There is a big debate going on now about the role of shadow banking in the current crisis. Charles Davi says Brad Delong and Mark Thoma have got it all wrong. From what I can see Delong and Thoma are on the right track. Now, different commentators seem to mean slightly different things by the term shadow banking. So let me state explicitly my views on the roots of the financial crisis.

First, there was a lot banking activities by institutions that had no access to the discount window and no lender of last resort. What does that mean? It means that retail loans were produced and distributed to investors without ever passing through what we would think of as a traditional bank. Fundamentally, a bank is a middle-man. It borrows money from one set of people and lends it another. The trick of banking is to convince the people its borrowing from that their money is safe and liquid, while at the same time lending the money out to the type risky, illiquid projects that actually cause the economy to grow. This sleight of hand is added by the Federal Reserve, who in exchange for regulatory oversight promises that it will guarantee the bank is always liquid. Given that the Fed prints the money, that’s a promise that you can well, take to the bank.

In our new world, however, a potential home buyer could walk into the offices of New Century. New Century would provide them with a loan. New Century would then sell that loan to Bear Stearns. Bear Stearns would then take that loan and packaged alongside other loans in a Collateralized Debt Obligation. Those CDOs would then be sold directly to investors. The trick in all of this is that the CDO did the work the traditional bank used to do. CDO’s as you probably know are cut up into traunches. When the investors send in their mortgage checks the traunches are paid off in a particular order. That order means that even if some of homeowners didn’t make their payments, the highest rated traunches would still get paid.

So the CDO has produced safe debt from a risky project. So, safe in fact that it can get a AAA rating that makes the debt liquid. So we have safe liquid debt from an unsafe and illiquid project. That is a bank. This bank, however, has no access to the discount window. So what happens when so many people are late on their payments that the market gets spooked and won’t buy your bond despite its AAA rating? Liquidity collapses, trades don’t occur and the market grinds to halt. From the point of view of the outside world, the “bank” was run. Investors can’t get their money back and potential borrowers can’t get loans.

Second, there was the use of off-balance sheet entities. From the point of view of investors this bank without a bank thing was a great idea. See, with a traditional bank the Federal Reserve wants to get caught up with annoying details like capital adequacy ratios. Being the spoil-sports that they are the Fed makes you put a lot of your own money in the game, so that you are not hanging the taxpayer out to dry if they have to step in and provide their liquidity guarantee. The more of your own money you put in the less leverage you have. The less leverage you have the less money you make. So, the Fed is really cramping your style here.

To make matters worse, now that non-banks can do banking you have a competitor on your front end, stealing business from you. So what is a bank to do? Well, since non-banks can get into this why don’t we just create a non-bank entity that is funded by the bank and gives its profits to the bank? This entity can engage in the same CDO transactions without those pesky regulators harshing your mellow. Yes, there are those who say that ultimately if the non-bank entity goes bad then the bank will lose its investment and therefore reduce its capital just as if the non-bank entity were actually a part of the bank. Sure, that may be technically “true” and “completely reflect the actual aggregate risk.” But, they are forgetting something – it’s totally legal.

Third, there is another layer to this where we get into something called Synthetic CDOs. Here we don’t even have to get rid of the loans. We can keep the loans and buy insurance against them going bad. If we structure the insurance just right it will have the same net effect as having chopped up the loans into parts based on their riskiness and then selling the parts. The insurance we buy is called Credit Default Swaps or CDS. As long as the companies we buy the CDS from a solvent and able to meet their obligations, there should be no problems.

The entire system created here operated outside of the traditional commercial banking and outside of the regulatory power of the Fed. Yet, it funneled an enormous amount of money into the real estate, car loan and to some extent even credit card markets. This in turn allowed consumers to take on more debt than was sustainable. When it all came crashing down there were three huge consequences. One, once the easy credit was gone the economy was faced with a sudden readjustment. Homebuilders were building too many homes. Automakers were making too many cars. Retailers were stocking too much merchandise. This sudden adjustment weighed on the economy.

Two, the riskiness of a mortgage depends crucially on the amount of equity the borrower has. The end of easy credit meant a reduction in the number of home buyers and a reduction in the price of homes. Falling home prices mean falling equity and hence rising riskiness. This increase in riskiness lead to even wider losses on mortgages and an even sharper reduction in the supply of mortgages.

Third, the collapse of some non-banks, in particular Lehman Brothers and Bear Stearns created a panic which reduced the availability of credit for everyone and sent the world economy into a tailspin. Lehman and Bear had engaged in this CDO business but they also had other more benign business. They were storied firms and they could and did borrow freely in the credit markets. Then everything went bad and investors became afraid that the money the lent to Lehman and Bear would not be paid back. Those fears were realized on September 15th 2008, when Lehman went bankrupt. Now the fear was not only not getting your money back from Lehman but getting your money back from people who may have in turn loaned your money to Lehman. The best course actions was to hoard all the cash you possibly could and buy US Treasury bonds. So Treasuries soared and everything else tanked. Credit dried up and companies wary of not being able to make payroll or meet suppliers began to lay off workers en masse.

This is the outline of the crisis as I see it and it stems mainly from the proliferation of the non-bank banks.