Taking some profits with Radian Group and MGIC

I took some of my position in Radian Group and MGIC off the table yesterday. I sold half of my position in MGIC and about 30% of my position in Radian.

The stocks have moved quite a lot in a short period of time. I was buying MGIC at under 80 cents a month ago. I bought Radian in the mid-$2’s in July. These are pretty big moves.

The fundamentals behind the stocks are improving. My thesis that the housing market is bottoming is showing more evidence of playing out. Delinquencies are falling, foreclosure rates are falling.

Yet while I remain optimistic that they can outrun their legacy book, the possibility remains that they won’t.

There were a couple of events today that outlined that uncertainty.

MGIC Conserves Cash

First, MGIC announced last night in an 8-K filing that they would be deferring the October interest payment of a junior convertible debenture.

The announcement isn’t as bad as it sounds. Deferment of interest payments is within the terms of the debt. This isn’t a default.

It does, however, underscore the uncertainty. MGIC is still in negotiations with Freddie Mac to settle a dispute on an amount of money large enough to make or break the company. We have heard nothing about how those negotiations are progressing. The deferment is likely just prudent cash flow management linked to Freddie’s requirement that MGIC downstream $200 million of capital from the holding company to the main insurance sub.  Still, we don’t know that for sure.

We also don’t know how much the eventual settlement with Freddie Mac will eat into statutory capital. Until we know that its difficult to handicap the probability of survival of MGIC. It was a lot easier to accept that uncertainty at $0.80 than it is at $1.60. So I pruned my bet down to the same dollar value it was at then.

Radian gets some attention

As regards Radian, a very good article was published on SeekingAlpha today that developed a case to short the stock.

I’ve read through the article a few times. It raises some valid concerns.  I learned a few new things and it forced me to research some others.  It is written from the perspective of a short seller so there are a few conclusions that are on drawn on circumstantial evidence and there are a couple of worst case scenarios have been extrapolated out, but hey, when you are short the stock that is your job.   Nothing was written that isn’t out of the realms of possibility. I want to step through the points made and address them.

The first point  is that the statutory capital held by Radian Guaranty (Radian Guaranty is the mortgage insurance subsidiary) is not all easily accessible by them. This is because some of the capital counted against Radian Guaranty’s risk in force is locked up in another sub, Radian Asset Assurance (RAA).  While RAA is wholly owned by Radian Guaranty, because it is itself an insurance sub, Radian Guaranty would have to get the permission of the insurance regulator to upstream capital. In the author’s opinion this wouldn’t happen.

While Radian Guaranty technically has $3.8 billion of capital, $1.5 billion of that capital is tied up in RAA (actually to be technically correct the statutory filing states that only about $1.15 billion is tied up at RAA but the rest of the amount is tied up at other insurance subs that suffer from the same problem).

I really don’t know how this would play out in the event Radian needs the cash. I’m also not so sure that its that big of a worry. It only matters if Radian needs the cash. And for that to happen, Radian has to pay more claims. If you take a negative view on Radian’s legacy book, you can paint an ugly scenario.  At current claim rates, it is less likely to be an event.

Without the $1.5 billion, Radian Guaranty has $2.3 billion in cash and investments on hand to pay claims. The company paid claims of $263 million in Q2 after paying $218 million in Q1. The company is expecting paid claims of a little over $300 million per quarter in Q3 and Q4. Against that Radian is earning about $215 million  in premiums and investment income each quarter and spending about $55 million each quarter in operations. So they are paying net cash of about $140 million per quarter. The losses would have to continue for some time for them to eat far enough into Radian’s claim paying resources for this to matter.

The article argues that Radian has been making up for the deficiency in cash by writing single premium business at below market rates, and by denying claims in order to stretch out the payments to a longer horizon.  By writing single premium business Radian is raising cash up front.  By denying claims Radian is delaying payments that would drain their capital.  The article suggested that once those denials taper off, claims paid per quarter will rise to $400 million.

MGIC said in their Q2 conference call that there was a competitor that was writing single premium business at rates they couldn’t compete with. I suspected that was Radian. It turns out it probably is.

Yet Radian isn’t writing an unusual amount of single premium business. Looking at their breakdown between single and monthly in their second quarter release, monthly premiums made up 67% of business written, which is actually more than it was in the same quarter last year. If Radian is undercutting on rates, they don’t appear to be pressing too hard.

As for the question of denials, its a topic I have written about myself. It remains the single biggest reason why I sold some stock today. The denial trend at Radian is without precedent. It is also concentrated within a single servicer.  I don’t have a good feeling for how it will play out.

The crux of the article’s argument is that the denials are going to come back to Radian, and when they do Radian is going to be hit with a much higher claims rate than they have been.  Yet while the author expressed confidence that Radian was using denials as a technique to both delay paying claims (he pointed out that the loan bucket in Radian’s pending claim category is almost double what MGIC’s is), and to limit claims on higher dollar value loans (Radian’s average claim paid has dropped from $55,000 to $50,000 ), I can only say that I do not know the motives, only that there are question marks.

I wrote in a previous piece that Radian has proven accurate in its estimates of rescissions and denials in the past. I also concluded that the evidence thus far suggests that the single servicer denials are resolving in line with Radian’s estimates. But I also concluded that it was too early to draw a firm conclusion on whether this trend would hold up, and that we would have to wait until the 3rd and 4th quarter before we could establish whether more or less of Radian’s denials were being put back on the company.

I also think the estimate of $400 million claims per quarter is a worst case scenario.  When I look at Radian’s monthly claims and rescissions and denials data for the last year, in order to average $400 million of paid claims per quarter I have to assume all of the rescissions and denials eventually result in claims paid, and that the average amount paid per claim rises from $50,000 to $55,000.   These assumptions are both consistent with the articles assertions, but I tend to be more optimistic about how the next few quarters are going to play out.

There are also some lessor points made that I won’t dwell on too much.  Radian’s risk in force has a higher concentration in California, and the article asserts that Radian has a more risky book in general. I checked the numbers and this is true. Radian has higher exposure in California and in Illinois, and the two companies have similar exposure in Florida.

However, if you look at the books by vintage, MGIC has a higher amount of risk in force in 2007 (24.4% versus 20.3% for Radian) and in 2006 (13.5% versus 9.3% for Radian). So it depends how you slice this apple.  I don’t think I can draw a conclusion one way or the other in this respect.

There were also some comments about Radian’s IBNR reserve but I don’t fully understand the argument being made there so I will just leave that one alone.

Overall I appreciated the article. It made me think hard about my investment in Radian. Even though I continue to draw more optimistic conclusions about Radian’s future, it made me reconsider the uncertainty in the stock, and how, in the interests of prudent risk management, I should probably take some profits. So that’s what I did.