January 09, 2006
What's the Fed waiting for?
The Fed is likely to stop raising rates soon. What will be the final signal that enough is enough?
Commentators seemed to read the minutes from the December FOMC meeting as signaling that we may be nearing the end of interest rate hikes. To summarize a few of the reactions:
the tone now seems to be that most of the heavy lifting is done.
I left the minutes with the sense that another rate hike at the end of the month is in the bag, but beyond that, future changes in policy are not automatic but instead data dependent.
The Fed announced their plans today to "only invert the Yield Curve somewhat," rather than the deeper and more extensive inversions some traders feared.
So what will cause the Fed to pause? We already know that it's not concerns about the employment effects of the energy price shock. For whatever reason, the Fed has decided that its mission in response to such disturbances is to worry about the inflation rather than the unemployment that such events may produce.
And I agree with Tim Duy that it will take more than just the disappointment in the December employment report, which, since it accompanied a revision of the November figures, left the two-month average at a still-respectable 206,000 new jobs per month.
And we've already seen that an inversion of that little old yield curve doesn't scare the Fed. As I noted earlier, the main reason that Greenspan seems unafraid of the yield curve inversion is a belief that it is only when the nominal interest rate is high relative to inflation that there is much concern of an economic downturn.
The graph at the right plots the nominal yield on 3-month Treasury bills minus the median change in the CPI components over the preceding 12 months, with U.S. recessions indicated by the shaded areas. This "real interest rate" series averaged 1.29% over the last 40 years, and was above that average value prior to each of the last five recessions. Up until October, this measure was below its average historical value, and indeed even negative during the overstimulus of the economy of 2001-2004.
But whatever comfort the Fed (or you) may have taken from this fact is about to be taken away. With the tbill rate now at 4.22% and the median CPI over the last 12 months of 2.4%, this measure would currently stand at 1.82%. If we get another 25 basis point bump at each of the next two FOMC meetings, or even more after that, not much comfort in that little factoid at all, I would say.
|Measure||Implied real rate|
|TIPS due 2007||2.50|
|TIPS due 2011||1.98|
|CPI (as of Nov 05)||0.72|
|CPI (as of Sep 05)||-0.48|
Ah, but the fun part about the "real interest rate" is that it's not real at all, but depends on investors' expectations of inflation, which nobody can actually measure. If you look at the rate on inflation-indexed Treasuries maturing in 2007 or 2011, you get an even higher current real interest rate. On the other hand, if you use the change in the actual CPI over the last 12 months of 3.5%, rather than the median CPI as I have done above, you calculate a real rate of only 0.72, leaving lots of room for the Fed to keep reaching up. For that matter, if you used the change in the actual CPI for the 12 months ended September 2005, you get a real rate of -0.48%-- another year of "measured pace", anybody?
Of course, that last observation-- that we would get a wildly different prescription for policy depending on whether we use data from September or November-- is one of the main reasons I hope that the Fed would be paying more attention to something like the median CPI. And, if the Fed-watchers quoted above have it right, that's what the Fed is doing.
Posted by James Hamilton at January 9, 2006 06:17 PMdigg this | reddit
Your chart also makes a different point which worries me quite a bit: the last time the monetary policy had been that stimulus as in the last 4 years was in the 1970s, which was followed by runaway inflation. Should I worry the long lags of inflation, or should I take the benign core inflation this year as evidence that the Fed gets away this time?
Posted by: pat at January 10, 2006 07:58 AM
Real interest rates recently were not as low (negative) as during the early 1970s (see graph at site above). There are other differences. So the circumstances are a little different now vs. the 1970s.
I lean toward a pause at 4.25%. Let's give Bernanke a chance to do one raise to 4.5% and see where to from there. I realize I am probably in the minority and that many may be expecting an increase to 4.5% before Bernanke takes over for Greenspan.
Posted by: nate at January 10, 2006 08:26 AM
When Bush has finished destroying the middle class?
Just a thought.
Posted by: save_the_rustbelt at January 10, 2006 08:36 AM
JDH: "What will be the final signal that enough is enough?"
When the Fed decides to hide M-3 from the American public.
Posted by: John at January 10, 2006 11:00 AM
I am not an economist to understand the implication. But, I think, graph has been read selectively here.
1. before '79 recession the graph is not even as high as it is now.
2. Between last two recessions, it has remained above the average line for quite some time.
3. also, between '82 and '91 recessions there were fluctuations above average line.
I do not disagree with a possibility of recession induced by monetary policy but, this graphical illustration doesnt satisfy a common reader.
Posted by: Pankaj at January 10, 2006 12:27 PM
Pankaj, I was not inferring from the graph that a high real rate is a sufficient condition for an economic recession, but rather was repeating the views of others that it is a necessary condition. As for my own views, you might take a look at some of my earlier posts such as  or .
Posted by: JDH at January 10, 2006 12:44 PM
Interesting that the magnitude of debt which must be financed and refinanced doesn't enter this "real interest rate" discussion. It seems to me Bernanke is right to worry about deflation, since there has been so much debt creation. Higher rates very simply mean debt is harder to pay off, which means more defaults, which means bank write-offs. If I've got it right, that is the path to spirals of deflation.
Posted by: Jeff G at January 10, 2006 09:46 PM
Another measure of inflation expectations is the price of gold. Since gold pays no interest the only way to make money with it is for its price to rise.
Recently the price of gold has reached 25-year highs. This would seem to indicate that "they" believe inflation is either here now or about to be.
Also, the rate of increase in M2 has accelerated in recent months. This and the gold price would indicate that further interest rate increases are coming.
Posted by: jim miller at January 11, 2006 04:57 AM
The dollar was up 15% in 2005 because oil is priced in dollars and oil was up big.
(Central banks demanded more dollars to be able to pay for oil.)
Iran plans to price its oil in Euros in March 2006.
Iran?s move could devalue the dollar.
A devalued dollar could destabilize the financial system: stock markets, banking, etc.
The Fed announced it will no longer report the money supply, M-3, in March 2006.
The Fed's decision to hide M-3 means it is about to be increased dramatically.
The Fed can ?monetize? assets by ?printing? more dollars to [try to] keep markets stable.
Historically, the stock market and gold behaved inversely ? but this has changed:
The stock market is up because the money supply is supporting the market -- all that money has to go somewhere.
Gold is up because too many Dollars is inflationary and M-3 is going through the roof:
Over six weeks, M-3 is up $177.8 billion or 28% annualized.
Posted by: John at January 11, 2006 07:32 AM