Guest Post: Handicapping the Stress Test, TCE data @ 3/31

Submitted by Rolfe Winkler, publisher of OptionARMageddon

Ahead of official announcements regarding stress test results, OA thought we’d publish our latest update for banks’ tangible common equity, a metric that is likely to figure prominently in the results.

A recent Reuters report said “U.S. regulators want the top 19 banks being stress-tested to have at least 3% [TCE].” In other words, regulators want leverage ratios below 33x.* Surreal, no? That the banking system has grown so bloated that 32x leverage can be considered “healthy?”

Using the 3% Test, the results for the nation’s nine largest banks are mixed…..four pass, five fail…..

And by the way, this is before “stressing” the balance sheet per future “adverse” scenarios. As you can see, most banks fail the test before they even sit for it…

(Click to enlarge)

To be clear, this is not a prediction of the government’s verdict. As Jack Ciesielski of The Analyst’s Accounting Observer points out: “there is no iconic definition of TCE. Treasury may come up with one of their own that takes into account questionable items” so that all the banks pass. That would be consistent with early reports….

The banks themselves have varied definitions of TCE. The measure is supposed to be the true acid test of bank capital, which means it should be calculated conservatively. ALL intangible assets have to be backed out.* To calculate the TCE Ratio in the highlighted column, I’m using a very conservative calculation.**

Banks’ own methodology for calculating TCE varies. The “% overstated” column is meant to show which ones have taken the most liberties. (Interested parties can contact OA via e-mail to purchase our data set containing more detailed info about each bank’s calculation methodology, as well as quarterly TCE data and Level 1/2/3 assets dating back to Q1 ’08.)

A huge caveat with this data is that these companies have off-balance sheet exposures. Some of them huge. And many have big chunks of “other assets” on balance sheet, some of which may be intangible. When there is disclosure for these, it varies. So to keep the data consistent, both are excluded. Hopefully the stress testers got a decent view of these risk buckets in order to factor them into results….though I wouldn’t count on it….

Incidentally, I added the right-most column in order to give readers a sense for the degree of vulnerability on the asset side of bank balance sheets. Level 2 (“mark to model”) and Level 3 (“mark to myth”) asset values are the ones over which management has the most discretion. These days management can’t be trusted so it’s our bet that those with the biggest discretionary buckets—JPM, BofA, C—are the ones sitting on the largest losses.

—————–

*All the banks fail to back out mortgage servicing rights from TCE, even though these are intangibles under GAAP. Some banks (BofA, Wells, Chase) have substantial MSRs, others (GS, MS) don’t. Some banks add back “def’d tax liabilities” related to intangibles (BNY Mellon, Wells, Chase, PNC). This is not conservative. Two banks (Wells, PNC) goosed TCE by reclassifying “noncontrolling minority interests” from liabilities to equity.

**TCE = shareholder’s equity (excluding noncontrolling minority interests) – goodwill – intangibles – preferred – MSRs.

Fair Disclosure: OA has short positions in companies mentioned in this post.

Print Friendly
Tweet about this on TwitterDigg thisShare on Reddit0Share on StumbleUpon0Share on Facebook0Share on LinkedIn0Share on Google+0Buffer this pageEmail this to someone

9 comments

  1. stoverny

    No matter how you slice it, it seems that Citi passing the stress test will be a completely transparent sham.

  2. tyaresun

    The 3% TCE is supposed to be adequate to meet the adverse scenario? That itself is questionable.

  3. Rolfe Winkler

    tyaresun…..great point. It’s surreal that 33x leverage is the new standard separating the weak from the strong….

  4. James B

    Rolfe,
    To follow up on tyaresun’s comment, is the standard supposed to be that banks have 3% TCE now, or that they can maintain 3% TCE under the stress test scenarios? The whole “stress test” part of the stress test would seem kind of irrelevant if they’re basing their pass-fail decisions on current ratios.

  5. Rolfe Winkler

    James…I think you’re right, that the point of the test is to see if they can maintain 3% TCE under “adverse” scenarios. I guess my point, which I should have made more directly, is that most of them fail the test going in…

  6. Elizabeth

    Isn’t it a little harsh to give zero weight to MSRs? They are equivalent to IO mortgage backed strips. Yes, highly volatile but not zero.

  7. cap vandal

    I think you are being a bit harsh, but no problem with a different pov.

    I don't get the problem with throwing out a deferred tax liability on an intangible asset if you toss out the asset. Basic matching.

    Also, your 2 big investment banks do the best on your revised metric. I don't think they are really in the same business, in spite of the effort to turn them into commercial banks to bail them out.

    A major point of TARP was to improve Tier I capital via preferred stock. If you want to look at capital ratios excl preferred stock, thats fine, but it isn't the only reality.

    As far as MSR's are concerned, I was VERY negative about them last year, since I thought that they would end up having negative value under a scenario with a lot of defaults, but that doesn't seem to be the case. At least Wells seems to be able to wring enough extra fee income to offset the extra expenses.

    If you look at goodwill, which we all agree isn't tangible, it is also associated in some cases with very low cost deposits. That's how it got there — buying profitable banks above book value. Low cost deposits have an economic value that some of the lower TCE banks have (Wells) and that the investment banks don't.

    I can't argue with the basic thesis that the banks could use more capital.

    Then you have your hierarchy of capital and we know that deposits aren't capital and vanilla common equity is — but there isn't much controversy regarding the priority of claims. Junior debt (preferred) and minority interests go before senior debt and deposits.

    The point of the whole exercise has been lost in rhetoric. We don't want undercapitalized banks because they lend too little (hoard cash) and are hoarding zombies or because they are reckless and lending too much — like zombie S&L's. People can't figure out which types of zombies we have.

    This leads me to believe that there are all sorts of assumptions that make no sense.

    The most interesting finding — with GS and MS on the top — leads me to believe that this is skewed to favor investment banking. Since they have large trading books, and they don't have deposit bases or the same types of fee income, they damn well need more capital.

  8. Hu Flung Pu

    **TCE = common shareholder’s equity (excluding noncontrolling minority interests) – goodwill – intangibles – preferred – MSRs

    ***************

    I realize that these banks need capital, but can we at least get the definitions straight?

    TCE = Tangible COMMON Equity. Thus, you don’t subtract out preferred (from common shareholders’ equity), as it wasn’t included in the first place.

    It’s hard enough to have an intelligent conversation on these issues when everyone actually has the nomenclature correct.

    Uggh.

  9. Rolfe Winkler

    Hu Flung Pu–don’t worry…the figure is the correct calculation as per your comment. my formula the way it was typed is incorrect. I start from shareholder’s equity and then back out preferred. I’m not backing it out twice.

    (It was 5AM when I completed this work, hopefully folks will pardon the typo!)

Comments are closed.