Monday, January 27, 2014

The Fed's Foward Guidance is Not Truly State Dependent

In my previous post I criticized the Fed's forward guidance without explicitly getting into the difference between time dependent and state dependent forward guidance. Time dependent forward guidance is when a central bank attempts to shape the expected path of interest rates without conditioning on the state of the economy. Sate dependent forward guidance does condition on the state of the economy. The latter approach is better and is what the Fed claims to have been doing since December, 2012 when it adopted the Evans Rule. Still, there is much to be desired about the Fed's forward guidance. First, the Fed's forward guidance focuses too much on the expected path of the interest rates and not enough on its target. In terms of my earlier analogy, the ship's captain worries excessively about expected path of the rudder and too little about the ship's destination. This can be seen in the Fed's qualification of its inflation and unemployment thresholds. Here, for example, is how the FOMC described it forward guidance at the December, 2013 meeting. 
To support continued progress toward maximum employment and price stability, the Committee today reaffirmed its view that a highly accommodative stance of monetary policy will remain appropriate for a considerable time after the asset purchase program ends and the economic recovery strengthens. The Committee also reaffirmed its expectation that the current exceptionally low target range for the federal funds rate of 0 to 1/4 percent will be appropriate at least as long as the unemployment rate remains above 6-1/2 percent, inflation between one and two years ahead is projected to be no more than a half percentage point above the Committee's 2 percent longer-run goal, and longer-term inflation expectations continue to be well anchored.

Note that the FOMC says its highly accommodative stance of monetary policy will persist for a "considerable time after" the economy recovers and that the current low rates will be appropriate "at least as long" as the breaching of its thresholds. The Fed is saying here that its accommodative stance, including the low policy interest rate, will not be shaped by the state of the economy alone. It also will be shaped by time. The Fed's forward guidance, then, is not truly state dependent. This makes the forward guidance time-inconsistent and not credible for the reasons laid out in my last post.

Second, the Fed's use of the unemployment rate as a threshold is problematic. Central banks should never target real variables unless they are certain they know its market clearing or 'natural rate' value. Determining natural rate values, however, can be challenging if not impossible to do in real time. Consequently, it is far better and easier for central banks to target nominal variables. I made the case it should be nominal GDP and that doing so would accomplish what forward guidance is trying to do. Jeff Frankel agrees:
The Federal Reserve and the Bank of England have each recently backed away from “forward guidance” that they had given earlier in the form of thresholds for the unemployment rate.   As a result of their changes in emphasis, they are both being accused of confusing the financial markets.
There was another way.  A year or two ago, many of us were suggesting that the monetary authorities could announce a target or threshold for Nominal GDP, instead of for inflation, real income, unemployment, or other alternatives.    Some of us explicitly warned that a threshold phrased in terms of the unemployment rate would be vulnerable to extraneous fluctuations such as workers exiting the work force, and argued that a nominal GDP threshold would be more robust with respect to such unforecastable developments.

Just over a year ago, for example, I wrote in favor of “a commitment to keep monetary policy easy so long as nominal GDP falls short of the target.  It would thus serve a purpose similar to the Fed’s December 12, 2012, announcement that it would keep interest rates low so long as the unemployment rate remains above 6.5% - but it would not suffer the imperfections of the unemployment number (particularly its inverse relationship with the labor force participation rate…).”  [Central Banks Can Phase in Nominal GDP Targets without Losing the Inflation Anchor blog, December 25th, 2012.]
This is yet another instance of a long-standing point: if central banks are to focus attention on a single variable, the choice of Nominal GDP  is more robust than the leading alternatives. A target or threshold is a far more useful way of communicating plans if one is unlikely to have to violate it or explain it away later.
It is time for the Fed to scrap forward guidance and adopt a NGDP level target.


  1. Forward interest rates guidance can't be used to project gDp. The FED’s target or transmission mechanism (interest rates), is indirect, varies widely over time, & in magnitude.

    Only rates-of-change in aggregate demand can be used to forecast gDp. As Greenspan pontificated in “The Map & the Territory”: “The laws of physics…once identified, rarely have to be revised”:

    Rates-of-change (roc’s) in monetary flows (our means-of-payment money times its transactions rate-of-turnover), equal roc’s in all transactions in Irving Fisher’s “equation of exchange”: (MVt=PT). Roc’s in nominal-gDp are a proxy for all transactions.

    The lags for monetary flows (MVt), i.e. the proxies for (1) real-growth, & for (2) inflation indices (for the last 100 years), have been mathematical constants

  2. For example, Alan Greenspan attempted to "tighten" monetary policy for 41 consecutive months (raising the FFR on 17 separate occasions or until Bernanke took over). But at no point did the rate-of-change in money flows (proxy for inflation), reverse its course.

    An "abrupt policy adjustment" could have been made at any juncture with virtually no economic disruption. Unfortunately the Federal Reserve doesn’t gauge the volume and timing of its open market operations in terms of the amount and desired rate of increase of member commercial banks legal reserves, but rather in terms of the levels of the federal funds rates.

    By using the wrong criteria (interest rates), rather than member bank legal reserves, in formulating and executing monetary policy, the Federal Reserve errored on both the upside & the downside of real-gDp and prices.

    Monetarism has never been tried. Paul Volcker's administration never deviated from Paul Meek's 1974 description of "open market operations" (using non-borrowed reserves).

    The "administered” or actual prices (in the case of the Great Recession - housing prices), would not be the "asked" prices, were they not “validated” by (MVt), i.e., “validated” by the world's Central Banks.

    Bankrupt you Bernanke confessed: "Like most crises, the recent episode had an identifiable trigger--in this case, the growing realization by market participants that subprime mortgages and certain other credits were seriously deficient in their underwriting and disclosures. As the economy slowed and HOUSING PRICES DECLINED, diverse financial institutions, including many of the largest and most internationally active firms, suffered credit losses that were clearly large but also hard for outsiders to assess. Pervasive uncertainty about the size and incidence of losses in turn led to sharp withdrawals of short-term funding from a wide range of institutions; these funding pressures precipitated fire sales, which contributed to sharp declines in asset prices and further losses"

  3. Good post. My view is that the Fed has provided some modestly useful forward guidance, but an amount that falls far short of true state-dependent forward guidance.

  4. Rates-of-change in MVt = roc's in the target - nominal-gDp. QE operations don't assure that the money stock will expand. In fact, if you perform the reconciliation (back into it), you will find that QE operations predominately are transacted with the commercial banking system (primary dealers), & not the non-bank financial intermediaries (where money would actually be created). This effect: "pushing on a string" was the direct result of paying interest on excess reserves.

    The distributed lags for money flows are not "long & variable". Even given the non-conforming data provided by the uncooperative Federal Reserve's technical staff, roc's in the proxy for inflation have been mathematical constants for the last 100 years.

    The roc in the proxy for inflation peaked in Feb during 2013 @ .603. This coincided with the peak in gasoline prices for the year. The roc subsequently fell until it bottomed at subpar levels in Dec @ .251. The Fed has the tools for an "abrupt policy adjustment". But that adjustment can't be carried out using an Evans rule.