Saturday, April 10, 2010

Depression, Double-Dip Recession or Lost Decade?

Over at Naked Capitalism last week, there was a guest post by Gonzalo Lira, a Chilean novelist, comparing the "extend-and-pretend" situation we're currently facing with the big banks to the "extend-and-pretend" plan Paul Volcker instituted in 1982 during the Latin American debt crisis.

Lira makes the point that the banks' balance sheets are in horrible shape, which they are, but at the same time, I think he's overstating the size of the holes.  Lira's post is decidedly gloomy, and it doesn't seem to square with all the positive news coming out about the economy last week. So I decided to take a look at Lira's claims.

After my examination, I think I decided that his doomsday claims have some merit, but not overwhelmingly so.  However, I also think the recent spate of positive news should be tempered.

To begin with I looked at Lira's claim that the suspension of mark-to-market accounting standards has made the bank balance sheets completely fictional.  With the news about the repurchase agreements Lehman, BofA and, more likely than not, the rest of the bulge brackets were using, it's quite possible their financial reporting is completely fictional.

But I proceeded with my analysis as if there was some semblance of truth to them (and I figured WellsFargo was the best barometer for that, as they're less like the bulge brackets than the other major banks).  

How did I start?  Well I went back to the stress tests and Mike Konczal's post where he pointed out that the four biggest banks (Citi, JPMorgan, BofA and Wells Fargo) had a combined $477 billion in subordinated loans (2nd Liens) on their books. The government stress tests estimated losses on those loans to be in the 13-15% range, but Konczal pointed out that they were essentially worthless because they're nearly all underwater.  And recent revisions to the stress-test estimates are now raising those estimates significantly. 

For the sake of this argument, I assumed there was at least some salvage value in those loans -- and if home prices rise a little in the recovery, perhaps some can be salvaged.  So I went with Konczal's worst-case scenario that puts losses on those loans at 60%.  That means the big four have to write down $286 billion on those loans. 

Well, contrary to Lira's claim, the banks actually have started writing off and writing down those loans.  In fact between the stress tests and their 2009 10K filings (9 months), they wrote off a combined $43 billion (Citi, BofA and Wells Fargo all wrote off about $8 billion and JPMorgan wrote off $17 billion).  Again, this all assumes that those loans weren't simply shifted off balance sheet.  And, in fact, some of the expensing on the banks' income statements reflects provisions for bad mortgage debt, so it would appear they're trying.  

What's troubling, however, is that even if they're being completely truthful, at their current prorated pace of write-downs ($64 billion annually) it will still take them more than 3-years to shed it all (and that's if the losses on those 2nd Liens are, at max, only 60% and that they'll be able to maintain their current rates of profit-making).

Which brings me back Lira's post. He mentioned that extend-and-pretend will only last as long as the banks can earn profits in the environment that has been artificially created by the Fed (this environment was best documented by Tyler Cowen). It's a statement with which I agree, because I don't believe the "artificial" environment will last for another 3-years. So it looks like the banks are, and will be, back in trouble at some point in the future, but I'm not sure that validates the belief in another, imminent crash either.  

Lira predicts a coming Great Depression but I don't believe the situation is that dire.  Are we looking at a "double-dip" recession? Quite possibly. But it's also possible we're looking at a Japan-style "lost decade" where the banks need ten years or more to write-off the bad debt on their books (this coming on top of the 2000s which were already a lost decade shouldn't make anyone happy, however).  

There is stability in the overnight markets, the banks have avoided a crash, and they are (slowly) earning their way out of the hole. What it all means is that a crash likely won't happen again, but we'll probably see a lot of the same as what we saw through most of 2009 and the first quarter of 2010 -- businesses not being able to get credit, slow-to-meager GDP growth and high unemployment.

This is because the banks' most likely route to profitability -- which also seems to be their current core competency -- is through trading operations, not through long-term, small-business loans (when Yves Smith waded into the comments of Lira's post, she pointed this out by saying, "The equity markets are heavily manipulated right now -- I hear this from virtually every institutional investor I know, including ones at the top of large and respected firms.")

Well of course the equity markets are flooded with cash.  The debt markets are flooded as well.  The major banks don't appear to know any other way to earn short-term profits large enough to cover the holes in their balance sheets.  Which is why we're likely facing a lost decade.  The banks won't begin to properly loan to a credit-starved economy until they fix their own internal finances -- which could be at least three years away.

No comments: