4 ms·
> Uh, Can you document that? http://files.ots.treas.gov/87171.pdf http://files.ots.treas.gov/87171.pdf That report comes from the Office of Thrift supervision
by anamax 18y ago
> Uh, Can you document that?
http://files.ots.treas.gov/87171.pdf http://files.ots.treas.gov/87171.pdf
That report comes from the Office of Thrift supervision (OTS). They started regulating AIG in 1999 when AIG applied to become a "Federal Savings Bank". (The application wasn't approved until 2000.) OTS regulates and supervises both FSBs and their holding companies.
The report also documents OTS cooperation with and the existence of AIG regulation by "a multitude of regulators in over 100 countries".
As the report documents, AIG's problem is liquidity. One of the paragraphs that I found interesting says: "AIGFP's CDS provide credit protection to counterparties on designated portfolios of loans or debt securities. AITFP provided such credit protection on a 'second loss' basis, underwhich it repeatedly reported and disclosed that its payment obligations would arise only after credit losses in the designated portfolion exceeded a specified threshold amount or level of 'first losses.' Also known as 'super senior', AITFP provided protection the layer of credit risk superior to the AAA risk layer. The AIGFP CDS were on the safest portion of the security from a credit perspective. In fact, even today, there have not been credit losses on the AAA risk layer."
FWIW, AIG became concerned about this biz in 2005 and stopped selling these instruments in late 2005 and started unwinding then. I'd agree that they did so too slowly.
> My impression is Alan Greenspan specifically nixed regulation of CDOs around 1999.
Greenspan wasn't ever in a position to make that call.
> Everything I've ever read claimed they were entirely unregulated. I would like to hear if that's not the case.
Yes, there are unregulated financial instruments. (Leases and ownership aren't regulated either.) However, the existence of such instruments does not imply that regulated institutions (such as AIG or its customers) can use them to escape supervision.
For banks and insurance companies, regulators define what counts as an asset, how it is valued, and regulate exposure to different kinds of assets. That applies to all assets, whether they're "regulated" or not.
Note that OTS says that the portfolios that AIG insured have not had losses, the asset valuation has held up even as the market cratered, so it's a little unclear why AIG is making payments. I suspect that they're caught in mark-to-market, which is a regulatory creation. (Assets that are throwing off cash are not worthless even if there are no buyers today.)
Did AIG screw up - yes. But there's no evidence to show that more regulation would have done better. More to the point, it is simply wrong to argue for more regulation on the theory that lack of regulation was a factor. It wasn't. If anything, regulation failed.
- joe_the_user 18y agoHmm, Thank you for the link and for expanding your point. I think a careful read of the document you link to shows that the only regulatory action that took place was the OTC becoming concerned in 2008 that AIG's previous derivative contracts had now become dangerous. They noticed nothing when the derivative contracts actually were issued. Also note the comment in the text: "the level of review and amount of resources needed to asses a complex structure such as AIG's is vastly deeper and more resource-intensive than what would be required for a less complex holding complex", which is something of statement that OTS really wasn't up to very much supervision of this "vastly complex" enterprise. But this is not really the main issue... >> My impression is Alan Greenspan specifically nixed regulation of CDOs around 1999. > Greenspan wasn't ever in a position to make that call. He was indeed not a regulator of that but his testimony to congress influenced the resulting lack-of-legislation on the issue of derivatives when the official who would have regulated them asked for the ability. > Yes, there are unregulated financial instruments. (Leases and ownership aren't regulated either.) However, the existence of such instruments does not imply that regulated institutions (such as AIG or its customers) can use them to escape supervision. > For banks and insurance companies, regulators define what counts as an asset, how it is valued, and regulate exposure to different kinds of assets. That applies to all assets, whether they're "regulated" or not. There indeed, is the rub. Being modeled as "very safe", my impression again, is that CDS were in fact often considered "off-balance-sheet-instruments". But that's again secondary... Now that we have gotten into the details of these things, I should say "there are regulations and there are regulations". Bank deposits are and have been rightfully regulated actively in terms of explicit, apriori limits on deposits because deposit expansion and contraction can have a powerful effect on the economy as a whole. Derivatives have only been regulated in terms of one institution or another keeping loose track of what's happening and not particularly limiting anything until things seem worrisome - which now turns out to be long after the horse has left the barn. What I am saying that derivatives should be regulated in a similar fashion to bank deposits, with limits to who can issue, how much can be issued, how much systemic risk they create and so-forth. This is again consistent with state's necessary regulatory role. State regulation basically require apriori legislation to prevent private actions which can cause harm beyond the ability of a private actor to repay the victims of that harm. The OTS might indeed have looked at the risk of failure which AIG's actions involved for its depositors, its stock holders and even for Uncle Sam in his role as deposit insurer. The OTS clearly didn't look at the systemic risk which AIG's issuance of derivatives created. What should be required, in the future, is such apriori limits on derivatives. Either no institution can take on risk beyond a certain multiple of capital or whatever other formula would work best. This is because, despite many claims to the contrary, unregulated derivatives have become what Warren Buffet calls "financial weapons of mass destruction" through their ability to first expand and then contract the money supply (or the velocity of money if you prefer). > Did AIG screw up - yes. But there's no evidence to show that more regulation would have done better. More to the point, it is simply wrong to argue for more regulation on the theory that lack of regulation was a factor. It wasn't. If anything, regulation failed. Consider that folks are not really concerned with a hypothetical failure of AIG in itself. By money-size, AIG wasn't "too big to fail". It is rather the systemic failure that would accompany AIG's failure that has people up in arms. You could argue that the OTS or other bodies already had the authority needed to place limits on issuance of derivatives. This might be true but the real point is that neither the OTS or other bodies are charged with directly reducing systemic risk because they can't calculate systemic risk. Indeed, consider that Large modern economies (or, as another example, the world's ecology) are complex beyond all mathematical models - despite economists' earlier protests to the contrary (protests that look pretty shoddy now). Because of this, regulation must involve simply forbiding those actions which create unquantifiable systemic risk (including but limited to unlimited dervative issuance, unlimited deposit expansion or discharge of known polutants into the environment). I saw an interview with Nassim Nicholas Taleb where he sketched a future version of capitalism where all investors put their money into either rather strongly regulated funds or very explicitly risky and unregulated funds, with only those who could show they could sustain the loss being allowed to jump into the risky funds.