American International Group Inc (AIG) 2007 Q4 法說會逐字稿

Welcome and thank you for standing by.

At this time, all participants are in a listen-only mode.

(OPERATOR INSTRUCTIONS).

Today's call is being recorded.

If you have any objections, you may disconnect at this time.

I would now like to turn the meeting over to Ms.

Charlene Hamrah.

Ma'am, you may begin.

Thank you very much.

Good morning and thank you for joining us today.

Before we begin, I would like to remind you that the remarks made today may contain projections concerning financial information and statements concerning future economic performance and events, plans and objectives relating to management, operations, products and services and assumptions underlying these projections and statements.

It is possible that AIG's actual results and financial condition may differ, possibly materially, from the anticipated results and financial condition indicated in these projections and statements.

Factors that could cause AIG's actual results to differ, possibly materially, from those in the specific projections and statements are discussed in Item 1A, Risk Factors, of AIG's annual report on Form 10-K for the year ended December 31, 2007.

AIG is not under any obligation and expressly disclaims any such obligation to update or alter its projections and other statements whether as a result of new information, future events or otherwise.

The information provided today may also contain certain non-GAAP financial measures.

The reconciliation of such measures to the comparable GAAP figures are included in the fourth-quarter 2007 financial supplement, which is available in the information -- investor information section of AIG's corporate website and now I would like to turn the call over to Martin Sullivan.

Thank you very much, Charlene and good morning, ladies and gentlemen.

As usual, I am joined this morning by a number of my senior management colleagues.

AIG's results in 2007 were clearly unsatisfactory.

This was a very challenging quarter and year.

Rapid deterioration in the U.S.

residential real estate and credit markets significantly affected several of our operations and investments.

This offset the strong results generated through the first half of the year.

As expected, a number of our businesses were adversely affected by their exposure to the domestic residential real estate market and we expect these businesses will continue to be challenged in the foreseeable future.

As you have seen, AIG reported a net loss of $5.3 billion in the fourth quarter of 2007 and an adjusted net loss of $3.2 billion.

Two large items adversely affected the quarter's results.

First, adjusted net income for the fourth quarter was reduced by the net unrealized market valuation loss of $11.12 billion pre-tax or $7.23 billion after-tax related to AIG financial products super senior credit default swap portfolio.

These losses occurred in the context of a significant widening of spreads on asset-backed securities, principally those related to US residential mortgages, the severe liquidity crisis affecting the structured finance markets and the effect of rating agency downgrades on the underlying collateral.

Second, net income was affected by $1.71 billion in after-tax net realized capital losses, as well as $418 million in after-tax other than temporary impairment charges related to AIGFPs available for sale investment securities.

Two of our other operations exposed to the US residential real estate market were also affected in the fourth quarter.

United Guaranty reported an operating loss of $348 million and American General Finance reported $9 million in fourth-quarter operating income, principally due to an increase in the provision for finance receivable losses and a decline in mortgage banking revenues.

We expect operating results in both these businesses to continue to be challenged.

With UGC, our best estimate is that future premiums on the existing in force book of both domestic first and second-lien risks will exceed future losses incurred.

However, losses will likely emerge in advance of premiums earned and we expect that negative operating results will persist throughout 2008 as a result of continued weakness of the US housing market.

While not immune to the downward cycle of the housing market, AGF's adherence to disciplined underwriting standards will continue to help maintain credit quality.

We also believe opportunities to acquire high-quality portfolios, similar to the pending acquisition of Equity One's branch loan portfolio, will continue.

We are obviously witnessing and living through extraordinary market conditions and we are trying, as are many others, to value very complex instruments.

These valuations are not mechanical.

They involve difficult estimates and judgments.

I can tell you that we have, at all times, brought our best judgment to bear in making these valuations.

I want to take a few minutes now to further comment on the valuation of AIGFP's credit default swap portfolio and the circumstances that led to the filing of our recent 8-K.

In response to strong investor interest in November, we decided to devote a previously scheduled investor conference on December 5 to the subject of AIG's exposure to the residential mortgage market.

During that webcast conference, the AIGFP team discussed the model and the methodologies that they use to value the super senior credit default swaps.

They also discussed that their estimated numbers accounted for the differences between what is described as, one, the cash prices for the underlying securities and two, the pricing of the super senior credit default swap derivatives.

This is the negative basis adjustment.

The AIGFP team also noted that their numbers also took into account structural supports that provided additional protection to AIGFP in adverse circumstances.

For example, cash flow diversion features.

This is the structural adjustment.

AIGFP did not detail the dollar amounts of these adjustments or the gross amount of the unrealized market valuation loss without those adjustments.

In early February, as we prepared our year-end financials, we concluded that we should clarify and expand our prior disclosures relating to the methodology and data inputs used to determine the values of the super senior credit default swap portfolio.

We also understood at that time that we would not be using a negative basis adjustment when we filed our 10-K.

When we made that determination, we decided to inform the market what the estimated November 30 number would be without the negative basis adjustment.

At the same time, our auditors concluded that the Company had a material weakness in the valuation process and the oversight of that process for the AIGFP super senior credit default swap portfolios.

We then promptly filed an 8-K describing all of this information in detail.

We have heard from some that our 8-K was confusing.

We did provide a lot of detail, but it was detailed because we were trying to ensure that investors could see exactly how the gross number related to the net number used in the December presentation through the adjustments that were referenced in that presentation.

We have already begun the process to remediate the material weakness identified by PWC.

Since becoming CEO, I have made increased transparency and improved disclosure a higher priority.

We have taken considerable steps in the right direction and we will continue to do so.

We believe the disclosure in the 10-K, the presentation we have just put on our website and the discussion my colleagues will lead you through shortly will help our investors better understand how the extreme market conditions are affecting our businesses.

In many respects, we are in unchartered waters.

In this context, let me make a few observations.

The mark we reported this quarter essentially values the super senior notes net of the benefit from the cash flow diversion features built into these transactions rather than the value of the credit derivatives that AIGFP has written to protect these notes.

This is clearly not representative of the risk AIGFP holds on the super senior credit default swap transactions, but we could not find observable data points in this highly disrupted and illiquid market to value the protection we provide.

This business was carefully underwritten and structured using models to ensure that attachment points would withstand highly-stressed economic conditions.

We continue to believe, however, that the unrealized market valuation losses on the super senior credit default swap portfolio are not indicative of the losses AIGFP may realize over time.

Under the terms of these credit derivatives, losses to AIG would result from the credit impairment of any bonds AIG would acquire in satisfying its swap obligations.

Based upon our most current analysis, we believe that any credit impairment losses realized over time by AIGFP would not be material to AIG's consolidated financial condition.

Although it is possible that they could be material to results of operations for an individual reporting period.

Except to the extent of any such realized credit impairment losses, AIG expects AIG's FP unrealized market valuation losses to reverse over the remaining life of the super senior credit default swap portfolio.

My colleagues will speak to the valuation methodologies in further details in just a few minutes.

Over many years, AIGFP has provided excellent returns, efficiently utilizing capital and managed its business to generate low volatility in its economic results.

It has also been a tremendous source of intellectual capital and innovation within AIG.

With our other financial services businesses, AIGFP complements and will continue to provide further diversification to our core insurance operations.

In that regard, Joe Cassano has decided with our concurrence that he would like to pursue opportunities outside of AIG and effective March 31 will be retiring from AIGFP.

Joe has been a very valuable member of the AIGFP senior management team for over 20 years.

He has had a great career with us and we wish him the very best in the future.

Effective April 1, Bill Dooley will assume interim responsibility for the day-to-day operation of AIGFP's business, along with his current duties.

Joe will be continuing as a consultant to AIGFP for the rest of the year, so we will continue to benefit from his expertise.

Joe has recruited and trained a very strong senior management team at AIGFP and I'm confident that AIGFP is in excellent hands pending the appointment of a new CEO.

Now I would like to turn to our investment portfolio.

As I mentioned earlier, the significant credit market disruption adversely affected that portfolio too.

In the fourth quarter, AIG recorded pre-tax charges of $3.3 billion for other than temporary declines in value, including $2.32 billion in pre-tax impairments related to AIG's investments in structured securities, including RMBS.

Most of these impairment charges resulted from the significant rapid declines in market values of certain residential mortgage-backed securities in the fourth quarter.

Even while retaining their investment-grade ratings, these securities were priced at a severe discount to book.

Despite our continuing intent and ability to hold them and despite structures that would indicate a substantial amount will continue to perform in accordance with their original terms, AIG concluded that it could not reasonably determine that the recovery period will be near term or the impairment temporary.

In the fourth quarter, we also recorded $2.54 billion after-tax or $3.8 billion pre-tax in unrealized appreciation of investments included in other comprehensive income.

Details are included in the investment presentation posted on our website.

With regard to our share repurchase program, during the fourth quarter, we repurchased over 21 million shares, bringing the total to 76.4 million purchased for the full year 2007.

An additional 12 million shares were purchased in 2008 through February 15.

Substantially all of the shares purchased in 2007 and in 2008 through February 15 were funded by high grade debt issuances.

We view this as an efficient use of capital as these transactions replaced high-cost common stock with more cost-efficient high grade securities.

AIG does not expect to purchase additional shares for the foreseeable future other than fulfilling commitments that existed at December 31.

We believe this is prudent in light of continued market volatility and credit market uncertainty, as well as our ongoing discussions with the rating agencies.

I know these issues are of great concern to all of AIG's stakeholders and is it is important that I emphasize again that we believe AIG can and will manage through them efficiently and effectively.

It is also important that we have not let these issues distract us from our focus on AIG's underlying businesses.

Before turning to what we are currently seeing in our specific businesses, let me just say that AIG is well-positioned to grow shareholder value despite the current turbulent environment.

We have a diverse portfolio of global businesses with the scale and world-class expertise that extends throughout the organization.

We also have a strong capital base and we are not raising additional capital.

We have accomplished a number of important objectives in 2007, including increasing distribution through acquisitions and new partnerships.

We acquired Wueba, a middle market commercial insurer in Germany and Matrix Direct, a direct marketing life insurance business in the US.

We also struck new distribution agreements with Japan Post and [Sinopak] in Taiwan.

In China, we secured a license for our Foreign General business to establish a wholly-owned foreign enterprise.

We also completed several new ventures, including establishing an asset management company in India and launching three new mutual funds there, breaking ground on the 5.4 million square feet international financial center in Seoul, obtaining permissions from the Qatar financial center to conduct individual and group life, as well as nonlife businesses and setting up a private pension fund administrator in Romania.

In 2007, we also made progress on streamlining our portfolio of businesses, including selling our stake in Allied World Assurance for an attractive return and completed the acquisition of the remainder of 21st Century, creating a personal lines insurer with approximately $5 billion of premium covering the most attractive segments and distribution channels.

While it is clear that certain of AIG's businesses did not perform up to expectations, others are.

I would now like to briefly touch on the performance of our major business segments.

DBG reported record results in 2007 with pre-tax operating income of $7.4 billion with a combined ratio of 85.52, a nearly 4.5 point improvement from the prior year.

Throughout the year, DBG posted improved underwriting results due to favorable loss trends and a higher net investment income.

Despite increased competition, DBG remained disciplined, taking advantage of opportunities offered by an extensive product set and expanding distribution platform in an innovative culture, responsible for an average of one new product or service launched per week in 2007.

These competitive advantages will enable DBG to identify attractive opportunities that meet our underwriting criteria in a softening property/casualty marketplace.

Issues related to the US residential mortgage market may affect our management and professional liability lines.

We have provided detailed information on potential exposures on page 12 of the financial supplement and needless to say, we are monitoring the issue closely.

At this point, it appears to be manageable.

Like DBG, Foreign General performed well, posting a combined ratio of 85.51 for the full year 2007.

Foreign General was also successful in expanding its personal lines presence in the high net worth segment with strong growth in United Kingdom.

In 2008 and beyond, Foreign General is particularly well-positioned for growth.

They continue to shift their personal lines product mix to higher-margin non-auto business and to expand our private client group operations into new growth markets such as Australia and Dubai.

We are the leading provider of financial lines in Continental Europe, a distinct advantage when serving the needs of top corporations while providing a platform to expand into other lines and market segments.

We will also continue expanding our operations in emerging markets.

In 2007, domestic personal lines reported a loss due to increased loss frequency, California wildfire losses and upfront costs related to the integration of 21st Century.

The 21st Century integration is progressing well and is on track to deliver significant cost savings, which we expect to begin emerging this year.

We have also begun marketing on a unified platform in the first quarter.

With regard to our life insurance and retirement service businesses, within the domestic life insurance operations, as previously discussed, we took aggressive steps to address investor-owned life insurance activity, which put pressure on sales in 2007.

In the fourth quarter, sales of Universal life picked up due to enhanced underwriting methods at the older ages and continued success from our innovative index Universal life product offering.

In addition, the domestic life team is committed to building a presence in the variable life product area with plans to expand distribution in this market segment.

With the addition of Matrix Direct, we are aggressively pursuing a direct response approach as part of our multi-distribution strategy.

We believe these and other initiatives to innovate product offerings and to expand distribution will improve sales in 2008.

AGLA, while not a growth-orientated company, has been able to improve the persistency and profitability of its business through a combination of better agent selection and training and a focus on expense efficiencies in distribution and the home office.

We did see continued strength in our payout annuity business, which is expected to continue into 2008 with favorable conditions in the terminal funding market and the opportunities to cross-sell structured settlement products with DBG, an example of our deliver the firm strategy.

In domestic retirement services, individual variable annuity fourth-quarter 2007 operating income was adversely affected by a change in actuarial estimates related to a system conversion and to a lesser extent a DAC unlocking.

However, this business reported record fee income and assets under management in 2007.

Sales growth should improve in 2008 due to AIG's SunAmerica's competitive live-in benefit features and initiatives to expand the wholesaler organization and wirehouse distribution.

Our fixed annuity businesses experienced continued outflows in 2007 as a large volume of five-year business came out of their surrender charge period.

Surrenders from business written during record sales periods are expected to continue at a higher level in 2008.

However, the current rate environment should provide for improved deposit and net flows through AIG Annuity's continued leadership in bank distribution.

AIG Retirement, also known as AIG VALIC, experienced double-digit deposit growth in 2007 and a steady increase in fee income and assets under management.

We expect that to continue in 2008 as independent distribution is being added to an already strong captive distribution force and our advisers continue to focus on the roll-over retirement market to retain assets.

AIG's foreign life insurance operations performed well overall.

Growth in first-year premium and single premiums were very strong, while bottom-line results were affected by actuarial changes in estimates, remediation activities and losses in the UK from variable annuities.

In Asia, we remain focused on meeting the growing demand for investment-orientated products through increasingly broad distribution channels, including bank assurance and independent producers.

Market conditions in Japan are expected to remain challenging.

However, on the strength of their multiproduct multidistribution platform, we expect to exceed industry growth in the long term.

The planned integration of AIG Star and AIG Edison, anticipated to be completed in 2009, will provide enhanced distribution opportunities and operating efficiencies.

As we progress in 2008, we expect to take advantage of the full deregulation of insurance product sales in banks and privatization of the Japan Post.

Central and eastern European operations performed well in 2007 and demographic and economic conditions in these countries provide excellent opportunities for growth in 2008 and beyond.

We are encouraged by the sales momentum and distribution expansion gained by foreign life in 2007.

In 2008, we expect to achieve good growth in foreign life.

However, further market dislocation will affect investment returns and diminish demand for certain products.

ILFC's excellent results reflect its status as the world's premier aircraft leasing franchise.

Despite the potential for an economic slowdown, the outlook for ILFC remains favorable as demand for its modern fuel-efficient fleet remains strong and all 73 of its 2008 new aircraft deliveries have been placed.

ILFC also concentrates on placing leases in strengthening regions such as Asia.

Asia and the Pacific region have grown to account for 27% of ILFC's revenues.

Our institutional and retail asset management business reported strong results this quarter as we continued to attract client assets and grow this business.

As the world's seventh largest institutional asset manager, total assets under management have grown to more than $750 billion and non-affiliated client assets under management have grown 26% in 2007 to approximately $94 billion, primarily as a result of our diverse robust product lineup across all asset classes.

While our institutional clients have always benefited from our strong local presence in developed and emerging markets, we have further enhanced our capabilities, serving the needs of local retail investors with funds launched in China, the Philippines and Taiwan.

With regard to our investments, domestic retirement services, as well as domestic general insurance and to a lesser extent, other lines of business will be affected by future returns from partnerships and other alternative investments.

As we have stated previously, particularly strong performance in recent periods is not indicative of future returns.

We expect the partnership asset classes to generate a 10% to 15% return over the long term, but in today's environment, even those returns could be challenging.

While 2007 was a very challenging year, we executed on several strategic initiatives and many of our businesses performed well.

The businesses that did not perform well to our expectations, we have concrete plans to address the issues and we expect improved performance in 2008.

As in the past two quarters, we are providing additional info informative disclosures.

This quarter, we are providing a presentation on AIG's businesses with exposure to the current credit market disruptions.

This presentation and additional supplemental materials are currently available on our website.

Now let me turn the call over to my colleagues.

First, Steve Bensinger.

Thank you very much, Martin and good morning.

Over the next few minutes, Win -- Win Neuger, Bob Lewis and I will update you on various developments and certain of our businesses during 2007.

First, I will discuss the super senior credit derivative business at AIGFP, including comments concerning the unrealized market valuation loss we have taken.

Second, Win will comment on selected components of AIG's insurance investment portfolios, concentrating on the US residential mortgage-related investments, commercial mortgage-backed securities, our exposures to monoline insurers and our exposures to CDOs and CLOs.

Finally, Bob will conclude with brief remarks on United Guaranty, our mortgage insurance business and American General Finance, our domestic consumer finance business.

We have posted two presentations to the Investor Relations section of our website.

During our discussion this morning, we will be referring to the one entitled Conference Call Credit Presentation.

The other one is entitled Conference Call Credit Presentation Supplemental Materials.

This presentation contains more detailed information regarding our businesses and portfolios for your information and review.

Now if you would turn to slide 4 in the Conference Call Credit Presentation, one of the businesses in which AIGFP is engaged is the business of writing credit derivatives to cover the risk of incurring realized credit losses on the most senior tranche in the capital structure of a securitization, referred to as the super senior risk layer.

The term super senior is used by market participants to denote very remote credit risk, but there is no single definition of the term.

Each transaction is bespoke, highly-customized and designed to withstand extreme stress and yet incur no expected loss.

Great care has been exercised in structuring the deals, starting with due diligence of the counterparty's motivation for entering into the transaction, positive selection of the assets, experience of the manager and definition of portfolio maintenance characteristics to name just a few.

Each underlying security is assigned an internal credit rating by AIGFP where possible based on the fundamental credit analysis and judgment of AIGFP's credit officers and on the ratings assigned to the collateral from the three rating agencies where available, along with a review of its current market spread.

AIGFP augments the super senior structuring process with its own actuarial models, using assumptions more conservative than those used by the rating agencies.

The minimum attachment point for the super senior portion of the portfolio is modeled as a minimum threshold above which there is no expected loss to AIGFP.

The final attachment point is negotiated to exceed the modeled attachment point and we will discuss typical transaction structures in a moment.

AIGFP writing this business in 1998 and to date, has not incurred a realized loss in any of its super senior transactions.

Now if you turn to slide 5, AIGFP has entered into super senior credit derivative transactions in three broad categories -- regulatory capital-motivated corporate and European residential mortgages, corporate arbitrage and multisector CDOs.

By net notional exposure, the largest category of the business is regulatory capital-motivated, almost $230 billion in corporate and $149.1 billion in European mortgages and is typically subject to both regulatory and contractual calls by the counterparties with the former starting in January 2008 when Basel II took effect in Europe.

These transactions were structured to provide the counterparty with capital relief in their regulatory jurisdiction and not structured to transfer significant credit risk.

Counterparties achieve lower capital charges by transferring a portion of their regulatory capital requirements to AIG.

As shown in the table, the expected maturity of the regulatory capital-motivated transactions, based only on the contractual call dates, is 1.2 and 2.3 years respectively, but many of these trades may be or are being terminated earlier than that as AIGFP's counterparties implement models compliant with the new Basel II accord.

As of February 26, 2008, $54 billion in notional exposures have either been terminated or are in the process of being terminated.

AIGFP's arbitrage-motivated corporate book represented $70.4 billion in notional exposure as of year-end 2007.

The underlying collateral in these deals is comprised of primarily investment-grade corporate debt and collateralized loan obligations.

AIGFP recorded an unrealized market valuation loss of $226 million or about 0.3% of the notional exposure in the fourth quarter as a result of general credit spread widening experienced in the market.

AIGFP's exposure to multisector CDOs in its super senior credit derivative portfolio, the third category of exposure, totaled $78.2 billion as of December 31, 2007, of which $61.4 billion had some level of exposure to subprime mortgages.

The attachment point is the most important feature of the risk mitigants built into these deals.

Of the deals with some subprime exposure, it ranges from an average of 15% on the high-grade deals to an average of 37% on the deals with mezzanine collateral.

At inception, the attachment points are always higher than the attachment points used by the rating agencies for the AAA rating equivalent.

Another important risk mitigant in the super senior structures of AIGFP is the priority of the cash flow waterfall to the super senior layer.

While we always sit at the top of the waterfall when it comes to being paid, this position is typically further enhanced by the existence of one or more overcollateralization or interest coverage tests that further direct available cash flows to amortize our position more rapidly if they are breached.

By way of illustration, and please turn to slide 6, you will see the makeup of a typical multisector CDO super senior transaction.

Various assets, such as residential and commercial mortgages, auto loans and student loans, are packaged together into asset-backed securities, RMBS, in the case of residential mortgages and CMBS in the case of commercial mortgages.

These securities are structured into tranches with various ratings from AAA down to an unrated equity tranche.

Typically 125 to 200 securities are purchased to form the collateral pool of a CDO.

So-called high-grade CDOs have high investment-grade collateral securities, typically rated AA or AAA at inception and so-called mezzanine CDOs consist of primarily low investment-grade with some subinvestment-grade securities in certain deals.

These portfolios of securities constitute the transaction gross notionals of the CDOs.

The CDO itself issues several tranches of debt securities from unrated equity to BB, BBB, A and finally junior AAA-rated, up to a more senior AAA-rated tranche called the super senior tranche.

Cash flow waterfalls dictate how principal and interest flows are allocated to the various tranches of the CDOs.

Various tests, including overcollateralization and interest coverage tests divert principal and/or interest flows to the more senior tranches of the CDOs.

AIGFP writes credit protection to the holders of the Super senior tranche of the CDO through the issuance of a credit derivative.

AIGFP's exposure to the CDO through the credit derivative is called the net notional exposure.

Now as we've stated in past communications, after thorough due diligence, analysis and modeling, AIGFP determines an acceptable attachment point or subordination level over which it is willing to write protection.

At inception AIGFP conducts extensive due diligence, including a thorough analysis of the historical and expected future performance of the collateral and the manager, each underlying obligor, asset servicer and originator.

Considerable effort is spent to avoid concentrations to single obligors, servicers and geographies.

AIGFP then utilizes proprietary data-driven actuarial models to analyze the fundamentals in the transactions.

The models are calibrated to be worse than the worst recession experienced post-World War II.

The models produce loss distributions by simulating the credit performance of the underlying obligations in the portfolio.

Also, if the CDO managers have latitude to substitute or exchange collateral in the pools, the transaction is modeled assuming the CDO managers select the worst possible portfolio within approved criteria.

The final results of AIGFP's due diligence in modeling are then used to negotiate attachment points for the super senior structure.

On slide 7, you can see that the AIGFP super senior business is subject to the oversight processes of AIG Enterprise Risk Management, which includes components of credit risk, market risk and operational risk management.

Super senior transactions entered into by AIGFP are subject to the approval of AIG's Chief Risk Officer and Chief Credit Officer under the delegations approved by AIG's credit risk committee.

Each transaction is independently subjected to an analytical review by credit risk management.

Every quarter, AIG ERM independently reviews the exposures, which are updated to show realized loss trends and current subordination levels.

In addition, AIGFP's actuarial models are rerun with updated credit ratings to show the model attachment points relative to the available subordination levels to confirm to what extent the transactions still constitute super senior risk of incurring any realized loss.

ERM identifies all transactions that show unexpected deterioration and will add these to AIG's internal watch list.

ERM also assesses whether any transactions could represent probable loss, thus potentially requiring the establishment of credit reserves of which there have been none to date.

It is also noteworthy at this point to mention that the credit processes at AIGFP, working in close collaboration with AIG ERM, began to see evidence that mortgage underwriting standards have declined and pulled back from this sector toward the end of 2005.

For that reason, the notional of 2006 and 2007 subprime collateral comprises a modest 4.9% of the total collateral pools underlying the entire portfolio of CDOs with credit protection at year-end 2007.

Furthermore, AIGFP's multisector CDOs have only a modest exposure to higher-risk, second-lien mortgages.

If you'd please turn to slide 8, recent credit quality deterioration in the subprime mortgage sector has led to market concerns about the credit quality of securities backed wholly or in part by subprime residential mortgages.

In order to cause an economic or realized loss to the super senior layer, mortgage delinquencies have to deteriorate to default losses and the default losses, in turn, must cause loss to the securities in the collateral pool underlying the CDO.

The securities losses have to accumulate to a level above the attachment point and reach the super senior layer after the waterfall diversions have been exhausted.

Contractually, the credit derivative protects the super senior CDO holder only for actual default losses in the underlying collateral; not the loss of market value of the collateral securities or the CDO itself.

To test the potential for unexpected realized losses to the super senior layers under prevailing conditions, both AIG's ERM and AIGFP have conducted risk analyses of the portfolio.

Moreover, AIG has conducted an analysis to assess the risk of incurring net realized losses over the remaining life of the portfolio.

In addition to analyses of each individual risk in the portfolio, AIG conducted certain ratings-based stress tests, which centered around further stressing of broad classes of the portfolio collateral from current rating levels.

The results of these stress tests indicated possible realized losses on a static basis since the assumptions in these stress tests assumed immediate realization of loss, but actual realized losses would only be experienced over time given the timing of losses incurred in the underlying portfolios and the timing of breaches in the subordination.

No benefit was taken in these stress tests for cash flow diversion features, recoveries upon default or other risk mitigant benefits.

Furthermore, the stress tests were applied at December 31, 2007 using the lowest ratings of the major rating agencies updated through January 3, 2008.

On this slide, we characterize the conditions of the severe stress scenarios and show the result in the graph.

Under this severe stress scenario, the realized loss would be approximately $900 million in contrast to the current unrealized market valuation loss of $11.25 billion.

Clearly, this market valuation estimate is biased by factors in addition to credit risk.

Although the super senior structures in AIGFP transactions only cover credit risk.

Based on these analyses and stress tests, AIG believes that any losses realized over time by AIGFP as a result of meeting its obligations under these derivatives will not be material to AIG's consolidated financial condition.

Although it is possible that such realized losses could be material to AIG's consolidated results of operations for any individual reporting period.

Please turn to slide 9, which indicates that AIGFP accounts for its super senior credit derivatives in accordance with FAS 133 and EITF 02-3.

We also considered the guidance in FAS 157 and the paper issued in October 2007 by the Center for Audit Quality entitled Measurements of Fair Value in Illiquid or Less Liquid Markets.

AIGFP does not recognize income and earnings at the inception of each transaction because the inputs to value these instruments are not derivable from observable market data.

Income is recognized over the life of the contract and as observable market data becomes available.

The fair valuation of AIGFP's super senior credit derivative portfolio is challenging given the bespoke nature of each transaction and the lack of market observable transactions or information.

In the absence of any observable market, in accordance with GAAP, AIGFP must estimate fair value using the assistance of models.

In estimating fair value under GAAP, AIGFP uses a combination of valuation models, principally the binomial expansion technique or BET, third-party prices, relevant market indices and where necessary, management's own judgment.

Through June 30, 2007, AIGFP concluded that there was minimal change in fair value since the inception of the derivatives as the super senior credit derivative were in essence put options significantly out of the money that are insensitive to normal changes in market credit spreads.

Now the table on slide 10 breaks down the notional amount of AIGFP's exposure by super senior, underlying type and their respective cumulative mark-to-market losses experienced in the third and fourth quarter.

For the $70.4 billion notional exposure for corporate arbitrage transactions, AIGFP valued these transactions using relevant market indices or third-party prices and reported an unrealized market valuation loss in the amount of $226 million for the fourth quarter.

At September 30, 2007, AIGFP employed the binomial expansion model to value this portfolio and it resulted in no noticeable change in fair value.

For regulatory capital trades, both corporate and European residential mortgage-related and please turn to slide 11 where you see it explains that these transactions were concluded with counterparties primarily to facilitate regulatory capital relief for them rather than to transfer credit risk to AIGFP.

As I said earlier, these transactions are expected to terminate within the next 12 to 18 months as the counterparties implement the new capital provisions under Basel II.

AIG conducted a comprehensive analysis of available information at year-end 2007, including the counterparties' motivation and behavior, the portfolio's performance, marketplace indicators and transaction-specific considerations.

As a result of this analysis, AIG believes that these regulatory-driven trades are appropriately valued at zero fair value as of December 31, 2007.

The most compelling market observable data to support this conclusion is the fact that $54 billion notional of transactions have been terminated in early 2008 without AIG being required to make any payments and in some cases, with AIG being paid a fee.

As can be recognized from the information on slide 12, the fair valuation of the super senior credit derivatives has become increasingly challenging given the limited availability of market observable information due to the lack of trading and price transparency in the structured finance market, particularly in the fourth quarter of 2007.

These market conditions have increased reliance on management estimates and judgments in arriving at an estimate of fair value for financial reporting purposes.

Furthermore, disparities in the valuation methodologies employed, the degree to which market data may be available to a market participant and the varying judgments reached by such participants when assessing volatile markets has increased the likelihood that the various parties to these instruments may arise at significantly different estimates of their fair values.

AIGFP's valuation methodologies and processes for the super senior credit derivative portfolio have evolved in response to the deteriorating market conditions and the lack of sufficient market observable information.

In the third quarter, as credit spreads started to widen considerably, AIGFP implemented a modified binomial expansion technique, the BET model, using credit spread inputs on generic ABS obtained from a third-party source and other inputs, like Moody's historical recovery rates.

The BET model also utilized diversity scores, weighted average lives and discount rates.

The model accounts for the specific features of each transaction such as portfolio amortization and tranche subordination.

Our valuations indicated an unrealized loss of $352 million at September 30, 2007.

As the market continued to deteriorate in October, AIGFP continued to refine its model.

It enhanced its existing data inputs by adjusting our RMBS and CDO credit spreads for the relative change in the ABX home equity index.

At November 30, AIGFP ran two versions of the BET model that had been used for the October estimate of the portfolio mark.

Method A was similar to the October version of the BET model, but in incorporated the net benefit of structural risk mitigants, principally the cash flow diversion benefits, that resulted in a reduction of the net unrealized market valuation loss.

Method B incorporated two new features.

First, during the month of November, AIGFP obtained third-party prices of the underlying collateral securities collected by CDO managers as of October 31, 2007.

These prices were used as inputs to the modified BET model to derive credit spreads more closely associated with the underlying collateral securities than the general spreads used in method A.

Second, method B incorporated a negative basis adjustment to reflect the fact that cash and synthetic instruments frequently trade at different levels with cash instruments normally at a wider spread than CDS or high-quality assets.

Under method B, the BET model is effectively a valuation model that estimates the fair value of the CDO, which AIGFP's credit derivative protects.

Using cash spreads as inputs to the model in substance assumes that the entire spread is comprised of only one risk, credit risk.

However, the spread achieved from holding a cash instrument incorporates not only credit risk, but also other risks such as funding cost differentials, liquidity risk and risk aversion costs.

In times of market stress, these and other risks tend to widen together with credit spread widening.

The wider the spread becomes from these other risks, the wider the negative basis.

You can think of negative spread as the difference between total spread and the credit spread.

When AIGFP entered into super senior CDS transactions on multisector CDOs, there was an observable negative basis proven by the fact that the major motivation of the deals was to earn a net positive spread between the spread earned by the investor of the CDO and the premium cost the investor paid to AIGFP for its credit derivative hedge.

Under method B, AIGFP estimated the benefit of this negative basis based upon best estimate assumptions provided by major market participants.

Under GAAP, sufficient observable evidence supporting the existence and quantification of negative basis for AIGFP's transactions is needed to take into account any negative basis in its fair value determination.

Currently, despite the fact that the ABS market is quite illiquid, entering into new transactions that demonstrate the existence of the negative basis is possible.

However, evidence of negative basis in AIGFP's existing transactions is currently unobservable in these market conditions.

Upon completion of its review of the evidence available, AIG concluded that recording a negative basis adjustment at this time is not consistent with GAAP fair value requirements.

Now as described on slide 13, AIGFP has filed a rigorous process to determine its best estimate of fair value for AIGFP's super senior credit derivatives on multisector CDOs at December 31, 2007.

This process is required because there are no observable market prices for the credit derivatives AIGFP has written.

Therefore, AIGFP utilizes a model to determine fair value similar to method B in the previous slide.

The method maximizes the use of third-party market observable inputs where possible.

There are five key components to the process.

First, AIGFP was able to acquire third-party prices on about 70% of the underlying collateral securities in the multisector CDOs.

In cases where AIGFP received multiple quotes, an averaging of prices was used.

Prices were reviewed for consistency across ratings and time.

Matrix pricing was used where third-party prices were unavailable.

Second, AIGFP benchmarked these third-party prices to independent pricing services and sources.

Third, other key inputs were obtained for the modified BET model.

For example, weighted average life of securities, diversity scores, discount curves and Moody's recovery rates.

Next, AIGFP performed a valuation review and stress testing of the modified BET results.

Finally, AIGFP obtained a number of super senior bond tranche quotes for the underlying CDOs or implied them from collateral calls from 12 major dealers to adjust the BET results where appropriate to make a best estimate of the exit value of the transactions.

No negative basis adjustment was utilized for the reasons I explained earlier.

In effect, the $11.25 billion unrealized market valuation loss we have recorded intrinsically assumes that we own the cash CDOs.

No credit is given to the synthetic nature of our obligations since the credit component of the spread widening is not presently observable.

In summary, AIG believes that its $11.25 billion best estimate of the unrealized market valuation loss represents fair value under GAAP.

AIG also believes that the results of its fundamental credit analysis and stress testing provide confidence that any realized losses in the portfolio, as I said, will be materially below the GAAP fair value estimates.

Let's turn to slide 14.

I will talk a little bit about economic capital.

In determining the available economic capital at AIG, so far in our development of the methodology, we have used GAAP capital as a proxy for available economic capital.

We do not feel it is appropriate to deduct the after-tax unrealized market valuation loss from available economic capital since if we did so, it would assume that the appropriate economic capital required for these obligations would be based upon the current distressed and illiquid capital markets.

Rather, AIG intends to hold what we believe are good risks on our books until the transactions mature.

Economically, we would assume the severe stress loss described earlier, plus the cost of capital charge, would constitute the current market consistent settlement value of the loss.

Slide 14 explains the net after-tax adjustment to GAAP we would apply to determine AIG's available economic capital as of December 31, 2007.

You can refer to the economic capital update memo we posted to our website for a thorough explanation of this process.

Turning to slide 15, in accordance with GAAP, AIG recognized the sizable unrealized market valuation loss in 2007 consequent to the severe market disruption and credit deterioration, particularly of subprime mortgage-backed collateral.

This market valuation loss represents management's best estimate of the exit value of this portfolio into the current illiquid and distressed market.

However, AIGFP underwrote its super senior credit derivative business to a zero loss standard incorporating conservative stress scenarios at inception.

Although there is likely to be continued volatility and perhaps further deterioration in the credit markets based upon AIG's analyses and stress tests, AIG does believe that any credit impairment losses realized over time by AIGFP will not be material to AIG's consolidated financial position nor to its excess economic capital position.

Although, as I stated, they could be material to an individual reporting period.

Finally, on slide 16, AIG recognizes that continued improvement in its internal controls is necessary and remediation of the identified material weakness will be a very high priority in 2008.

Over time, AIG intends to reduce its reliance on certain manual controls that have been established and to migrate models that have been developed in response to market events to a more robust production environment.

AIG is also currently developing new systems and processes, which will allow it to rely on front-end preventive and detective controls that will be more sustainable over the long term.

AIG is committed to making the significant investment necessary to make these improvements.

Now I will turn it over to Win to discuss our investment portfolios.

Thanks, Steve.

I will provide you with an update on AIG insurance investment's exposure to residential mortgages, commercial mortgage-backed securities, monoline insurers and collateralized debt obligations.

We posted a more detailed presentation on the website as Steve mentioned.

Here, also, we incurred significant negative realized and unrealized accounting losses during the fourth quarter.

The overall market dislocation had an impact on investment valuations, which resulted in realized capital losses in 2007 of $3.6 billion.

We also had reported an unrealized depreciation on investments of $8 billion, reducing accumulated other comprehensive income to a still positive $4.4 billion net of tax.

On slide 19, the components of the 2007 losses and declines are detailed.

Although all risk assets were impacted, the RMBS securities were most affected as shown in the far-right column.

I also want to point out that these combined realized and unrealized losses are approximately 2% of the total insurance investment portfolio and not all of the losses on the left-hand column are even related to these portfolios and non-RMBS losses amount to 1% of non-RMBS insurance investment portfolios.

Beginning with realized losses, more than 100% of the net number is accounted for by the $4.1 billion other than temporary impairment.

Of that OTTI loss, approximately $3.1 billion is expected to recover in value.

The $3.1 billion amount that we expect to recover is associated with either our inability to reasonably assert that certain severe declines in valuations are temporary, our lack of intent to hold to recovery or an adverse change in the timing, but not the amount of cash flows in structured securities.

Another $500 million is associated with other than temporary changes in currency valuations and that leaves the remaining approximately $500 million, which is associated with credit and structured securities impairments without expectation, of full recovery of principle.

Please note that actual transactions during the year generated net gains of $1.2 billion and only $30 million of losses were taken on sales of RMBS.

Moving to the $8 billion of unrealized depreciation for the year, $5.1 billion or 63% is associated with RMBS, mostly rated AAA and AA.

In addition, we have also seen declines in valuations of monoline insurers, commercial mortgage-backed securities and CDOs.

We continue to believe that these market value declines are driven predominantly by market conditions, not by significant changes to the risk of principal repayment.

I will now move to talk in more detail about the key segments of the RMBS portfolio and explain our current position and why we believe that we will be mostly repaid.

I am going to skip from slide 20 to 21 and talk about the breakdown of the $89.9 billion of RMBS exposure.

Our subprime RMBS holdings totaled $24.1 billion, down from $28.6 billion at the end of June and $25.9 billion at the end of September.

Alt-A RMBS holdings totaled $25.3 billion, also down from earlier periods.

Slide 22 displays further details of the RMBS portfolio by rating for each type of mortgage-backed security.

The slide also underscores the point that the overall RMBS portfolio is high quality with 91% rated AAA [or agency] and 7% rated AA.

On slides 23 and 24 are details of the subprime and Alt-A portfolios by vintage year, rating and weighted average expected life.

Slide 23 shows the details of AIG investment's $24.1 billion of subprime holdings.

87% of subprime is rated AAA and another 12% is rated AA.

Slide 24 provides details of our Alt-A holdings, totally $25.3 billion, of which 95% are rated AAA and 4% are rated AA.

An important component in understanding the risk in our RMBS holdings is the degree of credit enhancement.

On slide 25, we have presented our original and current average credit enhancements for the subprime 2006 vintage, the largest vintage year among our portfolio subprime holdings.

Although the market's loss expectations have increased, our portfolio has delevered significantly with an average credit enhancement of 29.6% for our AAA holdings and 21.5% for all holdings below AAA.

While subprime 2006 vintage loss estimates have risen into the high teens or even low 20%s, the combination of excess spread and current credit enhancement is providing the intended cushion.

Slide 26 shows comparable information for the Alt-A portfolio.

As shown on slide 27, the major rating agencies have been downgrading and placing many securities on negative watch during the past six months.

About $443 million of AIG's RMBS securities were downgraded in the fourth quarter of 2007 and an additional $3.6 billion were downgraded through February 25.

$9.7 billion of the portfolio is on negative watch as of that date.

Of these downgraded securities, $2.7 billion were subprime exposures and represented 11% of the total subprime holdings.

Taking into account all rating actions through yesterday, 93% of our subprime RMBS exposure is still rated AAA and AA.

It is important to note also that rating agency downgrades do not affect the structural priority of payments, so the securities originally rated AAA retain their payment priorities, including trigger or [turborights] irrespective of their current ratings.

Turning to commercial mortgage-backed securities or CMBS on slide 28, our holdings are $23.9 billion with close to 89% rated AAA or AA and only 0.2% below investment grade.

Moreover, $21 billion are traditional CMBS securities.

Of the 9% of the CMBS exposure that comes from resecuritizations of CMBS and commercial real estate or CRE CDOs, two-thirds of the loans underlying these securities are seasoned 25 months or more.

Slide 29 presents the CMBS portfolio's current delinquency profile compared to the market experience, showing positive comparisons with only 23 basis points of delinquencies on the portfolio.

Next, slide 30 shows our CDO portfolio -- shows that our CDO portfolio amounts to $4.6 billion, most of which are CDOs of corporate bank loans.

In general, the pricing of this portfolio has been adversely affected by current market conditions.

However, after fourth quarter write-downs, the portfolio has only $58 million of remaining ABS CDOs with some subprime exposure.

And only 1.6% of the holdings in the total portfolio were downgraded in 2007.

Based on our current analysis, the ultimate loss of principal in the CDO holding is expected to be minimal.

Turning to AIG's exposure to the monoline insurers on slide 31, the investment portfolios have $42.2 billion of exposure, of which more than 99% are financial guarantees from the monoline companies.

$31.4 billion of the monoline exposures supports the municipal bond portfolio, which has a high underlying credit quality of AA.

Slide 32, which shows -- slide 32 rather shows that 84% of the monoline exposure has an underlying rating of A or better.

In other words, the bonds that are being insured are rated A or better without the guarantee.

Financial guarantees are viewed by AIG purely as a secondary source of payment for all wrapped investments.

On slide 33 is the breakdown of exposure to each monoline insurer.

You'll see that 96% of the monoline exposure is to MBIA, FSA, MBAC and FGIC.

Slide 34 summarizes the characteristics of our RMBS, CMBS and ABS that are wrapped by the monolines.

The RMBS, CMBS positions with internal ratings below investment grade represent primarily second-lien and HELOC home equity loan pools that have experienced worse than anticipated performance.

Currently, there are 10 RMBS second-lien and home equity transactions totaling $380 million or less than 1% of AIG's total insured portfolio that are known to be receiving contractual payments through their financial guarantees.

Concluding with slide 35, since August 2007, the broader capital markets have emphasized liquidity and aversion to risk.

The US residential mortgage market has continued to deteriorate with limited financing opportunities for mortgage borrowers and substantial increases in lifetime loss expectations on 2006 and 2007 US mortgages.

This deterioration has increased our mark-to-market and downgrade risk.

However, our preference for RMBS exposures high in the capital structure continues to guide our current expectation that the risk of an ultimate loss to investment principal in these securities remains low.

We have also focused on monoline CMBS and CDOs in this presentation.

All to varying degrees have been focal points of market volatility in the past six months.

So our CMBS holdings have suffered from volatile market pricing, underlying fundamentals remain strong with very low delinquencies.

Furthermore, our CDO holdings have suffered minimal downgrades in 2007 and have little exposure to the current struggling subprime CDO market.

Because AIG focuses on fundamental credit analysis, our holdings did not rely on financial guarantees from monoline insurers as a primary source of repayment at the time of acquisition.

In this environment, in our insurance portfolios, we have been opportunistically increasing liquidity, further downgrading migrations seem fairly likely, but it is important to note that rating changes do not affect the structural protection and thus, we expect substantially full recovery of principal and interest for most of our investment holdings.

Furthermore, we are looking for opportunities to take advantage here as the markets create what are truly compelling opportunities in the market.

With that, I will turn it to Bob.

So moving on briefly to AIG's mortgage insurance subsidiary, United Guaranty in slide 37, UGC has been providing lenders mortgage insurance on first and second-lien mortgages since 1963.

As a broad market participant in a cyclical business, UGC's performance is highly correlated to the fortunes of the housing market.

The current housing downturn has affected performance and asset quality, but UGC continues to outperform the industry on average.

Several underwriting and eligibility adjustments being implemented by UGC and their lender customers are improving the quality of new business production.

The composition of UGC's portfolio has not changed significantly as shown on slide 38.

Loans to borrowers with FICO scores less than 620 constitute about 8.4% of UGC's domestic mortgage risk in force and 70% of their net risk in force has FICO scores greater than 660.

Furthermore, high-risk products such as interest-only and option ARMs remain less than 10% of the risk in force.

The 2007 vintage exposure was driven by higher utilization of mortgage insurance in the overall mortgage market.

However, UGC's marketshare of its traditional first-lien flow business did decline by 1.2 percentage points quarter-over-quarter and 2.3 percentage points year-over-year.

Turning to slide 39, the deterioration of the US housing market has affected all segments of the mortgage business, but the high LTV second-lien product is particularly sensitive and accounts for 51% of UGC's 2007 domestic mortgage net losses incurred.

Due to the accelerated claims cycle of second-lien mortgages, these net losses incurred should work through the portfolio much faster and peaked in 2007.

First-lien net losses incurred, however, are starting to have a significant effect on operating results and some further deterioration is expected in 2008.

Nevertheless, as of December 31, 2007, expected losses are significantly below the net risk in force.

Moreover, future premiums are projected to exceed future losses on the existing domestic mortgage portfolio despite an inherent mismatch in the timing of premium earnings and incurred losses.

UGC's decision to be only a minor participant in the higher risk bulk channel, which is predominately subprime, Alt-A and option ARM business, is the primary driver behind UGC's better delinquency performance on first-lien mortgages than the rest of the mortgage insurance industry as depicted by the graph on slide 40.

The graph is narrower -- the gap is narrower than historical levels due to a rapid deterioration in the overall market.

However, in January, the gap widened to 73 basis points from the 64 basis points in December.

As slides 42 and 41 describe, UGC continues to implement key risk initiatives to improve the quality of new business production in both the first and second-lien businesses, including tightening eligibility guidelines and increasing rates in select high-risk business segments.

There are a number of risk mitigating factors described on 43.

In light of the cyclical nature of the mortgage insurance business, UGC employs risk-sharing arrangements through captive reinsurance with most of its major lender customers.

It also purchases quota share reinsurance on portions of its subprime first-lien business and segments of its second-lien product.

UGC maintains an important exclusion for fraud on both its first and second-lien business and it has a geographically diverse book and continues to focus on insuring single-family owner-occupied residences.

In summary on slide 44, UGC is engaged in a highly cyclical business and downturns in the housing industry have negatively affected short-term results.

UGC continues to implement key risk initiatives to improve the quality of new business production.

However, UGC expects that the downward market cycle will continue to adversely affect its operating results for the foreseeable future and is likely to result in another significant operating loss in 2008.

Turning to slide 46, American General Finance is AIG's domestic consumer finance business.

AGF is a portfolio-based lender whose products include real estate, nonreal estate and retail sales financed loans.

It originates real estate loans through its 1600 branches.

AGF also originates and acquires loans through a centralized real estate operation.

Slide 47 depicts the slowdown in AGF's real estate production beginning in the third quarter of 2005.

As we have mentioned in previous calls, AGF did not relax underwriting standards when the market overheated in late '05 and '06.

Rather, they anticipated many of the real estate market issues and sacrificed growth for long-term credit quality and earnings stability.

The results of AGF's underwriting discipline has resulted in a mortgage portfolio with credit quality far superior to that of the subprime asset-backed security real estate market as shown on slide 48.

On slide 49, AGF's real estate 60 plus delinquency rate was 2.64% and its real estate net charge-off ratio was 0.47% at year-end '07.

The current housing downturn has resulted in delinquencies in losses that have risen from recent all-time lows and credit quality remains subject to future macroeconomic conditions.

Nevertheless, AGF's credit quality remains below its target ranges, which were set by AIG and AGF management years ago to denote sound credit quality parameters.

Slide 50 highlights key risk mitigants in the mortgage portfolio.

AGF has required full income verification on almost its entire real estate book.

In addition, 88% of the portfolios are fixed-rate mortgages, which have much lower delinquencies than losses than adjustable risk products.

A limited ARM customer base is qualified on a fully indexed and fully amortizing basis at origination and only 9% of the overall real estate loan portfolio is due to reset interest rates by year-end 2008.

Furthermore, AGF did not delegate underwriting on purchased loans and has not made option ARMs.

Essentially all the real estate loans are first mortgages and owner-occupied.

AGF recently announced the pending acquisition of $1.5 billion in outstanding balances of branch-based consumer loans of Equity One Inc., representing approximately 130,000 accounts.

As described on slide 51, this transaction is expected to close during the first quarter of '08.

AGF believes this is an excellent opportunity to augment organic growth with a large portfolio of first and second fixed-rate mortgages, consumer loans and retail receivables that is similar to AGF's customer profile and credit quality performance.

In summary on slide 52, at the end of the fourth quarter, the real estate portfolio remained at $19.5 billion, flat to the end of the third quarter.

AGF believes that the housing market will likely continue to deteriorate for the remainder of '08, but the Company's business model and underwriting approach are sound and will allow the Company to continue to pursue opportunities as they arise.

And finally, Martin.

Thank you, Bob.

I want to conclude by reinforcing that AIG remains a great company with unmatched competitive advantages, strong brand recognition and a unique global franchise.

We believe AIG is extremely well-positioned for the upcoming market opportunities.

We know that this has been a very long presentation, but there has been much to cover and I hope our disclosure has been helpful for you to understand these issues in more detail and in more depth.

We appreciate your patience and now we will be more than happy to take your questions.

(OPERATOR INSTRUCTIONS).

Jimmy Bhullar, JPMorgan.

Hi, good morning.

I just have a few questions.

The first one is for Edmund if he is there.

You mentioned that there is a chance that you have to contribute additional capital to your Taiwan business because of higher capital requirements.

If you could quantify that on what that could be.

And secondly, on your $14.5 billion to $19.5 billion estimate for excess capital, what is the potential that this number could decline when you revise your excess capital estimate once you use the updated 12/31 inputs?

I think it has been made that you are publishing the revised number?

And the final question is on your economic loss language for the capital markets business.

If you can discuss the rationale behind the change in language, how much of it is driven by a deterioration in the market, how much of it is just conservatism on your part and then just generally talk about the factors that could affect your realized loss estimate of $900 million, (inaudible) revised higher or lower over time?

Hi, Jimmy.

Edmund here.

I am here with the group.

I can answer your first question relating to the capital requirement in Taiwan, namely Nan Shan.

In fact, we have a very strong underlying performance in our life company in Taiwan.

Nan Shan is making operating income of $900 million for the year, but there is some new capital requirement that is regulated by the insurance commissioner and there is a capital charge to certain Taiwan investments and as a result, our RBC, risk-based capital, may be in case of investment performance below the requirement of 200%.

Right now, we don't have a problem.

We just plan ahead in case we are below that.

Then we may have a problem to increase our offshore investments as only if our RBC is above 200%.

They will approve a further increase in offshore investment from the 35% to 40%.

But based on the current situation, we do not need an injection of capital to support our local operation as it is very profitable and the financial is very strong.

We are thinking of different ways to see if we could further improve our RBC, including possible reinsurance and/or realize some capital gains from certain investment assets and so forth, but the last resort will be to probably thinking of is it preferred shares to enhance our further capital base in Taiwan, but that is really for a precaution.

Jimmy, it's Steve.

With regard to your second question on excess capital, the $14.5 billion to $19.5 billion range that we just published is based upon the roll-forward of our current modeling as I talked about earlier.

To what extent that might change when we do the full review on the December 31, 2007 numbers, it is very difficult for me right now to project that.

If we look at the past circumstances, the updates have been pretty consistent with the roll-forward, but it is really impossible for me to give you any real level of confidence on that.

Although I would say we feel pretty good about how we have derived these numbers at this point in time.

With regard to your third question on economic capital loss estimates -- I am sorry -- the loss estimates on the stress studies, I will turn it over to Bob Lewis.

Yes, Jimmy, you asked, given our conservative estimate of excess capital at the end of '07 of between $14.5 billion and $19.5 billion, you said when we actually run the numbers and validate whether you think that there would be a substantial change in that number.

No, we do not, but the initial assessment or estimate is based upon a roll-forward of certain assets and liabilities, which we, of course, need some time to fully model.

So there is no expectation that there would be a major adjustment.

That range does already take into account the results -- the financial results that we have reported.

Regarding your third question on the stresses, regarding -- on that, as far as the effect of further rating changes, I think you can see from the stress description on page 8 that we have taken into account, one, the current ratings that exist through January in our current estimate and then also we have stressed those ratings substantially from here and so that I think that stress that we provide there in detail shows a significant change of a potential quality in the portfolios that is really driven from the current analysis of expected delinquencies and loss development, as well as recovery rates or housing price appreciation.

So by accessing our own views and also those of third-party analysts, particularly the rating agencies who have used that in determining the stress tests that we have shown there.

Regarding how that estimate could move and what could move it, obviously no one knows exactly how deep and how long-lived this downturn will be.

However, given the assumptions of delinquencies and losses and housing price appreciations underneath that, we are quite comfortable that that stress really is a severe stress to our business and that that number is right now our current best estimate of a real severe stress loss.

Okay.

And are the $6.2 billion credits that you are taking or the benefit in your excess capital estimate, is that something that the rating agencies are okay with or is that -- are the rating agencies okay with that treatment?

Well, Jimmy, I think we have been pretty clear as we have discussed this over the last year that the rating agencies right now are not as far along as we are in terms of our own economic capital modeling, so the rating agencies are using their own testing.

Now I think you have seen from some of the publications over the past few weeks that a number of the rating agencies are doing their own modeling of our underlying super senior portfolios.

Some of them gave some initial views a couple of weeks ago when they changed our outlook or put the Company on rating watch or review.

So they are not -- they are not using the same methodologies we are, but we expect that as we continue to converse with them over the next few weeks and the immediate future that we will be showing them the rationale for our own modeling and comparing that with them to their own and providing them with the information they need to come to their own conclusions.

Okay.

Thank you.

Ron Bobman, Capital Returns.

Hi, good morning.

It would seem to me given the circumstances that the greatest area in your control to create value and make money is in getting better rates in the property and casualty area and I wanted to know what you are doing on that end.

Thank you.

Thanks, Ron.

Well, obviously during the fourth quarter and year-end, we continue to see rating pressures in the P&C environment and domestically, in the United States, overall, I think rates were down around 9% and I think internationally in our commercial business, rates were down about 11%.

The good news is that, so far again generally speaking, terms and conditions, i.e.

policy wordings and deductibles, holding reasonably well.

There are always, of course, examples where that is challenged, but both Kris Moor and Nick Walsh are here to give a little bit more color maybe by line where we are seeing continued rate pressures.

One area I am particularly concerned about is the aviation sector, which continues to see significant rating pressures over many years, but I will let Chris and Nick add a few words.

Yes, Ron, this is Kris.

Overall, there was 9%, but excluding the property and workers' comp for the fourth quarter, they were down 10% and for a whole year, about down 8%.

Property rates declined in the fourth quarter about 13%.

We are seeing similar in the first quarter so far.

Overall rates are down about 10% and property rates are down about 12%.

But with that said, we still feel that in almost all our lines of business except for a few that rates are plenty adequate and that there's a lot of opportunity out there for us.

The areas that we are watching very closely or looking at closely is the aviation area and also in some of the states for workers' comp, but besides that, we still see a lot of opportunity out there.

Ron, from a Foreign Gen standpoint, the comments on pricing are pretty much the same, but we have some terrific opportunities across the world.

We have excellent growth in Latin America where there are some pricing opportunities because of reducing local capacity.

We are doing extremely well in Continental Europe where we are primarily under scale, so there is lots of upside for us there and we are very excited about the Middle East and some of the emerging markets.

So it may be tough in some of the more conventional markets, but there is still lots of opportunities for Foreign Gen.

I appreciate the commentary and the discussion of opportunities.

I am surprised I don't hear any talk about rating action that you are taking as a leader in large swaths of groups or lines of business in P&C basically taking a leadership position to hold rates or take rates up.

Ron, this is Kris again.

It's a good comment.

And in areas -- there are always areas that have exposures that you feel -- that you are coming close or you are at the point where the rates have to change and obviously in today's environment, financial institutions is an area that rates we are pushing and leading the market and there are one or two others that are with us on that and increasing the rates in that area.

And in some other productlines, and if you go down very specifically, you will see spots where we are increasing rates.

Then in other areas, we see opportunities where the rates don't need to be increased.

So we look at that and we do lead the market in that area.

Andrew Kligerman, UBS.

Good morning.

A couple of questions more conceptually.

What was it that led to the conclusion by PriceWaterhouseCoopers that you have a material weakness?

Where did that -- when did that thinking come into play and with that, and correct me if I am wrong, I don't focus on the monoline insurers, but don't they use a fair amount of manual marks in their accounting, so that would be question one?

Part two is -- I know you have just touched a bit on your economic capital and the $14.5 billion to $19 billion I think it is and it is not clear what the rating agencies are thinking.

Is there a risk that they are going to ask you for more capital and maybe you could -- I assume your comments earlier indicated no, but could you give us a sense of what they are thinking on that front?

And then in terms of the Japanese market, it looked like there were a few pressure points.

You had -- let's see.

Japan's sales were down in the fixed annuity area about 10%.

You saw some weakness in the group premiums, about 9% and the personal accident area actually showed an 8% increase, also weakness in variable annuity sales though.

So maybe if you could -- and I don't mean this to be a long one, but maybe just a quick collar around each of these products real quick in Japan and what the environment was like?

Thank you, Andrew.

Well, I will take question number one.

I am sure Steve will chart tackle question two and Bob Clyde is on the line from Japan, so I will defer number three to him.

He can give you a lot more color.

Just on the first part of your first question, obviously given the extreme dislocation in the markets affecting obviously AIGFP's super senior credit default swaps, we were working during the fourth quarter to implement system and controls to accurately report our financial results.

While obviously we didn't succeed, obviously because we have the material weakness that evolved during the closed process in implementing and overseeing the sufficient valuation process before the end of the fiscal year, we obviously have addressed those issues in preparing our year-end numbers by obviously inserting compensating controls.

The good news and of course, in the bottom line is that we have a clean audit opinion and that the accountants have signed off on the numbers and as Steve mentioned earlier, we are obviously working very hard to remediate as soon as possible the material weakness.

Obviously, on the second part of your question, I obviously can't comment on the other companies, particularly the monolines as you suggest and how they have formulated their numbers.

Yes, then Martin, maybe just -- you just seemed so confident at that December Investor Day and it just makes me wonder what was PriceWaterhouseCoopers thinking at the time to let you go into that Investor Day and be so confident?

Maybe this stuff could have been corrected in time before the material weakness and actually on that note, when do you think it is possible that could get lifted?

Well, on the first part of your question there, Andrew, obviously those numbers that were presented on December 5 were unaudited.

And secondly, as I have indicated, we are going to work very diligently to remediate the material weakness as quickly as we possibly can and we will be working very hard to do that through the balance of 2008 and as soon as we are fully remediated, obviously we will advise you as soon as possible.

Okay, and then -- okay.

Andrew, on the question on the rating agencies, obviously I can't speak for them.

However, I think if you read the publications that all four of the major rating agencies made a couple of weeks ago after we filed the 8-K, I don't think that you will see any indication that they feel that we have capital issues discussed and that has not been the subject of any discussion with them so far.

Now again, I can't speak for them or for what their conclusions might be once they digest our year-end numbers, but so far that has not been a discussion topic.

Steve, spreads have widened a fair amount in January and February.

If you had to make the marks today, would you be looking at something similar again for the first quarter and then would the agencies get concerned?

Well, Andrew, I think a lot of what we have discussed with the rating agencies and what they have written has not only been focusing on the fair value marks for GAAP, but they are also looking at the economic stress test analysis and the underlying strength of the portfolio.

If you go back to the presentation that we discussed a few minutes ago, these super senior structures were designed to withstand very severe economic stress.

That is how they were underwritten and so far from an economic standpoint, they certainly seem to be performing as they were designed.

So I think the substance of that is also going to be a significant factor in the actions that the rating agencies take based upon the conversations we have had with them and also what they have actually already written.

So I think we just have to see how it evolves and again, I really can't speak for them other than to point you to what they have been saying publicly.

Steve, do you think the mark will be similar in the first quarter to what we have just seen in the fourth quarter?

Andrew, we really don't know that at this point in time.

I mean there has been more deterioration in the underlying market as a whole.

There has been, as you have stated, further spread widening.

What that precisely implies to our overall portfolio, we just don't know at this point in time.

The processes that we now employ that I went through with you are highly complex, they involve a great deal of market information that takes time to obtain.

We will have our best estimate mark when we published our first-quarter financials, but right now, I think it would be just too difficult to really give any kind of an answer to that precisely.

Okay.

Thanks, Andrew.

Bob, if I can ask you to respond to the third part of Andrew's question.

Yes, hi, Andrew.

Let me start with your questions about annuities.

The sales of fixed annuities were 11% below the fourth quarter of the prior year, but it improved by 6% versus the third quarter, which marked the third straight quarter of increasing sales due largely to the strength in the yen, the steepening of the yield curve and the recent equity market volatility.

And also fixed annuity sales for ALICO's face-to-face channel has shown some nice gains too with the new knockout [rider] option that we offer.

Now we initially introduced that in the face-to-face channel.

We also introduced in the bank assurance channel and fourth quarter and thus far, three banks have taken it up and we are on the process of launching a number of others.

With respect to variable annuities, we were 5% above fourth quarter, but we declined 22% versus the third quarter due to the turbulent equity markets.

And of course, the risk categorization applied to these products.

Because of the new financial instruments and exchange law, that has created a bit of a current going in the other direction, but the new lifetime guarantee minimum withdrawal benefit VA for life product saw really significant growth in the fourth quarter.

Sales grew 65% versus the third quarter and 53% versus the -- against the fourth quarter and that has helped -- that was helped in large part by a third-quarter launch by SMBC, one of the megabanks and a mid fourth-quarter launch by BTMU, another mega-bank.

And as a result, at the end of December, we had launched that product in 35 banks and we are the only VA -- it is the only VA product sold at all four megabanks incidentally.

And the sales of that GMWB for life VA has continued to increase every month.

Let me just move over to A&H sales, which in the fourth quarter, declined by 15.6% over prior year, mainly due to deliberate measures that we had to reduce our advertising spend in ALICO's direct marketing channel to improve profitability.

And to offset that decline, we had increased A&H sales through face-to-face channels.

As we look forward, we are anticipating a stronger outlook for A&H sales as concerns raised by the claims payment issue are starting to fade.

We are into the final phase of the bank deregulation, which has opened up new opportunities as well.

In fact, on the 22nd of December, as you know, there was the final phase of deregulation and ALICO started to sell four A&H products, including a new innovative single premium whole life FIH and 20 major banks, including four megabanks and again, we are the only provider of products in all four megabanks and the mandates in the banks are running ahead of our plans.

In fact, we had 24 tieups by the end of February and we expect 30 banks by the end of April.

We have several new product initiatives as well.

We have got a new cancer product that we will be launching in June that we are excited about.

It will be competitively priced.

It will be particularly important to the female and younger market segments.

We expect a new FIH product to be developed in December.

Just briefly, I would like -- you didn't ask about life, but let me just mention one thing about life.

Life insurance is down 12.3% in the fourth quarter.

Major driver of that negative trend was the suspension of the [lapse] supported increasing term product back in April of 2007 and following that suspension, our sales totally dried up.

If you exclude that increasing term product, sales were up 14% in the fourth quarter and 8.6% for the year and we had other products to overcome that impact of the suspension of LSIT and for example, we posted record sales of dollar products, which have profit margins that are significantly higher than our yen-based products, so that was a really positive outcome.

If we look forward in life, the outlook I think is very favorable.

Single premium dollar whole life products remain very strong in both face-to-face and bank channels.

ALICO launched a very innovative product called [ESDN] or [yen soto dollar naka], which is a product that is yen on the outside and dollar on the inside and the product has surpassed our expectations in the first couple of months and we have a suite of about six products now through the banks.

Bob, if I can, I am going to have to try and end your response there only because we have got a lot of people on the call.

Andrew, if you would like anymore information, obviously we can get back to you on that.

That was great.

Thank you very much indeed.

Ladies and gentlemen, we have got a number of callers obviously and if I may ask that you kindly limit maybe to one or two questions and so we can take as many calls as we possibly can.

Thank you very much indeed.

Nigel Daly, Morgan Stanley.

Great, thank you, good morning.

You commented that consumer finance delinquencies and charge-offs remain below your target ranges.

You are barely breaking even in these operations.

So perhaps you can provide some additional color on what is driving the significant pressure on earnings if it is not the performance of your loan portfolio.

Second, just a numbers question, if you can break down the mark-to-market losses between the high grade and the mezzanine model CDOs and subprime.

Nigel, we have Rick Geissinger with us, obviously runs our domestic consumer finance business, so I will ask him to respond.

What was driving the fourth quarter is basically three or four major factors.

We continue to have problems in our mortgage company.

We are taking a number of actions to correct that and we'll continue to do so.

We also added about $70 million to our allowance for loan losses.

The reason for that is that we came off historical lows for the Company in the summer of '06 and delinquency and charge-offs are up somewhat.

We are not immune to what is happening in credit markets, but we are still below the target ranges that we published in 1997 in all of our products.

So there is a little bit of margin spread, margin squeeze I should say and that is fundamentally what is driving the earnings in the fourth quarter.

Nigel, on the second part of your question, we are looking for that information and we will try to get back to you during the period of this call.

Okay, that would be great.

Thanks.

Nigel, just a clarification.

You were talking about the investment losses, correct?

The CDS mark-to-market loss, which you took in the quarter, just breaking it down between the high grade and the mezzanine.

Are you talking about on the super senior?

In the super senior, yes, that's correct.

I am not sure we will be able to have that information at this call, but we can try to provide that.

Great.

Thank you.

Josh Shanker, Citi.

Thank you.

Two questions.

The first one, I wanted to discuss the $900 million stress test scenario number, how that relates to the less than $600 million discussion in what seemed like a worst-case scenario described at the December 5 conference call.

Then I want to also ask about that $900 million.

During the commentary on the Q&A, we said that number was as of January 31.

I just want to confirm that that is actually as of December 31.

And finally, I wonder if you could give any commentary on the D&O subprime exposure.

I noticed you added some new disclosure there and I am curious to know if you can talk about number of potential policies affected or what kind of rate online per million of coverage we are talking about in that book of business?

Sorry, Josh.

Bob will respond to the first part of your question there and we have John Doyle with us who is responsible for our executive liability business and he will give you some color on the last part and I think Nick will chime in there with some information on the international side as well, but first Bob.

Good morning, Josh, Bob Lewis.

As far as the comparison of the $900 million and the $600 million, the $600 million -- I assume you're referring to one of the frequently asked questions in the December 5 presentation, which described in it what the actual stress, if you will, was defined as being and the $900 million, the stress there is identified for you on slide 8 of the presentation we just referred to.

So the actual components of the stress are slightly different.

I don't have the actual $600 million in front of me here, but the $900 million that we've just discussed is really, as we articulated there, what we feel from current situations and rating situations that are in the marketplace, what we feel is a severe stress from here and that is rolled up on the portfolio to this $900 million potential unexpected loss.

To be fair, the $600 million was writing off all 2005 subprime RMBS BBB and below in all '07 and '06 regardless of vintage, as well as ABS CDOs from those -- I think regardless of vintage and year, which seemed like an extreme stress.

Are you saying the $900 million is an even more extreme stress than that?

I think one can simply say it is a different stress.

We have done a number of stresses and the one we have articulated in the presentation today is the one that we feel, given where the current stress is being shown in the marketplace, which is in the subprime and Alt-A areas, that this is the stress that we feel is the most appropriate for our own internal purposes to use as a severe stress.

And that is as of December, not January?

As of December, incorporating, however, rating actions through January 3, so there were significant rating actions taken on January 3 and our stress is a stress from that.

Very good.

And on the D&O front?

Good morning, Josh.

As Martin mentioned, we continue to monitor the issue very, very closely and claim activity remains manageable at this time.

We have, as of a week ago and dating back to the beginning of 2007, we are on notice of claims or potential claims on 261 policies.

It is 334 potential notices.

And is there anyway to get any more granular about how that $347 million in premium is divided among various clients or whatnot?

Sure.

That number -- that number of claims, by the way, is a global number, so it is our worldwide operation.

$301 million of the $347 million in written premium is from our domestic operation, but I am not sure what other information you are looking for there.

Can you talk about rate online for D&O at that time?

Is there any way that we can translate that into $347 million covers how many clients or anything of that nature?

How many clients we have?

How many clients -- I mean how many possible clients are touched by -- when you say that you have totaled the total subprime potential exposure, how many clients does that actually cover?

I don't have that number in front of me, but we did an extensive ground-up review in the domestic insurance operation to come up with the profile of the accounts that are potentially exposed.

In the footnote, it notes all the various classes of business that we took a look at.

As Kris mentioned, we are pushing price in the financial institution area and prices vary widely from private to public risks, from small E&O risks to other parts of the portfolio.

So prices are very, very different from one segment to another.

Okay, well, if you guys can give further disclosure on that, I am sure everyone would appreciate it.

Thank you very much.

Larry Greenberg, Langen McAlenney.

Thank you and good morning.

On the [FP] side, I know that AIG guarantees the (technical difficulty) capital there.

Is there any possibility that the rating agencies will require (technical difficulty) capital into FP?

Larry, that -- again, that has not been a subject matter of discussion because of the guarantee that AIG provides to AIG financial products.

So they are relying on AIG's overall financial strength to support the operation.

So that has not been a subject of discussion, no.

Okay, great.

Have any other key management of FP left or are planning to leave beyond Joe?

(inaudible), Larry and they are getting back to work straight after this call.

Great.

And just a couple of small pieces.

What was the FX impact in the foreign life earnings?

For the quarter, it was 3%, 2% year-to-date.

It was a lift, it was a benefit on operating income bottom line.

We disclosed the impact on revenues and GAAP premiums in our stat supplement, but bottom line, it was 3% for the quarter, 2% year-to-date.

Great.

And you made reference to some unusual expenses in the Foreign Gen expense ratio.

It looks like it was about four points above normal, which is about $130 million.

Is that reasonable for me to assume?

Nick Walsh has got some color on that, Larry, in great detail.

For those of you who don't have the supplement in front of you, the fourth-quarter '07 expense ratio is 42.25 against a prior of 38.48, which is a difference of 3.77.

And the year is at 34.95.

The first comment is that the fourth-quarter expense ratio is always the highest of the year for seasonalization issues.

The reasons for the difference is primarily through the realignment of certain legal entities and integration costs.

A major part of that is what we call a [part seven transfer] of our organization in the UK from a US branch to a local subsidiary.

That started on time and will finish on time.

It is finished.

I will give you some numbers.

It involved 200 people for the majority of the year.

We sent out information packets to 2.2 million customers and that was 10.5 million pieces of paper and we received 47,000 inquiries, all of which were dealt with.

The other part of the integration is between our acquisition in Taiwan Central and the AIG operation.

Aside from that, Ascot had a -- that's our [Lloyd's] business -- had a spectacular year and the increased profit commission shows up as an impact on fourth-quarter expense ratio.

Aside from that, we have a continuing emphasis on the consumer business and part of our strategic imperative is to change the format of some of our commercial business we are targeting and I have spoken about this in previous events and we are targeting smaller businesses because we think we have an opportunity there and the commission costs and of course, the operating expenses around that are higher, but that is all according to our plans.

Jay Cohen, Merrill Lynch.

Yes, thank you.

Two questions.

First, on UGC, clearly you are suggesting that the '08 year is going to be another pretty bad one.

I am wondering if you could put any parameters around it.

Should we be expecting losses in '08 to essentially mimic what we saw in the fourth quarter of '07?

That is the first question.

And secondly, in a lot of your analysis, you make a reasonable distinction between the '06, '07 vintages and the '05 vintages, which I think up until this point, that has been pretty reasonable, but with real estate prices continuing to come down, do you see a risk or why don't you see a risk that the '05 year will look maybe as bad as '06 given that the real estate prices continue to fall?

Jay, Billy Nutt is with us, so I will ask Billy to respond to the first part of your question.

Sure.

Good morning, Jay.

First question , as we have all seen, the housing indicators are all trending negative and they are trending negative at an accelerating pace, particularly in the fourth quarter and likely to continue to do so.

We have pushed out our economic forecast to suggest that the housing market is going to continue to experience a lot of stress through '08 and probably will not bottom out until the first half of '09, if then.

So the combination of the deterioration in the housing market, combined with obviously a slowing economy, is going to put quite a bit of stress on our domestic portfolio.

And as Bob Lewis said, is likely to result in another significant operating loss in 2008.

Without giving any specifics, we would anticipate that that operating loss would be somewhere in the range of where we were in '07 to somewhat higher than that.

As regards the decline in home price appreciation and the impact on the '05 book, the '05 book was underwritten with more conservative underwriting criteria than the '06 book and the first half of '07 book, so it should perform better.

Although it will experience some additional stress as property values continue to decline and we are estimating a decline or forecasting a decline in property values of another 7% or 8% for 2008.

On the line with me is Len Sweeney, our Chief Risk Officer.

Len, would you like to add any color to the '05 and the prior

No, Billy, I think you said it well.

The '05 book, we have about $5 billion of risk in force.

Again, it is starting to see the same level of stress as the more recent books, but it had -- it did enjoy some early appreciation.

So again under stress, but probably not to the level of the '06 and '07 books.

Actually, with that distinction, I was thinking more on the credit derivative business wherein the stress tests, obviously you are stressing the 06, '07 years more intensely and I am wondering in really that business, why don't you do a similar stress on the '05 year?

Jay, I have got Kevin McGinn with us, our Chief Credit Officer, so I'm going to ask Kevin just to respond on that.

Yes, hi, Jay.

Yes, the rating agencies definitely and all the delinquency and default data suggest that the '06 and '07 losses are going to continue to climb and you're now seeing estimates as high as in the high teens, low 20s.

The '05 is definitely showing some deterioration, but the numbers are not nearly as bad and that is probably in large part because a lot of the resets have already happened in the '05 vintages.

The highest loss assumptions we are seeing in the '05s are really around 7.5.

The rating stress tests that we've ran essentially took that into -- specifically took that into account because we are expecting some more downgrades in the '05 vintages.

We think we are on top of that and we have essentially stressed it appropriately.

That's helpful.

That is a good point about the resets to.

Thank you.

Alain Karaoglan, Banc of America Securities.

Good morning.

A couple of questions on capital and on the property/casualty business.

From an enterprise risk management point of view, are you considering whether you should be in the AIG financial products business at all given the heartache that it has given to the stock and the mark-to-markets on the portfolio and the volatility to the book value?

And from an excess capital point of view, you mentioned $14.5 billion to $19 billion, but that seems to be a little inconsistent with the fact that we are stopping the share repurchases.

Would it be possible to see if you could give us the excess capital versus what the capital requirements from the rating agency at the rating that you would like would be if there would be any excess capital at that level?

On the first part of your question, as I said in my opening remarks, AIGFP has been a very important and continues to be a very important part of AIG.

It has produced very good returns over many, many years at a very good return on capital.

Obviously, like all of our businesses, everybody stays under constant review, but the businesses performed exceptionally well to date.

We are in, what I have described as unchartered waters and like everybody else, they will continue to be reviewed, but at the present moment, I think they add significantly to the diversification.

On the second part, Steve?

If I could follow-up just on that, but on the capital, Martin, isn't that a notional amount of capital that you are allocating that AIGFP wouldn't be able to operate on a standalone basis and so the return on capital are high, but they are not similar or comparable to the rest of the business?

It is notional, but it is also drawing on the implied guarantee of AIG.

So when AIGFP's capital position is being reviewed by the agencies or by its clients, I think they are looking at the entire financial strength of AIG.

Joe, would you like to add?

Alain, it's Joe Cassano.

How are you?

One of the things that we have done historically when we go through our own capital management within FP and measure the usage of capital with FP is we've -- there is basically a charge against the AIG capital book.

And when we have done that, up until this period of time where we have these unrealized losses right now, we have always been a very, very high performer in terms of return on capital.

If you can think back or if you can find the files where we gave a presentation in May where we posted some of the historical return on capital, it has actually been relatively good.

It is the case that right now with the unrealized losses, it is a completely different story.

But what I would -- one of the things we do look at and one of the things we have been presenting over the last six months is the fundamental positive attributes of the portfolio we have written and looking at that, we really -- and if you look at it on that basis, we think the return on capital is still very robust and when we get through this period, the Company and the team will continue -- will return back to the very, very positive types of returns that we have had in the past.

I hope that helps.

Alain, with respect to your second question on excess capital, again, the $14.5 billion to $19.5 billion range is our estimate based on our own internal economic capital modeling.

It does not reflect the rating agencies' views of our excess capital.

They all have their own different perspectives on what level, if any, of excess capital that we have and again, I really can't speak for them.

I think, as we go through the process in the near term of hopefully settling our ratings, which are now, as I said, some on negative outlook and two on rating watch and review, that we will have more of a dialogue on that to see where they are relative to us.

So for the time being, given the fact that our ratings right now are on negative outlook and in some cases on review, we think it is prudent to suspend any new share repurchase activity until we have a better view on capital with all four of the major rating agencies.

And on the property/casualty business, could you comment on the adverse reserve development on the 2000 to 2002 years that is still bothering you and not allowing the benefit of the recent year reserve releases to flow to the bottom line completely?

And on the personal line segment, if I did the math correctly, even excluding cats and adverse reserve development, I get a combined ratio of 112.

Would've anything additional happened in the quarter in personal lines to lead to such a high combined ratio?

Frank Douglas is here, so I'm going to ask him to respond to the first part and then Bob will respond, Bob Sandler, respond to the second part.

Yes, we have seen continued development as we have described really all year, it is not new this quarter, from accident years 2002 and prior.

In general, the development has been less adverse than it was in prior years.

I think if you look in the 10-K, you will see that 2002 and prior has about $1 billion less adverse development, about $800 million actually less than it did last year, so it is trending down, not as fast as we would like.

It is coming largely from excess casualty, to some degree from transatlantic and to some degree from workers' compensation.

Those three areas, which we have talked about I think throughout the year.

We are seeing a lot of latent claims still emerge from our excess casualty book that are a function of the soft market years and we have improved our terms and conditions and underwriting guidelines to the point where we just don't think you are going to see that kind of latent development from the more recent accident years and as those older years continue to wind down, we should see less of those surprises, but they did happen this year.

They probably won't go away tomorrow, but we are certainly expecting minimal adverse development.

Certainly as we go forward, we expect those numbers to diminish.

And our next question comes from --.

No, we have a response to the second part of the question.

Pardon me.

This is Bob Sandler.

Let me take you through some of the pieces.

You have mentioned some of them, but not all of the pieces that distort the quarter.

There is a lot of noise in the quarter.

There was $75 million roughly of wildfire losses that are included in those numbers from the private (technical difficulty) areas.

You have got about $33 million of integration -- merger integration costs coming out off the merger of AIG and 21st Century.

You have got about $36 million of adverse development in the quarter.

We increased our loss pick in the quarter -- for the year, the current accident year loss pick and that had about a $50 million impact on the first three quarters of 2007.

So if you look at those pieces in total and subtract them, you'd actually find the quarter is probably running more on the 104, 105 range if you do the math, which is not a brilliant quarter.

I would just remind you that the fourth quarter is typically -- seasonality generally affects that quarter.

Last year, for example, we ran about a 102 in the quarter, so we are running maybe 2, 2 .5 points higher than that.

And some of that is being caused by the newer business that we have been writing in 2007, it's a higher proportion of our in force than in previous periods.

That is particularly true of the 21st Century business outside of California.

So that business, first-year business does carry higher loss ratios typically and so while the 104, 105 is not a number we aspire to, I think the underlying book is pretty strong.

We still anticipate that the direct business is going to produce a pretty good underwriting profit in the year '08.

In the fourth quarter, we probably ran in the 97, 98 range on that direct book anyway when you account for some of this.

Thank you very much.

Eric Berg.

Thanks very much.

I have two questions, both for Steve Bensinger.

Steve, with respect to this line on page 8 that culminates with a discussion of a realizable loss of $900 million, can you just clarify --?

You were breaking up there.

Oh, can you hear me now?

Yes.

Steve, with respect to the slide on page 8 that culminates with a realized loss under the severe stress scenario of $900 million, can you just clarify what is being expressed in the right-hand side of the table, what the various percentages mean?

Yes, I think I will let Bob and Kevin give you more color on that one.

They are the architects of that stress test.

Well, if I understand -- this is Bob Lewis.

(inaudible) help us understand what this means.

You are talking about the severe stress criteria?

Yes.

Okay, I got it.

I think that it would be useful for Kevin McGinn, our Chief Credit Officer, who has been managing the entire stress scenario work that we do to describe and put that stress in overall context, which I think is your question.

Yes, we essentially stressed three categories of assets -- subprime, RMBS Alt-As and CDOs and going from the top, you see we essentially took all the 2007 vintage, anything below AAA and wrote it off 100% with no recoveries.

We took the -- we did the same thing for the second half of '06 and for the first half of '06, we essentially took 100% of anything A+ and below and 50% of the AAs and we thought that was the appropriate stress given all the rating activity that has happened around the '06 and '07 vintages.

In the '05, we essentially wrote off 50% of all the second half of '05 BBBs and below.

Now this is -- again, a lot of rating activity took place all the way through the fourth quarter, so we think we have basically captured that pretty well.

We took also 100% of anything BB+ and below from the first half of 2005.

We then wrote off 100% of all the, what we call, the inter-CDOs.

Those would be any CDO buckets within the CDOs themselves A+ or below regardless of vintage and regardless of type, whether they are high grade or mezzanine and finally, we took 100% of the -- we wrote off everything A+ and below of Alt-A for '06 and '07.

That basically takes you through the various categories there.

And again, just to reemphasize, Eric, that those stresses were performed as static stresses.

That's what that footnote says.

So it was assumed to result in immediate losses which they wouldn't, they would occur over time and also there is no benefit in this scenario for cash flow diversion features and other mitigants that AIGFP has structured into these portfolios.

So that is why we call it severe.

My second question, Steve, relates to your discussion in connection with the negative spread or the negative basis and I think you said -- I hope I have it right when I say that you said that one can think of this negative basis as the difference between the value of a CDO if it were uniquely a credit instrument versus the -- or spread on the CDO if it were only a credit instrument versus the actual spread that is observed given the multiple risks on a CDO liquidity aversion to risk and so forth.

Is that what you were saying?

What I was saying I think that's -- yes, I think you have got it.

Let me restate it.

Yes, please.

In a different way.

A bond has a total spread attached to it, a cash bond.

That spread has various components that I outlined credit, liquidity, and this market perhaps market aversion risk and others.

And the difference between the total spread on the bond and just a credit spread on the bond would be the negative basis.

Okay.

So here then is my -- I think I have it right and I understand and thanks for that follow-up explanation.

Here is my second and final question, if the dealers were able to identify this spread differential on December 5 and you felt like it was perfectly fine to use it then, what exactly happened between December 5 and when you issued your 8-K that no longer made the application of the spread differential appropriate under US GAAP?

It was a matter of observability.

As of the December 5 call, our colleagues at AIG Financial Products had indications as I stated for market participants on different levels of negative spreads.

And they made a judgment regarding how those spreads that they were -- that they gathered from the market were related to their book of business.

At year end upon an exhaustive review, what we learned is that in today's market which had deteriorated much more significantly that those spreads were no longer identifiable and therefore, we could not take credit for those in coming up with a fair value determination at year end.

Thank you.

Ladies and gentlemen, I think we have probably got time for one or two more questions.

Ian Gutterman, Adage Capital.

Hi, thanks for keeping the call going so long.

I just also want to clarify on the negative basis, really more from the material weakness side and just -- can you clarify, is the material weakness -- do you have to resolve the material weakness and take credit for the negative basis or is the material weakness as much about procedures in that you might be able to if the market gets more robust and you can identify the negative basis in a way that PWC agrees, can you start taking that earlier than the material weakness is resolved?

Yes, I don't really see them connected on a perspective basis.

I think the credit for negative basis will come about based upon market changes that will illuminate the spreads more clearly.

It is not connected to the material weakness.

The material weakness is surrounding resources, the additional controls and procedures we are putting in place and what I talked about earlier in terms of sustainability of those controls.

So I wouldn't connect them on a perspective basis.

Okay and is anything you are doing procedurally to try to use the limited information to get more clarity or is it really just have to be patient and wait for the market to start trading better?

Well, we did a lot of work around year-end to see if we could -- if there was market clarity and I don't think the opacity of the market has clarified any further since then.

In fact, I think it has probably even less liquid and more opaque than it has been.

So right now, certainly I wouldn't say that we see that in the immediate future.

Okay, and then again to clarify on the rating agency issues, to the extent -- is your sense that the concern is -- is it an overall perception of consistency of results or is it something vague like that or have they been more specific that we are worried about the CDS losses or the investment portfolio losses?

And I guess where I am going with this is to the extent the market seems to be most worried about the CDS, if that is really the issue, is it possible that the rating agency concerns will be more around FP and maybe at the corporate debt rating and less around P&C and life and subsidiaries?

Do you see where I am trying to go?

Whether we are talking about corporate excess capital or whether we are talking about possible rating, the P&C or life companies that actually would affect your business more.

I think different rating agencies have different views, so I can't -- it would take too long for me to recount each one of them, but some of the rating agencies are looking more at the AIG debt ratings.

Some are looking more at the underlying subsidiary financial strength ratings and so I would say everything right now is under review or under outlook and we are taking all of that very seriously in terms of providing all of them with as much information as we can and descriptions of what we have done to try to clarify it and ensure that we can try to stabilize this as soon as reasonably possible.

Okay.

In just one quick numbers question if you have it at the tip of your fingers.

You said there was some remaining repurchase that was already committed to you.

Do you have the dollar amount of what that will be coming through?

Yes, it is a little over $1 billion.

After that, we would have about $9 billion remaining in the current authorization that the Board provided, but as we stated, we don't intend to utilize that in the immediate future.

Right.

I just wanted to clarify that $1 billion.

Thank you so much.

Gary Ransom, Fox-Pitt Kelton.

Thank you.

I sneaked in here.

I wanted to ask about the business of the CDO wrapping and how all this mark-to-market that has occurred is at all changing your strategy and whether there are certain parts of that business you never want to do again or want to get into again when times improve and just how has it changed your thought process about the business itself?

We have Andrew Foster with us, Gary, so he will respond to that.

Yes, I think really since we pulled out of doing the subprime sectors back at the end of '05, beginning of '06, our focus has predominantly been in the regulatory capital space where we have continued to do business.

We did an additional $39 billion of business in the fourth quarter and that has really been the focus that has still been a very strong sector for us and again, that is probably declining as well with the onset of Basel II that was mentioned before.

So in general, it is a space that is going down in our priorities.

In the multisector and the arbitrage type deals especially, is that -- since that is the area where we have had these big marks, can you see a possibility in the future that you would get back into it?

Would this size of a mark more or less -- or the potential for the size of the mark we have seen more or less permanently take you out of that marketplace?

I think that marketplace in general is very much reduced anyway.

The new issue sets pace there where that sort of collateral is pretty much dead and has been for some time and is likely to come back at any time.

But going forward, I think we expect a lot of -- most of the losses to be reversed and we will always look at the different opportunities that are available to us and assess them on an ongoing basis.

All right.

Can I sneak in one little property/casualty question on loss ratio picks for the current accident year?

You already addressed it for personal lines.

Were there any other changes of note or significance in any of the other classes for the current year accident pick in the fourth quarter?

Yes, this is Frank Douglas.

Nothing significant in the fourth quarter.

What we have seen throughout the year is downward pressure from the favorable development.

As you know, in accident years 2006, 2005 and 2004, tended to offset the rate decreases that have earned into 2007.

The earned rate decrease was probably only about 5% though.

So obviously we are going to have to watch that going forward when the rate decrease is maybe a little bit larger, but for now, the answer was very little change was needed in loss picks, virtually none.

All right.

Thank you very much.

Thank you, Gary.

Ladies and gentlemen, let me apologize to those who didn't have the opportunity of asking their question.

If you would like to call Charlene, we will try and respond as quickly as we can.

Thank you very much indeed for all your patience.

It is much appreciated.

Thank you.

That concludes today's call.

Please disconnect your line at this time.