How fractional reserve banking led to the creation of CDOs, SIVs and 30-times gearing at the big banks...
LEGEND HAS IT that bank robber Willie Sutton once famously said he robbed banks because "That's where the money is," writes Aaron Krowne of Implode-Explode Heavy Industries Inc.
Understandably, most people still believe this aphorism today. Banks are assumed to hold "the money" in our economy. But unfortunately, this "conventional wisdom" couldn't be further from the truth – and the facts lie at the core of the current financial crisis.
Here is how most people assume the model works:
- They deposit their money in the bank (in checking or saving accounts or CDs);
- This is where banks get most of their capital (or for non-deposit broker-dealers, the money is assumed to come from investors and other equity-holders);
- Banks then supposedly lend out the money, dollar for dollar, and make a profit on the spread – the difference between what they pay depositors, and what they charge for loans.
But in fact, our system doesn't even pretend to work like this, although the authorities passively permit people to believe it.
Instead, we have a "fractional reserve banking system", which states that banks need only keep a small portion of cash on hand relative to lending. In the regulatory parlance, this is called "core capital" – but as we will see, this is a complete misnomer.
The logic behind a fractional reserve banking system is that bank depositors will only withdraw a small fraction of their money at any given time. After all, your money spends most of its time just sitting in a bank, somewhere. This leaves the majority of the money available to be re-lent, generating profits for the bank.
A reserve ratio is of 10:1 is considered a "reasonable" number by the authorities. And working under a 10:1 fractional reserve, a bank could lend out not simply 90% of the money it takes in deposits, but actually ten times the money deposited by repeatedly lending out 90% of deposits.
(Remember – when they lend out money, it just gets deposited back in the banking system; so banks in general can re-lend that 90% again and again...)
The end result of this system is that banks now have huge books of loans (considered assets) relative to their deposit and/or shareholder equity bases – even if they are faithfully following a fixed reserve requirement.
The big problem with fractional reserve lending is when depositors happen to withdraw more money than the bank has on hand. The bank becomes manifestly insolvent, and must close its doors (or in our present system, go into government receivership).
This effect typically feeds on itself, creating a "run on the banks" and spreading panic.
Interestingly, in our present system, the authorities consider a "run on the bank" the bad thing itself, rather than the fractional reserve lending (and the lack of actual cash it implies) that causes it. So as usual, the regulatory system addresses the symptoms, not the cause.
Now, all of the above might sound shaky enough, but our system has gone far beyond it. In 2007, as the mortgage crisis spread into a credit and banking crisis, it came to light that Wall Street banks were leveraged not 10:1, but in the ballpark of 30:1 on average. And even that is likely too cautious an estimate, because most banks hide considerable amounts of lending activities "off balance sheet", using essentially the same kind of vehicles Enron used to commit extensive fraud.
But according to the regulators, in effect, when the banks do it, it's okay. More on this in a moment.
The extreme leverage used across the banking system is not only dangerous because depositors might withdraw core funds, but also because the value of bank assets might fall to a degree that exceeds those funds. This also brings about insolvency!
For example, the hedge fund Long Term Capital Management (LTCM) failed in 1998. It was leveraged at least 30:1 (by some counts, higher, but let's assume just 30:1). That means it only had about 3.3% "core capital" from its investors (akin to bank depositors). LTCM then made bets all over the world regarding the movement of bond funds.
Unfortunately, in 1998 the Asian Financial Crisis struck, and lots of things happened that weren't supposed to happen according to LTCM's trading models. One day Russia defaulted on its sovereign debt, which the LTCM propeller-heads had assumed was impossible, and the value of its bets decreased by more than that 3%. That meant LTCM's gambling wiped out all the money lent to it, and then some.
All of LTCM's banking counterparties had to band together, at the behest of the Fed, and forgive the losses (taking a hit themselves). As Roger Lowenstein – author of the definitive book on the LTCM debacle, When Genius Failed – recently wrote in an update, it appears Wall Street and our bank regulators learned nothing from this episode. Since then, they have winked at the increase of leverage throughout the US banking system, as Wall Street's major banks and brokerage houses became just as leveraged as LTCM, if not more.
Perhaps they did learn a lesson, however –that they could get away with being irresponsible, and that backroom dealing and government intervention would cushion them from the consequences.
Today as I write this, Lehman Brothers and Washington Mutual are both facing just the same kind of collapse, caused by using cash deposits and investments to "gear up" and take a bet more many times their capital bases. Because of leverage in all its proliferating forms (including extreme fractional reserve lending), the banking system is virtually devoid of "capital". There is some cash, but by and large, much of what is counted as "capital" is a joke.
In fact, most people today (including regulators and bank CEOs) don't even remotely have a clue what capital is. Yet the concept is very simple. Capital is SOMETHING with INTRINSIC VALUE. This is why, for example, plant machinery and equipment with REAL PRODUCTIVE and USE VALUE are called "capital" in formal accounting terms. When the use value of these things wears down, they go through DEPRECIATION, which acknowledges that the capital is diminishing in a very real sense.
"Hard money" is also capital. I am including for rhetorical purposes cash, but since cash depreciates due to inflation, it is not as good as true hard monies, like Gold and silver.
And what is NOT capital? Entities that speculate and gamble based on leverage do not have "capital". This is plain to see translated into an everyday context.
If you borrow $10,000 as a cash advance from your credit card, then turn around and lend that $10,000 to a friend, it is nonsense to say you have or had any "capital". You may have a loan note from your friend, saying "I.O.U. $10,000 and will pay you 10% APR in interest in the mean time", but it is nonsense to consider that note an "asset".
Why? Because the credit card company can call you up any day and say "Hey, we're changing the terms of your credit", or just "We are closing your account because we need the money back" – and you're screwed. You're insolvent. Your friend probably doesn't have the money to give back to you right away.
Perhaps you even thought you were being nice by helping out your friend, "stimulating the economy" by making use of "spare credit". But when the chips were down, it became clear you over-extended yourself.
If instead, you had $10,000 in cash in savings, or $10,000 worth of Gold, and decided to lend THAT out, the situation would be entirely different. That would be lending based on real capital. Now, no one can call you up and demand a return of the money you lent out. You have the luxury of waiting until the maturity of the loan. You are the boss.
Lending based on credit vs. capital bears its ugly teeth in another manner. For "tradeable" assets, when a "cash call" comes in, the holder of the asset is FORCED to sell rather than having the luxury to sell at the planned time or price point. So instead of waiting for a gain on the asset, the lender-speculator may have to take a loss on immediate sale.
If this happens to enough lender-speculators all at the same time, they are likely to actually force down the market price of the assets of concern. This happened with subprime loans – especially Collateralized Debt Obligations (CDOs) – in a big way. Their tradeable values collapsed to cents on the dollar, just when banks needed to cash them out the most.
Yet banks universally refer to loans and speculative positions held on leverage (credit) as "assets". What kind of "asset" subjects us to insolvency because of illiquidity (failure to sell) or decline in asset value when we need to cash it out? Not an "asset" in any real sense – so don't fall for bank-speak. The very problem is that these holdings are more like obligations than assets, because they are not held against capital, and they are not themselves capital.
Finally, banks even go so far as to treat their equity (value of shares outstanding) as capital. But that is nonsense, since the value of shares can fall in a blink. Sure, when shares are initially sold, the bank gets precisely that much in cash ($100 million in shares sold at an IPO results in $100 million in cash raised for the bank). But it is not true that when the shares double in value that it is just like the bank having another $100 million to "play with". And once the initial $100 million is tied up in loans, it is no more fit as capital backing for additional loans than deposits are.
Many banks seem to have come to an informal reliance on their equity value as part of their "core capital". These very same banks find themselves in straits that are even more dire when their share value plummets and they need to issue new equity to tap the market for cash: even a relatively small amount will totally wipe out the existing ownership, and do little to solve their long-term insolvency problems.
So equity, too, is a form of credit, not capital. It is an obligation. Yet the regulators have bought-into and signed-off on all today's mis-interpretation and over-gearing. They have allowed a joke to be perpetuated in their treatment of "core capital". Even fractional reserves have been informally abandoned in favor of breaking down lending and speculative holdings into individual positions, and making sheer guesses as to how much cash is needed to back them. This is called the "capital adequacy" framework and it is advocated by the international bankers (see the BASEL accords, agreed by central banks and governments). It uses as its core tenet estimates of "risk" to determine how little "core capital" is needed to back positions (of course, the smaller the better).
But "risk" is estimated with an emphasis on past price volatility and questionable and arbitrary "ratings".
One of the first things any investor is told when they hand their money to Wall Street is that "past performance is no indicator of future results". Yet when it comes to estimating "risk", little more than past performance (with an emphasis on recent past) goes into the computation. Some estimates might be good, some might be bad, but they are all just guesses.
As for ratings, we've seen how the ratings agency cartel was corrupted due to poorly structured incentives, as well as all the above fallacies used in estimating risk. So their output was a combination of guesswork and corruption. Nevertheless, the difference between an "AAA" and "AA" or "BBB" rating had huge ramifications for the capital reserves that had to be held against the corresponding bank "assets". The holding requirements varied 10 to 1 or more.
In other words, if you could get an "AAA" rating slapped on a CDO instead of the "BBB" it may have deserved, you could perhaps hold only a penny on the dollar in reserves against it, instead of ten or twenty cents.
The problem here is not so much that one method of judging financial risk is so much better than the other, or that ratings agencies and other risk control officers were stupid. The problem is that it is impossible to truly know risk in advance, so the only sure method of insuring oneself against it is to have full capital backing for lending and speculation.
That is, no leverage.
To take on any leverage is to begin to magnify the consequences of getting the risk estimate wrong – and to spread these consequences to depositors and investors. In aggregate, the peril is spread to the entire financial system.
The "capital adequacy" framework has been damaging by encouraging a willful ignorance of the "big picture" – a bank's capital relative to all its committments – instead breaking the analysis down to a "micro" level. Then, you have thousands of individual cases to look at to determine the "adequacy" for each, and it is much easier to pass off some lame justification of why this or that loan needs only ten cents or even one cent on the dollar. I believe it is only with this kind of approach that banks could have been allowed to legitimately keep gigantic off-balance-sheet positions, such as in Special Investment Vehicles (SIVs) and "conduits", with virtually no capital backing them.
At peak, Citigroup had upwards of $2 trillion of such commitments, almost all off-the-balance-sheet and not counting towards capital holding requirements. But when the crisis hit, shareholders quickly felt real pain in terms of endless dilutions as the SIVs were brought back on balance sheet, and/or the assets contained therein had to be sold off at a steep loss.
The final cherry on top of this cake is that in our system the line is blurred between "core capital" as reliable cash versus the questionable sorts of assets discussed above. This distinction is highlighted by a current conundrum: that the value of Fannie and Freddie preferred shares has been almost completely wiped out, yet the regulators had encouraged many banks to hold these securities as their core capital.
It's bad enough to leverage core capital against ten or 100 times itself in loans – but when that "capital" itself can vanish overnight, your leverage becomes effectively infinite!
There are more subtle examples of this problem. For example, E-Trade put a lot of its customer's money-market money in SIVs that turned out to be backed by subprime assets, which one day went "poof!". So what they thought to be "good as cash" turned out to definitely not be.
More generally, banks widely hold US Treasury securities as "good collateral" – also reasoned to be "as good as cash". Plus, they pay interest. But should the US's sovereign debt take a hit, the value of these securities as tradeable assets would be threatened as well. And as we've seen, when you operate on credit instead of capital, tradeability can suddenly become important at the least opportune of times.
Banks, contrary to popular belief, have NOT been where the money is. Banks have been the biggest borrowers, not the creditors. They don't have capital, they have debt. They have been running on credit, which merely masquerades as capital. That is the genesis of this crisis and why it is refusing to simply "blow over".
It will not be over until leverage is dramatically reduced and economic activity becomes based more on capital than credit.
You might naturally ask, then, whether your money in the bank is at risk? As a first approximation I would answer "probably not" – because the government has made a commitment to disallow any retail deposit losses in the banking system (up to $100,000 for each account). If it has to, the government will print the money to return it to you, in the event your bank fails.
But this highlights a disturbing continuation of the above analysis. If the US government is backing the deposits of the entire banking system, and that entire banking system was over-leveraged and is devoid of capital, this implies that the US government is going to take a huge financial hit to back all of this.
Ominously, the FDIC has already opened communications with the Treasury about directly backing its finances. And if, as with Fannie and Freddie and Bear Stearns, the government is forced to monetize the backing of virtually the entire US banking system, then the concern will be that the dollars paid out will become worth less and less as more are printed and created on electronic accounts. In other words, it would become obvious that the US Dollar itself is no more an example of true capital than a subprime CDO, nor can it constitute capital. It is simply another form of credit – albeit the last to fail.
In that case, I would suggest it is not cash dollars you want to keep around as your own personal trove of "capital", but rather Gold and silver.