For the last 19 months, Fed policy has been on a mission: hiking the Fed funds rate from "accommodative" to the "neutral" neighborhood. Throughout the 350 basis-point tightening journey, from 1% to 4½% Fed funds, the FOMC religiously told us that they didn’t know where neutral was, but that they will know it when they get there. Clever rhetoric, but I’ve never really believed that they didn’t have some pretty firm idea of where the neutral neighborhood would be.
Indeed, before starting this tightening campaign, none other than Alan Greenspan himself said that before the Fed started tightening, it would need to have an idea of where it would end. Yes, he did say that, on January 9, 2004, in response to a question after his speech at the AEA meetings in San Diego:
"…we observe that every time we move, or even indicate that we might move – you can see the whole structure of the yield curve change. That is beginning to tell us that as we start on any particular pattern of monetary policy, we have to ask ourselves where does it end."
At the time, Mr. Greenspan was concerned about a repeat of 1994, when the initiation of a tightening campaign sent long-term rates soaring, as we the market discounted worst-case scenarios for the cycle’s terminal Fed funds rate. In today’s parlance, so-called financial conditions tightened even more than the Fed! With that as backdrop, Mr. Greenspan was telling us in San Diego that in the looming tightening cycle, starting from the 1% Fed funds rate prevailing at the time, the Fed would better manage forward rate expectations than it had in 1994.
As things turned out, of course, the 2005-2006 tightening cycle was not only not a repeat of 1994, but the exact opposite: long-term rates not only didn’t soar, but actually declined modestly, giving birth to the conundrum! Whether this was a good or bad outcome is debated intensely, setting up a parallel debate as to whether the "failure" of long rates to rise means that the "neutral" Fed funds rate is higher than what the Fed might have thought at the beginning of the tightening process. We can’t re-run history to find out.
What we can say is that a 4% terminal Fed funds rate should have been a very conventional forecast before the start of the tightening cycle, falling naturally out of a conventional Taylor Rule. I didn’t make it. But that was not because I didn’t know the Rule, but rather because I thought the FOMC would modify the conventional Taylor Rule, to reflect the fact that the Fed’s secular mission had shifted from winning a war against inflation to securing the peace of price stability. More on this in a bit.
But it was never a secret that the conventional Taylor Rule was screaming 4% as the terminal Fed funds rate. And as Mr. Greenspan hinted, I think that Fed policy makers have always had such a terminal neighborhood in mind, even as they have resolutely argued that they haven’t during the tightening journey.
Now that we are here, it would seem timely to review just what goes into the Taylor Rule. It’s really quite simple:
Fed funds = Equilibrium Real Short Rate + Actual Inflation + 0.5 (Actual Inflation – Target Inflation) +0.5 (Actual GDP – Potential GDP)
As you can quickly discern, if inflation is at target and GDP is at potential, the last two terms in the Taylor Rule "drop out," and the "neutral" nominal Fed funds rate equals the equilibrium real short rate plus the at-target inflation rate. Mr. Taylor assumed that the equilibrium real short rate is 2%. And we all know that the Fed’s shadow inflation target is 2% (max).
Thus, all we need to know to figure out what the nominal Fed funds rate "should be" is an assumption for the NAIRU (non-accelerating inflation rate of unemployment), so as to figure out actual GDP relative to potential GDP (the so-called output gap).
Yes, the dirty little secret of the output gap methodology is that it is, in practical terms, nothing more than a veil for the unemployment gap – the difference between actual unemployment and that which is "needed" to keep wage demands at bay. Yes, the output gap methodology is about maintaining an adequate "reserve army of the unemployed" to resist inflation, as Karl Marx intoned.
Indeed, the late, great Professor Okun gave us a rough and ready ratio (frequently known as his "law") to jump back and forth between the output gap and the unemployment gap: an output gap of 2½% of GDP translates to an unemployment gap of one percentage point. Thus, we can restate the Taylor Rule as:
Fed funds = Equilibrium Real Short Rate + Actual Inflation + 0.5 (Actual Inflation – Target Inflation) + 1.25 (NAIRU - Actual Unemployment)
Now all we gotta do is make an assumption for NAIRU! Consensus at the Fed has long been about 5%, so let’s go with that for a moment (even though philosophically and personally, I believe it is lower). Bottom line: what Taylor says is that at 2% inflation and 5% unemployment, the nominal Fed funds rate should be 4%.
I knew this when Fed funds was at 1%, and so did the Fed. What is more, the formula says that the nominal Fed funds rate "should be" 15 basis points higher than that 4% "neutral" level for every 10 basis points that the inflation rate is over the 2% target and 12.5 basis points higher for each one-tenth percentage point the unemployment rate is below 5%.
Thus, with the core PCE deflator presently running just under 2%, and with the unemployment rate three-tenths of a percentage point below 5%, today’s 4½% Fed funds rate "makes sense." To wit, it was forecast-able before the Fed starting the tightening process from 1% Fed funds.
Owning My Mistake
I didn’t make that forecast, of course. But not because I didn’t know what Taylor’s Rule said! Rather, I believed that the Fed should and would reduce its estimate of the equilibrium real short rate – to 0.5%-1%, versus Taylor’s 2% – to reflect that the Fed secular mission had shifted from winning a war against inflation to securing the peace of price stability. 1 I still believe in the "should" of that thesis. But clearly, the "would" was dead wrong. The Fed’s cyclical reaction function is amazingly the same as during the war against inflation!
Or, maybe not? While the Fed has followed an off-the-shelf Taylor Rule in a fashion that would make former Fed Governor Larry Meyer proud (Larry and I are good friends, even though we are on the opposite ends of the hawk-dove aviary!), the term risk premium has totally collapsed, begetting the conundrum. Thus, we don’t know if the Fed would have "had to go" to 4½%, if long rates had risen in sympathy with the Fed funds rate, even if proportionally much less.
If only long rates had gone up 100 basis points as the Fed was marching upward, taking the ten-year yield to 5½%, maybe the Fed would have stopped the march long ago! Maybe, just maybe, current froth in property markets – the proximate cause of devil-may-care consumer spending – would be a lot less frothy. We’ll never know, of course, because again, we can’t re-run history.
What we do know is that the Fed is behaving as if the secular environment has not changed, while the term risk premium is behaving as if it has. We also know that an inverted yield curve is not a normal state of affairs in a capitalist economy/financial system. Thus, as discussed last month 2, we also know that the current configuration of the yield curve will not endure indefinitely.
In the near term, wiggles in the curve will be "data dependent," as the Fed tells us time and time again. And for now, to figure out what the Fed, and thus the curve, will do, Occam’s Razor says that you should "filter" the data through the conventional Taylor Rule laid out above. Beyond the near-term, however, I believe it is very much "game on" as to how the Fed’s reaction function will evolve.
Ironically, Larry Meyer and I joke about this, as he tells me I’ve been right on long rates despite being wrong on short rates, while I tell him that he’s been right on short rates because he’s been wrong on long rates. Humility, we fervently agree, should be the order of the day.
Our secular view here at PIMCO is straight forward: in a globalized world with a deficiency of ex-USA global aggregate demand, real interest rates should be lower across the maturity spectrum than during the era of a secular war against inflation. So far, we’ve been very right on intermediate and long-term real rates, but very wrong on short-term real rates.
Water under the bridge, you might argue. And in a narrow sense, you would be right. But in a larger sense, you would be wrong. Whether or not the Fed’s reaction function will change, even though it hasn’t so far, is a profoundly important secular question, one which will be answered only over a complete cycle of both tightening and easing.
My key forecasting mistake before the current tightening campaign began was to assume that the Fed would proactively change its reaction function, rather than let time and events prove a change to be necessary. So far, it hasn’t. But that doesn’t mean it won’t, once the lagged fruits of its tightening campaign are more fully ripe.
In the fullness of time, we believe strongly that today’s Fed funds rate will prove to be not neutral, but restrictive, as evolving weakness in residential property activity self-feeds in reflexive fashion. 3 To be sure, time may not yet be cyclically full. But, ironically, the more the Fed pushes the conventional Taylor Rule envelope, the greater the probability that our secular forecast is right!
When the cyclical turn comes, it will be a wicked turn, our guts say, as conventional policy gives way to unconventional property market weakness. It will be time to think and act unconventionally, while listening to a few bars of Mary Chapin Carpenter:
Show a little passion, baby
Show a little style
Show the knack for knowing when
And the gift for knowing how.
February 7, 2006
1"Needed: Central Bankers With Far Away Eyes," Fed Focus, August 2003
2"Is The Price Right?," Fed Focus, January 2006, /LeftNav/Late+Breaking+Commentary/FF/2006/FF+January+2006.htm
3"How To Break the Irony of Too Much Success," Fed Focus, October 2005, /LeftNav/Late+Breaking+Commentary/FF/2005/FF+October+2005.htm