For the second time in a row, the minutes of the monetary policy committee meeting of the Federal Reserve really has raised concern. Discussions that are published in this document show that some members of the FOMC have asked to stop the Quantitative Easing strategy. (Quantitative Easing is a program in which the Federal Reserve buys USD 85bn of assets (MBS for 40 and TBonds for 45) each month). Some of them would like to stop it at the end of the year. This discussion is based on the economic scenario for the US economy. If the robustness of the US economy and its capacity to create new jobs are perceived as sufficient then asset purchases could be reduced and then stopped. It is not necessary to go too far and to buy assets for too long if the economy is going better as this purchase strategy could create disequilibria that could at the end harm the economy.
Those supporting the QE strategy are not convinced that the economy is robust enough to stop the purchases. We clearly feel here their reluctance to change their priorities too quickly. They probably have in mind what has happened in 1937/1938 when after a strong growth episode that started in 1933 the Treasury and the Fed wanted to become less accommodative. It was too early and growth stopped. They don’t want to be in the same situation currently as they feel that the economic situation is not that strong. Looking at the economic situation there were positive signs but none that could place the economy on a pace that converges to full employment. GDP level is still far from its potential trend and employment rate is still high (In its latest study on the Federal budget; the Congressional Budget Office (CBO) expects that GDP level will converge to potential GDP level in 2017).
If the economic momentum does not change too much, these discussions on the Fed strategy will continue for a while. On this point it will be interesting to see if the seasonality seen in 2011 and 2012 will continue in 2013. In 2011 and 2012 employment figures at the turn of the year were strong, as in 2013, but weaker in spring. This profile will drive discussions in the coming months. If there is this seasonality (with low employment figures in spring) the exit strategy will almost disappear from the minutes before coming back later this year….
All this is conditioned by uncertainty of the economic outlook. Tim Duy (@TimDuy) summarizes perfectly this point by saying that we do not know how to discriminate between these two propositions “How good does the data need to be to end QE” and “How bas does the data need to be to continue QE” (its comment is here)
We are currently in a corridor that is limited by this two proposals going from one side to the other. After disappointing March figures on employment, retail sales or ISM and after IMF downgrade forecast for US growth in 2013, it is too early to say if the economy is out of these two boundaries.
We can also have a different approach on the Fed strategy. We can calculate the Fed reaction function using macro variables to explain its behavior. That’s what I’ve done and what is shown on the chart below. I’ve made an estimate of this reaction function from January 1990 to December 2007 (monthly data). Following Mankiw I used the difference between core inflation and unemployment rate as the explanatory variable. After 2008 I’ve done a simulation with the parameters that have been estimated and the real observed values for core inflation and unemployment rate. We see on the chart that from April 2009 the Fed interest rate should have been negative to fit the Fed reaction function. As it is not possible to have negative interest rates, the US central bank has used another instrument: quantitative easing. We see that in February 2013, the simulated interest rate is still negative. It’s probably too early to remove quantitative easing.
One point has to be added. One thing is to stop QE another is to increase interest rate. The Federal Reserve said that it will not increase them as far as unemployment rate is above 6.5% with a stable inflation rate and well anchored inflation expectations. As unemployment rate is currently 7.6% the margin with 6.5% is narrow. In fact as participation rate is very low convergence will probably not be linear (participation rate is the ratio of (employment + unemployment) to population of 16 years and over). Participation rate is at 63.3% the lowest level since May 1979. This means that, even if employment is growing, a lot of unemployed leave the labor market as they expect they will not find a job rapidly. But usually when the cycle becomes stronger their expectations change and they are back. At this moment the unemployment rate will rise.
Nevertheless we will still have a lot of discussion on the exit strategy and the moment where the US central bank will increase its interest rates. What we have to keep in mind is that when the Fed will stop its asset purchases, investors will expect a change in Fed interest rates, because the Fed will stop QE when the economic prospect will be good, adding a premium on long term interest rates. The US economy will then have to be strong enough to absorb this shock without changing its trajectory. That’s the real challenge for the US central bankers