The above chart is a graphical depiction of how the Fed's balance sheet (in simplified form) has changed over the past six years, from a time in early 2007 when the economy was growing normally (although the housing correction was just beginning), to the situation today. The major changes: on the liability side, relatively normal growth of currency in circulation (see below for details), and gigantic growth in bank reserves; on the asset side, the elimination of T-bills, strong growth in T-bonds, and unprecedented growth and holdings of MBS and Agency securities. Put another way, the Fed has purchased about $2.36 trillion of T-bonds, MBS, and Agency securities, shed about $280 billion of T-bills, and in the process created $420 billion of new currency and $1.7 trillion of new bank reserves.
As the chart above shows, currency has grown at a 6.8% annualized rate since early 2007, the same rate that we have seen for the past 20 years. Currency actually grew at a faster rate from 1993 through 2003 than it has in the past six years, and the former was a period when inflation averaged 2.5%. In the past six years, inflation has averaged about 2.2%. So the recent growth in currency is nothing out of the ordinary, and does not necessarily imply higher inflation than what we have seen in recent decades. It is also important to note that the Fed only supplies currency on demand, in exchange for bank reserves. Thus, the Fed cannot "force-feed" currency to the world. The world holds only as much currency as it wants to hold, so growth in currency is not necessarily inflationary at all.
In addition to currency in circulation, the other major component of the Federal Reserve's liabilities is bank reserves, shown in the chart above. Here we see enormous growth, orders of magnitude in excess of historical experience. In fact, the Fed has expanded bank reserves by a factor of 18 in the past six years. Isn't this the same as "printing money?" No, because bank reserves aren't "money" and can't be spent anywhere. Bank reserves exist only as a liability on the Fed's balance sheet. Banks can exchange their reserves for currency, and currency can be spent, but as we saw above, currency has not grown by any unusual amount. Banks have instead been content to hold on to the majority of the reserves the Fed has created. This is due to the fact that the Fed started paying interest on reserves in 2008, which makes reserves functionally equivalent to T-bills, and the fact that banks have wanted to increase their capital and fortify their balance sheets, and bank reserves fit that bill.
In effect, the Fed has been doing nothing more than swapping newly-created bank reserves for bonds and currency, and the world and the banking system have been happy to participate in the swap. The Fed has taken duration risk out of the market and supplied low-risk and highly liquid assets in return. In a post last December, I described how in a way the Fed has been acting like the world's biggest hedge fund.
Banks can exchange their reserves for currency if the public demands it, or banks can hold on to their reserves if they want to increase their capital and fortify their balance sheets. The only other thing banks can do with their reserves is to support an increase in their deposits. In our fractional reserve banking system, banks must hold about one dollar of reserves for each ten dollars of deposits. Banks are the only ones who can "print money," and that is done by extending credit to borrowers. Given the increase in bank reserves, banks could theoretically have increased their lending and their deposits (which currently total about $8.7 trillion) by a factor of 18, or over $150 trillion. Obviously, nothing of the sort has happened.
As the chart above shows, the M2 measure of the money supply has grown only slightly more than 6% per year for the past 18 years. Since early 2007, M2 has grown at an annualized rate of 6.5%, about the same as currency. Again, there is no sign here of any usually fast growth as a result of the Fed's enormous expansion of bank reserves.
The proof is in the pudding: inflation by any measure has been averaging between 2 and 2.5% for the past six years, with no signs of any acceleration. We are thus forced to the conclusion that the Fed has not allowed any undue expansion of the money supply. Whatever growth in the amount of money there has been has only been slightly in excess of the world's increased demand for money and money equivalents, and that is why inflation has been relatively low.
As the chart above shows, the world's demand for M2 money has been extraordinarily strong in recent years. Since early 2007, M2 has grown 26% more than nominal GDP, with the result that the ratio of M2 to nominal GDP has reached its highest level since the late 1950s. Monetary policy has been enormously accommodative, but at the same time money demand has been exceptionally strong. It follows that if the Fed hadn't been so accommodative, we could have suffered deflation and/or a weak economy, since the economy would have been starved for liquidity.
Since November 2008, the fastest-growing component of the U.S. money supply has been savings deposits, which have increased 68%, from $4 trillion to $6.7 trillion. The enormous growth in savings deposits has almost certainly not been driven by their attractive yields (which have been close to zero), so it must be due to an overwhelming desire for liquidity and safety. Consumers have been deleveraging and increasing their money balances to an exceptional degree because they have become more risk-averse. Most of the growth in the money supply has been due to an increased demand for money and safety.
And just as consumers have worked hard to increase their money balances, banks have been very willing to hold on to the extra bank reserves the Fed has created, as they have struggled to boost their reserves and fortify their balance sheets.
In a sense, the Fed has spent most of the past 5-6 years responding to an unprecedented increase in risk aversion, which has manifested itself in a tremendous increase in the demand for cash, cash equivalents, deposits, and bank reserves. This increased demand for safety and liquidity has been the flip side of an equally unprecedented decline in confidence and an increase in risk aversion.
All of this leads to the following conclusion: since the Fed's quantitative easing efforts to date have been only sufficient to satisfy a huge increase in the demand for cash, cash equivalents, and bank reserves, what will happen when the demand for safe and liquid assets declines? When confidence in the future returns? When risk aversion declines? Will the Fed be able to reverse its quantitative easing efforts in a timely fashion, before an excess of reserves and a declining demand for cash pushes inflation higher? As I mentioned in a post last December,
... the biggest risk we all face as a result of the Fed's unprecedented experiment in quantitative easing is the return of confidence and the decline of risk aversion. If there comes a time when banks no longer want to hold trillions of dollars worth of excess bank reserves for whatever reason (e.g., the interest rate the Fed is paying is no longer attractive, or the banks feel comfortable using their reserves to ramp up lending, or the public no longer wants to keep many of trillions of dollars in bank savings deposits), that is when things will get "interesting."
Stephen Williamson discusses this same issue in great detail here, but from a slightly different perspective. The common thread we share is that the thing to worry about going forward is a declining demand for cash, cash equivalents, and bank reserves, since that could result in higher inflation if the Fed fails to take offsetting measures (e.g., draining reserves and/or increasing the interest it pays on reserves) in a timely and potentially aggressive fashion. There is no a priori reason to think they can't, but there is certainly reason to be worried, since we are sailing in uncharted monetary waters.
Unfortunately, it's ironic and paradoxical that the Fed's efforts to supply additional reserves to the banking system—at a time when the banks have many times more reserves than they could possibly want—may be doing just the opposite of what the Fed intends. Instead of boosting confidence, the Fed may be contributing to worries about the future course of inflation and interest rates. More reserves are not going to increase bank lending, since banks already face zero constraint on that score. Will more reserves increase confidence, or will they just increase concerns about the future?
It's not clear at this point what will happen as the Fed increases its purchases of Treasuries and MBS and creates more bank reserves in the process. But as long as uncertainty exists—and the Cyprus bank deposit fiasco threatens to increase risk aversion in the Eurozone—the demand for reserves is likely to remain strong and inflation is likely to remain subdued. As for the economy, it looks like it is slowly healing and taking care of itself.