by twolf
Remember that time we gave AIG, say, $150bn. Boy that was fun, wasn’t it? AIG’s 10-Q for the second quarter is now available, so let’s see how much our 80% of the company is worth right now. Not much, says one analyst.
First the good news. AIG reported a profit in the second quarter of $1.8bn, compared to a loss in the first quarter of $5.1bn. See p.5, which requires some arithmetic. Look at the loss line, not the loss attributable to AIG line, we don’t care why they lost money. Second, it looks like the company has a positive balance sheet net worth, some $58bn at June 30 (p. 7). Balance sheets historically are fairly conservative, so we can hope it’s actually a bit better.
The credit default swap portfolio continues to be a problem. AIG recognized losses on that portfolio of $2.4bn, which had previously been included as unrealized losses. That’s actually good news, because it wasn’t worse than previously thought. The portfolio improved in value somewhat, $636mn in the second quarter, and $184mn for the first six months. There is a discussion of the calculation of these values on page 29. The estimates have used consistent methodology so they are comparable results, but as I have previously noted, Wall Street models are suspect. These may be especially suspect:
AIGFP uses a modified version of the Binomial Expansion Technique (BET) model to value its credit default swap portfolio written on super senior tranches of multi-sector collateralized debt obligations (CDOs) of ABS [asset backed securities] …. The BET model was developed in 1996 by a major rating agency to generate expected loss estimates for CDO tranches and derive a credit rating for those tranches, and has been widely used ever since.
A rating agency did the model? Hmmmm. I’d like a bit more justification.
The CDS portfolio includes three sectors. The regulatory capital sector was written to help banks, primarily European, reduce their need for capital. AIG continues to expect little loss on that portfolio, and expects most of it to terminate at no cost as regulatory changes take effect. This portfolio is down another $57bn in net notional amount in the first half.
AIG has worked off the more dangerous arbitrage sector from a net notional amount of $63bn to $50bn, which is good, and estimates that its liability has decreased. The portfolio is complicated and difficult to evaluate. See page 60 for a discussion. The single-name portfolio is small, with a net notional value of $3bn, which is the maximum exposure.
The new problem is weakness in the insurance companies controlled by AIG. Premium revenue is down, and AIG has had to borrow $2.4bn from the Treasury to pump up their capital. Page 91. That is a bad sign. The original idea was that we would come out of this mess by selling off some of those companies, and weakness means that they may not be as valuable. Indeed, AIG reports premiums of $36.6bn in the first half, down $5.8bn in the first half of 2008. This 13.7% decrease is attributed to economic circumstances, but AIG admits that its brands are tarnished, and that concerns about its financial strength may be a problem as well. Page 108.
This leads to the question: what are we likely to recover. One analyst at Citigroup says there is a 70% chance there is not one penny of equity in AIG.
In a note to clients [July 9], Citi Investment Research analyst Joshua Shanker said the continued risk of more credit default swap losses and its management’s eagerness to sell off businesses at a low value jeopardizes AIG’s equity position.
I haven’t found Shanker’s paper on the interwebs, so I don’t know how he evaluated the either the CDS risks or the prices AIG is getting for sales of its insurance assets. We could be a long-term investor in Chartis and other newly branded insurance companies for the foreseeable future. On the other hand, GM says it will do an initial public offering within a year. We can hope to be as lucky with our investments as private equity funds expect to be with theirs.
Related posts:
- Former Insurance Industry Exec: Baucus Plan a “Gift” to Insurance Companies
- Republicans Move to Permit Credit Card Companies to Jack Up Their Rates for the Next Several Weeks
- Medicaid for All: An Alternative To Subsidizing Insurance Companies
- Max Tax Allows Insurance Companies to Suck Cash Directly from Treasury
- Only the Insurance Companies Want a Level Playing Field
Spotlight
Support this site!
Subscribe to the newsletter
Advertise on Firedoglake
Send
us your tips
Make us your homepage
About Firedoglake
Advanced search
BWA-HA-hahahaha!
Seriously though, I don’t think I will hold my breath on waiting for the taxpayer/owners to get a deal comparable to what the private equity get, even though the private equity wouldn’t be getting diddly-squat without the taxpayer dollars.
I was wondering if their profits were real or just paper – or possibly only because we were bailing them out.
Thanks!
“Remember that time we gave AIG, say, $150bn. Boy that was fun, wasn’t it?”
Especially that “no Vaseline” part. That enhancement would undoubtedly have been another $100bn, give or take a few bucks.
Thanks for this summary. I still have a bad feeling about this.
I’d prefer to hear that there are some serious fraud investigations that will result in something more serious than the dainty wrist slaps that have been meted out thus far.
Bob in HI
Imagine if that money had gone to Detroit instead. The recession really would be over by now, and GM would be rolling the first Chevy Volts into the showrooms.
Book Salon a couple of flights upstairs with Chris Mooney’s Unscientific American hosted by JD Stemwedel
The revelation today that the Obama administration has worked out a sweetheart deal with big pharma should be the straw that tips the scales. It’s obvious whose side he’s on. We should have known the day after the election when he appointed Rham Emanuel COS. It’s time for progressives to let him know he has lost trust and support. The U.S. doesn’t need 8 years of Bush Lite. Time to start thinking about making him a one termer and finding a challenger in the primaries. Here’s an interesting essay.
http://www.infowars.com/progre…..rat-party/
These people are ridiculously easy to pull apart (demonstrate their methods is invalid and its use negligent), here’s a BET explanation:
http://pointlessly.blogspot.co…..nsion.html
“
- First, the collateral pool is mapped to a hypothetical portfolio of N uncorrelated assets (the diversity score).
- The default probability of assets in the hypothetical portfolio is then calculated with WARF (Weighted Average Rating Factor).
- Since the default event of each asset is uncorrelated, the probability of J defaults in the portfolio simply follows binomial distribution:”
The first statement is an assumption “portfolio of N uncorrelated assets.” In the economic system, there are no uncorrelated assets, as we are bitterly discovering.
Two recently provide factors that provide correlation among assets are:
- House Prices
- Unemployment
We’re done with this BET risk management technique.
As was recognized in the BIS (bank of international settlement) paper:
http://www.bis.org/publ/work163.pdf
“However, if the actual intra-sector correlation is greater than 20%,
Moody’s old approach runs the risk of underestimating expected losses and over-rating notes by applying a DS that is too high.”
and
“On the other hand, it would appear that the BET underestimates EL when the collateral assets are correlated, with the degree of underestimation rising in the subordination level.”
No shit: “degree of underestimation rising in the subordination level” and what did our wonderful mashers of the universe do? Sold a huge number of 80%+20% (The 20% was a second loan subordinated to the 80% loan), and these loans were directly correlated by being paid by the same homeowner.
Synoia, thanks for that explanation. Your comments and links are really helpful to me as I try to work with what I used to know about statistics for these posts.
The only things this method has going for it are that all of the financial elites use it, so all the players are probably using similar estimates in their planning; and it provides some consistency across time, so the estimates at different dates at least use the same methodology.
AIG says it has tweaked the assumptions, see page 29 of the 10-Q, link above. If you have time, I’d appreciate your thoughts on their discussion of this calculation for the multi-sector CDO portfolio, particularly whether their assumptions make sense.
I agree that using BET’s is problematic in that the underlying ‘risk’ analysis (implicitly) assumes that a worse case scanario (e.g. unparallelled failures in mortgage payments as a result of a collapse in housing prices) could NOT be factored in as there was no (recent) historic parallel. This is the typical failure of the ‘random-walk’ paradigm that all future ourcomes are discounted by current prices. Keynes showed in his General Theory (1936) that market ‘perfection’ required that optimal market clearance (=> no ‘unemployed’ resources) could not be achieved as correct discounting across all markets, at all times, was impossible. Thus government intervention to prevent sub-optimal clearing (=> unemployed resources) is necessary.
IMHO, an easier explanation of this current turn might be to look at what the insurance subsidiaries actually comprise. Value (discounted net revenue streams => stock price) is the sum of both premium income and market returns. Inurance companies will typically try to match expected cash flows, e.g. by purchasing a bond against an annuity (price of bond annuity receipt, bond interest annuity payouts) & making the spread. Whilst Treasury bond prices have gone through the roof, the widening Treasury/Corporate spreads (& indeed the general impossibility/volatility of valuing corporates) have substantially increased insurance co. risks. In addition, burgeoning cash flows have become MUCH more difficult for the insurance companies to invest given the above-mentioned valuation issues and historically low interest rates. It is not surprising that under such uncertainties, especially given more suspicious regulators post Sept ‘08, that ‘conservative’ pricing has become the ‘norm’.
In 2010 and 2012 I’m voting straight against the incumbent, regardless of the party of their opponent.