About


The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.


« My Pet Peeve | Main | Why There's No Money In Monetary Policy »

January 18, 2006


Don’t Like the CPI Trend? Help May Be on the Way.

NOTE TO MACROBLOG READERS: Dave is playing hooky today, allowing me another shot at his blog.  The comments are mine, not his, and not those of the Federal Reserve Bank of Cleveland or the Reserve Board of Governors.

~ Michael F. Bryan

For a refreshing change, the December CPI report had a bit of good news this morning for both consumers and inflation worry-warts like me.  For consumers, many of the items that posted price declines in November continued to do so in December, including energy, men’s apparel, footwear, and new and used vehicles.  To these were added falling prices on women’s apparel, beverages, dairy products, and bakery goods.  In fact, when you add them up (and I did), 35 percent of the consumer’s market basket showed either no change or a price decline last month, compared with about 25 percent a month earlier.  Likewise, only a little more than 13 percent of the CPI basket posted price hikes in excess of 5 percent (annualized) in December, compared with about 24 percent in November.  So whatever sleep I lost in the days immediately following the November CPI report, I hope to recoup starting this evening. 

Sure, the “core” measures of inflation, reported in the following table, weren’t great, but they certainly didn’t worsen, seeming content to remain in the neighborhood of 2 to 2½ percent.  (More detail can be found in the Cleveland Fed’s median CPI press release.)  Any whiff in the November data that prices were poised to break the upper 2½ percent threshold seems to have vanished in December. 

Percent Change, Last 1 Month

Jul

Aug

Sept

Oct

Nov

Dec

CPI

0.5

0.5

1.2

0.2

-0.6

-0.1

CPI Ex. Food/Energy

0.1

0.1

0.1

0.2

0.2

0.2

MEDIAN CPI

0.2

0.2

0.1

0.2

0.2

0.2

Percent Change, Last 12 Months

Jul

Aug

Sep

Oct

Nov

Dec

CPI

3.2

3.6

4.7

4.3

3.5

3.4

CPI Ex. Food/Energy

2.1

2.1

2.0

2.1

2.1

2.2

MEDIAN CPI

2.3

2.3

2.3

2.3

2.4

2.5

Now, if you’re like me, you might be more comfortable if the CPI trend was closer to the lower end of that 2- to 2 1/2-percent range.  Well, we might be getting some help from the BLS beginning with next month’s CPI report. 

Here’s a puzzle for you: If a good’s price goes up, and you don’t buy it, has your cost of living risen?  [Actually, this is a question that most economics majors are asked on an examination at some point in their undergraduate studies.]  The intent of the question is to get students to think about the theoretical difficulty of measuring changes to our cost of living.  Here are two rules in measuring cost of living, and if they seem in conflict, well, that’s because they are.  Rule one: When you gauge price changes, you must fix the market basket.  One can’t make any conclusion about the cost of living by comparing the price of the oranges I bought yesterday to the price of apples I bought today.  And rule two: Once you fix the market basket, it immediately becomes outdated! 

It turns out that there is no right answer to the question above.  You can either price the oranges you don’t buy today, or the apples you didn’t buy yesterday.  Most economists have concluded that it’s most useful to value the items you bought in an earlier period using the prices of today—as long as that earlier period wasn’t too long ago.  Next month, the Bureau of Labor Statistics will update the market basket on which they compute CPI changes.  The basket will reflect consumer expenditure patterns over the 2003-2004 period, compared to the 2001-2002 basket of goods they use today.  Can the market basket really change that much in just a few years and, moreover, is it likely that the new market basket give us a higher or lower reading?  Well, yes, the market basket can change quite a lot.  And I’m not just thinking about ipods, Xboxes, and high-definition TVs.  Whenever prices move, people change their expenditure patterns—generally moving away from the goods going up in price in favor of goods where prices are relatively more favorable.  So yes, we can anticipate that when the market basket is updated, changes in the cost of living will be more moderate than they would have otherwise been.  How big a change is this likely to be?  We’ll know next month.  But to give you an idea, consider the figure here, which compares the “core” CPI to the “core” CPI on a chain-weighted basis. 

Chained_cpi

The chain-weighted CPI is an experimental measure which updates the consumer market basket more frequently that the conventional CPI.  Over time, the chained CPI has averaged about 4/10ths of a percentage point below the conventional measure, an indication that updating the basket could have a significant effect on our CPI reading.  So, if you are like me, and would like to see the CPI trend closer to 2 percent than 2 ½ percent—the BLS might just have what you are looking for in next month’s report.

January 18, 2006 | Permalink

TrackBack

TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8341c834f53ef00d8346cb62253ef

Listed below are links to blogs that reference Don’t Like the CPI Trend? Help May Be on the Way. :

Comments

"So, if you are like me, and would like to see the CPI trend closer to 2 percent than 2 ½ percent—the BLS might just have what you are looking for in next month’s report."

More good news! My car is getting better gas mileage now that I'm measuring it in rods to the hogshead.

Are you an inflation hawk, or just a reported-inflation hawk? I suppose if what is actually important is people's inflation expectations, then shifting the measurement basis might have an effect on its own. In that case, maybe we could avoid all this econometric nonsense and let the BLS just make up the numbers (and keep them small, of course).

Posted by: eddie | January 19, 2006 at 01:46 AM

It makes NO difference whether one buys apples or oranges, he/she's buying fruit - a commodity basket that provides the common value of its member items. Why do Economists love to focus on trivialities that clearly detracts from their ability to accomplish their stated objectives? I guess if your tenured & a collector of information, rather than an interpreter of information it makes some sense. But, then I ask: what value is this person adding? CPI has been so overly manipulated & contorted over so many years it is now WORTHLESS. The adopted Boskin recommendations was the final assault on reason. Economists who perpetuate the sham should be embarrassed. Sycophants all.

Posted by: bailey | January 19, 2006 at 08:01 AM

Eddie, of course you are correct. The idea that we are getting closer to price stability by "fixing" the price index was a bit of tongue and cheek on my part. But I thought some might like a heads-up on the change. Now, an important reason an inflation hawk like me supports a CPI trend just a bit north of zero is a healthy appreciation of some of the long-term measurement problems (biases) seemingly inherent in the computation of our cost-of-living. The BLS has been very diligent in addressing these well-known problems, including updating the consumers market basket more frequently, introducing new goods/shopping outlets earlier, and devising more innovative ways to adjust for changes in product quality. Regarding the arbitrary nature of some measurement standards and the problems that causes, I think this is exactly the problem that inflation inflicts on our economy and I have written about it here: http://www.clevelandfed.org/research/com97/1001.pdf (By the way, the rods to hogsheads reference was great, thanks.)

Bailey, I think it IS crucial that a cost-of-living index compare apples to apples and oranges to oranges. True, this example trivializes the problem, but would we really claim that someone who used to buy a subcompact and now drives and SUV has experienced a "cost-of-living" increase? I think not. This is someone who has experienced a "standard of living increase." To accurately measure the cost-of-living, we need to hold the standard of living constant. And since that standard is constantly in flux (and generally improving), this means that the reference market basket can't help but to be outdated (and thus over-state the true cost-of-living) almost as soon as it is set. Now, if your point is that inflation (as distinct from the cost-of-living)is common to all goods, apples and oranges, and that the problem of inflation measurement boils down to a signal extraction problem, I agree completely (and you and I are in a minority.) My particular views on that issue can be read here: http://www.clevelandfed.org/Research/com2002/0515.pdf

Posted by: Mike | January 19, 2006 at 09:13 AM

Mike, I appreciate your i/p, consideration & availability, I've read your work (& even recommended it to a few friends.) Let me voice a few concerns for you to toy with or discard. I certainly don't expect a response to this rambling.
1. CPI as a cost of living measurement has been modified into absurdity & I believe no one dislikes the measurement more than Fed analysts. For me, adoption of the Boskin recommendations in the face of Abraham's matter-of-fact response was the final disclaimer to CPI's value. (I don't believe anyone can build a representative sample of our population that would come within two points of it.) And, as an inflation indicator the CPI is even worse. Yet, it's rolled out every time it SUPPORTS a monetary decision (without the appropriate notation of the French Revolution jurist who proclaimed: first the decision is made, then the facts are gathered.)
2. Large denomination time deposits portion of non-m2 M3 was +12% for 2005. I've yet to read a reasoned explanation why the Fed no longer sees M3 worth reporting. (I've heard of AG's ediface building so cost-savings is out.) If someone sells a house & puts the (now) modest proceeds of $1 million into a long-term bank note, does the money evaporate?
3. The Fed is supposed to be an INDEPENDENT body protecting the well-being of our banking system. Yet, it advocated for the repeal (instead of modification with further regulation) of Glass-Steagall. It opted to keep ff rates extremely low for an extended period of time, during which it watched its banks (with GSE guarantees) print & funnell enormous amounts of money using dramatically lowered credit standards. I can make no sense that the Fed refused to espouse sound economic principals in response to the Bush Administration's highly questionable fiscal proposals that are reshaping our future. And, the Fed denies credit has any association with "money" at a time when most people believe credit IS money. Its policy position, that it can only react to bubble excesses after the bubble has popped, may prove to be calamitous.
In short, I believe the Fed's partisanship, old-ways attitude & inability to encourage new approaches to evaluating & monitoring our economy has left us without an "honest broker" to protect our economic viability. My frustration is with tenured Academic Economists who've refused to call the Fed to account. "Bad Standards" indeed.

Posted by: bailey | January 19, 2006 at 01:43 PM

If half the people lose medical coverage and prices are doubled on those that do to cover those that don't, has healthcare doubled in cost or remained unchanged?

Posted by: Lord | January 19, 2006 at 02:36 PM

Mike: glad to be of service. :)

If I understand you, you prefer the CPI just north of zero rather than sitting squarely on zero because you think the CPI overstates inflation. If you were King Of The Fed, is there an inflation metric you would have your lackeys calculate for you that you would then try to keep at zero? Chained CPI is better than conventional CPI, sure, but is there something even better we should be looking at?

Posted by: eddie | January 19, 2006 at 03:33 PM

Eddie, yes, you understand me correctly. I used to think the measurement biases in the CPI were somewhat small—much smaller than the Boskin Commission put them. I have softened on that some, mostly on the observation that new methodologies seem to have had a large impact on the reported CPI (well, certainly much more than I would have guessed.) And the potentially largest biases remain, namely the mis-measurement of quality changes.

Regarding my promotion to King of the Fed, thanks, but I think I could probably live with either of the available retail price measures. It may be true that the PCE market basket is a more accurate reflection of household expenditure than the CPI, but in the end, the difference between the two measures just isn’t large enough for me to have a great preference between them. Instead, I think I would have my staff spend their energies on measuring inflationary expectations better. For a central bank, it isn’t so much the behavior of inflation that causes problems – for every person hurt by inflation there must be someone who is a recipient of an inflation windfall. But once bit by inflation, people will alter their behavior in anticipation that they could suffer an inflation loss again at some future date and this is where inflation starts to erode our prosperity. I’d like to know a lot more about how the public forms their inflation expectations. My current work is on how households respond to inflation expectations surveys (some of the results of which I report in http://www.clevelandfed.org/Research/Com2001/1015.pdf and http://www.clevelandfed.org/Research/Com2001/1101.pdf.) Alas Eddie, I’m not the King of the Fed.

Lord, I know your question was rhetorical, but let me answer anyhow. First, the CPI only measures a household’s out-of-pocket expenditures, so if the health care benefits people are losing were previously paid for by employers, it won’t be felt in the CPI. But it would eventually change the weight on medical care in the PCE price measure because that statistic includes things that employers pay for on the employee’s behalf. But more to the substance of your point. Using today’s market basket and pricing it in an earlier period (which is known as a Paasche price index http://stats.oecd.org/glossary/detail.asp?ID=1995) will understate the “true” cost of living for exactly the reason you suggest. It involves some measure of suffering on the part of consumers who, due to higher prices, no longer purchase goods they previously enjoyed. So, pick your basket, but know this--it's going to be imperfect in one way or another.

Posted by: Mike | January 20, 2006 at 10:08 AM

OK Mike, you write clearly & I appreciate you sitting in. It would be silly to have access to you & not ask a few questions, here are mine: You've written that central banks "can rid the world of inflation by carefully managing the money supply". I infer from this the Fed has a firm definition of "money" it relies upon. Where can I find it & has it changed in the last 20 years? (The definition I've relied upon is m3, but the Fed recently sent a clear message to forget that interpretation).
If inflation is a monetary phenomena, why doesn't the Fed stop referencing the CPI & advise us to track the increase in "money" that's resulted from fed policy & then factor in its velocity? Our population has a long-term growth rate, why (after adjusting for global demand) shouldn't our money supply?

Posted by: bailey | January 20, 2006 at 12:38 PM

Mike, thanks for the pointers, very interesting stuff. If it ever comes to it, you've got my vote for King.

I'm squarely in your camp when it comes to inflation and the goals of monetary policy. If your research is showing that people persistently overestimate inflation, and we want to minimize inflation expectations, wouldn't that best be served by shooting for zero inflation as measured by the BLS, no matter how they choose to measure it?

Rational actors (mythical though they may be) won't care what the inflation rate is, as long as they believe it to be stable and predictable. Targeting zero probably does this best. Irrational actors, which the surveys suggest our households are filled with, don't have accurate knowledge of the inflation rate and probably wouldn't adjust their behavior if they did. I suspect the typical household member may guess about inflation if you ask them, but completely ignore inflation when making time-preference decisions. As such, it doesn't matter to their decision making what the inflation rate is, but they'll be best served if inflation is zero.

To the degree that people are partly rational, then they may form more accurate guesses about the actual inflation rate if the BLS can announce every month that, once again, the inflation rate is within a half-point of zero either way.

Posted by: eddie | January 20, 2006 at 01:23 PM

Wow. I am such a stupid sheep. Thanks for setting me straight. Here I was buying gasoline, housing, private health care, saving for college tuition and stupidly paying all the taxes the nice people with the big guns were "suggesting" I owed "voluntarily" all the while insuring at ever increasing expense against disaster. I've been paying 10-15% more year over year on those items when all this time I should have been spending my money on those important core value items such as 27 inch analog televisions and other necessities and avoiding inflation.

Hey, ignore all those embarassing issues; answer me this; who do you know that would stand up and defend "equivalent rent" in a public forum without security personell ready to use force against the angry mob. Face it, when you don't personally feel comfortable telling people with $3500/mo housing costs that they are actually paying an uniflationary $1200 it isn't a matter of differing interpretations, it is a matter of the Fed lying to save their reputation regards the economy. Hey, there's an idea, telling the truth. Can I put together my own CPI and change at the first whim? It will not be ex-energy (three month moving average compared to y-oy-y instead) and it will include three more things; present value of govt obligations (commitments), total health care prorated by PAYING customers, and an HEW component. You know, HEW that Federal agency that has no basis in law.

Do I sound bitter or angry? Not at all. When I was born in 1958 I eagerly signed the same contract as all my peers. You know, the one that said I wouldn't get a fair shake and would always be a few years late to the party. Could be worse, I've seen the contract you've forced my kids to sign.

The sad part is none of this would be necessary if the Fed had not lied about inflation to protect the Federal budget and instead had come clean about real inflation and disavowed the unstated promise of protecting SoSec and other social wealth transfer schemes.

Oh, and in case you were wondering. My dad sat me down in about 1968 at the ripe old age of 9 and explained how SocSec was a lie. None of that conspiracy stuff, he told me that the retirement age was going to go up as I approached benefit time. Gosh Mr. Fed and what has happened to full benefits vesting since my dumb ole dad said this?

Clearly you don't understand. You live in one world and the people hurt by your overt denial live in another. Thanks for listening.

Posted by: Robert Cote | January 20, 2006 at 09:56 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search



Recent Posts


Archives


Categories


Powered by TypePad