We’ve got a little problem in the economy. Tiny really. Nothing to worry about.

Just that money supply is falling and that could endanger the entire economic recovery.

The Federal government and the Federal Reserve are pumping money into the economy as fast as they can and yet the supply of money in the economy has started to fall.

Think of it this way: The driver has filled up the tank with gas but when he stomps on the accelerator the car sputters forward because only a trickle of gas is reaching the engine.

Right now everybody’s got their foot pressed down on the monetary accelerator as hard as they can, but in the last four weeks the supply of money reaching the economy has actually declined.

If the trend continues or worse accelerates, any economic recovery could sputter to a halt. (And if you need a reminder that a recovery can go into reverse, try my soothing story on the recession of 1937 https://jubakpicks.com/2009/09/22/we-have-nothing-to-fear-but-a-replay-of-1937-itself/ .)

The Federal Reserve is buying mortgage-backed securities ($1.25 trillion of the little fellows) and debt from Fannie Mae and Freddie Mac ($200 billion), expanding its lending to banks by keeping interest rates low (as close to 0% as makes no never mind), and buying up U.S. Treasuries.

All that, according to the text books, should be flooding the economy with money. Which is exactly what you’re supposed to do to get the economy running again and to avoid turning the Great Recession into a re-run of the Great Depression.

And during the early stages of the financial crisis those policy actions did exactly what they do in the text books. M2, the broadest measure of the money supply the Federal Reserve still tracks, climbed from $7.36 trillion on October 1 2007 to $7.88 trillion on October 6, 2008, and to $8.39 trillion on June 22, 2009, according to the St. Louis Federal Reserve Bank.

That’s an extra $1 trillion to fund loans and credit card bills and plant expansion and state borrowing and… well, just about anything the economy needs money for.

And since each dollar of that extra trillion dollars gets used over and over again by the economy, the effect is even larger than that huge sum itself. What economists call the M2 multiplier has ranged between 8 and 12 for most of the period from 1959 to 2009. So that $1 trillion has the effect of $8 to $12 trillion racing around the economy.

Even in the huge $14 trillion plus U.S. economy that should be enough to jump start economic activity and to raise justifiable fears of runaway future inflation.

In normal times anyway. But the numbers coming out of the Federal Reserve say these aren’t normal times.

Despite everything the Federal Reserve—and don’t forget the $787 billion stimulus package passed by Congress—has done to pump money into the economy, in the four weeks to September 14, money supply (measured by M2 again) has actually declined.

And the multiplier has fallen off a cliff. And that’s because what’s called the velocity of money, the speed with which a dollar moves through the economy, has fallen.

 That’s not unexpected. During the Great Depression the velocity of money fell by 22%. In tough times people–everyone from consumers to bankers–sits on more money longer.

But this isn’t good, folks. It’s a problem big enough to jeopardize the recovery that the economy seems to be building.

 Look at what’s happened to M2 since it hit $8.39 trillion on June 22. By July 20 M2 had dropped to $8.34 trillion. (That’s a decline of $50 billion.)

By August 24 it was down to $8.28 trillion. (A decline of $110 billion.)

M2 recovered somewhat the next few weeks so that it stood at $8.30 trillion on September 14, the last data point from the St. Louis Fed. But the trend is ominous.

Economists who study this data use a four-week moving average to eliminate some of the week to week noise. At the worse point in the decline—the four-weeks ending on August 24—M2 was dropping at annualized rate of 12%. That’s the kind of contraction you get in a financial panic. Not the kind of growth you want to see as you’re trying to guide an economy to recovery.

Wait, if you factor in the drop in the velocity of money and in the M2 multiplier, the situation is even worse. Remember I told you that the normal multiplier from 1959 to 2009 was in the range of 8 to 12. But in the financial crisis the M2 multiplier, according to the Federal Reserve, dropped to close to 4. And it hasn’t yet bounced back.

So in the last few weeks money supply has dropped at a rate fast enough to derail the recovery and the velocity of money has remained stuck at the slow speeds of a financial crisis.

The good news is that it’s pretty clear what the problem is. The bad news is that it’s not at all clear what to do to fix it.

The problem is that the banks still aren’t lending. They’re sitting on a huge proportion of all the money that the Fed is pumping into the economy and because the money they’re sitting on isn’t moving at all this is also putting the brakes on the velocity of money.

Look at what M2, that measure of the money supply, actually measures. It’s the sum of all the currency in circulation plus all the money in demand deposits (money that can be withdrawn at demand without notice such as the dollars in checking accounts) at banks, plus savings accounts, plus time deposits (CDs and the like) under $100,000, plus money market funds, plus overnight repurchase agreements and Eurodollars at banks. (I think I’ve got all the big components.)

But you’ll notice that it doesn’t include the reserves that banks keep in their own vaults or at the Federal Reserve.

In other words all the money that the Federal Reserve has pumped into the banking system only reaches the money supply that drives the economy if banks lend it out. If they make a loan to buy a house or to build a factory, that loan turns into paychecks, orders from and payments to suppliers, and profits for builders. Some of that money gets spent again and again, producing the multiplier effect of any increase in the money supply.

 But all of it, perhaps only fleetingly, winds up in checking accounts, savings accounts, business accounts, and other instruments that are measured as part of M2.

If the money sits in a vault, it doesn’t count.

How do we know that’s what’s happening? By looking at another data series from the St. Louis Fed. (If you want to check any of this data yourself, start here with the St. Louis Fed’s series on M2 http://research.stlouisfed.org/fred2/series/BASE )

This one tracks how much money commercial banks have lent out.

As you’d expect banks cut back lending in the financial crisis and then ever so gradually increased lending when it became clear that the sky wasn’t falling. Total lending at commercial banks climbed from $8.70 trillion in the week of October 17, 2007 to $9.31 trillion in the week of December 24, 2008. That’s a very healthy $600 billion increase in just about two months. 

But then things stopped getting better. They didn’t get worse—except for the weeks around the March 9 stock market low when bank credit fell back below $9.3 trillion. But bank credit stayed stuck at that level.

Until July. That’s when it started to slide back. In July bank credit dropped below the $9.3 trillion level. In the week ended August 19 it dropped below the $9.2 trillion level. And it has stayed there for the first half of September. In the week ended September 16, 2009 total bank credit came to $9.12 trillion.

That’s not a huge decline from the $9.3 trillion level but it is definitely a move in the wrong direction at a time when the economy needs to see bank lending growing not shrinking or standing still.

Why are banks cutting back on their lending rather than increasing it? (I don’t mean to pick on banks here. I think other types of financial institutions are doing much the same thing. It’s just that the data on banks is so much better than it is in other parts of the financial industry.)

Part of it is that they don’t trust the recovery. With unemployment still rising and consumers still not spending being cautious about extending new credit doesn’t seem particularly outrageous.

But that’s not the major part of the problem. Banks aren’t lending because they still haven’t cleaned up their balance sheets. Many of them are still in the process of writing down credit card debt or mortgages or commercial loans—not to mention the complex derivatives they’ve still got in portfolios. They know that regulators have suddenly turned strict about capital ratios. There are only two ways to raise your capital to asset ratio (and remember for a bank an asset is a loan, money that someone has borrowed.) You can either go out into the still very hostile public financial markets and pay the pound of flesh that investors still want before they buy into an equity offering. Or you can improve the ratio by working on the denominator. Cut the amount you’ve got out in loans and your capital ratio goes up without the need to raise any more capital.

Banks know too that tougher capital requirements are coming not too far down the road. The recently concluded G20 economic meeting in Pittsburgh agreed that the various national regulators would go back home and increase the amount of capital that the banks they regulate need. For 2010 and 2011 banks face a steady ratcheting up of capital to asset ratios. They don’t want to make the challenge of meeting those new rules any tougher by increasing their loan books now.

None of this really comes as a surprise to the Federal Reserve and the world’s other central banks. They knew that they potentially faced exactly this problem. But they hoped that by unleashing a campaign of financial shock and awe that involved heavy fire from all monetary and fiscal weapons they’d be able to convince banks that it was only sensible to lend more rather than sit on their money.

It’s not clear that this policy has failed. The worrying trends really include just four weeks or so of data. But the short-term trend is worth worrying about because having fired all their guns, the world’s central banks really don’t have much left in their arsenal to roll out in a new assault.

Watch the data. We’re not out of the woods yet.