attractive growth opportunities such as Accident and Health.

A&H represents more than $3 billion in gross premiums written and is one of our best-performing businesses.

We believe A&H will be a strong engine of future profitable growth because global demographic trends and demand for A&H products match up well with our product offerings and global footprint.

Now I'd like to comment briefly on the overall rate environment and some broader market observations.

We continued to see meaningful rate increases across the board in the third quarter, a trend that has accelerated throughout the year.

In some lines, rate improvement has been at the highest level in over a decade.

The overall third quarter rate improvement for General Insurance, excluding Validus and Glatfelter, was in the high single to low double digits.

North America commercial rates increased in the low double digits compared to a high single-digit rate increase in the second quarter.

International commercial rates increased in the low to mid-single digits on average across all geographies, consistent with the second quarter.

Outside the specific businesses I highlighted earlier.

In North America, we're seeing meaningful rate increases in the mid-20% range in Admitted Excess Casualty and Energy.

In international rate increases continued to accelerate in the U.K. driven by approximately 20 points of improvement in D&O, over 30 points of improvement in Marine and low double-digit improvements in Energy.

These increases are being driven by proactively addressing loss cost trend inflation as well as our unique position as a primary and lead market across the commercial insurance landscape.

As we've noted on prior calls, loss trends due to the tort environment and social inflation continue to be key areas of focus for us.

We recognize this growing trend a while ago, and we continue to monitor the changing landscape and respond to risks as they evolve.

Given recent news reports about settlement activity, I want to touch on developments concerning litigation related to opioids, which has been filed primarily by state and local governments against manufacturers, distributors and retailers.

While it's too early to predict or quantify the outcome of all the litigation or the insurance that may apply, we've been very closely tracking these developments, so that we can address these complicated risks appropriately as they continue to evolve.

As Brian noted, we recently announced a plan to launch an innovative syndicate with Lloyd's, focused on the U.S. high net worth portfolio.

We expect this syndicate will support new and renewing business starting in the first quarter of 2020.

The specialist syndicate, Syndicate 2019, is expected to be Lloyd's largest-ever single new entrant with premium of up to $1 billion.

Lloyd's unique capital structure provides us with increased flexibility and efficiency and enables us to attract new and diverse long-term capital partners to support profitable growth where opportunities exist.

Before turning the call over to Kevin, I'd like to make a few comments on AIG 200.

As Brian noted, AIG 200 is our global, multiyear effort to reposition AIG as a top-performing company.

We are focusing on programs that involve transformational change to our infrastructure and underwriting operations as well as developing a new data architecture, all of which will be designed to achieve best-in-class operations that deliver value through scale and simplification.

While AIG 200's primary purpose is not cost-cutting, we do expect to achieve a reduced expense base over time and, more importantly, a much better experience for our distribution partners, clients, policyholders and colleagues.

AIG 200 reflects our commitment to continuous improvement as we pursue operational excellence and fortify our competitive position.

We're very pleased with the high level of engagement from colleagues across the company who are committed to shaping the future of AIG.

Now I'll turn the call over to Kevin.

Thank you, Peter, and good morning, everyone.

Life and Retirement recorded adjusted pretax income of $646 million for the quarter and adjusted return on attributed common equity of 10.1%.

Excluding the annual actuarial assumption update, adjusted pretax income was $789 million and adjusted return on attributed common equity was 12.5%, within our expectations.

Adjusted pretax income decreased by $67 million from the prior year quarter.

The primary drivers of the difference were the annual actuarial assumption update, which accounted for $45 million of the decrease and elevated mortality.

These unfavorable impacts were partially offset by the lower interest rate environment, driving increased call and tender income and higher returns on fair value option securities as well as other net investment income adjustments.

There was nothing in our assumption review nor mortality experience has suggested a change in the inherent profitability of our products nor a need to change our pricing strategy.

Also, it's worth noting that from a pretax income perspective, the overall impact from the assumption update was a positive $20 million.

Year-to-date, our adjusted pretax income was $2.6 billion, and adjusted return on attributed common equity was 14%.

We are pleased with our results to date but recognize the challenges and headwinds the low interest rate environment presents, along with the potential for increased volatility in equity markets given global trade and geopolitical concerns.

The declining equity markets would, among other things, negatively impact fees as well as deferred acquisition cost amortization.

Declining interest rates would typically result in higher returns on fair value option securities, although the impact on net investment income could be uneven and would depend on the timing and degree of interest rate movement.

At interest rate levels as of the end of the third quarter, our current expectation is for base net spreads to decline by approximately 1 to 3 basis points per quarter through the end of next year.

Finally, from a statutory perspective, we expect to continue to generate solid earnings and for our year-end 2019 risk-based capital levels to be higher than our strong year-end 2018 levels.

Our top line results continue to reflect our ongoing strategy to leverage our broad product portfolio and diversify distribution network to satisfy customer needs.

During the quarter, we grew Index Annuities sales in Individual Retirement and executed opportunistic transactions in Institutional Markets.

We expect lower levels of sales for certain product lines in the fourth quarter due to lower interest rates and the uncertain environment.

We will remain disciplined with respect to product pricing and features and continue to deploy capital to available attractive new business opportunities.

For Individual Retirement, premiums and deposits increased slightly due to growth in Index Annuities sales.

Although Fixed Annuities sales were basically flat from the prior year quarter, they have declined significantly from first quarter levels.

We expect lower sales of Fixed Annuities in the prevailing interest rate environment.

We achieved positive net flows, excluding Retail Mutual Funds, which is a comparatively small part of our earnings.

Total assets under administration increased driven by strong equity market performance and growth of annuity deposits during the first half of the year.

At Group Retirement, premiums and deposits were lower than the prior year quarter driven by a decrease in individual product sales related to lower crediting rates.

Net flows improved from the prior year quarter due to lower group surrender activity but still remained negative.

Although the timing of group acquisitions and individual contributions will result in quarter-over-quarter variances in deposits, we expect surrenders and other withdrawals to continue to drive negative net flows.

It is also important to note that the financial impact of outflows will vary based on product characteristics.

Despite facing negative net flows for a period of time, we've continued to produce solid earnings for this business as assets under administration have continued to grow.

For our Life Insurance business, total premiums and deposits increased due to international sales.

Our U.S. Life sales declined as we deemphasize guaranteed universal life sales in the current interest rate environment, and indexed universal life sales remain under pressure.

While mortality was elevated during the quarter, it should be viewed in the context of generally favorable mortality trends we have seen over the last 2 years.

For Institutional Markets, premiums and deposits increased due to a large GIC issuance and select transactions in our Pension Risk Transfer business.

We have continued to opportunistically grow our asset base, and our Institutional Markets business continues to be well positioned to capitalize on available growth while remaining focused on achieving targeted returns.

Across our businesses, we are continuing to invest as needed to prepare for the evolving regulatory and accounting landscape, including the SEC's Regulation Best Interests, New York State's Regulation 187, NAIC's Variable Annuity framework and FASB LDTI, among others.

We will take the opportunity to leverage these ongoing investments to further improve our efficiency and competitive position.

To close, we remain committed to our ongoing strategy to leverage our broad product expertise and distribution footprint to deploy capital to the most attractive opportunities, which we believe positions us well to help meet growing needs for protection, retirement savings and lifetime income solutions.

Now I will turn it over to Mark.

Thank you, Kevin, and good morning, all.

AIG's adjusted after-tax earnings per share was $0.56 for the quarter compared to a negative $0.34 per share in the corresponding quarter of 2018, representing a $0.90 per share improvement.

Adjusted book value per share, which excludes AOCI and the DTA, increased 1.2% sequentially from second quarter and increased 4.8% relative to year-end 2018.

As Brian mentioned, adjusted return on equity was an annualized 4.1% for the quarter and 8.6% on a 9-month year-to-date annualized basis.

Keeping to that theme, the individual segments achieved the following annualized 9-month year-to-date returns on attributed equity: General insurance, 9.5%; Life and Retirement, 14%; and Legacy at 4.6%.

Net investment income, or NII, in the third quarter was nearly $3.5 billion on an adjusted pretax income basis and $3.4 billion on a GAAP basis.

This was nearly identical to the third quarter of 2018 on both an adjusted and GAAP basis and sequentially, was $260 million lower on an adjusted pretax income basis and $337 million lower on a GAAP basis.

This quarter, investment income from all fixed income securities was virtually identical to that of the third quarter of 2018, whereas returns from the hedge fund and private equity composite was down materially to an annualized 4.6% versus the previous 6-month year-to-date return of approximately 17.6%.

Turning to General Insurance.

Brian and Peter commented on the accident quarter results.

So contrastingly, the calendar quarter combined ratio of 103.7% reflects 7.5 points of CAT losses, as Peter pointed out, with Japan representing approximately 60% of the global CAT losses.

The combination of gross line reunderwriting, together with our improved reinsurance program, has resulted in a material reduction to General Insurance net CAT ratios, from 45.5% of premium in the third quarter of 2017 to 22% even in the third quarter of 2018 to the 7.5% this quarter, an 84% reduction over a 2-year period.

Additionally, the prior period development ratio was negligible for the quarter net of ADC recoveries and ADC amortization.

Broadly speaking, calendar quarter underwriting gain was nearly $1.5 billion higher than the third quarter of 2018, with North America contributing $802 million of underwriting gain improvement and international operations providing $675 million of underwriting gain improvement.

Also, both commercial and personal lines improved their accident quarter underwriting margins third quarter over third quarter.

It's also important to comment on the General Insurance performance on a year-to-date 9-month basis versus the first 9 months of 2018.

On that basis, the year-to-date current accident year combined ratio, excluding CATs, improved 390 basis points, with the loss ratio contribution being a 240 basis point improvement, another 200 basis point improvement from the GOE ratio with a partial offset of a marginal 50 basis point increase due to the acquisition ratio.

As respect to volume, year-to-date reported net earned premiums are nearly flat with last year on a U.S. dollar basis.

But the continued success of the improved gross underwriting strategy, together with the increasing ceded earned premiums from reinsurance treaties place, are expected to reduce the fourth quarter net earned premiums by approximately 5% sequentially barring any additional reinsurance purchases.

Now I want to take the opportunity to reinforce the magnitude of the portfolio reshaping that Peter outlined in his prepared remarks.

This level of change is unprecedented in my 42 years in this business, but there are nuances, however, that should not go unnoticed.

For example, reducing Lexington Casualty total limits by an impressive 58% in the quarter, while simultaneously increasing or achieving an average rate increase in excess of 30% is only part of the story.

The 62% increase in submission volume signals the igniting of the wholesale distribution channel, along with its emphasis on smaller accounts that have more localized exposures and lower capacity needs.

The result in spectrum of risk quality has also broadened, thereby improving overall rate adequacy as well.

Peter also commented on General Insurance's achieved rate increases for the quarter, and helping to put the trajectory of those rate increases in context, 2 lines will be highlighted.

First, U.S. Directors and Officers Liability, or D&O, combining Primary and Excess coverages, accelerated between the fourth quarter of 2018 and the third quarter of 2019, inclusive as follows: 9% increase in the fourth quarter of 2018, 11% in the first quarter of 2019, 17% in the second quarter of 2019 and then 28% this quarter.

Secondly, Lexington Casualty business rate increases were similarly increasing as follows: 6% in the fourth quarter of 2018, increasing to 8% in the first quarter of 2019, 20% in the next quarter and then 30% -- 31% this quarter.

Furthermore, we were extremely pleased that our leadership position in driving rate extended even further beyond the United States, with Canada achieving a 9% weighted overall increase, the U.K. with a 10% increase and Continental Europe achieving 6% in the quarter.

Turning to prior year development.

Although actual versus expected loss emergence was reviewed and incorporated globally, the major areas receiving deeper reserve dives this quarter were, in decreasing order of reserve size, U.S. Primary Workers' Compensation, U.K. and Europe Casualty lines, U.K. and Europe Financial Lines, U.S. Commercial Property and U.S. Financial Lines Errors and Omissions, or E&O, exposures.

The negligible prior period development referenced earlier benefited from $58 million of favorable impact from the ADC amortization of the deferred gain, and therefore, adjusting for this, the development was unfavorable by $55 million.

Unpacking this further, the pre-ADC prior period development before any ADC recoveries was favorable by $74 million.

This potentially counterintuitive result stems from the sliding by accident year.

$226 million of net favorable development emanated from accident years 2015 and prior, and approximately $152 million of net unfavorable development stemming from accident years 2016 through 2018.

The accident year 2015 and prior net favorable development emanated primarily from U.S. Workers' Compensation, whereas accident year 2016 through 2018 net unfavorable developments stem primarily from U.S. Financial E&O lines and U.K. European Casualty and Financial Lines, with a partial offset of a favorable development from European Property and Specialty and European and Japan personal lines.

Digging further, approximately 1/3 of this aggregate $152 million of net unfavorable development for accident years 2016 to 2018 stem from the isolated impact of 1 large long-term agreement affecting all 3 accident years.

Approximately another 1/3 was centered in the U.S. Architects and Engineers book on both the primary and excess basis.

And we reacted to this due to an increased frequency of moderate severity claims so as to reflect negative trends more quickly than positive trends.

The remaining 1/3 reflected various U.K. and European impacts in Casualty and Financial Lines driven by a smattering of large claims, such as in French and Italian motor business and Irish employers' liability.

Overall, though, the pre- and post-ADC prior period development represents approximately 1/10 of 1% of carried reserves.

As for accident year 2019 implications from the net unfavorable development, we don't see any carryforward issues with a 1 long-term agreement, whereas the U.S. Architects and Engineers was deemed to have a minor roll-forward impact as with some of the European indication.

As a result, these deep dives, which, year-to-date represents 65% of total loss reserves, saw a 0.5 point loss ratio increase to accident year 2019, which reflected about 0.2 points for the current accident quarter with another 0.3 points of catch-up for the first 2 quarters of 2019.

This was virtually offset by the earned impact of stronger rate increases being achieved on the subject lines of business than anticipated within the original 2019 budgeted loss ratio.

Turning to the Life and Retirement segment.

The annualized 9-month return on attributed common equity of 14% would be unchanged after adjusting for both the actuarial assumption impact and the removal of a large beneficial IPO gain discussed last quarter since they virtually negated each other.

It's also informative to note that on a 9-month year-to-date basis, the adjusted after-tax income is nearly identical at $2 billion even compared to 2018.

Net spreads on the Variable and Index Annuity composite were up marginally from last quarter and for Fixed Annuities, was virtually flat with the sequential quarter.

Surrender rates on a year-to-date basis were flat for the Variable and Index Annuity composite and slightly better for Fixed Annuities within the Individual Retirement and 70 basis points better for Group Retirement.

Life Insurance experienced moderately increased mortality relative to the first half of the year with overall lapse rates fairly consistent with a year ago.

And Institutional Markets had roughly $350 million of pension risk transfer deals and also have roughly $375 million in debt issuance, as Kevin noted.

Lastly, assets under administration increased 1.1% sequentially and grew 10.4% through year-end 2018.

Turning to Legacy.

Adjusted pretax income was down slightly to $93 million on a sequential basis but up from $84 million in the third quarter of 2018.

Like Life and Retirement, Legacy's results reflected an adverse impact of the annual actuarial assumption update of $30 million.

Legacy net investment income on a year-to-date basis was flat with 2018 at $1.8 billion, and the annualized return for the quarter was 4.4%.

As respect to tax, the estimated effective tax rate was 22.6% for the year, applicable to adjusted pretax income and was 25.3% for the quarter, inclusive of discrete items.

As you know, the effective tax rate is updated each quarter using actual year-to-date results and then supplemented by the forecast of the remaining quarter.

As always, the tax rate is heavily influenced each quarter by the geographic distribution of income by tax jurisdiction.

We did not repurchase any shares in the third quarter, so our Board authorization remains at $2 billion.

And moving to leverage, as compared to year-end 2018, our total debt and preferreds to total capital ratio has now improved to 26.1% this quarter, which represents a 320 basis point improvement relative to year-end.

We also had a bond tranche redemption of $1 billion in mid-July that had been prefunded by our March 2019 debt raising.

This debt overlap at the end of the second quarter plus growth in earnings drove the leverage ratio improvement to 26.1%.

Adjusted book value per share increased 4.8% from year-end 2018, and GAAP book value per share increased 15.1% at year-end 2018, benefiting from approximately $7 billion of year-to-date AOCI gains.

And with that, I'll turn it back over to Brian.

Thank you, Mark.

So let's go to the questions and answers.

So operator, please go ahead.

(Operator Instructions) We'll take our first question from Paul Newsome with Sandler O'Neill.

I guess I'd like to focus in on the underlying combined ratio's improvement.

And is there any sense of how we can think about the magnitude as we look prospectively into, not necessarily next quarter, but prospectively, obviously, there was sort of a big bang this year with pre serious changes in reinsurance.

But how does that roll-forward look into 2020?

Mark -- maybe, Mark, do you want to try this?

Well, clearly, you have a lot of things going on at the same time, which I understand, makes it somewhat difficult to try to penetrate.

You have all the gross changes that Peter has itemized, which are fairly massive.

You have an increasing accelerated rate change on a written basis that will earn in to 2020.

And you have some level of loss trend that the industry is discussing that could perhaps go a little bit the other way.

So by not getting into chapter reverse, I think it's fair to say there'll be some level of improvement into 2020.

But the magnitude, let alone the reinsurance purchases, which affect your mix of business, really all need to come to bear.

So at this point, I'd say there's good -- you should think about improvement, but I'm not prepared to get into the magnitude.

Peter?

Brian, one thing I would just add to agree with everything Mark just outlined.

But the other piece is that we've been designing reinsurance structures that reflect the portfolio that we have.

And so with all the massive changes that we've seen in the Property book, I mean, that's what we're assessing over the next, call it, 45 days in terms of putting together some of the reinsurance structures that reflect the risk of the portfolio today.

So that will evolve as we enter into 1/1.

Yes.

The only other thing I'd say, Paul, like I've given you a data point, which is our expectation for a return on equity, which, by the end of 2021, we want to get over 10%.

So in order for that to happen, clearly, the GI, key portion of it, has to continue to improve.

And I think that gives some indication of what we think is going to happen.

A lot of the work that gets done is in rate, underwriting, et cetera, but it also is the expense levels that GI has, and those expense levels are high.

We have to get to those as well.

So yes, we'd love to be a little clear on that.

But all I would say to you is we're committed to continuous improvement in the combined ratios to get it to what should be world-class levels.

That's going to take some time, but we're committed to doing it over multiple years.

Go ahead, Paul.

Okay, you can do 1 more.

I was just going to ask, on the Life side, any thoughts on -- or comments on the SEC investigation?

That's obviously very topical to investors.

Yes.

Listen, I think you can appreciate that we just were not able to go into any detail about ongoing regulatory inquiries beyond what you would read in our 10-Q disclosure.

That's -- that document is going to be on file soon.

So I'd just point you to that for the additional information.

But thank you.

We'll take our next question from Tom Gallagher with Evercore ISI.

I'll just ask 2 quick ones.

Given the increase in Specialty losses in North American in commercial this quarter occurred while you've been growing top line pretty well, are you comfortable you've been growing that line profitably?

That's my first question.

And then, Peter, your comments on Hagibis losses for Q4, I just want to make sure I have the numbers right.

I was getting AIG's net retention being a maximum of around $75 million and Validus max of $155 million.

Is that a good way to think about 4Q based on what you know today?

Yes.

Let's do it backwards.

So let's do the Hagibis first.

Yes.

So the -- correct.

I mean the maximum -- the first part -- I just want to clarify the Validus Re.

The first part is that because of the per occurrence and the way the international aggregate and our global aggregate works that the maximum we would have on Hagibis is $75 million, assuming the forecast we've had for Faxai and others end up having the gross loss that we anticipate.

So I think that $75 million is a very good number.

What I said in terms of the Validus is that their retro would attach with another $155 million or thereabouts of retro loss.

So you should think about -- as it goes above $155 million, we have significant retro protection in place.

Yes.

On the specialty, let me start, but I think Peter needs to give you the most clarity on it.

The first thing we said was North American Specialty was higher.

The International was lower.

Net-net was actually pretty good, but it just happens to be the geographies here.

And frankly, we manage that business on a global basis.

We think of it as a truly global line.

You're writing aviation across the world.

I'm not sure it's fair to classify it as something we've been growing dramatically.

And maybe, Peter, you could give a little bit more color on it.

Yes.

No, I think you really outlined it well, Brian.

I think the only other thing is we've been getting rate in the Specialty classes.

When we look at a global performance in the quarter, we were very pleased.

Then when we look at the global performance, what we've seen year-to-date, it's been a very strong performer and expect it to continue to have good performance in the future.

So -- but what I said in the prepared remarks is that we're looking at limit management, how we're driving rate and how we're underwriting it.

But overall, it's a very good book.

Do you want to add?

Just one thing, if I could.

On Peter's, I think good explanation to your question about the maximum.

Just keep in mind that the $75 million, as he outlined it, in a likelihood sense, is more likely than Validus' additive number at the collection of ceding companies with different attachment points, and that remains to be seen.

I wouldn't take those 2 as additive necessarily.

Our next question will come from Josh Shanker with Deutsche Bank.

I'll be quick.

They're sort of similar, one Life, one P&C.

On the P&C, the improvement in commercial is outstanding when you take the medium and large loss in the crop numbers out of it.

I'm just wondering, with the new orientation of the book, is there seasonality in the accident year loss ratio in North American commercial that Q3 would be better than other quarters?

Well, look, accident quarters are so little, small that the volatility of a single-accident quarter, I think, would exceed any kind of seasonality.

I suppose you could come up with some in Property where -- with wind activities and things like that, you could have a seasonal movement.

But I just think you can't look at this an accident quarter as something that's very, very predictable.

There's a lot of volatility in a particular quarter.

That's why we report it because we have to -- I really point to the 9 months as a better indication of movement, okay?

Okay.

I appreciate that.

And then on Life, Kevin said that the mortality in the Life business should be viewed in the context that you've had very good mortality for the last couple of years.

Does that mean that Life has been over-earning just that one subsegment, and we should consider that in our forward model?

Well, Kevin?

Yes.

Thanks, Josh.

No, I think that the underlying message is that the third quarter was an anomaly.

We have seen our mortality well within pricing, in fact, that for 8 of the last 10 quarters within pricing.

And so the message is that we don't see this quarter as a suggestion that our ongoing trend for mortality is likely to change.

Going next to Brian Meredith with UBS.

One quick clarification number and then another one.

Peter, I think you said that the events going forward now are going to have a $20 million retention.

So i.e., the California fires going on right now, whether they get big or small, whatever, it shouldn't be that big of an issue?

What I was referring to was '20 is after Hagibis, that's our maximum retention in our International business.

Oh, International.

Got you.

Yes.

I mean, Brian, in terms of North America, we still have significant cover.

I mean we have some buy-downs that are California-specific that attach at $50 million.

We have a lot of occurrence cover, and we have a lot of aggregate.

So I mean, it's really hard to predict in terms of North America with the wildfires, but we have ample protection, and we'll see how it all evolves.

Great.

And then my next question is, there's been a lot of discussion this quarter about the tort inflation environment, particularly with respect to general liability lines.

I'm just curious, can you maybe give your perspective on it as well as what are the severity assumptions that you're assuming in your loss picks and your kind of reserving actions today?

Oh, I think, Mark, it's probably more yours.

Yes, happy to.

Hello, Brian.

Thanks for the question.

I think as we mentioned last quarter, we have, I think, some pretty steep loss trend pure premium assumptions and a lot of the lines of business subject to what you're concerned about.

Some of them in the mid- to upper single-digit trends already.

So that's already been kind of baked in on how we look at things, translates into pricing models, so forth and so on.

I think on even a broader basis, I mean, our view of those loss trends is based upon a lot of our own information and how that's moved over time between accident years.

And when you kind of think about the overlay of the macroeconomic environment, with everything Peter described, about moving up in attachment points, chopping the limits dramatically, I tend to look on a portfolio basis of all that reshaping, becoming a natural inflation hedge, irrespective of whether it's economic or social.

So I think the book characteristics are very protective in that respect.

Yes.

I can't emphasize enough.

Thanks, Mark.

About the -- whether it's a deductible or it's an attachment point or it's layering done in some of the Excess, whether it's D&O or other Excesses, all of those things are more powerful than the rate in terms of adjusting for inflation.

Next question will come from Jimmy Bhullar with JPMorgan.

I had a question first just on the lack of buybacks so far this year.

So given that you've got the AIG 200 initiative coming up, and I'm assuming you're still interested in acquisitions, should we assume that buybacks are probably going to be fairly light until maybe the leverage ratio has improved further?

And then, I had a question also on just your flows on the Life and Retirement side, specifically at VALIC, they seem pretty weak.

And if you could just comment on what's going on there and your outlook.

Okay.

Kevin can take the flows.

Let me address the buybacks.

So yes, we didn't buy back this quarter.

In my prepared remarks, I emphasized that I want to invest -- well, first of all, we want to deleverage, so that's an important thing to keep in mind.

And I want to reinvest in the business.

Some of that is the AIG 200 still being quantified.

And the other thing to keep in mind is when you're in a catastrophe quarter, you wouldn't normally buy back anyway.

And whether it's the third quarter or the fourth quarter, we've seen the fourth quarter also produce some catastrophes, I just think it would be prudent not to buy back in any case.

But the emphasis is on the other aspects of capital management that I referred to.

Kevin, why don't you talk about the flows?

Yes.

Sure.

So thanks, Jimmy.

So the situation at VALIC, we have been in a negative flow environment over the last couple of years.

I think it's a reflection of the fact that the new case acquisition for a period of time had tailed off.

We've seen an improvement in that recently, but this is a natural effect of the aging of the portfolio.

The VALIC has been a leading participant in this business for a long time.

And as either in plan participants roll out of the plan or as people start to utilize the underlying benefits that they have that certainly is part of what leads to the outflows.

In addition to that, particularly in the health care industry, over the last couple of years, there has been a lot of M&A activity.

And in cases where planned consolidations occur, in some cases, we're on the winning end of those; in some cases, we're on the losing end of those.

And we've reported those sort of large outflows as and when they occur.

So we're continuing to be successful in new plan acquisition.

We continue to see strong periodic deposits and nonperiodic deposits.

And we're likely to continue to see some negative flows in this business because of that natural aging effect in the portfolio.

But I think it's also important to recognize that the dollar value of the flows is uneven.

If they're higher guaranteed minimum interest rate flows, that is less of an impact on future earnings than more recent products.

And finally, the assets under management of this portfolio, which is the source of earnings in the business, have continued to grow along with the equity market.

So we're outpacing the negative flows with growth in assets under management.

We'll go next to Elyse Greenspan with Wells Fargo.

My first question.

You guys provided a lot of good information on the pricing environment, high single to low double-digit increases, getting a lot of rate in your commercial book.

I guess I'm trying to think through the rate versus trend.

If you could help us think where kind of loss trend is on a composite basis.

And then how we could think about -- I think you guys said 0.5 point of rate versus trend this quarter, how we could think about that building up as when it comes into earn over the course of the next year?

Mark?

Well, it's a great question.

See if I can give you a great answer.

So first, a very generic comment.

When you get a 12%, that's a North America number, a 12% weighted average written rate increase, whether your loss trend assumption is 4, 6 or 8, you're getting expansion.

And I'm picking those out just illustratively.

So we don't really get into what our weighted averages across the board because it really does vary by quarter with the massive changes Peter described, but we're viewing that as a beneficial 2020 benefit.

And because the way things go in, it's going to be more second half associated with the rate of increase of the rate changes.

So it's expansion on a gross basis, but you're looking at our results on a net basis.

And that interaction between it is very difficult to measure and probably from you on the outside looking even more so.

But I'll just keep it at least that you should expect expansion.

Okay?

Okay.

That's -- yes, that's helpful.

And then my second question kind of ties back to the comment you made about gross versus net.

You guys have been buying a lot of -- changing your reinsurance strategies in a different -- in addition to reunderwriting the whole book.

As we think about 2020, are there still big reinsurance projects or something that you have in mind?

Or we -- should we think about kind of a stable-ish retention year-over-year, meaning we could start seeing some expansion on the net premiums line?

Peter, would you?

Thanks, Elyse.

So a couple of things.

One is, we will be consistent with making sure that we're looking across the portfolio for reduction of volatility and outsized line size.

So there could be segments such as financial lines that we'll look at of doing something, perhaps, as we enter 2020.

And then the most important part is looking at the CAT because we've made so much dramatic change in terms of the gross limit and CAT exposed and the strategic projects that we're doing with Lloyd's.

So again, there's a lot of moving pieces and the reinsurance, and what we decide to put together for 2020 will reflect the portfolio that we have in force.

And so there will be some changes.

I can't describe those today because we're in the middle of it.

And -- but we'll be very transparent as we hit the sort of fourth quarter call.

Our final question will come from Erik Bass with Autonomous.

Can you provide more details on the drivers of your annual assumption review updates?

And particularly, any changes you made to your long-term interest rate assumption and if there's any impact on go-forward earnings from that?

Well, Mark?

Okay.

No.

I mean, Kevin.

Kevin is better at this.

Yes.

Sure.

Sure.

Thanks, Erik.

First, let's talk about what we didn't change.

We did not change our long-term interest rate assumption.

Our assumption -- each product has its own parameters to derive the final assumptions.

But generally, we assume a reversion to the mean over 10 years to 3.5% for the 10-year treasury.

Now to the rest of the review, as you know, we have a strong diversified portfolio between Fixed, Index and Variable Annuity as well as Life Insurance.

And each of those lines responds to economic changes differently, and experience emerges separately.

So the overwriting situation, rates are lower this year, which inevitably results in reduced lapse expectations for the Fixed Annuities where we write up the DAC, and that increases sort of APTI.

And then in index, one of our newer products, we're just seeing some experience emerge and where we updated lapse rates.

And then in the guaranteed living benefits, as you know, we account for those as an embedded derivative below the line.

There were 2 different effects on that this year.

The first was updating our lapse model, and the second was updating our mortality.

And both the fee income as well as the benefit usage goes below the line, which is why you see that economic movement below the line.

And then finally, in the Life Insurance business, there is the effect of interest rates on policyholder behavior, not dissimilar to Fixed Annuities.

And we've updated some of our premium projection capability and reinsurance calculations.

So net-net, I think that these are largely changes associated with what we've seen happen in the market.

No dramatic changes relative to the sizes of our reserves.

And as I stated in my prepared remarks, we do not see anything that changes our understanding of the inherent profitability of the business nor a need to change our pricing strategy.

So this is Ryan Tunis.

I just had 1 follow-up on the P&C side for Peter.

And so -- in your opening statements, you talked a little bit about the opioids and how it's still too early.

Just hoping you could maybe expand a little bit on what you guys are looking for there and what we should be thinking about in terms of timing of any potential reserve action or anything along those lines.

Well, first of all, I think, look at reserve action.

I'd like Mark to talk about that first.

Sure.

Well, interestingly, AIG, over time, has been building up mass tort reserves, not that we really talk about it, but a meaningful number associated with in our aggregate reserve levels across all accident years.

When you get into some of these complex coverage issues you're talking about, it's really difficult to even talk about it because we don't know the theory of liability that's going to come down.

You don't know how it's going to be triggered, and therefore, you don't know exactly how it may also be play out.

But we have a good amount of net mass tort reserves, and we still have roughly $6.5 billion in the ADC that is still available for this.

So we feel, from a reserve standpoint and given what we know today, in pretty good shape.

Yes.

Look, I think that probably is a good conclusion to the opioid question.

So let me just -- thanks, Erik.

Let me close by just -- and telling everyone how pleased I am with our progress to achieve long-term, sustainable and profitable growth for AIG.

And I want to thank our clients, colleagues, shareholders, industry partners and all the stakeholders for their continued support.

Thank you, and have a great day.

This does conclude today's call.

Thank you for your participation.