In the public world of insurance, there is no large carrier with an underwriting record worse than AIG’s over the past decade.  It’s been a dramatic mess.

The History

A very short reminder of the company’s past might ring a few bells in the head of the (two) people who might read this.  I shouldn’t name names, but…   In 1962, Maurice Greenberg was named the head of AIG’s North American holdings – anointed to the position by AIG’s founder, Cornelius Starr.  Six years later, Starr eventually picked Greenberg as his successor.  Baloney ensued.  As to when the baloney started inside of AIG remains a bit murky, though it appears Greenberg started this house of cards and subsequently over-rid all his past successes and virtues by turning a global insurer into a quasi “hedge fund” obsessed with toxic subprime mortgages all the while cultivating a cavalier attitude toward risk underwriting.  Good ole’ Maurice was ousted in 2005 due to shoddy accounting practices.  All hell broke loose two years later.

After a $182 billion taxpayer-funded bailout, AIG re-IPO’d in 2011.  By 2014, AIG had repaid its massive debts to the federal government, but in the process either had to or decided to rid itself of most of the crown jewels (on the cheap).  In the early years of the recovery from the financial crisis, AIG was led by Bob Benmosche who kept the business alive through these massive asset sales.  But Benmosche died from cancer in 2015 and was succeeded by an investment banker (more specifically a derivatives meister) in Peter Hancock with no insurance experience.  AIG’s board was probably hoping for a similar experience to Berkshire’s with Ajit Jain, but fate (or Peter) wouldn’t allow it.  Under Peter, prior underwriting sins continued to manifest and a brutal cost-cutting program undercut moral and had no material effect on underwriting improvement.  Unsurprisingly, Peter was canned in May of 2016.  Reserve charges on old vintages continued to materialize throughout Bob’s and Peter’s respective reigns (long live Maurice!) and even worse, AIG was releasing reserves on recent vintages hinting that AIG’s risk underwriting was making little progress.  In short, AIG was doing a decent job of living up to the title first levied toward it in 2008 as “The Most Hated Company in America”.  All of this was an obvious detriment to the growth in business value over recent years, though something is changing.

The Pivot

In 2017, AIG coughed up $10.2 billion in cash to Warren Buffet at Berkshire Hathaway in exchange for soaking up 80% of future losses on policies written for accident years prior to 2016.  When Berkshire took this deal, they calculated the liability for unpaid losses and loss adjustment expenses at $16.4 billion.  At the time, Warren thought the time value of money made this an attractive wager.  So far, AIG looks to be having the last laugh.  Berkshire has since upped this liability by $1.8 billion to a total of $18.2 billion.  This was a large and important pivot in AIG’s trajectory as it allows AIG’s current management to focus on the future and it limits investors downside from vintage exposures.

The Ringer

In early 2017, AIG hired a seasoned insurance executive in Brian Duperreault to step in as CEO.  As a young buck, Mr. Duperreault was at AIG for two decades, leaving in 1994 to lead ACE, a niche insurer with a single product.  Under Duperreault, ACE morphed into a best of breed P&C insurer with combined ratios 9 points better than industry peers.  After hand selecting his number two to replace him as CEO in 2004, he semi-retired to become ACE’s Executive Chairman.  Though in 2008, Mr. Duperreault returned to the insurance world as the CEO and resurrector of Marsh & McLennan.  M&M’s sales collapsed as the business went through a series of missteps that involved a broker rigging scandal and a drawn-out fight with Eliot Spitzer that put a ding in customer’s trust.  Mr. Duperreault stabilized the business during the regulatory issues, then through the credit crisis, was able to improve margins and put a boost in profitability.  Mr. Duperreault is well-known in the industry, and he probably ought to be in the capital markets.  During his decade reign at ACE, shareholders realized over a 500% return and beat the S&P by over 260%.  While at Marsh & McLennan, Mr. Duperreault turned the business around and outperformed the S&P by just over 20%.  A pretty decent history.

The Future

Today, Mr. Duperreault has been at the helm for about a year and a half and he’s working toward a cultural reset.  Most obviously, he’s been cleaning house and poaching talent.  Among the most comforting and interesting poaches is AIG’s relatively new Cheif Underwriter in Tom Bolt, a decades-long Berkshire Hathaway Executive.  Duperreault also convinced his old number two, Peter Zaffino (then, the CEO of Marsh), to join him as his current number two and Chief Operating Officer.  Mr. Zaffino had no good reason to leave his helm as CEO of Marsh except to believe that he’ll (soon) take over the helm at AIG.  Today, Mr. Duperreault is 70 years old and is the new face of AIG.  My take is that he’s there primarily as a talent gatherer and culture builder.  I expect him to build a strong team around him, then ride off into the sunset in 4-5 years after the business is on solid footing.  Mr. Zaffino is most likely the operator and implementor.  He has a strong incentive to ensure a solid performance and locking in his future Chief Executive status.

Among other hires, Mr. Duperreault has also poached:

Duperreault plans to grow the business, not shrink it as his predecessors have.  In this growing vein, AIG just closed on its first major acquisition in 17 years spending $5.6 billion on Validus Re.  Validus has achieved a combined ratio below 90% for the past decade or so and brings a suite of interesting risk modeling techniques (I know that sounds like LTCM) along with it.  Hopefully, Validus and its historical underwriting skill can contribute to the talent and process at AIG.  AIG will also be utilizing a higher level of reinsurance vs its past self and therefore expects less volatility in underwriting results, particularly related to catastrophe coverages.

Who is AIG today?

What I won’t do in this post is explain AIG’s business or balance sheet in detail, so I’ll just touch on them quickly and encourage my two readers to do more of their own work in understanding the business.  First, the good:  AIG has a life and retirement business that earns mid-teen ROE’s and has a lot of internal admiration.  All parties involved seem to like this business and expect it to earn similar returns even if the stock market stops going up and to the right continuously (we’ll see).  AIG also has a strong balance sheet with about $4 billion of holding company cash and upper single-digit billions in debt capacity.  Financial debt to total capital stands at about 27% and risk-based capital ratios are also strong at 480% for the life insurer and 409% for the P&C business.  And the bad?:  The P&C business has remained the maimed ugly duckling.  Underwriting has been horrific, and vintages of all shapes and sizes haunt the business.  2018 has been dubbed “the year of the underwriter” in an effort to change the culture and shine some light on an area desperately in need of new processes.  A lot of spring cleaning has been going on.  With that negativity said, Mr. Duperreault has guided to a combined ratio of 100 (breakeven) coming into 2019.  That would be terrific…though time will tell.  Further, a newly formed entity within AIG called DSA Re is managing another large legacy book of runoff business and it sold 20% of this business very quickly to Carlyle (the Rubenstein PE firm) at an implied 100% of book value.  AIG has about $6 billion of “other” legacy/vintage exposure left on the books that it thinks will be manageable.

The Value

Today, AIG trades at $52.50 and book value is at $68.40.  Obviously, a turnaround by Mr. Duperreault and team would consist of more solidified underwriting leading to a shrinking combined ratio.  If AIG were to reach company-wide ROE’s of 10%, the business would probably be worth book value.  Today, with the shares at $52.50, the company trades at 77% of book value (excluding accumulated other comprehensive income or AOCI), which is equivalent to about a 9.5% earnings yield to today’s buyer of the equity.  Another way to think about this is to assume a stable earnings yield awarded by the market.  If ROE’s moved to 10%, then the shares might re-rate to a price to book of 105% giving the equity an earnings yield of 9.5%.  Though if the company shows solid improvements in general insurance loss ratios, 100% of book value is probably conservative.  The investment could be an attractive one.  Moving from today’s price of $52.50 to today’s book value of $68.40 isn’t great, but isn’t horrible.  If book value grows AND the company improves is combined ratio then the upside looks a bit better.  But I think there’s a better way to get exposure.

In 2011, concurrent with the re-IPO, 10-year warrants were issued relating to the TARP.  As opposed to normal derivative securities, the strike price on these warrants get a downward adjustment for corporate actions benefitting the common stock.  So, with dividends paid, shares repurchases, and a few other scenarios, the strike price of these warrants fall.  Today, these warrants trade for $14.50 and have a strike price of about $43.69, so they’re in the money.  Warrants are pretty simple, they work just like call options.  In other words, purchasing the warrants today gives one the right to buy shares in January of 2021 at $43.69.  Because AIG will likely pay dividends and buy back stock in the mean-time, a conservative strike price on the warrants at the expiration will be $43.  So buying the warrants today is equivalent to buying the stock at $57.5 in 2.5 years ($14.50+$52.50).  Put another way, you can buy AIG stock at 84% of today’s book value 2.5 years from now.  If you believe that AIG might trade up to book value or be on a path to fixing its past within 2.5 years, the warrants yield some attractive returns.

The below chart shows some possible return scenarios for the warrants.

aig warrant returnsAs the chart states, the left axis is the rate that AIG might compound today’s book value over the next 2.5 years (until warrant expiration).  Across the top axis are ranges for the price to book value of AIG’s shares come January of 2021.  First, let’s focus on what would happen if shares did not re-rate at all.  We can see that if AIG grew its book value by 3% a year and ended up trading at roughly the same multiple to book value as today (78%), we wouldn’t lose money.  Those both seem like reasonably conservative scenarios considering GDP is growing at over that 3% rate and today’s share price seems burdened with pessimism.  A little bit of returning optimism could do wonders for the warrants.

Returning to the chart above and focusing on my highlighted colors, I’ve created a key as shown below.  This chart below details return scenarios and the likelihood of those returns (according to me).

aig warrant probabilities

Of course, if AIG simply re-rates to book value, the warrants have massive upside.  Most importantly, I don’t see the downside as anywhere near as large.  I won’t confuse you with more screenshots of excel, as you can run through various scenarios using this chart and your own thoughts on AIG.  Though after thinking through this a few different ways, I think the downside is most likely -25%, with the upside being 100%.  Of course, they could be a zero (they are warrants), so a reminder to size them correctly (I’m thinking maybe 1/3rd or 1/4 size of a normal position).

I’ll add one more kicker.  Duperreault was hired with stock options valued at a total of $16.1 million.  He gets a third of that after three years, but two-thirds of that amount won’t vest unless AIG’s shares are $10 or more above the $61.82 share price when he was hired.  $71.82 on the shares would be about 100% upside on the warrants.

My final warning/thought is to not get love-drunk on the possible upside.  If AIG can’t show some progress on the general insurance underwriting, the little optimism still remaining in the shares will almost surely turn to pessimism.


***As is always my disclaimer, do your own work and verify the numbers.  This is not investment advice, just general musings***



3 thoughts on “AIG”

  1. I’ve added quite a bit to the common stock over the last few weeks at 60 to 62% of book value, though have done nothing to the warrants I own (with almost exactly 2 years to expiration, the time horizon is too short for me to act in size despite the apparent value outlined above). For further reasons covered above, book value is stronger and with less downside risk than it has been over the past five years – yet we’re easily at a five year low in P/B. I don’t know what exactly the upside is, but downside seems extremely low at 60% of book.


  2. Some significant progress now shining through at AIG. Insurers are doing very well across the board, and the rising tide probably has something to do with AIG doing well. Q1 results came through in very solid. I will say that up to now, any quarter where AIG hasn’t increased reserves has portended ugly things to come in future quarters. I’m betting that reserve strengthening baloney is (mostly) behind us. Results here:


  3. Today I closed our exposure to AIG. Although a nice return, it did not work out as I had planned (though what does!?). Our gain from the warrants was only slight as the business took about a year longer to turn around than I had hoped. The longer-than-planned turnaround matters because the warrants have a finite time horizon. The good news is that it has (basically) turned around. AIG looks on track to sure up their underwriting and has cleaned house. Fortunately, we did purchase a lot of common stock at $42 in December ’18 and January of ’19 as mentioned in the comments above. The common stock delivered significant gains in 7 months (37%). So the return on our total committed capital was well into the high teens with average exposure of about 6 months. Not bad, just different than I had originally envisioned. The reason I am closing today is that I can only conservatively underwrite a mid-single digit CAGR over the next three years on the common stock. Nor can I justify a change in thesis which involved buying a deeply discounted insurer at a significant markdown to book value with the belief that Brian Duperrault could turn it around. That thesis has all but born out and with our thesis playing out, I can’t justify a thesis drift. I simply can’t underwrite the business well enough from this point nor do I want warrant exposure at today’s price considering time left until expiration.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s