Mark-to-market nonsense

Reading time: 9 – 14 minutes

The first thing I learned as a bank trainee, many years ago, was that all banks are insolvent — it is the nature of banking.

The banking business consists of borrowing money you have to repay today to lend to someone else who will pay you back tomorrow.

If your creditors demand their money today, you won’t be able to pay — because you’re a bank and banks are insolvent.

Lessons of the New Deal

Throughout the 19th century and up until 1933, hundreds of banks regularly failed every six years or so, whenever there was an economic recession.

During the New Deal, the Federal Deposit Insurance Company was set up to ensure depositors’ money, thereby effectively making banks “solvent” by government intervention.

Run on Northern Rock Bank, Birmingham, UK, in 2007

In 1933, demand deposits were the major source of bank funding. With a government guarantee on deposits, bank runs became rare.

The Glass-Steagall Act kept banks away from the dangers of investment banking.

Strict margin requirements on securities loans dramatically reduced risks on what were the “toxic assets” of those days.

In 1950, checking deposits made up 70% of US bank liabilities, while small time and savings deposits made up most of the rest.

In other words, the government guaranteed virtually all bank liabilities.

US commercial banks were safe.

Just as important, banks were engaged in very few different kinds of operations; they were much, much easier to manage and understand than today.

But times change.

Not your grandfather’s bank

By Q4 2007, private checking deposits were less than 5% of total liabilities of US commercial banks.

Interbank liabilities with respect to federal funds and repurchase agreements were 164% of US private checking deposits.

Glass-Steagall had been repealed.

Banks were engaging in blatant speculation in exotic instruments on their trading desks.

The effective safety measures that protected institutions that were — by definition — fundamentally insolvent, had been allowed to wither and become irrelevant.

Furthermore, and most importantly, most US bank liabilities today are no longer guaranteed by the US government.

The safe banks of the 1950s are no more.

Make-believe capitalists

Although bankers are well aware of the “dirty little secret” of their insolvency, they still oppose “government intervention” and “over-regulation”.

People often pretend to be what they are not.

Bankers know that they rely on government to stay in business, but like to pretend that they should be allowed to operate as free-spirited entrepreneurs in the “private sector”.

The Crisis of 2008 was simply a modern reenactment of the bank runs of the 19th and early 20th century.

Different liabilities, different times. Except that banking is now exponentially more complicated and has become virtually impossible to understand.

How to read a bank balance sheet

The second thing I learned as a bank trainee, back in the good old days, was that it is impossible to learn much of anything useful about a bank by reading its balance sheet.

Whether a bank is strong or not, depends upon the quality of its assets — information that is woefully absent from financial statements.

Once we understand that banks are insolvent and require government guarantees to continue in business, the question becomes what percentage of liabilities have government guarantees and which assets are good, or whether the government will eventually have to step in, liquidating the bank and paying off depositors, leaving zilch for shareholders.

Accounting can’t make water run uphill

Now accounting rules don’t require banks to list their loans on their financial statements, nor to indicate terms and covenants of each financial asset.

You’re not told the credit rating of borrowers, or even the weighted average of credit ratings (even if you put faith in credit ratings).

Information in such detail is not even feasible.

Accounting rules can make the impossible seem real ...

Nor do accounting rules require that auditors actually evaluate the likelihood that a bank will get its money back on a particular loan, or any group of loans.

Bank auditors don’t visit bank borrowers one by one and look them in the eye to see if they seem credit worthy, or whether they are drunken bums that have just blown the bank’s money at the racetrack.

Moreover, large international banks are extremely complicated, composed of thousands of subsidiaries, affiliates, side ventures, and special deals, and subject to the risk that some obscure trader in one of hundreds of countries where they do business may be hatching a scheme that will bring the bank to its knees tomorrow morning — operations of these banks are so complex that it is impossible to contrive a document that could possibly “protect investors” by supplying all relevant and material information.

Now, auditors don’t like to admit that bank balance sheets are bullsh*t, and that that the millions spent on auditing — for the benefit of “investors” — might just as well have been donated to charity.

So they try to find some way to get someone else to give an opinion on how much bank assets might be worth — at least some part of the assets.

This is where the “mark-to-market” rules come in — based on belief in the Efficient Market Hypothesis.

This discredited theory suggests that market price reflects intrinsic value.

Mark-to-market nonsense

Historically, banks haven’t been in the business of selling their loans to others.

Instead they hold loans until they come due and then credit the proceeds to cash, interest earned, or loan losses.

Accountants were not able to impose mark-to-market rules on ordinary loans for the simple reason that there was no market to serve as a benchmark.

But with modern, complicated banking, some assets actually do have markets that might be used as a measure of value.

Surely, some assert, it would be in “investors’ interests” to adjust the value of bank assets to changes in market price.

So accountants came up with a “mark-to-market” rule that says that, unless a bank intends to hold an asset until maturity, it must indicate the value of the asset in terms of market value.

Why ‘mark-to-market’ is misleading

Mark-to-market accounting is misleading for many reasons, but perhaps the most important is that market prices are not the value at which all assets of a certain type could be liquidated tomorrow, but merely the price at which a very small number of securities in circulation happen to have been sold yesterday.

Furthermore, unless a bank is obligated to sell asset by a certain time in the immediate future, but instead has the option to hold assets to maturity or until market prices get better, the current value of assets may be irrelevant both to bank investors and creditors.

Everyone can't convert to cash at the same time

If a bank holds $50 million in bonds that had $5,000 traded at 100 yesterday, it doesn’t follow that the bank will be able to sell all its bonds tomorrow at this price.

All securities markets — even the deepest and most liquid  — only trade a small portion of securities outstanding on any one day.

No security market on earth is able to liquidate all holdings of securities outstanding today, at yesterday’s price.

If all securities are liquidated at the same time, and there are no buyers, the market value is zero!

Now, in some cases, mark-to-market accounting is useful, although still misleading.

For example, mutual funds that promise to redeem shares on any day based on the market value of the portfolio as of the close of business, obviously must use mark-to-market accounting because they have a legal (or commercial) obligation to redeem shares at that price.

However, banks are not required to pay creditors based on the market value of bank assets, nor do bank shareholders have any claim on the bank based on the current market value of bank assets.

Furthermore, evaluating some bank assets on the basis of mark-to-market, and others on the basis of historical costs, means that total value of bank assets (apples plus oranges) is a gobbledygook number that defies comprehension.

If we start by admitting that all banks are, by their nature, insolvent, the key question we should be asking is “How much of bank liabilities are guaranteed by the government?”

In the 1950s, the answer was, “Almost all”.

Today, the answer is “Very little”.

Mr. Geithner’s Stress Tests

Timothy Geithner, the US Secretary of the Treasury and point man of the Obama administration’s efforts of resolving the economic crisis, has announced that he is getting ready to analyze the finances of US banks by subject them to some kind of “stress test”.

Does Mr. Geithner know what he's doing?

There is less said about these stress tests recently; perhaps Secretary Geithner has given up on the idea.

Now, I can save the American taxpayers a lot of money by pre-releasing the results of this test at no cost:

1. All banks are insolvent, although some have assets of lower quality than others.
2. All banks fail the “stress test” unless the government guarantees liabilities.
3. Banks that have the largest portion of liabilities (i.e., deposits) guaranteed by the government are the strongest. (Mainly, small community banks.)
4. Banks that have the simplest, smallest “product line” are the least likely to fail. (Mainly, small community banks)
5. Most major international banks are not “too big to fail” — they are “too big to manage”.

Learning from the past

In other words, the way to save the banking system and reduce “systemic risk” is to require the big banks to split up and drastically simplify their operations, with the government providing guarantees for virtually all liabilities.

Liabilities which can’t be guaranteed by the government should be eliminated.

Too strong, you say?

Well, it worked in 1933 and the fundamentals of bank safety haven’t really changed that much.

What is different is that the government has allowed too much of the bank balance sheet to avoid regulation.

Of course, if the government guarantees virtually all bank liabilities, as in the distant (but safer) past, there will be pressure on banks to eliminate the absurd salaries and bonuses paid to bank executives.

Impossible, you say?  Can’t get experienced bankers without paying a fortune?

Well, consider this: In May 1970, Walter Wriston became chairman of Citicorp — arguably the world’s premier banker at that time.

His compensation was just $257,820 a year, just $585 more than he had earned as president and CEO (Wriston: Walter Wriston, Citibank, and the Rise and Fall of American Financial Supremacy. Page 307).