Calafia Beach Pundit

The Federal Reserve's quantitative easing program began over two years ago, and it has since triggered waves of concern and countless forecasts (including mine) of rising inflation. To date, the fears have not been validated by any meaningful rise in the official inflation statistics—indeed, some measures of core inflation have fallen to record lows, as I noted in this post the other day. How can the Fed have engineered an explosive and unprecedented increase in its balance sheet without there being any noticeable impact on the economy and on inflation? The short answer is that quantitative easing has not yet resulted in any meaningful increase in the amount of money in the economy. 

What follows is a rather long but illustrated survey and discussion of the important evidence surrounding the Fed's experiment in quantitative easing. Although there's no evidence of any meaningful monetary expansion resulting from QE2, there does appear to be enough excess money relative to money demand to cause a future rise in inflation.

(Note that many of the charts shown here use a semi-log scale for the y-axis. This makes it easy to see how the growth rate of monetary aggregates has deviated over time relative to trend, since straight-line growth on a log scale is equivalent to a geometric rate of growth on a nominal scale.)

This first chart goes to the core of the expansion of the Fed's balance sheet, which has been accomplished by the purchase of over $1 trillion of MBS and Treasury securities. When the Fed buys an asset of whatever kind, it pays for it by crediting a bank with "reserves." Reserves are not money in the ordinary sense, but they can be used by banks to increase their lending and thus to increase the amount of money in the economy. Banks are required to hold about $1 in reserves for every $10 in deposits in our fractional reserve banking system, so an extra trillion in reserves could theoretically allow an almost unimaginable increase in the amount of money in the economy.

Since quantitative easing first got underway, the problem the Fed has faced is that either banks are unwilling to lend more, or borrowers are unwilling to borrow, or both. The vast majority of the reserves created through the Fed's quantitative easing program sit idle on the Fed's balance sheet as "excess" reserves, as shown in the above chart. $1 trillion of excess reserves is stark testimony to the ineffectualness of QE, but also to the fact that QE has eliminated any possibility that the economy is suffering from a shortage of money.

This next chart shows currency in circulation. Currency can only increase if banks ask the Fed to exchange their reserves for currency, and it takes $1 of reserves to get $1 of currency. Currency in circulation, while one of the components of the money supply, can also be thought of as a measure of money demand, since people only hold dollar bills (which pay no interest) if they serve some purpose (i.e., to facilitate transactions or as a store of value). The Fed can't force more hundred dollar bills into the economy than people are willing to hold, because unwanted dollars are easily returned to banks in exchange for deposits. As the chart shows, the growth of currency in circulation has been unusually slow for most of the past decade, up until the financial panic of late 2008 when demand for dollars surged. Plus, the ups and downs of the dollar have tended to coincide with the rise and fall of the dollar's value, as shown in the next chart. (The recent divergence of the red and blue lines is an unusual development that bears watching.) I should also note that a large portion of U.S. currency in circulation is held overseas (most likely over half), so currency growth can be a misleading indicator of conditions in the U.S. economy. But to summarize: the relatively slow growth of currency and the dollar's general weakness in the past decade are signs of declining dollar demand, and not a sign that the Fed has been too tight or that there is a shortage of money.

The next chart (below) shows the Monetary Base, or "high-powered money." The base consists of bank reserves plus currency in circulation (i.e., the sum of the first and third charts). This is the part of the money supply that the Fed can control directly by buying and selling securities. Note that the base has expanded in a fashion similar to that of reserves; both grew exponentially in the first phase of quantitative easing.

The next chart (below) is the M2 measure of the money supply. M2 consists of M1 (currency in circulation and checking accounts), plus savings deposits, small denomination time deposits, and retail money market funds. (I'm not including M1 in this exercise because its composition and behavior over time has varied due to changes in banking regulations and procedures such as sweep accounts. These changes have caused M1 to grow at a very slow rate since 1995, due to money shifting out of M1 categories into M2 categories. M2 effectively incorporates all these changes and thus has been a much more stable definition of money over long periods.)

Note from the chart above that M2 has grown on average about 6% per year since 1995. That's a little faster than the 5% annual growth of nominal GDP over the same period. But if we assume that over half of the growth in currency (a component of M2) was due to voracious overseas demand for $100 bills, then M2 growth has been quite similar to the growth in nominal GDP since 1995. With money growing at about the same rate as growth in the nominal economy, there is no obvious reason to think that the Fed or the banks have been creating too much money. This is not a case, in other words, of "too much money chasing too few goods."

Note also how M2 growth surged in late 2008 in response to the economic and financial crisis.  Fears of an imminent collapse of the world's financial system caused a huge, safe-haven-driven increase in the demand for money of all forms. People and businesses wanted to hold more money in their checking accounts, and they wanted to deleverage in order to reduce their risk. (Paying back loans is equivalent to increasing one's holdings of money.) Since then, M2 growth has slowed back down to its long-term trend. Nowhere do we see any sign of an unusual expansion of the money supply that might have resulted from the huge increase in the monetary base over the past two years.

So, up to this point we can conclude that although the Fed has injected over $1 trillion of reserves into the banking system, this has not resulted in any unusual expansion of the amount of money sloshing around the economy. Banks apparently have not been aggressive lenders, and borrowers have, in aggregate, not been eager to borrow. The extra reserves the Fed has created are not being utilized by banks to increase the amount of money in the economy by more than we might otherwise expect to see. Since the Fed now pays interest on reserves, the banking system effectively views reserves as similar to T-bills: they are risk-free instruments that carry a modest yield. The Fed has essentially swapped T-bills that it has created for bonds. The Fed is now bearing a substantial amount of yield-curve risk (and the market is thus bearing less), since higher bond yields could result in large losses to the Fed's balance sheet. Collectively, banks are sitting on a mountain of low-interest-paying cash, presumably in order to fortify the health of their balance sheets.

Now we turn to market-based indicators of the balance between money supply and money demand. While it is easy for the Fed to add up all the "money" in the economy (e.g., M2), it is quite another thing to know whether the amount of money that has been supplied to the economy is sufficient to satisfy the economy's demand for money. No one knows how to quantify money demand except after the face. But, as Art Laffer frequently reminds us, we can infer what the demand for money is doing by simply observing the behavior of market-based measures of the price of money. For example, if we know there has been a bumper crop of apples, and we know that the price of apples has been falling, then we can conclude that the demand for apples has not been as strong as the supply of apples.

The two charts above show the market-based value of the dollar relative to other currencies. The first one uses nominal values and a basket of only major currencies, while the second one uses inflation-adjusted values and a very large basket of trade-weighted currencies. Both show that the dollar is at or near the low end of its historical valuation range. This can only mean that the amount of dollars being supplied to the world under the Fed's guidance has exceeded the world's demand for dollars. Demand for dollars has been generally weak, while the Fed's willingness to supply dollars has been very generous. An over-supply of dollars, in turn, sets us up for the classic "too much money chasing too few goods" problem.

The chart above shows the dollar's value relative to gold. Gold prices have been in a strong uptrend (i.e., the dollar has been steadily losing value) ever since the Fed started easing in early 2001. As with the dollar, this strongly suggests that the Fed has been over-supplying dollars to the world. The market finds itself with unwanted dollars, and that encourages people to exchange some of their dollars for gold, since gold has been a trusted store of value throughout the ages, as well as being the favorite of speculators betting that quantitative easing will eventually prove inflationary.

The chart above shows the dollar's value relative to a basket of non-energy, industrial commodities. This too suggests that there is an over-supply of dollars in the world—too many dollars chasing a limited supply of commodities, pushing their prices ever-higher.

The chart above compares the slope of the long end of the Treasury yield curve to the market's forward-looking inflation expectations, as embedded in the pricing of TIPS and Treasuries. A steep yield curve (the long end of the curve is almost as steep as it has ever been) is a classic indicator of easy money, and easy money is what eventually gives us higher inflation. TIPS are priced to an acceleration in the CPI from the current 1.2% to 2.9% within the next 5 years or so. That's not a very scary number, to be sure, but it is definitely higher than the 2.3% annualized rate of inflation over the past 10 years.

Will QE2, which promises to create an additional $600 billion of reserves, make banks any more willing to lend, or borrowers more anxious to borrow, or will those extra reserves also just sit idle at the Fed? We can only wait and see, but I have to believe that at some point banks' willingness to sit on an ever-growing pile of T-bill substitutes paying almost no interest will begin to wane. Banks likely will be increasingly tempted to put their reserves to work, taking advantage of credit spreads which are still historically wide and a yield curve that is historically steep. 

But even if banks do begin lending more and M2 expands more rapidly, will that boost growth as the Fed is hoping it will, or will it simply fuel more inflation? I take the view that easy money can only boost growth if a prior lack of money is holding the economy back. You can't print your way to prosperity, but a central bank can help its economy by pursuing an appropriate monetary strategy. As is evident from the above discussion, however, that is manifestly not the case today—we are not suffering from a lack of money, so more money will only be inflationary. Meanwhile, the Fed is attempting to boost the supply of dollars at a time when the demand for dollars is relatively weak—not exactly the conditions that make for a sturdy equilibrium. As the next chart shows, the velocity of M2 has been relatively low since the depths of the recent recession, and it has plenty of room to increase if money demand falls further and people decide to spend some or all of the cash that they have been hoarding. Rising money velocity could boost the economy temporarily, but it could also fuel rising prices. 

Supporters of QE2, such as Mark Thoma, argue that no harm has been done yet, and that the Fed can quickly and easily reverse its quantitative easing if signs of trouble appear. That may be true, but it is also the case that the Fed is attempting a balancing act that could deteriorate faster than its ability to react. Plus, the potential gains from QE seem ephemeral at best, while the downside risks seem potentially huge. 

I would hope that arguments such as I have advanced here will continue to put pressure on the Fed to back off from a full implementation of QE2. The sooner the better. Monetary policy is not a good tool for micro-managing the economy, and the Fed shouldn't even attempt it, especially now.

UPDATE: Robert Barro has a nice essay on QE2 here that touches on many of the same themes. He questions the effectiveness of QE2 as expansionary policy, and he worries that the Fed does not have a good or easy exit strategy.