AIG M&A Strategy: Dumb & Dumber

Well, I had something else almost ready to go for this week but that can wait because there’s some big insurance news to address.

AIG announced two significant transactions last week – one for a 35% stake in Convex and the other for renewal rights to the much maligned Everest Insurance portfolio.

If you can’t figure out my views of these deals from the title of this note, let me say it clearly. I think these are bad deals for AIG.

Let’s look at them each individually.

Convex

AIG took a 35% minority stake in Convex. There is nothing particularly objectionable to that. Stephen Catlin has a well earned reputation and Convex appears to be operating well.

The problem is they paid nearly 2X book for it! And didn’t even get control!

How exactly does this transaction work out for AIG long term? They get no cash flow until they sell (unless Convex pays large stat dividends rather than reinvest in the business) and they have limited ability to sell because who wants to buy a 35% stake?

So, realistically, AIG is locked into this investment until the control owner, Onex, decides it wants to sell. That’s not a good position to be in.

More importantly, what are the odds an ultimate sale (or IPO) will be done at 2X book or better? Pretty darn low unless you can convince the Japanese to buy in. It seems unlikely there were alternate bidders at this level or Convex likely would have sold to them.

Thus, the financial math for AIG is about the intersection of a growing book value and a declining multiple. For example, let’s say Convex grows book 12%/yr (which would be an excellent result) and is eventually sold in ten years at 1.5X. The annual return on investment for AIG would be 9%.

That’s acceptable but not exceptional. Presumably, they could have made more deploying that capital into their own underwriting. Or maybe not, considering the second deal!

Everest Insurance

For its next act, AIG decided to something really indefensible. It paid $300M to buy renewal rights to the money losing Everest retail insurance book.

The value of renewal rights to a book of business that has likely lost money since inception rhymes with Nero, who is fiddling as AIG’s money burns.

While we don’t have details on the retail book’s performance, we can observe the entirety of the Everest Insurance book has been a long term money loser and they decided to keep the wholesale, which suggests the retail segment is even worse.

I’m sure there are a few contracts that might be worth having, but the entire book? Are you kidding???

Before examining how bad this deal is, let me first ask was there really a competing bid? Why couldn’t you walk away, force Everest to run the business off, and go to market to win the few accounts you wanted?

Renewal Right or Obligation?

So let’s set the table a bit. The subject book of premium was about $2B gross. In a typical renewal rights deal, a buyer would look to keep maybe 25% of the book and let the rest go. Thus, if you paid a 10% commission on $500M, this book should have gone for $50M.

So why is AIG paying $300M? Because their plan is to renew most of the book! No, for real. That’s the plan. They have said that in the press and Everest’s 8-K (see below) confirms it.

So, AIG is spending $340M (there are $40M of other expenses it is also reimbursing Everest for EDIT: I misread…it’s actually $120M, not $40M so the total is $420M!) to add nearly $2B of loss making premium! And while I’m sure they’ll get price increases, yada yada yada, there is no way the math makes sense.

Everest did the public a favor by disclosing in their 8-K that AIG paid $301M for the renewal rights based on a full 15% commission. Thus, we can fairly easily back into the $2B basis on which AIG will pay commission.

In other words, they’re paying 15% on the entire book, even the awful business. This is not how renewal rights deals are supposed to work. You buy the right to renew, but not the obligation, and you pay only on the accounts you actually, you know, renew.

I can’t recall ever seeing a renewal rights deal where the intent is to renew the entire book for the obvious reason that the book is being sold because it is unprofitable.

Path to Profitability?

To demonstrate, let’s do some simple math. If the book was (at best) producing a 70% LR and you are paying 15% renewal rights and paying the regular commission to the broker (this is not direct business) so about another 15% and then you have a G&A load of say an additional 15%, you’re at a 115 CR in year one!

Sure, that falls to 100 in year two since you no longer have to pay Everest and, hopefully, you can bring the loss ratio down some over time (but remember, nobody else wanted this business which is how Everest won it in the first place), but there is an awful lot of work to do to earn back that $300M.

If you prefer to look at it another way, they could have let the business go to market, undercut competitors by 10%, and that would have only cost $150M (assuming they reinsured 25% of the book).

If this were a book that used to be good but had been mismanaged or were in specialty lines that were hard to break into or maybe allowed you access to customers you otherwise couldn’t reach, we could at least have a discussion about justifying overpaying to get a foot in the door.

This is none of those. It’s global retail insurance that nobody else in the market wanted and brokers were thrilled to dump it on Everest.

I will conclude where I started. The renewal rights are worth a word that rhymes with hero, which is how I would describe whoever at Everest engineered this favorable exit for them.

Additional Thoughts

A few more thoughts on M&A before I go…

• There are certain people you never buy a business from because they are always smart sellers. Jack Byrne (and his successors at White Mountains), Stephen Sills, John Charman, and…Stephen Catlin. Buyer beware.

• You can’t spell Convex without Onex, but you can spell it without A I and G. On the other hand, you can spell it with Onex and CV (Starr), so maybe the Starr Foundation should have teamed up with Onex? They don’t have public shareholders to answer to and have a long time horizon.

• The last time AIG paid this much for a Bermuda based hybrid reinsurer (Validus), they lost most of the people and eventually sold it for less than they paid. Maybe forgoing control helps avoid the former problem, but, as noted, lacking control brings its own issues.
  
  So maybe the lesson is don’t buy into reinsurers? How many times (IPC, Transatlantic) does AIG have to learn that it’s not good at owning reinsurance businesses before they give up?

• You have to earn the right to do something dumb. Have other big insurers made dumb deals that eventually get swept under the rug? Of course. But they have enough going well that investors will look the other way.
  
  AIG has spent the last 20 years trying to regain credibility. They have been making good progress, but that’s where the focus should remain. The path to success is continuous underwriting improvement, not going on detours chasing dreams about being the AIG of last century again.