Friday, January 24, 2014

Forward Guidance is Hard. So is Navigating a Ship By Focusing on the Expected Path of its Rudder

Forward guidance took a beating this week. First, San Francisco Fed President John Williams said its use is often oversimplified, easily misinterpreted, and time-inconsistent. Next, an important Blooomberg editorial argued that forward guidance should be dropped because it lacks credibility and at best confuses investors. Finally, the Bank of England decided to abandon forward guidance altogether once it realized its unemployment threshold would be reached well before a full economic recovery.

This spate of bad news for forward guidance should not be surprising. For it is like trying to navigate a ship by focusing on the expected path of its rudder. No ship captain can credibly commit to a certain rudder path across the ocean because water currents and wind conditions are bound to change. A captain, however, can credibly commit to a destination and do whatever it takes to get there. In this case, the rudder is not the focus and is adjusted as needed to offset the unexpected changes in current and wind conditions so that the ship stays on course.  

Similarly, a central bank cannot credibly commit to a certain path for its policy rate if it wants to hit its destination. For most central banks that means full employment and stable inflation. Forward guidance commits the central bank to a path of low policy rates beyond this destination in order to get more stimulus today. That is akin to a captain saying it will keep his rudder and ship on a certain path even after they make it across the ocean. No one would take that claim seriously. The ship has to stop when it hits the shore. Likewise, a central bank cannot commit to a flexible inflation target and at the same time credibly commit to a path of policy rates that would violate this monetary regime. This is why markets are now challenging the FOMC's forward guidance on interest rates. They see an economic recover that will force the Fed to raise interest rates faster than the FOMC says it will.  

This creates a dilemma for central banks. How can they credibly commit to the monetary policy needed for a more robust recovery while still remaining committed to long-run price stability? If forward guidance cannot cut it, what will? The answer for the captain of the ship was to focus on the destination. The same is true for central banks. Focus on the destination or target of monetary policy.

The target, however, needs to be the right one. Market monetarists have been making the case that a nominal GDP level target (NGDPLT) is just the one. It allows the short-run monetary policy flexibility on both sides of the business cycle while still anchoring long-term inflation. It would solve the dilemma now facing central banks that use forward guidance. Here is how I described it before:

Under a NGDP level target a central bank would commit to keeping aggregate nominal spending on some targeted growth path, say 5%.  Such a rule would, therefore, anchor long-run inflation expectations.  It would also allow for aggressive catch-up growth (or contraction) in NGDP so that past misses in aggregate nominal spending growth would not cause NGDP to permanently deviate from its targeted growth path.  In other words, past NGDP growth mistakes would be corrected.  The following figure illustrates this idea:

The black line has NGDP growing at a 5% annualized rate.  Then, at time t a negative aggregate demand (AD) shock causes NGDP to contract through time t+1.  There is now an a NGDP shortfall.  To make up for it, the Fed must actually grow NGDP  significantly faster than 5% to return aggregate nominal spending to its targeted level.  For example, if NGDP fell 6% between t and t+1 it is now 11% under its trend.  Next period the Fed must make up for the 11% shortfall plus the regular 5% growth for that period.  In short, the Fed would need to grow NGDP about 16% between t+1 and t+2 to get back to trend.  This temporary burst in NGDP would probably make the inflation critics nervous, but they shouldn't be. There might be temporarily higher inflation as part of the rapid NGDP growth, but over the long-run a NGDP level target would settle back at 5% growth.  Nominal expectations would be firmly anchored.  
Note that by stabilizing nominal spending growth, the Fed would also be stabilizing nominal income growth. That leads to a related point:
Under a NGDP level target any temporary easing (or tightening) in the short-run is very time consistent, because it is returning nominal income to the Fed's long-run growth path target.  That's kind of the point of level targeting, it coordinates short-run policy moves with long-run policy objectives. Under a NGDP level target, everyone understands the "catch-up" nominal income growth is temporary and tied to an objective. The public would also understand that once the target path was hit, nominal income growth would slow to trend. They would come to expect it.  In other words, there are no inconsistencies between the short-run and long-run.
So we market monetarists believe NGDPLT would accomplish what central banks are now trying to do with forward guidance. It appears that others are now taking a second look at NGDPLT for this reason. For example, San Francisco Federal Reserve Bank President John Williams had this to say in his critique of forward guidance (my bold):
One lesson from the recent past is the difficulty in anchoring policy expectations when the short-rate is at the ZLB. Although quantitative forward guidance has proven a useful tool, it suffers from a number of limitations. Experience has shown that it is impossible to convey the full reach of factors that influence the future course of policy. As a result, forward guidance ends up being overly simplified and prone to misinterpretation. Moreover, forward guidance several years in advance may not be credible, especially in light of the change in policymakers over time. In theory, alternative frameworks such as nominal GDP targeting, if fully understood by the public, could help resolve these communication difficulties
An even stronger endorsement of NGDPLT was made by the Bloomberg editorial that criticized forward guidance (my bold):
The world’s most powerful central banks are struggling with their approach to “forward guidance” -- what they tell investors about their plans for monetary policy. A practice meant to give the markets more clarity is causing confusion...

In all this, there’s an underlying dilemma for the central banks: If forward guidance is to provide additional stimulus, the central banks have to change investors’ understanding of how interest rates will respond to economic conditions. Yet, for the sake of credibility, the Fed and the others want investors to see monetary policy as steady and consistent. Much of the effort to get forward guidance right is a doomed effort to have it both ways...

A more radical alternative is also worth considering: replacing inflation targets with targets for growth in total demand, also known as nominal gross domestic product. This would allow for temporary overshoots in inflation in response to periods of very low demand, thus providing extra short-term stimulus, without unsettling long-term inflation expectations. And unlike ad hoc targets and thresholds, it’s a framework that stays in place throughout the economic cycle.   
Well said. Foward guidance is hard, NGDPLT is easier. Here is hoping the Janet Yellen will have her 'Volker Moment' and be the Fed chair who ushers in NGDPLT.

Update: If you think this analysis does not apply to the Fed since its forward guidance is state dependent, think again. In a new post I argue there that even now with the Evans Rule the Fed's forward guidance is not truly state dependent.
1Technically, the federal funds rate is an intermediate target and bank reserves are the operating instrument.


  1. David, great post. I "paid tribute":

  2. Very good post David. Nautical metaphors are the best!

  3. Economics is a field where assumptions define the behavior.

  4. Excellent blogging...just make sure that NGDP targets are robust...left to their own devices, central banks will suffocate real GDP while hitting low NGDP targets