AIG - American International Group

[Shane]

Peter - I know Berkowitz is worth hundreds of millions, but you would very likely lose this bet.  Yes he has accomplishments, I don't like to assume past performance predicts future performance.  There are plenty of examples of great long-term track records occurring before long stretches of poor performance.

 

It seems people are replying only to defend Berkowitz but not answer the post?

 

I should be clear that I don't care whether Berkowitz is a good investor or not.  I took comfort in knowing he was in the investment at first and now I have some doubts.  I'm just trying to pick the brains of investors on this message board as I think some very smart people post here.  After revisiting this investment I am having doubts as I am unable to find a compelling upside.  No offense is meant to anyone, I am being sincere.

[Shane]

Peter - I'm having difficulty getting my point across here.  Let's please stop talking about Berkowitz?

 

I'm not done deciding on what I think about this investment...

[ShahKhezri]

Shane Smith, interesting anecdote about Berkowitz. Did your friends provide any specifics?

 

Chartis' expense ratio hovers around 30%, which is pretty decent for its size and scope. My problem with AIG is that the opportunity costs are fairly high with the best insurance companies in the world offering nice deals. A steady eddie like Traveler's (although maybe ShahKezri can offer dissenting opinion) is trading under book with a low expense ratio and shareholder friendly policies. They have repurchased 50 M + shares over the last year!

 

Basically, when you buy insurance, the individual subsidiary that you purchase it from is usually rated under multiple company's.  At least, that's how we did it with in the Public Sector division.  We used three company's under 1 umbrella (the division), based on the qualities of the insured, under 40% loss-ratio would be put into the best company with the most allowable credits (25% off) and so on.  The AAA of AIG allowed for lowered pricing in certain Company's (that's assumption).  Growth in BV, I agree with the numbers board members here are sensitizing (7-9% growth)...anything +11% and I think the appraisal is aggressive as that kind of growth will come with AAA and that shouldn't be an assumption. 

 

Company is cheap, I'm getting around to the idea more and more.  Discussion has been helpful, quantitatively it makes sense (thanks Eric).  Qualitatively, I just have to more comfortable.  Management is doing the right things, although:

 

American International Group Chief Executive Robert Benmosche has complained to senior executives at investment banks about the "unfavorable research" of the insurer's stock, the Wall Street Journal said, citing people familiar with the matter.

Benmosche's complaint suggested that some analysts do not fully understand the company and its value, the Journal said.

AIG is considering which four banks should lead another large offering of the U.S. government's shares later this year, the paper said.

 

I don't like that, kind of a threat, conflict of interest, stock is down more than 50%, so what? 

[Shane]

ShahKhezri - Can you explain further why you think a 7-9% growth in book value is reasonable?

 

What I don't really understand is how one can assume this without seeing them put together consistent positive earnings.  They were positive last quarter but have gone in the black and back in the red before.

[txlaw]

ShahKhezri - Can you explain further why you think a 7-9% growth in book value is reasonable?

 

What I don't really understand is how one can assume this without seeing them put together consistent positive earnings.  They were positive last quarter but have gone in the black and back in the red before.

 

Shane, the only thing I would say is that you really need to take a look at the supplemental financials that are disclosed by AIG, try to figure out the structure of the company to figure out what the AIG Holdco really is comprised of, and then figure out where earnings and losses come from.  AIG is a complex beast with non-core assets that people don't really talk about.

 

And then you also have to take into account the current environment for insurance companies versus what the environment will look like in a few years. 

[ERICOPOLY]

ShahKhezri - Can you explain further why you think a 7-9% growth in book value is reasonable?

 

What I don't really understand is how one can assume this without seeing them put together consistent positive earnings.  They were positive last quarter but have gone in the black and back in the red before.

 

Historical industry average growth -- that's why it's reasonable.

[ShahKhezri]

ShahKhezri - Can you explain further why you think a 7-9% growth in book value is reasonable?

 

Sure, I think as Tx mentioned, the supplemental financials show what core vs. non-core assets consist of, there is definitely a drag there. 

 

Soft market is an issue and you just have to look where the puck is going.  For example, "premium audits" are used in casualty lines, say at the beginning of the year your exposure units are 20 police officers, at the end of the year you have 18, the insured gets paid back on the lower exposure units insured.  At the end of the day, you need growth in the economy, employment is number 1, because exposure units drive premiums on the casualty side. 

 

I called three friends last week that are on the underwriting side, Oil and Gas, Public Sector and Commercial and property rates are firm, casualty is weak. 

 

Break up that 7-9% growth into segments.  Say you use 7% to be conservative:

 

3% transformation of core vs. non-core

2% insurance market from soft to less soft

2% as a result of low GDP growth (if you don't think unemployment is going to go over 10%).

 

Not to mention the DTA's that can drive growth.

 

There is going to be heavy consolidation in the industry, I think Pickens once said it's cheaper to buy oil on the NYSE, same holds true for PnC, cheaper to buy premiums on the NYSE. 

[ERICOPOLY]

American International Group Chief Executive Robert Benmosche has complained to senior executives at investment banks about the "unfavorable research" of the insurer's stock, the Wall Street Journal said, citing people familiar with the matter.

Benmosche's complaint suggested that some analysts do not fully understand the company and its value, the Journal said.

AIG is considering which four banks should lead another large offering of the U.S. government's shares later this year, the paper said.

 

I don't like that, kind of a threat, conflict of interest, stock is down more than 50%, so what?

 

He should put out more presentations to investors on AIG's investor relations website.  Citigroup does a much better job at this -- you've lost all trust and scared the hell out of the world, to build it back takes a bit of extra effort.  Spoon feed it.  Organize the information and push it out there in power point.

 

Or not, instead complain when people don't "get it".

[enoch01]

 

He should put out more presentations to investors on AIG's investor relations website.  Citigroup does a much better job at this -- you've lost all trust and scared the hell out of the world, to build it back takes a bit of extra effort.  Spoon feed it.  Organize the information and push it out there in power point.

 

Or not, instead complain when people don't "get it".

 

Although it is immaterial to business performance, they should consider changing the name of the company.

[biaggio]

After following the BOA thread, I though I would re read this thread + look at company presentations, to better understand and learn how some of the smarter guys here think. I am also intrigued that someone like BB would have 23% of his assets invested.

 

Is there anyway to simplify this idea so even someone with a 12 year old mentality (like myself) can understand it. Here is  my version:

 

From my reading over  the last 30 minutes:

 

AIG used to be one of the great U.S. corporations, led by a good jockey (H Greenberg), showed growth consistently and as a result sold at a premium to book value.

 

Got into trouble after new management came in---they did not know what they were doing, allowed  unknowingly some cowboys to bet the house on high risk stuff (credit default swaps, CDO's  etc) which led to...

 

Had to be bailed out by U.S government.

 

New manager who seems to be doing a good job (according to what I read, i would have no idea otherwise)

 

Guys who did not know what they were doing are out. I assume all those people who worked at the various subsideries that built a great business are still there.

 

All there high risk stuff has been unwound.

 

They are currently operating in a soft insurance market, so they are not turning a profit +/or they are booking large reserves just so that they don t have another financial catastrophe in the future. Insurance market is not going to be soft forever.

 

When they do have a profit they have $25 billion in tax credits or past NOL's  so they wont have to pay taxes for a while.

 

Bottom line -we re looking at a very large company with a past history of good operation, that is being highly scrutinized by regulators, trading at 50% of BV when they should be selling at least at BV

 

Why is it selling this low:

-because the government has to sell a huge amount of common, so there is no urgency to buy, as they know it will be low for a while

-the consensus does not trust the company

-the consensus does not realize that all the risky stuff has been taken out by government (I am assuming this)

-I think so many people were burned by AIG, that they are not even looking at this i.e how would it look if they  were burned twice

-financial companies are out of favor (good place to look).

 

Seems to be a dollar bill for 50 cents?

 

Appreciate any feedback.

 

Disclosure:I have no money in AIG. as it is not something I would be into or know about but looking to expand my circle(dot) of competence.

[biaggio]

http://sec.gov/Archives/edgar/data/5272/000104746911001283/a2202141z10-k.htm

 

I was looking at the 10k noted above.

 

Any thoughts re loss reserve triangle on page 10.

 

It looks like they are not reserving enough. i.e they are/were aggressive...am I looking at that right?

 

 

[prunes]

Did anyone ever find the investor presentation referenced at the beginning of the thread?

[prunes]

What are the key impediments to AIG achieving an industry average ROE?

 

From what I've read, much of the company's poor performance has been due to charge-offs related to excess insurance where it's more difficult to estimate liabilities. This business line being phased out, correct? What would Chartis's numbers look like if this service line wasn't included?

 

Seems like AIG has long tail risks. How will we truly know if it is reserving properly? I know this will improve to a certain extent once Chartis simplifies the services it offers. What else?

 

The federal government wants to get repaid in full. Is this incentive in conflict with the need for full disclosure with regard to AIG's reserve estimates? [i.e. could there be an incentive for the Government to want AIG to have rosier earnings?]

[BargainValueHunter]

This may be why AIG is crashing...

 

http://www.cnbc.com/id/45418963?__source=yahoo

 

In a year when a lot of U.S. financial firms were shrinking their exposure to Europe, American International Group increased its exposure to European sovereign debt by over $700 million.

 

AIG now says it has $4.4 billion of exposure to European sovereign debt, compared with $3.7 billion at the end of 2010. It also has a whopping $5.8 billion of exposure to European banks. That creates a total European exposure of $10.2 billion.

 

“AIG believes that its combined credit risk exposures to sovereign governments and financial institutions in the euro zone are manageable risks given the type of exposure and the quality and size of the issuers,” the company said in its last quarterly filing.

 

Much of that increased exposure came from buying German bonds. AIG’s exposure to Germany’s debt grew from $1.1 billion to $1.9 billion, according to an SEC filing. The company also has an additional $957 million of exposure to German banks.

 

But AIG also grew its exposure to Spain, from $257 million to $294 million. It also disclosed $779 million of exposure to Spanish banks.

 

Exposure to France also grew, to $1.171 billion from $1.134 billion. Exposure to French banks sits at $933 million.

 

Exposure to the sovereign debt of the Netherlands went from $341 million to $436 million. Bank exposure is $2.2 billion for the Netherlands. Exposure to Finland went from $34 million to $88 million.

 

Exposure to Portugal went from $6 million to $7 million, with no exposure to Portuguese banks.

 

AIG did, however, dramatically reduce exposure to Italy, from $448 million to just $114 million. Exposure to Italy’s bank is $293 million. AIG also doesn’t own any of the bonds issued by Greece or Ireland.

[PlanMaestro]

Chartis presentation including some historic data

 

http://www.scribd.com/doc/80840540/AIG-Chartis

 

[racemize]

Here's the GAO data on AIG.  It is somewhat old, but I'm going through it now.

 

http://www.gao.gov/search?q=AIG

[PlanMaestro]

From the December 2011 Goldman Sachs conference

 

A number of your competitors complained vigorously over the last couple of years that while the company was in trouble, you were mispricing risk. And my question is, it sounds like you've changed that definitively, but what assurance can you give investors that you've adequately gone back and reserved properly for the risk that was not priced appropriately during the crisis?

 

<A - Peter D. Hancock>:

 

Well, I think that the first thing is that our competitors would say that, wouldn't they? But it sparked a lot of inquiry, as you can imagine. So, there was an independent inquiry on that very fact led by the Pennsylvania Department of Insurance, where they got outside actuaries to examine those charges, and they found no evidence of that whatsoever.

 

I would say that we have appointed a new chief actuary at the beginning of this year at Chartis level, came from Allianz. So, fresh set of eyes.

 

We have created a new function at the AIG level of a Chief Actuary to look over both SunAmerica and Chartis to verify those.

 

We added 28 pages of disclosures to the Schedule P so that you can reach your own conclusions.

 

We are committed to adding even further disclosure in terms of loss triangles to make it easier for people to reach their own conclusions.

 

We accelerated the pace of third party scrutiny by outside actuaries, so that it's not as slow a cycle.

 

And finally, we just cut back aggressively on businesses that were providing the greatest challenges in terms of reserving, the number I mentioned in terms of reduced business, so there's just less at stake from that.

 

Where I would grant some ground to the charge is that, during the midst of the crisis, the industry lost their favorite leader on pricing. And so I've got a lot of feedback that when starting in January, started to ask for rate, there was, I think, a lot of relief from the industry that finally, we are back driving the bus in terms of getting a proper return on risk.

 

Because I think the inadequate pricing for risk is not an AIG-specific issue. I think it's an industry issue. And I think that if you look at the numbers of many of our competitors, their calendar year results are being flattered by a lot by reserve releases, not through current pricing.

 

So I think that as an industry, we need to be better as a whole at understanding the risks that we are taking, getting properly compensated for them and stressing value over volume, which is really the heart of our philosophy. But I think that we've done everything we can possibly do, but we don't expect to get fully credited for that until development demonstrates that we are right on that. But good question, thank you for that.

[biaggio]

PlanM, time will tell re reserving + pricing of their products. I don t know that there is any way for us to verify, is there? Have to have trust in management.

 

I like FFH model in that management has all their wealth at stake and own a sizable share of the company.

[vinod1]

I have calculated the loss reserve development from a policy year perspective to answer the question: for policies written in a particular year how did the loss reserve develop?

 

This is developed directly from AIG 10-K Loss Reserve Triangle. The advantage with the above method is that it separates loss reserve changes for each policy year rather than combining current year estimates with past year revisions. The first table is an intermediate table to get to the second table which is of interest. The highlighted data indicates the loss reserve development after "x" number of years for policies written on that year. If you look at this I do not see a major concerns - 5 policy years where they had over-reserved (at least per current estimates, which might turn to under-reserving in future) and 4 policy years in which they had under-reserved. Two of the years in which they under-reserved are 2001 and 2002 - pretty much as one would expect.

 

Note that negative numbers are good as the data is presented to show under-reserving.

 

Vinod

Loss_Reserve_-_Calculated.pdf

[vinod1]

I have calculated the loss reserve development from a policy year perspective to answer the question: for policies written in a particular year how did the loss reserve develop?

 

This is developed directly from AIG 10-K Loss Reserve Triangle. The advantage with the above method is that it separates loss reserve changes for each policy year rather than combining current year estimates with past year revisions. The first table is an intermediate table to get to the second table which is of interest. The highlighted data indicates the loss reserve development after "x" number of years for policies written on that year. If you look at this I do not see a major concerns - 5 policy years where they had over-reserved (at least per current estimates, which might turn to under-reserving in future) and 4 policy years in which they had under-reserved. Two of the years in which they under-reserved are 2001 and 2002 - pretty much as one would expect.

 

Note that negative numbers are good as the data is presented to show under-reserving.

 

Vinod

 

In 2008 they did under-reserve by $575 million as of current estimates but it is not a big enough number to worry. Most of their under-reserving had been for policies written prior to 2001. In fact, $23 billion of reserves had to be added for policies written before 2001 in the last 10 years.

 

Overall my point is that I am not seeing major obvious red flags in AIG's reserving. Of course, we all know that we can only be confident after a long long time.

 

Vinod

[BargainValueHunter]

From the December 2011 Goldman Sachs conference

 

A number of your competitors complained vigorously over the last couple of years that while the company was in trouble, you were mispricing risk. And my question is, it sounds like you've changed that definitively, but what assurance can you give investors that you've adequately gone back and reserved properly for the risk that was not priced appropriately during the crisis?

 

<A - Peter D. Hancock>:

 

Well, I think that the first thing is that our competitors would say that, wouldn't they? But it sparked a lot of inquiry, as you can imagine. So, there was an independent inquiry on that very fact led by the Pennsylvania Department of Insurance, where they got outside actuaries to examine those charges, and they found no evidence of that whatsoever.

 

I would say that we have appointed a new chief actuary at the beginning of this year at Chartis level, came from Allianz. So, fresh set of eyes.

 

We have created a new function at the AIG level of a Chief Actuary to look over both SunAmerica and Chartis to verify those.

 

We added 28 pages of disclosures to the Schedule P so that you can reach your own conclusions.

 

We are committed to adding even further disclosure in terms of loss triangles to make it easier for people to reach their own conclusions.

 

We accelerated the pace of third party scrutiny by outside actuaries, so that it's not as slow a cycle.

 

And finally, we just cut back aggressively on businesses that were providing the greatest challenges in terms of reserving, the number I mentioned in terms of reduced business, so there's just less at stake from that.

 

Where I would grant some ground to the charge is that, during the midst of the crisis, the industry lost their favorite leader on pricing. And so I've got a lot of feedback that when starting in January, started to ask for rate, there was, I think, a lot of relief from the industry that finally, we are back driving the bus in terms of getting a proper return on risk.

 

Because I think the inadequate pricing for risk is not an AIG-specific issue. I think it's an industry issue. And I think that if you look at the numbers of many of our competitors, their calendar year results are being flattered by a lot by reserve releases, not through current pricing.

 

So I think that as an industry, we need to be better as a whole at understanding the risks that we are taking, getting properly compensated for them and stressing value over volume, which is really the heart of our philosophy. But I think that we've done everything we can possibly do, but we don't expect to get fully credited for that until development demonstrates that we are right on that. But good question, thank you for that.

 

Good work on AIG, PlanMaestro :)

[biaggio]

Vinod, thanks for posting tables.

 

I would like to understand these triangles better.

 

Typically would you see something in the past 10 years numbers if they were being conservative? 

 

Does anyone recall the loss reserves/triangles for the insurance company acquired by FFH which led to the 7 lean years-was there any indication in their triangles that would have predicted that they underpriced their products or did not reserve enough (I am talking about the numbers for Crum & Forster and TIG before they were acquired by FFH ). Or was it just that they were in a very soft market with poor pricing, combined with catastrophe of 9/11 etc.

[racemize]

I have calculated the loss reserve development from a policy year perspective to answer the question: for policies written in a particular year how did the loss reserve develop?

 

This is developed directly from AIG 10-K Loss Reserve Triangle. The advantage with the above method is that it separates loss reserve changes for each policy year rather than combining current year estimates with past year revisions. The first table is an intermediate table to get to the second table which is of interest. The highlighted data indicates the loss reserve development after "x" number of years for policies written on that year. If you look at this I do not see a major concerns - 5 policy years where they had over-reserved (at least per current estimates, which might turn to under-reserving in future) and 4 policy years in which they had under-reserved. Two of the years in which they under-reserved are 2001 and 2002 - pretty much as one would expect.

 

Note that negative numbers are good as the data is presented to show under-reserving.

 

Vinod

 

This seems very good, but I'm not quite following how the numbers are calculated--could you shed some light on it for me (it seems like I'm missing something simple).

[PlanMaestro]

I would like to understand these triangles better.

 

Me too! Thanks Vinod.

 

David Merkel did some triangles analysis comparing AIG with other companies. I made some comments on it. My understanding is that actuaries would say that triangles are not enough and that the 2008 numbers are a mess considering all the things going on at the time.

 

http://variantperceptions.wordpress.com/2010/06/05/thinking-about-investing-in-insurance-companies/

 

I was thinking more on soft factors,

 

* regulators up their necks (FED ready, Pennsylvania DOI review).

* new management (Chartis CEO and actuaries) that had all the incentives to go clean if there were problems.

* all recent steps show an emphasis on cleaning up the balance sheet (asbestos deal with Buffett, Chartis large one time increase of reserves).

* significant reduction of leverage at the Parent level indicating risk aversion.

 

But it would be great to know more on the hard data side. And thanks to the person that linked to the GAO docs, ready to dive into them

 

Plan

 

 

[biaggio]

thanks PlanM -I bookmarked the site + alephblog sites