Mark-to-Market, OTTI, Loan Loss Reserves, ETC.

"Accounting Arcana"


 "It's fascinating that the not-so-tiny matter of a
$30 billion loss comes down to accounting arcana,
but it does."

I copied that quote down last night when half asleep and this morning I can't recall where I got it. Now I can't find it. My best guess is the My apologies to the author.

The article was about the unfortunate suicide of a Freddie Mac official, despite the fact that he had recently won a ruling from the SEC worth $30 billion. In my previous post, "Stressing Over Bank Stress Tests," I had tried to make a similar point: How accounting rules, which often seem arbitrary and not well suited to the situation at hand can make a huge difference and how bankers' challenging the results of the recent stress tests have about as good a chance of being "correct" as do the bank supervisors. I didn't feel like I had made my argument clear enough; so, last night I realized that what was missing was the term "accounting arcana."

Accounting arcana can make a huge difference. In my example, I cite reports that The Fed thought Bank of America might need $10 billion more capital while Bank of America thought it might have half a billion more than necessary. In that example, accounting arcane had $10.5 billion on the line. This morning papers indicate that the perceived capital need may be as much as $35 billion. That's a lot of money riding on "one turn of pitch and toss," otherwise known as a regulatory ruling.

Accounting arcana can cover so many issues that the bank might prevail on most of them but be done in by another one. As I've noted previously, there are many accounting issues related to marking assets to market, when impairment is "other than temporary," whether the impaired asset is in a "hold to maturity" account or an "available for sale" account and whether it's possible to change initial classifications. Then there is the whole issue of how much capital must be set aside in loan loss reserves or in reserves for securities mark-down. That is not as clear cut as it sounds either. When traditional risk weights devised in simpler times are applied to some of today's exotic securities, nonsensical answers sometimes result, such as a downgraded mortgage-back security requiring reserves greater than its total value.

Anyway, I wish I had thought of "accounting arcana" earlier. It helps get to what I meant to say. Whether arcane or not, accounting is becoming as toxic as some of the toxic assets being accounted for.

I just heard, once again, on TV this morning that the first step it solving the banking crisis is just to shut down the insolvent banks.  Okay, but my point is simply that knowing which banks are insolvent is not simple. There are many accounting issues that can tip banks from one side of the insolvency line to the other. And the "correct" answer is not at all obvious.

Comments (2)

Trackback URL | Comments RSS Feed

  1. Nemo says:

    Minor correction: The $10.5 billion range you cited yesterday was for Citigroup. The bank needing $35 billion (according to the leaks) is BofA. Which does not change your point.

    As for seizing banks, solvency is only one question. A more interesting one, in my view, is whether “too big to fail is too big to exist”. In my opinion, the answer is yes, but I would be interested to know yours. (Or in general what your take is on the “too big to fail” problem.)

  2. Bob McTeer says:

    You may be right. All these banks are running together now. However, my recollection is that it was BOA with a need of $10 billion, although the next day or the day after that the number had grown to $34 billion. Anyway, as you indicated, which bank it was doesn't matter to the point, which is that these numbers have so many questionable assumptions behind them that there is no clarity as to what is really needed.