Mark-to-market is going to go the way of the dodo so let’s at least give it at least a decent eulogy.

The meme is that marks are at the heart of our banking problem. A convention dreamed up by diabolical accountants intent on making financial institutions statements purer than Caesars wife. Get rid of the convention, go back to historical cost accounting and presto our banks are fine. 

To listen to the conversation you would think that most bank assets are subject to this unfair treatment. Otherwise, why would it be such a big deal. Well the truth is that it’s pretty much a tempest in a teapot.

David Reilly at Bloomberg did a little research. He found two intriguing factoids.

Of the $8.46 trillion in assets held by the 12 largest banks in the KBW Bank Index, only 29 percent is marked to market prices, according to my analysis of company data. General Electric Co., meanwhile, said last week that just 2 percent of assets were marked to market at its General Electric Capital Corp. subsidiary, which is similar in size to the sixth-biggest U.S. bank.

In other words, 70% of the loans on these banks’ books are carried at historical cost. The bank is free to make its own judgments about whether and to what extent they might be impaired and to make appropriate additions to reserves to account for those decisions.

Now mark-to-market is pretty much impossible to argue against intellectually if one is talking about applying it to trading portfolios. If you’re running a book and trading in and out of it the only way to realistically assess your financial condition at any point in time is to mark the securities in that book to the market. At the same time, if you’re making loans that you intend to hold til maturity marking them makes little sense. The process of reserving against losses in that portfolio is the logical accounting method. That does leave you at the mercy of the banker’s judgment but, hey, no world is perfect.

So caught in this trap what’s the argument for changing the rules for the banks? It’s the thesis that there is no market for these securities so marking them is illogical. Well, Mr. Reilly has some data on that.

Plus, the basics of marking-to-market often get confused. Consider the refrain that banks can’t mark assets to market prices because markets are frozen. Of the 12 banks in the KBW that I reviewed, only 5 percent of total assets on average were designated as being hard to value because market-based prices weren’t available.

These so-called mark-to-myth assets represented 58 percent on average of total shareholders’ equity among the banks. They fall, though, to just 16 percent of equity if the Big Four banks — Bank of America Corp., Citigroup Inc., JPMorgan Chase & Co. and Wells Fargo & Co. — are excluded.

Is it starting to make sense that this is an issue that’s related to a very small number of banks? And why might that be. Mr. Reilly hits the nail pretty much on the head. The big guys went into investment banking in a big way and piled up a lot of assets — trading assets — just like the real investment banks. And how do investment banks have to value this stuff? Mark-to-market.

Two conclusions come to my mind.

First, except for a handful of banks, this is not an issue. Remember this the next time the talking heads on CNBC start going off on this subject and say we have to get rid of mark-to-market in order to get banks lending again. It ain’t true.

Second, the big guys need it to go away because they are most likely insolvent under the convention. Note that doesn’t mean they can’t generate tons of cash flow. Whether all that money is enough to cover the losses though is the real question. They require the flexibility and yes artifice of historical cost to have a fighting chance of getting through this.

So as we see mark-to-market accounting either emasculated or killed outright remember that he was a good guy. He spoke the truth and you know how dangerous that can be.