My question: “Why is the Fed using CPCE versus another measure of inflation?” The very good news is that there is answer, and it comes from a very "reliable' source – The Federal Reserve. A detailed analysis on this topic was conveniently made public just a month ago.
Alan Detmeister produced a doosey of a report. Ya gotta love the title page:
This beast runs 25 pages, it includes tons of charts and references. It compares the utility of using Core PCE to a dozen other inflation yardsticks. There are easy-to-understand formulas to support the conclusion:
Guess what? Good old Alan makes a compelling argument. Anyone who tries to question the use of CPCE is going to get hit over the head with this report. This is just one of the many charts that prove (according to Detmeister) that CPCE is the only way to go when considering monetary policy:
There’s just one teeny problem with Alan’s work. He did all of that comparing and studying using data from pre-2010. Using that information, CPCE lines up very well as a consistent barometer of inflation. But the analysis falls to shit when you look beyond 2009. CPCE took a nose-dive after 2009 (versus CPI and Core CPI):
Information on CPCE and the other measures of inflation is available monthly. There’s no reason (that I can think of) why the Fed chose to deliberately omit two years of data that would conflict with the “desired’ conclusion. To me, it looks like the authors manipulated the report.
I think the Fed made a mistake targeting inflation. It's now stuck with the choice. It can’t go back on this without looking awfully stupid. The policy of allowing CPCE to determine the direction of monetary policy will last the until the end of Bernanke’s term at the Fed. Then it will be abandoned in favor of more pragmatic approaches to decision making.
I think there is enough monetary juice in the global system for there to be a risk of inflation north of 2%. We shall see. I think Bernanke is going to get his balls squeezed. He deserves that fate, he put them in a vise. As of today, he no longer has choices. He’s made himself a slave to a single dopey statistic.
The markets are the best measures of how people perceived today's announcements from the Fed. The dollar pissed on the Fed in general, the gold market hit Bernanke square in the face with an ingot.
Posts
Good post.
ReplyDeleteIs the guys name really "debt"meister? That's too much!
DeleteThanks Bruce, nice post. Looks like major inflation on the way...
ReplyDeletefor rookies like myself:
http://en.wikipedia.org/wiki/Personal_consumption_expenditures_price_index
"...This price index method assumes that the consumer has made allowances for changes in relative prices. That is to say, they have substituted from goods whose prices are rising to goods whose prices are stable or falling."
http://www.bea.gov/faq/index.cfm?faq_id=518
"The redefined "core" PCE price index will be more consistent with the most frequently cited purpose of a "core" price index: to strip away volatile components to reveal "underlying" inflation. Food services prices are not volatile while prices of purchased food, such as fresh vegetables, are."
Do you realize that in your last, "gotcha!" figure, that you're plotting the CPCE against the Consumer Price Index? That is, you're not plotting "Core Inflation Rate" against "Regular Inflation Rate." You're plotting the rate of change against the total. It's like travelling in a car and plotting "miles per hour" against "distance traveled," and then going, "HEY, these diverge! The speedometer must be a conspiracy!"
ReplyDeletePost is on point
ReplyDeletehttp://www.youtube.com/watch?v=GbOWiJ94Xvg
ReplyDeleteBernanke. Perhaps not the sharpest tool in the shed?
The Fed has gone 'all in' supporting asset prices, especially equities. I think they've pretty much given up on housing (MBS was the subject of QE1), as housing depends on employment, and they are smart enough to know that the manifold easing to date has done next to nothing to create jobs, and jobs is the only thing that will help housing recover. I think this is the Fed's only real goal at this point -- making sure stock markets don't tank. If the market began a serious correction, it would not be long before an overt QE3 was right back on the table. Of course right now there is stealth QE3 -- ZIRP etc is effectively stealth QE3, just more dispersed. If interest rates stay low long enough banks will eventually earn their way back to solvency.
ReplyDeleteI also think this is why there has been no serious talk of austerity in the US (as opposed to Europe) -- everyone knows QE3 will be implemented if the markets tank.
So the Fed 'put' is still strongly in place. With that going on, shorting the market remains very dangerous.
Ben S. Bernanke laid the groundwork for a third round of large-scale asset purchases should unemployment remain higher than the Federal Reserve would like while inflation falls below a newly-established target. The Federal Open Market Committee “recognizes the hardships imposed by high and persistent unemployment in an underperforming economy, and it is prepared to provide further monetary accommodation,” Bernanke said yesterday at a press conference in Washington.
ReplyDeleteQuestion: Where, or what, is the evidence that "monetary accommodation", aka 'bond-buying', adds jobs?
Nice catch, Bruce. Thanks for laying out the intellectual bankruptcy of the Fed. All we can do now is sit around like Wotan and wait for the Gotterdammerung. It might take awhile (like the damn opera) but it will come.
ReplyDeleteBruce,
DeleteThanks for the link to my working paper, a couple of comments about this post though:
1. This paper doesn't address your question of whether the Fed should target PCE inflation or CPI inflation. It only examines different constructs of "core" PCE inflation (such as excluding food and energy. the trimmed mean, variance weighting, and a few others along with Michigan inflation expectations) and ask the question: are they useful in predicting future inflation or tracking the current rate of inflation purged of transitory noise. I find they are. Among the best are the trimmed mean inflation rate over the past 9 or so months, but you could also use a 24-month (or so) moving of overall PCE inflation.
2. The current PCE inflation rate used in the paper stops in 2009 so that benchmarks could be constructed to compare it to. The centered 36-month moving average of overall PCE inflation that you show the figure above needs 18 months before and after the current month to be constructed. Another benchmark used in the paper (overall PCE in the 12 months after the next 12 months) requires 24 months of data after the last observation of "current" PCE inflation for construction. Overall I used the most recent sample as I could given the need for data to construct benchmarks (and the few week lag between when I ran the analysis and the release of the working paper).
3. In the figure you show above a merging of the lines is not important. The level is what matters in this figure. It shows that the error in matching a centered 36-month moving average of overall PCE inflation are generally lower when using some measure of "core" inflation (where "core" is broadly interpreted) than when using overall PCE inflation. The trimmed mean (the green line), Michigan 12-month ahead inflation expectations (the blue line), and PCE inflation excluding food and energy (the red line) all tend to have lower errors at tracking a long centered moving average of overall PCE inflation than does a short backward-looking moving average of overall PCE inflation (the black line).
4. The length of the equation you show is partly as a result of typing out things like "unemployment rate" and "nairu" and "imports". I find it more convenient to understand an equation when those are spelled out rather than using u, u*, and i, etc, as is often done in economic papers. The rest of the equation is simply a linear regression.
If you have any other thoughts on the paper, feel free to let me know.
Alan
Oh, I should have also mentioned that FOMC announced their inflation goal in terms of overall PCE inflation. (See the fourth paragraph of http://www.federalreserve.gov/newsevents/press/monetary/20120125c.htm ). I don't see "core" or "excluding food and energy" anywhere in their document. So, I'm not sure the relevance of my paper to any arguments about the FOMC's inflation target.
DeleteAlan,
Deleteso i'm looking at graphs 3 and 4. Don't know what index is tied to each color for graph 3 (actually, you explain it in point 3 above - thanks) but note it is 'overall' PCE...
Then in graph 4 note that the orange line is core PCE...
so i guess the question is what does the overall PCE line look like if added to graph 4?
I'm not sure the figures you are looking at. In the paper overall PCE prices (aka headline PCE, topline PCE prices, the PCE price index) is always shown in black. In Bruce's figure above (the last figure above) that is not from my paper. The orange line is an inflation rate (PCE prices excluding food and energy) while the other two lines are price levels (the CPI and the CPI excluding food and energy).
Deleteso here's what may be some pertinent questions:
Delete- generally, and only for the sake of this discussion, fed could choose either cpi, core cpi, overall PCE, core PCE. The Fed is now using overall PCE which would include what is considered the more volatile components - food and energy. If they wanted a lower inflation target, they would've chosen core PCE. But instead, they chose overall PCE which, keeping things constant, should result in a higher inflation rate than core PCE. So, really, they're not trying to hide anything. Thoughts Alan?
- What you're saying is using 36 moving average provides most accurate results, generally speaking, so that's why you chose that over say a 24 month MA. You used the most up to date info.
- your paper focuses in on various types/constructs of core PCE...nothing about CPI and especially nothing about whether one should use PCE over CPI...
- so in other words, does overall CPE understate inflation when compared to CPI (not core CPI) - isn't that the crucial question here? Two totally different Indexes. CPI was nixed b/c it didn't take into account substitution.
Has there been a study done comparing CPI with overall PCE?
should rephrase the first point:
Deleteif the fed is saying (in bk's words) so long as overall PCE stays below 2%, they'll have their foot planted on the pedal...they'd, imo, be able to keep their foot floored for a longer period of time using core PCE rather than overall PCE...instead, by choosing overall PCE, they've taken the more conservative approach...
Dear Alan Detmeister,
ReplyDeleteI see you have removed your report from the "Board of Governors of the Federal Reserve System" web page. http://www.federalreserve.gov/pubs/feds/2011/201156/201156abs.html%20
....says page not found, last updated April 14, 2011. You are one tricky fellow!
To all: the article ist STILL in its place, the link provided is bad.
ReplyDeletePlease REMOVE the last 3 characters (%20), and voila, like magic, the article appears ... ;-)
The guy's name is actually Detmeister?
ReplyDeleteI thought this was a spoof at first.
The family name was in fact Debtmeister, but the "b" was dropped during the depression years.
ReplyDeleteMR. KRASTING,
ReplyDeleteYOUR SPEECH IS QUITE PLAIN AND, HOW SHALL I
SAY, TO THE POINT! HOWEVER, ONE MUST UNDER-
STAND IN A WORLD RUN BY THE ROTHSCHILDS,
ROCKEFELLERS, AND SECRET SOCIETIES, IT IS
THE MAJORITY VOTING PUBLIC'S MINDS THAT ARE
THE SLOP JARS FOR ALL THAT PISSING!
EDT
CHICAGO, ILLINOIS
i don't follow too many blogs but i get the impression it's rare when an author joins in on the comments section. So to Mr. Detmeister, thanks. I don't know 1/100th of what you're report's about but if you are in fact the author, well done for showing up.
ReplyDeleteHe provides additional views on his paper and seems to welcome feedback/insight.
Instead, we have bloggers making fun of his last name. Are we really 5 years old?
I have a question - will CPCE understate the 'true' inflation rate moving forward? b/c when i read the definitions above, it seems CPCE removes volatile items like food which, you know, is increasing for everyone.
Is this just another way to understate inflation?
PCE excluding food and energy has had a downward bias (on average about 1/4 percentage point too low relative to overall PCE inflation) for most of the time since about 1999. This is also the period when oil prices have moved up strongly. From the early 1980s to about 1997 the bias went the other direction and PCE excluding food and energy generally was higher than overall PCE inflation.
DeleteMoving forward, who knows? It is really a guess on what energy prices will do. Food prices in general are very volatile month to month, but usually play less of a role in biasing the index than energy prices.
Hello Alan.
DeletePlease see my note below.
bk
Inflation is not going to return until housing picks up substantially or until the Fed prints enough money to make up for the trillions of dollars the banking and housing bubbles removed from circulation by defaulting homeowner loans. New loans create new money and conversely when a borrower defaults that money is removed from circulation. Its still a case of supply and demand and that is why the dollar still has value despite all the money printing.
ReplyDeleteThis is a shoot from the hip comment, but it feels right to me. If Food & Energy's contribution to CPI are being weighted down by the other components why not exclude them? Excluding only makes sense if they food & energy contribute noise and are ultimately reflected in the other variables, but if the direction is largely up and are not being picked up elsewhere(with lag), then the Fed's thinking is suspect.
ReplyDeleteMy bills indicate the direction is only up...
ReplyDeletegood that the author showed up....Doesn't change the fact that current data is ignored...without that it's
ReplyDeleteeff all useless.
But kudos to AD for having a spine.
I want to thank Alan Detmeister for contributing to the comments on this piece. I did not anticipate that someone from the Fed would respond. I have many articles on the activities of the Fed, this is the first time a Fed official has contributed to the process.
ReplyDeleteThe year end number for PCE was 2.3%. The Core PCE was 1.6%. The Fed will use the core number. If they did not, they would already have to tighten based on Bernanke's pledge to keep "inflation" at sub 2%.
So core it is, and I maintain that this is the most favorable metric that the Fed could have used to justify its monetary policy.
I think we will see CPI (includes food and energy) at 4% while core PCI lags at less than 2%. In other words, most of the 300mm citizens will pay the price.
If it were correct that monetary policy could cure unemployment, this might be a decent trade off. But the Fed can't fix unemployment that is structural in its nature. It is creating a bubble to fix a problem that it can't fix.
Alan, on that formula, a bit of a cheap shot, I admit. But you should understand that a significant amount of my readers and folks in general simply don't trust an answer that is based on a "techno" formula.
Your article was not intended for that audience. You were shooting for economists who live and breathe your formulas. That's why blogs are important.
I can produce a half dozen formulas upon which big decisions were made. The most significant conclusions by US and Fed economists (pre 2008):
-The US could never experience a nationwide recession in housing.
-Declines in RE could never exceed 10% in a year in any region.
Well, we got both, and here we are.
I know that folks at the Fed read this Blog. A message for them:
You have a communication and an image problem. There is no one looking over your shoulder, so many of us have a natural distrust of you.
When Mr. Detmeister comes forward and participates in a forum like this, it does make a difference. For just for a few seconds you look like you're human, sort of like the rest of us.
I again thank Alan for participating, I hope that the Fed encourages more of this.
Bruce Krasting
Bruce,
DeleteIn the future, please STOP making cheap shots - ever, full stop. Your articles are simply too good to need any.
Either you honestly feel you have a serious point to make on whatever the subject is, or you don't. If not, skip it.
The world is already filled up -indeed, all the way to the brink- with people making cheap shots. Yet what we desperately need are people who engage in, and maintain, mature conversations among grownups.
The one exception allowed could perhaps be officially certified & standardized humor.
Bruce,
ReplyDeleteHow do we know this is really who you think it is, Alan Detmeister?
Bruce wrote:
ReplyDelete"The year end number for PCE was 2.3%. The Core PCE was 1.6%. The Fed will use the core number. If they did not, they would already have to tighten based on Bernanke's pledge to keep "inflation" at sub 2%.
So core it is, and I maintain that this is the most favorable metric that the Fed could have used to justify its monetary policy.
I think we will see CPI (includes food and energy) at 4% while core PCI lags at less than 2%. In other words, most of the 300mm citizens will pay the price."
well, that would seem to answer my questions at 6:28pm/6:44pm on Jan 27.
Alan, in your link to the fed's news article, paragraph 4, it does not specifically state overall or core...so if the #'s Bruce uses are correct above, would that not indicate the fed is using the most favorable measuring stick and hence, the prices of stuff I need to buy ie. food/energy, have not only gone up but will also very likely continue to rise moving forward?
thoughts?
Bruce,
ReplyDeleteExcellent post and commentary. Whilst it is wonderful to have AD's contribution, I look forward to the day you get BB to respond.
Bernanke should be in Guantanamo. He's forestalling the collape of Amerika under the present adminstration.
ReplyDeleteIn the words of Shorthouse, "Nothing but infinite pity is sufficient for the infinite pathos of human life.
ReplyDeleteI came across your website yesterday and you have really motivated us to get started on crocheting again! I had been simply wondering had you been in a position to mail myself your routine for this specific beret (it really is cute!) since hyperlink doesn't seem to be employed by myself..? I would really enjoy it!
ReplyDeleteAlan Detmeister! That is a pseudonom for Alan Greenspan. Debt-meister...LOL
ReplyDelete