If a company has a deal to be acquired for $3 per share, and the deal is supposed to close by the end of the summer, what is it telling you when in early May the stock is trading below $2? Particularly when the company recently released its delayed 10-K with ominous comments about its ability to do business?
Sounds like the deal’s in pretty big trouble, right?
The company is Tower Group, a once-fast-growing property/casualty reinsurer that got over its skis by growing too fast via acquisition, particularly with workers’ compensation business that went sour. Back in January Tower Group agreed to be acquired by ACP Re, a privately held insurer, for $3. Last summer the stock was around $20.
Complicating matters, but also making them more interesting for investors, is Tower Group’s 5% convertible bonds, due to mature in September. As Tower delayed its filings last fall, recognizing that its reserves were inadequate to cover its exposures, the convertibles went from a seemingly safe cash substitute with a big coupon to a highly speculative play. They traded down into the 80s last fall—which may not sound speculative until you remember that the bonds were supposed to pay off at par in a year.
A cursory look at the balance sheet made some investors think Tower had sufficient resources to make good on the bonds. This overlooked, however, the basic reality that Tower’s convertibles reside at the holding-company level, while the cash belongs to Tower’s regulated insurance subsidiaries. Those subsidiaries have severe restrictions on their ability to dividend cash to the parent—especially with their ratings being downgraded, as they have been.
Tower’s recent 10-K filing is not pretty. The excerpt that tells all: “There is substantial doubt about the Company’s ability to remain as a going concern. As such, any potential investor or lender may be unlikely or unwilling to provide additional capital or loans to the company.” Meanwhile, ACP appears to have the right to back out of the deal in the event of material adverse developments with Tower. That’s how the market seems to be reading things.
One possibility is a restriking of the deal at a lower price. This helps explain why the convertible bonds were still, as of this morning, quoted in the low-to-mid 90s. Presumably, bidders for the convertibles think there will be a deal, even if at a lower price, and either the deal will trigger clauses requiring the convertible to be paid or the buyer will simply make good on them as a matter of course.
At 93 cents on the dollar, where the bonds are roughly quoted, buyers would get a 27% annualized return based on the mid-September maturity. If you think the original stock deal will go through, Tower shares at 1.93 imply an annual return of better than 230%!
Perhaps the right way to look at this is to extrapolate a reduced deal price from the convertibles. Just after the deal was announced, the stock traded around $2.94, implying about a 3% return, while the bonds were around $97, suggesting about 9%. But the stock probably had a small component of hope for a higher bid, while the bonds were effectively capped at par. Several weeks later, the bonds settled around 90 and the stock around 2.50. The relationship made more sense there—about 33% annualized for the stock and 28% for the bonds. Clearly a high-risk deal, in any event.
Given that steady-state level, and given the bonds’ current return of about 27%, where does this suggest a deal might be re-struck? Given the delay in the 10-k and the disturbing contents, we’ll assume that shareholders need a bigger boost from what the bonds offer. Let’s call it 50%.
To make an annualized 50% from current levels, Tower shareholders would be figuring on getting paid $2.25 in late summer when the bonds mature. Given the quote in the convertibles and the overall risk profile, that’s one guess as to where the deal might be re-struck. But guess is the operative word. There is a reason why they call it risk arbitrage.
EDIT: Original piece submitted May 7, 2014 at 1:21 pm. On May 8, 2014, Tower and ACP announced that the deal had been re-struck at $2.50 per share.