In the current environment, I am an ardent supporter of those who would resist calls to suspend fair value accounting rules. But, when I was at the SEC, I had a front-row seat on what was perhaps one of the most brazen abuses of fair value accounting in history. I was reminded of it by Joseph Stiglitz's recent commentary on CNN.com, in which he characterized the mortgage securitization craze as just another pyramid scheme. Keep that in mind as I tell you the story of Stephen Hoffenberg's $400 million fraud.
Hoffenberg controlled Towers Financial Corporation whose business ostensibly was to purchase junk receivables for just a few cents on the dollar from hospitals, credit card companies, phone companies, etc. Claiming to be able to draw water from stones, Towers would blithely write up the value of the receivables (not strictly 'fair value,' but functionally the same) to create the illusion of collateral and earnings. Investors would be enticed by high interest rates, stellar financial results, apparently healthy cash flow, and a successful track record of paying off prior investors. But like all pyramid schemes, the returns to investors came from only one real source: new investors-cum–victims. The collection operations themselves were pretty much a sham.
All of this occurred in the late '80s and early '90s before business journalists could no longer allow themselves to remain ignorant of accounting matters—save, perhaps the occasional exposé in Barron's. The Justice Department was even worse, eschewing the drudge work of explaining complex financial fraud charges to juries in favor of prosecuting drug pushers. Hoffenberg also flew under the SEC's radar for years by avoiding (perhaps, evading) registration of its debt and equity offerings (think hedge funds).
Hoffenberg might have been able to actually get away with his scheme if he had only known when to quit. But, as is often the case, greed and ego knew no bounds for him. He eventually accumulated enough of other people's money to buy the New York Post from its bankrupt owner, and was practically knighted by the state's governor, Mario Cuomo (by coincidence, the father of the guy whom John McCain now says should be the next SEC chairman). Being on the verge of obtaining permanent financing of his acquisition of the Post by using his ill-gotten gains as collateral finally lit a fire under the regulators; if Hoffenberg were allowed to close the deal, the government's case could turn into an "investigation of a citizen above suspicion" in the eyes of the public.
Less known about the case, but adding to my own deep sense of injustice was that Hoffenberg was able to successfully parry the SEC for a time by trotting out all manner of exorbitantly paid 'expert witnesses' to bless his accounting and business model. Exhibit A was Sidney Davidson, an emeritus professor of the University of Chicago's business school and former member of the Accounting Principles Board. Was Davidson also taken in by Hoffenberg? I think not; Davidson was also allegedly (my information is second hand here) on the payroll of Charles Keating's Lincoln Savings and Loan, one of the great implosions of the S&L era. One of Davidson's 'consulting assignments' was to convince Ernst & Young that the partner who smelled a rat should be taken off the engagement.
Back to the Present
The point of dragging Davidson into this is to show how big money can talk; and it is doing a lot of talk about fair value accounting lately. Invariably, it seems, a very ugly side of human nature seems to come out as a pyramid scheme inevitably starts to lose steam. Those whose investment may be hung out to dry curse the doubters for 'ruining it for everyone else.' That's how I see the proponents of suspending, or even doing away with, the current set of rules we have in place for recognizing certain financial assets at fair value, and taking the gains losses to income.
But, how could fair value accounting be the device by which one scheme was kept alive, yet could have prevented another? Like the Hoffenberg case, there is no question that the two main ingredients of the current fraud were lack of transparency into what was going on, and accounting tricks to give the illusion that all was well. The difference is that in the case of our present extreme unction, it was the ability to hide actual losses (as opposed to create fictitious gains) by not using fair value accounting for junk assets. The answer for the apparent paradox lies in a significant flaw in 'fair value' accounting.
Fair Value is Not Perfect, But It Replacement Cost Would Have Been a Lot Better
First, the limitation of fair value. As I have long argued, I believe in comprehensive application of some form of 'current value,' but 'fair value' ain't it. Fair value is based on exit prices – the highest amount something could be sold for in some market, and there may well be more than one, or even no market to choose from. But 'current replacement cost' is based on the lowest amount that you would have to pay to replace the thing you already have. If the standard for measuring bad loans were current replacement cost, there would have been no way that even Sidney Davidson could have defended a higher replacement cost than the price Hoffenberg had just paid, no matter how much he, the putative master collector, thought he could extract.
The logic of replacement cost also, I believe, takes out the critics of current values at the knees, and also gives them a little something–a higher valuation than fair value. In illiquid financial markets, the question of how much it would cost you to replace a stream of risky cash flows with another contract(s) that provide the same flows with the same risk could be a lot greater than what you could sell them for. Replacement cost relies on the proposition that there will always be a seller if you offer enough money. Stated another way, replacement cost accounting would say that the point on the bid-ask spread to value a security would be the lowest accepted 'ask' price from among all sellers, which in illiquid markets could be significantly higher than the 'bid' price.
Even the proponents of fair value measurements would have to admit that FAS 157 equivocates on where fair value lies on the bid-ask spread. Under replacement cost accounting, the answer is unequivocal. If you want more information on the logical superiority of replacement cost, see here and here for two earlier posts in which I explain the correspondence between replacement cost measures and economic concepts of wealth.
But setting aside the differences between fair value and replacement cost, there are other important observations to make about the mess we are in, and who is proposing what to get us out. First, those who would argue that fair value rules should be suspended during this economic crisis are no different to me than the first potential investors in the pyramid scheme who express concern that there might not be a seat for them when the music stops.
Second, if fair value accounting had been more comprehensively applied, and not less as some would like to see, than at least the bubble would have burst before it became as large as it did; financial institutions would have been forced to take more losses earlier for fair valuing loans and held to maturity investments. For this, the FASB has to take much of the blame for the loopholes they permitted when crafting FAS 115.
Third, it appears that the fair value debate, no matter how cynically one may regard the critics, is finally making it clear that we have to de-link regulatory capital measurement from reporting to investors. If regulators want to rely on a fair value balance sheet, fine. If not, they should make their own, thank you very much.
And another big difference between Towers and the current crisis is that Hoffenberg got 20 years. Today's CEOs are smart enough to take their money and run.