Mishkin lays out (and minimizes) the problems facing the policy choices of the Fed in a recent paper.
On page 21-22 "OVER THE CLIFF: FROM THE SUBPRIME TO THE GLOBAL FINANCIAL CRISIS"
Shrinking Central Bank Balance Sheets
Actions by central banks to contain the global financial crisis resulted in huge expansions of their balance sheets. The expansion of balance sheets arising from liquidity provision is typically easy to reverse because most of the liquidity facilities have provided loans at interest rates that are higher than market rates during normal times. As financial markets return to normal, market participants are no longer willing to borrow at above-market ranks, this source of balance sheet expansion naturally reverses itself as the financial system recovers—which is exactly what has happened.
The asset market purchases of long-term mortgage-backed securities are not selfliquidating in this way. Over $1 trillion of the mortgage-backed securities have maturities of ten years or more. Thus, a strategy of just letting them run off will leave the Federal Reserve in this market for a long time, which raises several issues. First, by holding these securities the Federal Reserve will be exposed to both credit and interest rate risk. Second, the presence of private securities on the Federal Reserve balance sheet means that the Fed has become directly involve in perhaps the most politicized financial market in the United States. The public and Congress may begin to hold the Fed accountable for what happens specifically to mortgage rates, rather than to interest rates in general. Politicians may tend to see the Fed as institutionally responsible for developments in the housing markets.
Can the Fed extricate itself from this situation by selling the mortgage-backed securities? The experience of the end of the purchase program for mortgage-backed securities at the end of March 2010 is encouraging. For some months before this date, the Fed had been in essence the sole buyer in this market. However, given that financial markets had stabilized and that the end of the purchase program was well publicized, the Fed’s exit from the market did not cause any disruption. The spreads of mortgage-backed securities over Treasury bills did not rise after April1, 2010. This experience suggests that if the Fed announces a program of asset sales well in advance and financial markets are functioning normally, it should be able to liquidate its positions. Of course, if this turns out not to be the case, then the Fed could discontinue its sales and announce that its sales are contingent on the market continuing to function normally.
A final concern sometimes raised is that the expansion in the monetary base will necessarily be inflationary, but this is unlikely to be the case in the current environment. The reason is that banks are perfectly happy to hold huge amounts of excess reserves—thus essentially neutralizing the effect this money would have on demand or the price level— as long as they are paid interest on the reserves, as is now the case. However, purchase of long-term government bonds has raised concerns that the Fed is willing to accommodate profligate fiscal policy by monetizing government debt, and this does have the potential to unanchor inflation expectations, which could have inflationary consequences in the future."
I am less sanguine about the Fed's ability to manage expectations. The anchoring of expectations depend on the Fed (reversing)
liquidating the holdings it acquired under QE. It is therefore likely that the Fed will have to sell its bond holdings at precisely the time when inflationary expectations are rising and therefore the yield curve is moving up. This will push yields up even faster (in a nonlinear fashion i) and will almost certainly be a significant drag on economic activity. Should this event occur in the next 24 months before the economy has had a chance to stabilize the Fed will have to make a Sophie's choice and may be tempted into thinking it can fool all of the people all of the time. My guess is that when eventually the Fed will be forced to make some very unpleasant trade-offs and bond market volatility and tail event outcomes are very likely. What he Fed will choose is not apparent so it makes playing the scenario difficult.
To quote the Godfather
Michael Corleone: Just when I thought I was out... they pull me back in.