Last week we learned that for price stability, the money supply needed to grow with the economy. But to make things more challenging, changing the velocity of money also changes the money supply needed for price stability.

Today, I’m going to take a quick look at how the fractional reserve banking system works and see how loan insurance (e.g. CDOs) affects the money supply.

Most of our banking system requires 10% reserves (some countries have lower amounts). This means that if a bank has $100, it can lend $90 while keeping the $10 in a “safe” asset. Since the people borrowing that money (or the people who receive it in a purchase) probably put it in a bank, that $90 will then allow another $81 to be lent out. This cycle continues until there is $1,000 of loans with $100 acting as “reserves”.

In other words, with a 10% reserve requirement, banks can increase the money supply 10 fold. If the reserve requirement was 20% then it would only be a 5 fold increase. When you hear about credit tightening and banks not lending, this means they are not making full use of their money supply generating capabilities which will have the effect of reducing the supply of money. The federal government then has to try and increase the money supply to compensate.

Of course when the economy improves the banks start lending again and the federal government needs to remove some of that money supply to keep it from expanding too quickly. That gives you a quick overview of what has typically happened over the business cycle for the last 100 years.

Our latest cycle had a new twist however. Let’s go back to our first bank with $100 that gives out $90 in loans. Imagine that the bank can insure $10 of loans for $1. So for $9 it could insure all $90 of the loans it had made. Since those $90 are now considered “safe” (they are insured after all). The bank can treat them as reserves and can lend another $81. Notice the pattern is the same as our fractional reserve system and that this insurance is allowing the bank to increase leverage by 10x. So the instead of a 10x money supply increase we get a 100x money supply increase. This wasn’t the actual cost of the insurance of course, but you can see how the insurance allows the bank to create more leverage and add more money into the money supply.

The federal government was slow to catch on to this practice and didn’t realize this shadow money supply was what was pushing housing prices up.

Now imagine what happens if all this insurance a bank has bought is discovered to be worthless. This would cause all the banks that had bought insurance to be under reserved which would mean they would have to start calling in their loans (like the worlds worst margin call). Because the problem was system wide, this would spiral into a massive contraction of the money supply and an instant depression.

This is why the federal government stepped in to save AIG and to make good on all those insurance contracts. It allowed all the banks to keep reporting that they had sufficient reserves. So the federal government is doing everything they can to make banks as profitable as possible so that they can rebuild their true reserves and become less reliant on the insurance based reserves.

While the banks are doing this however, they have very little interest in creating a lot of new lending because that just increases the reserves they need to get back to a safe position. As we talked about earlier, this puts additional downward pressure on the money supply. This is why there is so much talk about deflation at at time when the federal government is throwing money around like a drunken sailor.

And finally we get to the punchline. While the economy is down (i.e., low velocity of money) and banks are rebuilding their reserves (i.e., low leverage) it will be virtually impossible for the money supply and inflation to grow significantly. However, when those conditions change, the money supply could explode as velocity increases and banks start creating money again and find new innovations to go beyond the traditional 10x limits. And if the federal government does not respond quickly enough (which it rarely does), big inflation will arrive.

Unfortunately, while this dynamic seems quite clear, the timing of when we will flip from one state to the other is completely opaque. My broken crystal ball says the economy is unlikely to take off over the next 12 months and the earliest the flip might occur would be in 2012 (presidential election year), but it could take MUCH longer to occur (see Japan).