This is one of the many times that I wish RealMoney.com had not changed its file structure, losing virtually all content prior to 2008. (It is also a reason that I am glad I started blogging. It’s more difficult to lose this content.) When I was a stock analyst at Hovde Capital Advisors, I made 2 humongous blunders. I wrote about them fairly extensively at RealMoney as the situation unfolded, so if I had those posts, it would make the following article better. As it is, I am going to have to go from memory, because both companies are no longer in business. Here we go:
Sustainable competitive advantage is difficult to find in insurance. Proprietary methods are as good as the employees creating and using them, and they can leave when they would like to. This applies to underwriting, investing, and expense management. What else is there in an insurance company? There are back end processes of valuation and cash flow management, but those financial reporting processes serve to inform the front end of how an insurer operates.
One area that had and continues to have sustainable competitive advantage is life reinsurance. An global oligopoly of companies grew organically and through acquisitions to become dominant in life reinsurance. Their knowledge and mortality databases make them far more knowledgeable the life insurers that seek to pass some of the risk of the death of their policyholders to them. They can be very profitable and stable. I already owned shares of RGA for Hovde, and in 2005 wanted to expand the position by buying some of the cheaper and more junior company Scottish Re.
Scottish Re had only been in business since 1998, versus RGA since 1973. These were the only pure play life reinsurers in the world. Scottish Re had grown organically and through acquisition to become the #5 member of the oligopoly. The top 5 life reinsurers controlled 80% of the global market. I made the case to the team at Hovde, and we took a medium-sized position.
The first thing I should have noticed was the high level of complexity of the holding company structure. Unlike RGA, they operated to a high degree in a wide number of offshore tax and insurance haven domiciles — notably Bermuda, Ireland, Cayman Islands, and others. Second, their ownership diagrams rivaled AIG for complexity, and their market capitalization was less than 2% of AIG’s at the time. [Note: balance sheet complexity did not bode well for AIG either — down 98% since then, but it beats Scottish Re going out at zero.]
The second thing I should have noticed was the high degree of underwriting leverage. Relative to RGA, it reinsured much more life risk relative to the size of its balance sheet.
The third thing I should have noticed was the cleverness of some of the financing methods of Scottish Re — securitization was uncommon in life reinsurance, and they were doing it successfully.
The final thing that I should have noticed was that earnings quality was poor. They usually made their earnings, but often because their tax rate was so low… and the deferred tax assets were a large part of book value. (Note: deferred tax assets only have value if you are going to have pretax income in the future. That was soon not to be.)
In 2005, Scottish Re won the auction for buying up another member of the oligopoly, ING Life Re. I asked the CFO of RGA why they didn’t buy it, and his comment was that he didn’t think anyone would pay more than they bid. That should have led me to sell, but I didn’t. The price of Scottish Re drifted down, until August 3, 2006, when they announced second quarter earnings, reporting a huge loss, writing off a large portion of their deferred tax assets, and the stock price dropped 75% in one day. I eventually wrote about that at RealMoney, noting it was the single worst day in the hedge funds history, and it was due to my errors. You can also read my questions/comments from the conference call here (pages 50-53).
If you look at the RealMoney article, you might note that we tripled our position at around $6.90 after the disaster. That took a lot of guts, and we didn’t know it then, but it was the wrong thing to do. The stock rallied all the way up to $10 or so. If it hit $11, we were going to sell out. That was not to be.
I spent hours and hours going through obscure insurance filings. I analyzed every document that I could get my hands on including the rating agency analyses, because they had access to inside data in aggregate that no one else had outside of the company. The one consistent thing that I learned was that insolvency was unlikely — which would later prove wrong.
The stock price fell and fell all the way down to $3, with rumors of insolvency swirling, when Mass Mutual and Cerberus rode to the rescue on November 27, 2006, buying 69% of the company for a paltry $600 million in convertible preferred stock. At that point, I finally got it right. All of my prior research had some value, because when I read through the documents that day and saw the liquidity raised relative to the amount of ownership handed over. Given the data that they now handed out, I concluded that Scottish Re was worth $1/share, and possibly zero.
But there was a relief rally that day, and we sold into it. We ended up selling about 4% of the total market cap of Scottish Re that day at a price of $6.25.
The bright side of the whole matter was that we could have lost a lot more. Scottish Re was eventually worth zero, and Mass Mutual and Cerberus took significant losses, as did the remaining shareholders.
As it was, the fault was all mine — my colleagues at Hovde deserved none of the blame.
The Lesson Learned
One year later, I wrote a note to the late Greg Newton who wrote the notable blog, Naked Shorts, when he was critical of Cerberus (they had a lot of failures in that era). This was the summary that I gave him on Scottish Re:
Cerberus got into SCT @ $3; it’s now around $2. For me, on the bright side, when their deal with SCT was announced, I quickly went through the data, and recommended selling. We got out @ $6.25. That limited our losses, but it was still my biggest failure when I was at Hovde. The mixture of leverage, alien domiciled subsidiaries, reinsurance underwriting leverage, plus complex and novel securitization structures was pure poison. I was mesmerized by the seemingly cheap valuation and actuarial studies that indicated that mortality experience was a little better than expected. I violated my leverage and simplicity rules on that one.
He gave me a very kind response, better than I deserved. As it was Scottish Re went dark, delisting in May 2008, and trading for about a nickel per share at the last 10K in July of 2008. It eventually went to zero.
The biggest lesson is to do the research better on illiquid and opaque financial companies, or, avoid them entirely. Complexity and leverage there are typically not rewarded. I’d like to say that I fully learned my lesson there, but I got whacked again by the same lesson on a personal investment later in 2008. That’s a subject for a later article.
I have one more bad equity investment from my hedge fund days, and I will write about that sometime soon, to end this part of the series.
Full disclosure: still long RGA for my clients and me