Deconstructing QE2: it's less than meets the eye

Today the FOMC told us they are prepared to buy an extra $600 billion of Treasuries over the next 8 months—about $75 billion per month. They did not commit, however, to buying this much: "The Committee will regularly review the pace of its securities purchases and the overall size of the asset-purchase program in light of incoming information and will adjust the program as needed to best foster maximum employment and price stability."

In other words, the Fed is not going to make a big monster purchase of Treasuries that would flood the system with new money. They are going to dribble it out a little at a time, and might even decide to end the program after a month or two, should the economy and/or inflation show signs of picking up. This is almost exactly what I was hoping would happen.

With $75 billion in new purchases per month coming on top of the purchases needed to offset maturing MBS, the Fed will be buying roughly $110 billion of Treasuries a month, which, curiously, happens to be almost exactly the amount needed to fund the Federal government's budget deficit—the federal deficit over the past 12 months was $1.295 trillion, or $108 billion per month. Hmm.

If this were Argentina in the 1970s and 80s, such a program would undoubtedly lead to a huge increase in inflation—it's debt monetization, pure and simple. Back then, Argentina's government essentially ordered its central bank to print up enough money to cover the government's budget deficit. The government paid its bills with newly-printed cash money, and that money became like a hot potato that people tried to spend as fast as possible. (Holding on to currency that loses a significant portion of its value every day is equivalent to paying a tax to the government, so people naturally try to spend the money as quickly as possible, in the process acquiring things that won't lose their value—such as dollars or gold or euros or commodities or real estate.) I should know, because I lived there in the 70s and vacationed there often in the 80s. In the four years I lived there in the late 70s, inflation averaged about 7% per month, or about 125% per year. During one 3-week stay in the mid-80s, I got the distinct privilege (for an economist) of witnessing a staggering 200% rise in the price level. That works out to about 9,000% per year, enough to qualify as true hyperinflation.

But this isn't Argentina. The Fed is not going to print up a billion $100 bills every month and ship them over to Treasury. Instead, the Fed is going to buy $75 billion of Treasury securities each month on the open market, and it is going to pay for those purchases by crediting banks with newly created reserves. This is the way the Fed financed its first quantitative easing, with the purchase of $1.3 trillion of Treasuries and MBS—bank reserves increased by $1.3 trillion. Note, however, that even if the Fed buys $75 billion of Treasuries for the next 8 months, that will barely add up to 4% of GDP. Printing an amount of money equal to 4% of GDP is not going to greatly upset the inflation applecart.

Moreover, the reserves the Fed creates to buy the Treasuries can't be spent the way $100 bills can; they only exist at the Fed. They can only turn into "spendable" money if the banking system uses the extra reserves to create new loans (thus increasing bank deposits) or turns some of those extra reserves into currency. And either way, that process depends on the system wanting to borrow more money and/or hold more currency.

With QE1, we know that the banking system was content to let almost all of the $1.3 trillion in new reserves sit at the Fed, where they conveniently earn a bit of interest. Banks were extremely risk averse, and their demand for rock-solid cash-like assets was huge, so they were happy to hold on to all the new reserves. The rest of the world was also very risk averse, and there was a lot of deleveraging going on. In other words, the demand for new loans was very low at a time when banks were very reluctant to lend and the demand for money was very high. Result? The Fed ended up satisfying almost everyone, and the financial crisis passed without any meaningful inflationary consequences.

With QE2, we may find that the newly created reserves end up being more than the banking system is content to hold. If so, banks may use some or all of the new reserves to increase deposits and further expand the money supply (note: M2 already has grown by almost $1 trillion since mid-September '08, and one dollar of reserves can support up to $10 in new deposits). Or it may be that the world decides to exchange some portion of the new reserves for currency (note: currency in circulation already has increased by $130 billion since mid-Sep. '08, and one dollar of reserves can support one dollar of new currency). So we'll need to watch M2 and currency in circulation even more studiously going forward, in order to gauge how QE2 is impacting things.

With today's clever announcement, the FOMC has not only indicated that it is willing to undertake a substantial QE2 program if the need arises, but it is also giving itself some time to make sure that nothing gets greatly out of hand. There is no need to panic—yet.