Why Berkshire’s cutting back on reinsurance

By Felix Salmon
July 13, 2009

Scott Patterson reports that Berkshire Hathaway wrote less than half the amount of reinsurance in 2008 as it did in 2006, and that 2009 will be substantially lower still. He relates this development to Berkshire’s credit rating, which was downgraded from triple-A by Moody’s in April:

In its credit opinion, Moody’s also cited the potential volatility of Berkshire’s “catastrophe-exposed business.” Berkshire is a major investor in Moody’s Corp., parent of the ratings group…

A higher book value and a rising cash stockpile could eventually lead to a reinstatement of Berkshire’s Aaa rating, though Moody’s isn’t likely to reverse itself soon.

If and when Berkshire wins back a higher rating, that could pave the way for the company to move aggressively back into the property catastrophe market.

It strikes me that the really important thing here isn’t Berkshire’s credit rating, so much as its CDS spreads. Reinsurers used to feel that they needed a triple-A rating because that connoted utter safety: you could reinsure your catastrophe risk with Berkshire safe in the knowledge that the risk of Berkshire being unable to meet its obligations was significantly lower than the risk of, say, a hurricane hitting New York.

Now that the value of a triple-A rating has plunged, however, in the wake of many formerly triple-A-rated securities defaulting over the past year or two, insurers are going to feel much more need to hedge their counterparty risk when it comes to their reinsurance contracts. As a result, the cost of reinsurance isn’t just the cost of the premiums any more: it’s the cost of the premiums plus the cost of buying credit protection on the reinsurer.

Conversely, from Berkshire’s point of view, every dollar that Berkshire reinsures is another dollar of demand for Berkshire credit protection, and the upward pressure on Berkshire’s CDS spreads will remain. And thanks to delta hedging and capital-structure arbitrage, sellers of Berkshire credit protection will ultimately end up depressing the Berkshire share price.

Berkshire’s shares are looking pretty cheap these days: at $85,000 apiece, they’re lower than they were five years ago. But five years ago, the idea that Berkshire’s insurance products would have to be discounted by its counterparty risk would have been unthinkable. Without its triple-A moat, Berkshire looks much less special than it did back then.

Update: My commenters are saying that insurers don’t hedge their counterparty risk to reinsurers. Either you trust a reinsurer or you don’t; if you do, you don’t hedge counterparty risk, and if you don’t, you don’t do any business with them at all. Maybe insurance regulators should be looking into this.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
7 comments so far

I am pretty sure this explanation is false.
I work in the insurance / reinsurance industry & pretty much no-one hedges counterparty reinsurance risk, let alone with a name like Berkshire:
- Sad to say it, but the way insurance accounting works (a very esoteric subject) means that reinsurance assets are booked at par notwithstanding the reinsurer’s CDS spread. Buying a CDS hence would just add to reported volatility, even if economic / real risk is reduced.
- I would severely doubt that a Berkshire CDS would be payable in the event of their default. Its a world ending event – would the CDS counterpart(ies) still be around? I doubt it.

Posted by James Goodchild | Report as abusive

They wrote half as much in 08 than 06 because prices had come down for cat risk. Nothing to do with their credit rating.

I agree with James’ comment. Reinsurers’ credit quality is usually a difference of type rather than one of degree; you’re either an acceptable risk or not. For some lines of business, contracts don’t settle for years (even decades) and only the best-rated reinsurers are on the ‘security list’ and so allowed to quote.

Once you’re on the list, though, you don’t get any added economic benefit from further strenghtening.

BRK is still as highly rated as you’re going to get, so they will write whatever they want, if they deem it profitable enough.

Posted by Dave | Report as abusive

It seems like this happens every few years. There is a huge disaster and all the insurers become insolvent and retreat from the cat market except for Berkshire. Gradually, everyone forgets about the disaster and comes back in to the market, squeezing margins. Just wait until the next big hurricane – Berkshire will be back in the market again.

Posted by Kyle S | Report as abusive

It’s not even clear you could buy CDS protection on Berkshire’s operating (insurance) companies. Reinsurance receivables can’t be delivered or auctioned in a cds settlement proceeding. Operating companies rarely issue much debt. Protection referencing the holding companies would be a poor hedge. Indeed monoline bond insurers are the rare insurance operating companies where there exist a liquid market for credit derivatives.

Furthermore CDS protection captures very different credit characteristics than insurer financial strength, which is what a primary insurer looks for in a reinsurer. In particular catastrophe reinsurance pretty much has a 1-year, and often even shorter June to October “tail.” It’s hard to imagine that Berkshire’s 3-month to 1-year credit risk is really less than AAA.

Finally for catastrophe reinsurance primary insurers do not receive any accounting credit until an even occurs. If there was a Berkshire underwriting meltdown, it would be known before any accounting credit was granted to the primary insurers for their reinsured losses.

Posted by MrLomez | Report as abusive

Another funny characteristic of insurer financial strength is that policyholders’ claims are senior to credit claims (that’s why the insurer financial strength rating is always higher than the bond rating).

The problem is that when an insurer approaches insolvency, claims liabilities are up for negotiation but the bonds are not. Insurers don’t actually default, they go into ‘runoff’, which means that they stop accepting new business and negotiate down all of their liabilities.

Biggest example? Lloyd’s of London, which packed up its old asbestos liabilities into Equitas, then reinsured the whole thing with Berkshire. No credit event, but the policyholders still got screwed.

CDS doesn’t even come close to working for insurers. Reinsurance brokers have a run at this problem every few years, but the event definition shuts them down every time.

Posted by Dave | Report as abusive

Reinsurer credit risk is accounted for to some extent in both regulatory and ratings models for insurer and reinsurer capital.

Some insurers have bought third party protection on their reinsurer risk. Aspen and Hanover Re are the best known examples. The Aspen deal is recognised as the clearest hedge of reinsurer credit risk and the most innovative.

I wrote some commentary on the issues around transferring credit risk on reinsurance recoverables from some quite extensive experience in the area. It is called “Run off with the recoverables” and it is available from http://www.bravepartners.com

sorry, Chris, I’m not sure the document you linked us to quite does the trick.

http://slingeek.wordpress.com/2009/07/23  /rating-berkshire/

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/