Tiresome QE III and bond rates update

I have an embarrassing confession to make.  I forgot the date QE III was announced.

I decided to graph the 3 year Tresury constant maturity interest rate.  The logic is that I don’t believe the current FOMC can plausibly precommit to policy more than around three years from now.  Most members’ terms end in January 2016. Bernanke’s term ends 2014 and there is no way Republicans will allow an up or down vote on reconfirmation (I guess he will be acting Fed chairman for a while).  So I made a graph from Sept 1 2012 on.  I didn’t remember that QE III was announced on September 13th.  I looked at the graph.  I had no idea whenthe huge signal about future monetary policy became public.

To make the graph less boring and more like my older discussion, I added the 5 year constant maturity rate.  This shows a modest decline on the 13th entirely reversed on the 14th.

The reason I am back on this topic (aside from the fact that I am obsessive) is that I am very sure I haven’t explained properly why I look at exactly these numbers.  I am trying to assess the forward guidance effect of QE III and not at all the portfolio balance effects.  This means that I absolutely am not claiming QE III was a flop.  I don’t think that, I think it is working very well.  I try to explain myself after the jump.

By forward guidance I mean signals about future conventional monetary policy, that is signals about the future target federal funds rate.  Basically, one argument for QE is that it is a dramatic way to convince people that the FOMC will keep the federal funds rate below 0.25% for a long time.  This can affect output right now in large part because if they keep the rate low until inflation rises medium term real interest rates will be lower.  The Krugman Woodford solution for monetary policy in a liquidity trap is to credibly commit to causing higher inflation therefore causing higher expected inflation right now.

But this works by changing forecasts of future conventional monetary policy, that is future short term rates.  Medium term rates are equal to the expected geometric average short term rate plus a risk premium.  The risk premium could be negative in theory, but it is generally positive as is shown by the fact that the yield curve slopes up.  Given a three year rate of 0.37 % short term rates expected for the next three years must be very very low.  Note the current 3 month rate is 0.1% not exactly zero.  My view is that the FOMC has reached the limit of what it can achieve via forward guidance.  Abstract models consider the case of no precommitment and total precommitment forever.  But they are depending on the current FOMC signalling what future FOMCs quite probably with different members will do.

In any case, to me the data are clear.  Expected average short term rates over the next 3 and 5 years were very very low before QE III was announced.   Then they didn’t change. There is no sign that bond investors’ views about short term interest rates during the term of current members of the FOMC changed at all.

Again this does not mean that QE III had no effect.  Massive purchases of MBS are not just a signal of future monetary policy.  They change the amount of mortgage default risk which private agents must bear.  This should increase the price (reduce the yield) of MBS.  This should make issuing mortgages more attractive and promote refinancing and home construction.

All of this can cause higher expected inflation not because of changed forecasts of conventional policy (the federal funds rate) but because of the direct effects of the massive QE III purchases.

For those who have read this far, I stress that this is a pointless purely academic dispute.  I agree with Woodford and Krugman that QE III is much better designed than QE II because a) the Fed is purchases MBS not Treasuries and b) the Fed says it will continue to purchase MBS as long as necessary.  They think improvement b is more important than improvement a.  I think improvement a is more important than improvement b.  But we agree about what the Fed should do which is both  buy risky assets and signal in any way it can that it will accept higher inflation.

update: Bruce Krasting asks why I wrote that QE III is working.  It is partly because, after some frustrating googling, I know that MBS yields are at a record low (but I don’t know how much they decreased).  It was mostly because I knew that the 5 year TIPS rate (the market real interest rate) dropped sharply after QE III was announced.  But when I went to Fred to make the graph, I discovered to my surprise and dismay that the 5 year constant maturity TIPS rate has gone up again and is only about 17 basis points below it’s September 12th close.