Saturday, August 27, 2011

Real Rates, Inflation and the Monetary Policy Stance (fg)

I totally agree with amv that the level of nominal interest rates per se has nothing to say about the stance of monetary policy.

In order to address the stance, we need a reference model against we can judge monetary policy actions. John Taylor proposed such a beginners'-rule where the equilibrium, Wicksellian rate is assumed to be constant and monetary policy is advised to change the policy rate more than one-by-one when inflation and/or output move. Rule-based reaction functions evolved out of this research with natural rates being specified as time-varying and determined by real factors (see for instance Laubach/Williams (2003), Measuring the Natural Rate of Interest, Review of Economics & Statistics, 85). Other benchmarks include checking the path of nominal GDP whether monetary policy is tight or loose.

One important transmission mechanism, according to standard theory, is the long-term real interest rate to which aggregate demand reacts most sensitive. If monetary policy has no leverage power to affect the long-term real interest, then monetary policy is ineffective. As long as there is an impulse, the financial, ex-ante real interest rate as measured by in the TIPS market deviates from the equilibrium, natural, Wicksellian rate.

Gerlach/Moretti (© argue that in the long-run, monetary policy cannot influence steady state equilibrium and, thus, the natural rate of interest. D'accord! If there is a one-time move in the policy rate, then, this effect should cancel out. As the natural rate is determined by productivity and thrift in goods market and due to no-arbitrage considerations (efficient market hypothesis), the real return on real capital sould be equal to the inflation-adjusted return on financial assets which cannot be affected by monetary policy in equilibrium.

Two points are worth making:

(1) Monetary policy operates on a regular basis. Consequenlty, in each point in time, market expectations about the future stance of policy are imbedded in the nominal as well as real yield curve, they do not cancel out in the medium term depending on the duration of monetary transmission; but they do converge to a time-varying level that is assumed to prevail in the long-run: the long-run inflation-adjusted rate which is a market belief of market participants (including the central bank). This rate, again, should be equal to the natural rate by no-arbitrage consideration in equilibrium. When jointly and consistently investigating nominal and real yields which are, again, linked by the efficient market hypothesis, we can decompose the long rate and the long-term forward rate into its components. It becomes clear that (i) most of the declince in the real long rate has been the result of the decrease in the financial market risk premium (ii) the risk-neutral expectation of market participants remains constant so that expectations of the Wicksellian rate remain constant, too (2%, iterate forward and you get the long-run version) and (iii) indeed, there was a decline in the risk-neutral real rate in 2000.

Figures are from Backus 2007, Brooking Papers on Economic Activity, and further estimates are kindly provided by Hördahl et al. 2010, ECB working paper.

(2) Gerlach/Moretti claim that
Real interest rates fell for reasons unrelated to monetary policy.
I am a bit sceptical on that unidirectional statement for a variety of reasons. Firstly, research shows that monetary policy affects the behavior towards risk through lowering both the quantitiy of risk and the market price of risk (risk appetite). Therefore, monetary policy contributed to the decline in the long rate through falling risk premia. Secondly, real rates massively fell when the FED communicated that it was likely to stay accommodative for a considerable period of time (not mirrored by policy moves). Such communication might have fostered the market view of low policy rates in the future thereby promoting low long-term interest rates and further risk-taking. Thirdly, if studies in the spirit of Taylor are conducted with time-varying natural rates, there still remains a negative real rate gap for the US for the period 2003-2006. Forthly, inflation did pick up speed in the US with above average nominal GDP growth. Break-even and survey measures of inflation did pick up in July 2003 and inflation was increasing in the euro area and the US (Clarida, 2010). Headline inflation in the US rose to 4.5% (CPI inflation even to more than 5.5%) in 2008 and core inflation picked up to 2.6%. Fithly, a transmission mechanism that allows for spill-overs to financial markets and back to the real economy shows that reacting inversely to risk premium dynamics is optimal for monetary policy that wants to stabilize inflation and output. Moreover, it is hard to imagine that goods markets can be in full equilibrium, and hence growth can be sustainable, in the presence of the observed credit boom in the run-up to 2007. The subsequent full-blown financial crisis suggests that the unusually rapid credit expansion was a sign that market rates were below the natural rate that did not show up too much in goods inflation but in asset price inflation by way of plumping discount rates (risk-neutral part as well as the risk premium) which are used to price financial assets and capital goods.