May 20th saw the release of the FDIC’s Quarterly Banking Profile (QBP) for the first quarter of 2010, trumpeting the headline: Industry Net Income Improves to a Two-Year High of $18 Billion. As we’ve shown in previous quarters, the banks have been serially under-reserving for losses in order to show headline profits, despite non-performing assets (NPAs) that continue to rise. We expected a similar story in the current quarter, and indeed we are told that provisions fell in the first quarter in the face of increasing NPAs. In fact, for the first time since 2005, provisions didn’t even cover charge-offs for the period.
But the story has changed a little bit. Wait until you see this one. As Hunter S. Thompson said: “Buy the ticket, take the ride.”
First let’s look at our favorite QBP graph (click to enlarge).
As the graph shows, there has been continued credit deterioration, but the coverage ratio improved thanks to the large increase in reserves. The reserve against loan losses showed a build of $34.5 billion during this quarter, by far the largest single quarter build ever recorded (even larger than 2Q 2008’s $23.3 billion). But wait, the FDIC press release stated that the banks recorded profits of $18 billion as a result of reduced provisions against loan losses. And we already know that charge-offs exceeded provisions during the quarter. Maybe you’re asking the same question we are: so how did the banks manage to build reserves, when provisions fell short of charge-offs?
Spoiler alert: it’s related to the implementation of new accounting standards. In simple terms, the adoption of FAS 166/167 requires that beginning January 1, 2010, companies must bring certain off-balance sheet entities onto their balance sheets. We’ve mentioned this before as it relates to the weekly Fed H.8 report, so we won’t go into more detail here.
According to the FDIC QBP press release (emphasis is our own):
The large jump in reported reserves was associated with the requirements of FASB 166 and 167, as affected institutions converted equity capital directly into reserves. The increased reserves caused the industry’s “coverage ratio” of reserves to noncurrent loans and leases to increase for the first time in 16 quarters, from 58.3 percent to 64.2 percent, even though slightly fewer than half of all insured institutions (49.2 percent) improved their coverage ratios during the quarter.
This reduction in equity capital clearly didn’t flow through the income statement, as reserve builds normally would. That’s because in the adoption of FAS 166/167, companies aren’t required to run these kinds of losses through the current year’s income statement; instead, a one-time adjustment is made in retained earnings on the balance sheet. The rules don’t require restatement of prior period earnings, but strangely enough, there were substantial revisions to previously reported numbers. Remember last quarter’s happy headline of $914 million in profits? That’s been revised away and now stands at a loss of $1.3 billion. The other 3 quarters in 2009 feature nearly $3.5 billion in additional downward revisions to previously reported earnings. We have to assume that these previous restatements are unrelated to the accounting change, but we are simply unsure and there are no notes about previous restatements in the release. The QBP mentions how the implementation of FAS 166/167 affected various other cash flows (interest income, interest expense, etc), but only has this to say about net income:
Application of the accounting changes had no significant effect on the year-over-year increase in the industry’s reported net income; lower provisions for loan losses and reduced expenses for goodwill impairment were the main sources of the improvement in industry earnings.
We’re not sure how to respond to that, except to say that this is a one-time adjustment to the reserve and we would caution against extrapolating future reserve builds of this nature. We would assume that any future reserve building would come out of current period earnings.
Lest we forget our usual FDIC reserve math machinations: even after this quarter’s stealth reserve build, it will take $146.4 billion of future earnings to simply get the coverage ratio back to 100%. That’s one hell of a headwind.