This week's FOMC meeting is shaping up to be quite the event. Not for the actually policy result itself, which is widely expected to be unchanged, but for the subsequent press conference with Federal Reserve Chairman Ben Bernanke. The Fed's communication strategy has clearly unraveled in recent weeks, and Bernanke has an opportunity to regain control. But will he be able to do so, or will he leave even more confusion in his wake?
Start with the basics, the statement itself. The Fed is not going to change the pace of asset purchases this week. Recent Fedspeak has made clear that it remains too early to reduce monetary accommodation. The statement will probably be relatively unchanged. I anticipate that it takes note of some moderately weaker data since the last FOMC meeting, as well as lower-than-anticipated inflation. Neither, however, is sufficient to drive asset purchases higher. It will be interesting to see how much the Fed emphasizes the fiscal contraction. If the statement shifts to the side of "fiscal contraction appears to be having little impact on private activity," the implication would be that they are looking through the fiscal drag on headline GDP numbers. That obviously sets the stage for reducing asset purchases sooner than later.
Next come the closely watched forecasts. Near-term forecasts for inflation and perhaps GDP growth may be softer, but also watch the longer-term forecasts and especially any change in the expected path of unemployment. The latter, I think, is critical in setting an end to asset purchases - the Fed will want to be draw QE to a close prior to hitting the 6.5% threshold at which point it has said it will evaluate rate policy.
Finally comes the press conference. Here is where the real action should take place. Market participants have become increasingly concerned about ending asset purchases. I believe that market participants are having trouble understanding the implications, or lack thereof, of altering the pace of quantitative easing on interest rate policy. Bernanke will attempt to clear the way for ending quantitative easing while attempting to divorce that decision from any subsequent decision on rate hikes.
This, of course, will be easier said than done. I think a critical problem is that the Fed does not want to be pre-committed to some policy path, but at the same time does not want to surprise markets. It wants to avoid a repeat of 1994, with the sharp spike in yields viewed as a communications failure. The Fed believed it would resolve this divide by shifting the focus to the data. It was wrong.
Ylan Mui at the Washington Post describes the nature of the problem:
Investors increasingly have focused on predicting the moment the Fed will start to pull back on its massive stimulus program...
...It’s the type of parlor game the Fed had hoped to avoid. Instead, it has tried to convince the markets that the date is less important than the data.
Fed officials deliberately chose not to attach a time frame to their easy-money policies when developing their forward guidance for the public last year.
Stop right here. I think it is important to note that Fed policymakers are the ones who started the ball rolling on the importance of the date over the data. Specifically, at the beginning of April San Fransisco Federal Reserve President John Williams defined a time line for ending QE given the current path of data. At that point the conversation shifted from data to date. Other policymakers followed suit.
Does that mean that Williams made an error? Not necessarily. I think it is impossible to communicate the path of policy without making market participants aware of the associated timeline. Once you describe your view of the data and your forecast, by default you will define the expected time for the policy change. In effect, the Fed can't have it both ways. It can't jointly pretend the date doesn't matter while at the same time clearly communicating the path of policy. If it doesn't communicate the path/timing, then the eventual policy move will trigger an overreaction. If it does communicate the path/timing, but doesn't explain how that path fits in the context of the data, then it also risks an overreaction.
The latter is the position the Fed now finds itself in. Williams let the cat out of the bag, that the Fed had a timeline in mind. But it is challenging to see how the data fits into the timeline. Back to Mui:
The goal was to help investors come to better conclusions on their own by revealing the public data that Fed officials use as guideposts. In theory, that means interest rates would hew more closely to incoming data than to Fed pronouncements.
But, as it turns out, there are many ways to parse the numbers.
“They kind of threw out these conditions,” said Michael Feroli, chief U.S. economist at JPMorgan. “They’re telling us something but not telling us something.”
Honestly, it is difficult to make the case for ending asset purchases on the data alone. It is neither clear that the labor market is stronger and sustainable nor that asset purchases should be cut in the face of falling inflation. In fact, I think you can argue that the Fed is moving the goalposts to some less defined objective. As noted above, I think the path of the unemployment rate plays a role. But so too does financial stability concerns. And also general discomfort on the part of policymakers about the size of the balance sheet. Indeed, the December conversion from Operation Twist to asset purchases may have been simply insurance against a fiscal disaster that did not materialize. Ultimately, the shift to tapering talk was too abrupt given the data, and that raises the possibility that some undisclosed factor is at work.
Now, if we don't understand what is driving the decision to scale back asset purchases, then it is likely we also don't understand how the data will impact subsequent interest rate decisions. Again, back to Mui:
Part of the problem has been muddy economic data that do not provide a clear signal of where the recovery is headed. The Fed also has left itself plenty of wiggle room to interpret the data. It did not define what “substantial improvement” would be required to dial back its $85 billion-a-month in bond purchases, and it has suggested that it could leave interest rates untouched even after its unemployment or inflation thresholds are met.
If the Fed's plans for ending quantitative easing are opaque relative to the data, then what is are we to expect when the unemployment threshold approaches. An equally opaque policy response? Is the Fed going to change the goalposts again? When does it shift from unemployment and inflation to concerns about financial stability? And how does it propose to pretend that you can commit to some data dependent path without implying a related time line?
Bottom Line: This FOMC meeting is about the Fed regaining - or further losing - control over its communication strategy. Bernanke will attempt to detail how exactly the data flow is supportive of scaling back asset purchases in the next few months (I believe the Fed prefers September) while at the same time disassociating asset purchases from interest rate policy. I think it is important that market participants believe that the shift to tapering talk was entirely data dependent and not influenced by some other factors. Otherwise, they will doubt the supposed data dependent thresholds for rate policy. And the Fed is going to have to come to terms with the reality that the instant it starts to anticipate a change in policy, it starts a clock ticking. Making the distinction between date and data is not as easy as it sounds.