Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Monday, August 31, 2015

Arguments for and against lift off

Should the Fed raise its policy rate this September or not? Seems like a lot of people want to know. For those not following the discussion closely, let me try to summarize what I think are the main arguments for and against a September rate increase.

The main argument for postponing "lift off" goes as follows. The Fed has a mandate to stabilize inflation and unemployment around a set of targets: 2% for inflation and (say) 5% for unemployment. We are presently a bit below the inflation target and a bit above the unemployment rate target. IF one believes that raising the policy rate will move inflation downward and unemployment upward, why would one want to do so right now? How can it make sense to undertake an action that is likely to move both inflation and unemployment further away from their targets? Better hold off for now and await incoming data. There is absolutely no sign of inflation either right now or in the future. If anything, market-based measures of expected inflation are falling.

The main argument for lift off goes as follows. While inflation and unemployment are presently away from their targets (and not by much), this does not mean that ZIRP is consistent with keeping these variables near their targets in the near and medium term. ZIRP has been helpful in bringing unemployment down, but its trajectory is such that it may very well fall below its "natural" rate. IF one believes in the Phillips curve, then undershooting the unemployment rate target will manifest itself as inflationary pressure. And while inflation is presently low, there are reasons to believe this to be transitory. If it is, and if unemployment continues to fall, the Fed may find itself with inflation running above target. At that point the Fed would have to  raise its policy rate much more aggressively than it is contemplating now. Better to raise a modest 25 bp in September (at press conference) rather than wait for December or later (there is no press conference scheduled for the October FOMC meeting). The Fed can keep its policy rate low, or even reverse course as economic conditions dictate. This is still a very easy monetary policy. And it is not unreasonable for the Fed to act in a manner that prevents it from falling "behind the curve" (as it has arguably done in the past).

Of course, both of these views are predicated on essentially the same theoretical (essentially New Keynesian) framework. If you don't buy into this framework (for what it's worth, I do not), you're likely to have a different set of policy recommendations. Feel free to propose yours below!


  1. Does it make sense to explain that inflation expectations are low because ZIRP implies the Fed forecasting lower growth into the future? So that raising the rate next month signals to the market the the Fed is confident in future growth... which should raise inflation expectations and then inflation?

    Not sure if I have it right.

    1. I am aware of a few people who make this argument. In principle, it makes sense if one believes that the FOMC consistently has access to better information than the market. But if this is true, then why not just release the information and make it public? In truth, the necessary information advantage does not exist. Therefore, the argument is suspect, in my view.

    2. I guess I'm influenced by the Bank of Canada policy this year. Their January rate cut seemed to surprise everyone, coupled with Poloz' statement about the 'atrocious' 1st Quarter here.

      I suppose that information advantage is pretty temporary, but it's the main reason I had this on my mind. Thanks!

  2. David, in your opinion, is it possible that the answer to "Should the Fed raise its policy rate this September or not?" is "It hardly matters?" I'm not saying that's what I believe, I'm just curious if that's on the table in your brain regarding possible answers.

    "Feel free to propose yours below!"
    Oh boy... things are going worse over there than I thought! Now you're soliciting the general public for ideas about what to do?!?! Will you be forwarding our suggestions to Yellen, or will she be sending us follow up emails directly? ;^)

    "If you don't buy into this framework (for what it's worth, I do not)..."
    Do you buy into any framework? If so, does it have a name?

    1. In the grand scheme of things, it probably will not matter much. I do find Tim Duy's take on the matter interesting though:

      I asked the question because people are often willing to criticize (which is good, of course), but when asked what they would do instead and why, it leads them to pause and think.

      My own framework is heavily weighted these days to the "shortage of safe assets for a host of different reasons" hypothesis. The main implication of this theory is that the problem of low inflation is largely out of the control of the Fed, at least, unless the fiscal authority is willing to accommodate with additional deficit-financed tax cuts.

    2. Thanks David. Say, in case you didn't know, Noah was asking for an explanation of your comment here:

  3. "If you don't buy into this framework (for what it's worth, I do not), you're likely to have a different set of policy recommendations. Feel free to propose yours below!"

    Depends. Do you think "monetary policy" is about interest rates?

  4. I find Orphanides' recent argument for a rule-based approach compelling:

    In it he states:

    "If policymakers wish to ensure full employment and price stability over time, they cannot afford to permit immense policy accommodation in the system once full employment is reached. If they did, it would not be feasible to withdraw that accommodation on time without either generating inflation or tightening so abruptly it could cause a recession. For this reason, policy normalization ought to commence long before reasonable estimates suggest full employment has been restored." (at p. 182)

    His allusion to policy lag is particularly relevant in the current context. I agree that liftoff in October or December rather than September will have little substantive difference in the near term; however, if the FOMC continues to wait for red flags in the data before enacting policy change, it may get caught flat-footed, which could lead to a lumpy or steeply-sloped normalization path.

  5. The Federal Reserve indeed appears to have lost the ability to nudge inflation rates. Perhaps the core problem is that the Fed and supporting economists view low inflation as the problem.

    Why not reset the inflation target to 0% with a 2% range from -1% to +1%?

    Or is fuzziness and a little confusion necessary to make the price system promote growth?