The Accountants Didn’t Let Us Down

An article appearing in the New York Times today (via Yahoo Finance) really upset me.  The article, entitled The Accountants Misled Us Into Crisis, is as blatant an example as I’ve ever seen of irresponsible, poorly-researched journalism.  Despite his impressive credentials, here is a man who clearly doesn’t understand the nature of markets, the fundamentals of underlying asset value, the finer points of statistical inference, or the significance of rare events.  But then, journalists (and accountants) rarely do.

Accounting Wasn’t The Problem

Anytime a rare or random event happens (and yes, sometimes events truly are random), the media rushes to assign a cause.  The interesting thing about it is that if the event is truly random, then by definition there is no readily-assignable cause.  We humans live in a world dominated by unpredictability in which the most innocuous of coincidences can cascade into toward disaster, or astounding success, at the blink of an eye.  Simply put, often times there’s nothing you can do to predict or prevent it.  Needless to say, this fact doesn’t sit well with people in general or journalists in particular, who are paid to see patterns where none exist.

As I sit typing this, a market recap on Yahoo Finance reads

“The stock market started the session with a loss…as several major foreign markets were knocked lower by profit takers.“

The question you should immediately ask on reading this, of course, is “what makes you so sure it was profit takers?”  The market could have been lower for any of tens of thousands (if not millions) of reasons, or even for no reason at all.  A 0.7% loss is well within a standard deviation of the average daily volatility.  It is entirely possible, if not outright probable, the 0.7% drop was due to nothing more than statistical noise and therefore doesn’t require any explanation at all.  Even were the drop to have some deeper meaning, it’s unlikely the uncoordinated buying and selling of tens of thousands of market players all over the world would magically result in a single over-arching purpose:  profit taking.  A 0.7% move requires no explanation.  A 10% more is a different story…

My purpose is merely to point out that that the financial crisis and subsequent recession could have been nobody’s fault.  It could have been nothing but pure bad luck:  a black swan.  Remember, an event with one in a trillion chance has to happen sometime, and there’s a first time for everything.  It flies contrary to human nature to admit there was nothing we could do to prevent the most recent recession, but you can’t rule out the possibility.  To do so would be illogical.

Back To Accounting

Am I implying by the above paragraphs I think the accounting and financial reporting systems in this nation are perfect?  Not at all, I’m merely saying I don’t think they were a significant contributer to our current situation.  Here’s why.

In his article, Norris quotes Robert H. Herz, the chairman of the Financial Accounting Standards Board as saying

“There were important aspects of our entire financial system that were operating like a Wild West show, huge unregulated opaque markets.”

Herz may be correct in stating the markets were unregulated, but they were certainly not opaque.  Investors knew exactly what was going on, or at the very least could have found out with a bit of due diligence.  Even the layman knew bad mortgages were being packaged as AAA-rated securities well before the crash (If you don’t believe me, check out the Craigslist money and Morningstar forums before the crash.  There is ample evidence of a priori knowledge.).  There was no big secret.  That these securities carried a much-higher risk of default was well-known even among smaller, less-sophisticated investors well in advance.

Investors knew what was going on.  Many were in on it.  But they all thought they could make a quick buck and get out just before the crash.  That’s how all bubbles work, after all.  Savvy investors almost always know the crash is coming but feel confident in their ability to get out before the music stops.  Some of them do.  Many don’t, but that doesn’t imply they didn’t understand the nature of the risks they were exposed to.  Psychology tells us that the rarer the event, the more we tend to underestimate its frequency (probability of happening).  That is, an event that happens half the time is pretty easy to predict whereas an event that happens only once out of a thousand is prone to being written off as impossible.  Oops.

Market Value Does NOT Always Equal Fair Value

Norris then goes on to make a statement that clearly shows his ignorance of the nature of assets, lambasting the banks for their fight against mark-to-market accounting by saying

“The banks have argued that market values can be misleading, and that their own estimates of the eventual cash flow from assets are more realistic than what…others…will now pay for those assets. The rules already allowed them to ignore so called “distress sales” in assessing fair value.“

The banks, of course, are precisely correct both from a logical and accounting perspective.  That Norris implies otherwise troubles me.  By definition, distressed sales should logically be ignored when assessing fair value because distress sales fail the requirements of a mutually-informed, arms-length, unmotivated market transaction.  In order for a transaction to be “fair” and thus indicative of an asset’s true value, both sides of the transaction must be well-informed, unmotivated (that is, they have no pressing need to sell in order to meet some other obligation), and arms-length (i.e. not selling to your nephew).  A distress sale, of course, utterly fails this requirement since the selling party has no choice but to sell at any price.  Since it is not a fair market transaction, it cannot be used to discern fair value.  This is Economics 101, something Norris doesn’t appear to be all that knowledgeable on.

In a period of mass distress selling such as happened in mid-2008, it is entirely correct and logical to assume that current market values do not reflect true asset values (again, since they aren’t fair market transactions they can’t logically count) and that educated estimates of eventual future cash flows are a much better estimate of a security’s true intrinsic value.  It’s not perfect, but it’s likely to be a better estimate than market value.  Remember, markets can only be expected to set prices accurately when the vast majority of transactions meet the criteria for fair market value transactions.  When they don’t (and they didn’t during the crash), the market’s price-setting mechanism breaks and usually becomes wildly volatile.

To reiterate, accounting didn’t fail us.  The goal of accounting is to provide an objective measure of a firm’s current economic situation, which is quite different from providing a realistic measure.  Since any accounting system will have to make arbitrary distinctions and over-simply complex economic situations for the benefit of distilling them down to a number, realism in accounting figures is impossible.  Furthermore, I believe attempting to make accounting numbers is a flat-out bad idea and makes them far less useful to the trained analyst.  It is not the accountant’s job to make his financial statements realistic, only to report the numbers objectively under some internally consistent and logical set of rules.  Leave the interpretation to me, please.