1. "A few" who say asset purchases will likely be warranted until end 2013.
2. "A few others" who agree with the need for considerable asset purchases but put no specific time frame or total on the amount.
3. "Several others" who say it is "appropriate to slow or to stop purchases well before the end of 2013, citing concerns about financial stability or the size of the balance sheet."
The identity of these "several others" is an important mystery for the market.
The biggest concern for those banking on a full year's QE program is that at least some of them come from either the board of governors or are centrists, both of whom have supported Fed Chairman Ben Bernanke in the past several years. The phrase "several others" in the minutes usually means three or four members.
On Saturday, at the AmericanEconomic Association annual conference in San Diego, Philadelphia Fed President Charles Plosser suggested he could be part of this group, saying that if scaling back the size of the balance sheet doesn't go well, "we may do more damage to the economy in terms of instability, or a rise in unemployment."
It is not especially worrisome for Bernanke if Plosser is a member, nor would it suggest an early end to the program. Plosser has been a frequent critic of the Fed's QE policy and he does not vote until 2014.
More troubling would be if it included one or more of the Fed governors, who typically vote with the chairman.
It remains unclear if Bullard and Kansas City Fed President Esther George, both new voters this year,will support QE. Many believe George will not, and instead will follow in the tradition of former Kansas City Fed President Tom Hoenig.
Other new voters,Chicago's Charlie Evans and Boston's Eric Rosengren, are unlikely to be among those who believe concerns about unwinding the balance sheet should trump the need for more asset purchases.
There are three specific concerns about the balance sheet. The first is that a large balance sheet will create inflation. Second, the Fed could take losses on its balance sheet if it has to sell assets amid rising interest rates. When the interest rate on a fixed income security goes up, the principal amount goes down. So when the Fed itself tries to raise interest rates, it will be responsible for creating losses on its own portfolio.
Finally, the concern is that in order to keep those excess reserves from creating inflation, the Fed will have to pay high levels of interest to the banks to keep them from being loaned out.This is not only counterproductive (the interest it pays will just end up increasing excessive reserves) but looks bad politically for the Fed.
Depending upon how high interest rates rise, the payments could end up being in the tens of billions of dollars—payments that the Fed will make to banks ostensibly to control inflation, but also to make sure they do not make loans.
All of that will mean that the payments that the Fed has been making to the Treasury each year—from the income generated by its portfolio— will dwindle and could end up being nothing. Last year, the Fed paid $77 billion in what is called seignorage to the Treasury in 2012.
Those on the other side in the debate counter that they are now concerned the committee could ultimately arrive at a decision on QE that has less to do with economics and more to do with optics.
They point out that the large balance sheet has so far not resulted in inflation and that the ability to pay interest on reserves is a key tool that should allow the Fed to induce the banks to keep that money from becoming loans and potentially fueling inflation.
They point out that strict so-called monetarists (who see a direct connection between the size of the Fed's balance sheet and inflation) have been wrong for most of the past four years.
If the size of the Fed's balance sheet does not in itself cause inflation, then the Fed could potentially sell no assets or only very few, avoiding losses on the portfolio. The balance sheet would shrink naturally as assets matured and were not replaced.
Losses do not trouble this group, since they point out the Fed has paid the Treasury an average of $67 billion over the past three years and that the Fed is not designed to profit from its balance sheet.
The decline in seignorage to theTreasury would only amount to a restoration of the status quo before the financial crisis when the Fed's payments were minimal, averaging about $25 billion in the ten years before the crisis.