In mid-2008, hedge fund manager David Einhorn took notice after Lehman Brothers made either an egregious typo or lied through its teeth. Lehman's "Level 3 assets" presented in its conference call were different from those disclosed in its SEC filings. Einhorn confronted Lehman management, who essentially told him to shut up and mind his own business. Lehman went bankrupt four months later.
And that's just one example of Wall Street's quest to blast a hole between truth and reality. It gets better.
Just before its death, Lehman used something called "repo 105" transactions to make chunks of its balance sheet disappear during the days it reported to shareholders, only to reappear days later.
In a repo loan, a bank borrows money from an investor. The bank backs that loan by lending the investor assets and promising to repurchase them at a set time at a slightly higher price (hence "repo"). The bank doesn't really sell these assets; it just posts them as temporary collateral, so for reporting purposes they're still part of its balance sheet.
But accounting rules say this is only the case if the collateral is worth between 98% and 102% of the repo loan's value. So all Lehman had to do was overcollateralize the loan by 105% (hence "repo 105"), and voila, the assets could be classified as "sold," even though by definition it was obligated to repurchase them in the future. As long as that future was after Lehman reported results, management could spew some cock-and-bull story about getting rid of toxic assets and reducing leverage. None of which was true.
Surprised? I wouldn't be. These accounting pranks happen more often than you might think. Here are other games banks play to mess with reality.
1. The quarterly pump 'n' dump
The balance sheet doesn't show results over a given period. It just shows what a company looks like at a specific point in time. Investors only get four of these snapshots per year -- on the last day of each quarter. That leaves 361 days for management to go nuts behind closed doors.
Remember the ridiculous 30-to-1 leverage ratio stories we heard about? Forget those. Former investment banker William Cohan explains in The House of Cards how it really works:
Before the [leverage] ratio was published at the end of each quarter, investment banks would take the necessary steps to sell enough of the assets to get the leverage down to a more "acceptable" ratio.
In the meantime, leverage "often approached 50:1 during the middle of the quarter," Cohan writes. Yeehaw!
2. Rewriting history
In late 2008, Goldman Sachs (NYSE: GS ) switched from a fiscal year ending in November to one ending Dec. 31. In doing so, December 2008 was literally eliminated from its Q1 2009 results. It became an orphan month. Not surprisingly, Goldman stuffed as much crap into this orphaned month as it possibly could. In Q1 2009, Goldman reported net income of $1.81 billion ... kind of. The orphaned December generated a $780 million loss. This completely important fact was buried on page 10 of its press release without a single mention in the much-trumpeted headline.
Goldman's CEO later bragged that "Our results reflect the strength and diversity of our client franchise, the resilience of our business model and the dedication and focus of our people."
That, and awesome amounts of arrogance.
3. Heads, I win. Tails, go fly a kite.
Wells Fargo (NYSE: WFC ) posted a $4 billion pre-tax profit in Q4 2009, nearly half of which came from writing up the value of its mortgage-servicing rights (MSRs).
There's nothing wrong with writing up MSRs. But Wells Fargo marked up both its MSRs and the corresponding hedges that are designed to (and historically did) move in the opposite direction -- by $1.1 billion and $830 million, respectively.
How'd that happen? Well, the hedges are typically Level 1 or Level 2, meaning their value is based on market prices, or something close. The MSRs, however, are Level 3, meaning management can value them based on whatever twisted logic it sees fit.
If both the MSRs and the hedges increased in value at the same time, then management's valuation estimate directly contradicts the view given by the rest of the market. Someone's probably wrong. Guess who.
4. The bankruptcy bonanza
A group of clever accounting regulators once got together and decided that being flung toward bankruptcy was to henceforth be known as "record profits."
This whack-job rule goes like this: When the market thinks a bank is about to fail, it sells its debt. The bank can then theoretically repurchase that debt on the cheap, discharging a liability at less than par value. Ergo, huge profits! Best of all, it can do this even if it doesn't actually repurchase the debt. Just theorize it and call it good.
On Sept. 16, 2008, Morgan Stanley (NYSE: MS ) was in the midst of a bank run that nearly cost it its life. In an attempt to quell the fear, it released earnings showing a $1.4 billion profit. But every penny of that came from "the widening of Morgan Stanley's credit spreads on certain long-term debt" In other words, investors betting it was about to fail.
Citigroup (NYSE: C ) and Bank of America (NYSE: BAC ) pulled similar stunts. In April 2009, Citigroup reported a $1.6 billion profit only after booking $2.5 billion in gains from its own debt getting blown to shreds amid insolvency fears. B of A then followed with a $4.2 billion profit, $2.2 billion of which came from the erosion of its own bonds. Management accepted praise for a job well done.
Your move, dear investor
These accounting tricks aren't the scariest part. What's scary is that absolutely nothing has changed since all this occurred. The rules are the same. The laws are untouched. And management is still incentivized to tweak short-term results at any cost. Until there's a dramatic regulatory overhaul, it's crazy to assume this stuff won't happen again -- or isn't happening right now.
That's why I'm still avoiding banks like the plague.
Be sure to check back every Tuesday and Friday for Morgan Housel's columns.