Money supply update

In partial rebuttal to my post yesterday, which asserted that the combination of huge deficits and a significant shortening of the maturity profile of Treasury debt posed the biggest proximate risk of rising inflation, monetary policy is still a critical issue, since at the end of the day inflation only happens if central banks allow it. In the article I referenced, Cochrane was not trying to argue that monetary policy has been inflationary; rather, he was arguing that it is quite possible that the Fed would be unable to respond adequately to a massive decline in money demand that could, in turn, be amplified by the very short-term nature of Federal borrowing needs. He recognizes that the Fed has provided all the kindling (i.e., a huge increase in bank reserves) that would be necessary for an inflationary conflagration, and what he focused on in his article was the origin of the spark that might set off a rising inflation fire. 

So what follows is simply an update of where the monetary policy fundamentals stand today. The dry fuel for rising inflation is still there, but as yet it poses no immediate threat beyond the gradual increase in inflation we have seen in the past year, which I think is likely to continue.

Bank reserves are the raw material for the money supply, since banks must hold reserves against their deposits. Our fractional reserve monetary system allows banks to create new deposits (new money) if they can back up those deposits with some reserves. Beginning in Sept. '08, the Fed launched its first quantitative easing program, which ended up injecting about $1 trillion of bank reserves, more than 10 times the reserves that the Fed had created in its entire history. All things being equal (which they are most certainly not), this massive reserve expansion theoretically could have resulted in an equally massive increase in money, which almost surely would have led to a serious bout of higher inflation. Not content with simply scaring the bejeesus out of Fed watchers everywhere—myself included—the Fed embarked on a second quantitative easing program in October of last year, boosting the amount of reserves by another $600 billion. All of this reserve expansion was fueled by the purchase of $1.6 trillion of Treasury and Agency debt. Among other things, this means that the Fed essentially financed the federal government's deficit last year, a classic case of "debt monetization," the mere thought of which would have sent the world into a panic gold-buying binge. Oh, wait, isn't that what has already happened these past several years?

Most of the reserves that were created by open market purchases are still sitting idle, however, in the form of Excess Reserves held at the Fed. Banks are apparently content with holding onto these reserves, since they pay 0.25%, which is more than twice the yield that reserves earn in the overnight market (currently 0.1%), rather than using them to make new loans which take the form of increased deposits. To put it another way, the Fed has undertaken a massive reserve expansion in order to accommodate the banking system's intense desire for risk-free, short-term securities. As the chart above shows, Required Reserves have approximately doubled since the first quantitative easing program, from $45 billion to $93 billion, and $10 billion of this increase has come in just the past month. So only a tiny fraction of the $1.6 trillion reserve expansion has been used by banks to increase the money supply. That tiny fraction has grown significantly, however, and it reflects a rather significant—albeit of lesser magnitude—increase in the money supply.

Most of the impetus to the expansion of the M2 money supply comes not from the Fed or the banks, however. Money has increased as a result of the world's increased demand for money; to date, the Fed's actions have been primarily designed to ensure that increased demands for money could be accommodated by the banking system—that there would be no risk of any shortage of risk-free liquidity. The chart above makes it clear that the recent, unprecedented growth in the money supply is similar to what we have seen during financial market panics of the past. The amount of money in the economy has surged, not because the Fed has run the printing presses overtime, but because the world has desperately demanded more dollar cash as a hedge to the risk of European sovereign debt defaults. Accommodating a sharp rise in money demand is not in itself inflationary.

From a long-term perspective, the latest increase in M2 is not all that unusual, nor is it frightening from a monetarist's inflation perspective. M2 is only about $400 billion, or 4% above its long-term trend, which has been 6% annualized growth over the past 15 years, during which time inflation has averaged about 2.5% per year. It's the equivalent of a minor blip on the monetary radar screen. We've seen such blips in the past, and they have not resulted in dire or inflationary consequences.

Nevertheless, there are still reasons to be concerned about the risk of higher inflation. Unlike other times when money supply growth has surged, today the dollar is scraping the bottom of its valuation barrel. The dollar was much stronger in 2001 and at the height of the 2008 panic than it is today. A strong dollar is prima facie evidence of a relative shortage of dollars, so today's historically weak dollar suggests at the very least that whatever the Fed has done has resulted in a relative over-supply of dollars, and that is an essential ingredient for inflation.

During previous episodes of strongly rising money growth, commodities were either extremely weak (2001) or suffered massive price declines (2008). This also suggests that there is a relative abundance of dollars in the world today, and that commodities, even after falling over 10% since last April, still represent inflationary pressures that can eventually find their way into the prices of other goods and services.

Finally, although the chart above (5-yr, 5-yr forward inflation expectations) shows that the market's inflation expectations have fallen about 80 bps in the past two months, a similar decline occurred last summer, after which inflation by all measures ended up accelerating. Inflation expectations today are still much higher than they were at the end of 2008, and that also suggests that monetary policy is relatively accommodative.

The thrust of Cochrane's argument was that the greatly shortened maturity profile, and massive size, of outstanding Treasury debt could provoke a significant decline in money demand if the world began to doubt the U.S. government's ability to tame its almost-out-of-control spending. Trillions of dollars could potentially try to shift out of short-term Treasuries and into other goods and assets, and this would have the effect of greatly increasing the price level.

As this last chart (which uses a conservative estimate of only 2.5% annualized GDP growth in the current quarter) shows, money demand has skyrocketed since the onset of the 2008 recession. If money demand were to revert to the level that prevailed prior to 2008, the added M2 velocity (velocity being the inverse of money demand) could add 20% to nominal GDP growth (most of which would likely be in the form of higher inflation) in coming years.

Would the Fed be able to withdraw all the reserves it has added in time to prevent such a surge of inflation from occurring if money demand starts declining? That is the question which everyone is asking, and which is keeping people awake at night.