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.

« The February PPI -- Hot Or Not? | Main | The BoJ Rate Hike Watch »

March 22, 2006

Is The CPI A Government Conspiracy?

It seems that a lot of people think so, and several readers have pointed me in the direction of Mr. John Williams, whose website appears to be devoted to the proposition that there liars, damned liars, statistics, and the lying damned liars who construct them.  Williams' section on the conspiracy that is the consumer price index has several of the usual complaints -- the use of quality adjustments, the invoking of "core" inflation, and so on -- but what he seems most exercised about is the shift in methodology from "arithmetic mean" to "geometric mean" calculations that the Bureau of Labor Statistics implemented in 1999:

In the early 1990s, press reports began surfacing as to how the CPI really was significantly overstating inflation. If only the CPI inflation rate could be reduced, it was argued, then entitlements, such as social security, would not increase as much each year, and that would help to bring the budget deficit under control. Behind this movement were financial luminaries Michael Boskin, then chief economist to the first Bush administration, and Alan Greenspan, Chairman of the Board of Governors of the Federal Reserve System...

Up until the Boskin/Greenspan agendum surfaced, the CPI was measured using the costs of a fixed basket of goods, a fairly simple and straightforward concept. The identical basket of goods would be priced at prevailing market costs for each period, and the period-to-period change in the cost of that market basket represented the rate of inflation in terms of maintaining a constant standard of living...

The Boskin/Greenspan concept violated the intent and common usage of the inflation index. The CPI was considered sacrosanct within the Department of Labor, given the number of contractual relationships that were anchored to it. The CPI was one number that never was to be revised, given its widespread usage.

Shortly after Clinton took control of the White House, however, attitudes changed. The BLS initially did not institute a new CPI measurement using a variable-basket of goods that allowed substitution of hamburger for steak, but rather tried to approximate the effect by changing the weighting of goods in the CPI fixed basket. Over a period of several years, straight arithmetic weighting of the CPI components was shifted to a geometric weighting. The Boskin/Greenspan benefit of a geometric weighting was that it automatically gave a lower weighting to CPI components that were rising in price, and a higher weighting to those items dropping in price.

Once the system had been shifted fully to geometric weighting, the net effect was to reduce reported CPI on an annual, or year-over-year basis, by 2.7% from what it would have been based on the traditional weighting methodology. The results have been dramatic. The compounding effect since the early-1990s has reduced annual cost of living adjustments in social security by a total of 30%.

The material on Williams' website is a bit dated (appearing first in 2004), but his take on the price statistics appears to have a lot of fans, being cited as recently as yesterday in readers' comments to an article by David Wessel.

So what is this arithmetic-mean/geometric-mean thing all about?  The Bureau of Labor Statistics explains;

In contrast to the fixed quantity weights of the current CPI formula, the geometric mean estimator employs a set of fixed expenditure proportions as weights to be used in averaging the prices of individual items within a CPI basic index. Fixing the relative expenditure proportions rather than the relative quantities implies that consumers can alter the quantities of goods and services they buy, albeit within the narrow range of a CPI category, when the relative prices of those goods and services change. It is, in part, this property of the geometric mean estimator that led to the Boskin Commission recommendation of its use in the CPI.

In other words, the crime of a geometric weighting scheme is to assume that people change their behavior when prices change.  If you are looking for a cover-up, you won't find it at the BLS, who helpfully explain:

The CPI is constructed as an aggregation of basic indexes computed for approximately 200 item categories, such as "ice cream and related products," in each of 38 geographic areas. Within each of these index components, or strata, prices for specific items in a sample of outlets (stores) are combined to produce a basic index. As noted above, the geometric mean formula will be used only to average prices within the item-area strata. Consequently, the use of the formula will address only the issue of consumer substitution within strata.

What this means is that the methodology employed by the BLS conspires to do nothing more than implement the common sense proposition that consumers substitute high-priced goods for similar lower-priced goods:

Substitution can take several forms corresponding to the types of item- and outlet-specific prices used to construct the basic indexes:

  • Substitution among brands of products, for example, between brands of ice cream;
  • Substitution among product sizes, for example, between pint and quart packages of ice cream;
  • Substitution among outlets, for example, between a brand of ice cream sold at two different stores;
  • Substitution across time, for example, between purchasing ice cream during the first or second week of the month;
  • Substitution among types of items within the category, for example, between ice cream and frozen yogurt;
  • Substitution among specific items in different index categories, for example, between ice cream and cupcakes.


...overall substitution across categories, such as between ice cream products in general and apples in general, is not addressed by the geometric mean formula. The geometric mean formula will not be used to combine the basic indexes in the CPI, like those for ice cream products and apples, into the overall index. In the same way, the use of the geometric mean formula within categories does not address the issue of whether consumers can, or do, respond to a general increase in the price of ice cream products by, for example, forgoing dessert.

Product categories without close substitutes are, in fact, treated in the old-fashioned way that Williams and his fans seem to prefer.

The new formula, the geometric mean estimator, will be used in index categories that comprise approximately 61 percent of total consumer spending represented by the CPI-U. The remaining index categories, which are shown in the attached table, will continue to be calculated as they are currently.

No doubt about it, measurement is hard, and there are a lot of judgment calls that have to be made.  But when it comes to assessing the motivation for those calls, I'm with David Wessel:

Ken Kesler writes:

Whether or not the CPI is accurate is beside the point. I remember hearing an ex-senator (I cannot remember who) say that there is not one number that comes out of Washington that has not been politically massaged.

David Wessel responds:

I know that is a popular view. I've personally found the technicians at agencies like the Bureau of Labor Statistics to be straight shooters who do their best to cope with the complex methodological issues inherent in measuring our large and dynamic economy.

And, for my money, they do it well.

March 22, 2006 in Inflation , This, That, and the Other | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Is The CPI A Government Conspiracy? :

» Was the Calculation of the CPI Altered to Reduce Entitlement Spending? from Economist's View
Did the government change the calculation of the CPI from an arithmetic mean to a geometric mean for valid economic and statistical reasons, or was it a conspiracy designed to save money on entitlement spending?: macroblog: Is The CPI A [Read More]

Tracked on Mar 23, 2006 6:36:27 AM


Statistically half the people you know should be telling you that inflation is lower than reported right? Must be that those people don't blog. Blogging causes inflation!

I'd agree that the technicians try very hard and do a transparent and consistent job even as the methodology constantly changes but the result is so contrived it doesn't tell us about inflation it only ends up telling us about aggregate geometric weighted market basket fluctuations. The BLS does a bang up job of reporting the AGWMBF just don't call it inflation.

Posted by: Robert Cote | March 23, 2006 at 06:34 AM

Dave. First, I hate it when you take commenter assaults personally, so let me pause to say thanks one more time. You deserve a WHOLE lot of credit, professional respect & community thanks for the time, & thought you put into this Blog. You don't run from anything, & personally, I appreciate how VERY GOOD you've become at sifting through our rambling efforts to communicate to address arguable specifics.
BUT, please, forget conspiracy theories; this is not about good guys & bad guys, it's about checks & balances.
I've defended Gov't staff against politically motivated attacks for a long time. For me, K. Abraham was THE consumate pro, in the Boskin hearings she personified the professionalism & demeanor of Gov't staff. That said, Kesler couldn't be more wrong when he says: "Whether or not the CPI is accurate is beside the point." The CPI IS the pont, because EVERYONE misrepresents & misuses it. AND, Wessel's equally wrong when he lays off final responsibility for the sad state of our Gov't stats on staff. Staff doesn't set the agenda, it makes recommendations & then does what it's told. (The simple truth is, NO ONE's more aware of the importance & the shortfalls of Gov't statistical measuring & gathering methods than Fed staff.)
BB represents a tremendous opportunity for the Fed & those of us who believe the checks & balances that made our country special have been severely threatened, and that our economy's on the WRONG TRACK & is picking up speed. The Fed's dilemma is not the 10 year yield, it's how BB can reframe the argument to do what's needed to get us out of this mess. I don't know if what he has to do is even possible, the odds are long against him. But, if anyone's up to the task, I think he's the guy.

Posted by: bailey | March 23, 2006 at 09:31 AM


Could you clarify for those of us who are ignorant? I didn't understand the point of your post, other than the economy sucks, gov't stats suck, and there's no checks and balances in our gov't. Oh, and Ben Bernanke is going to fix this 'mess'. Please enlighten me w/ some specifics.

Posted by: cb | March 23, 2006 at 12:24 PM

CB, Be glad to give it a try. First, the economy doesn't suck for everyone, just those in the lower 3/5 of it, plus those who haven't owned a home for a few years & those who don't have a healthy retirement plan (separate from SS & Corps). It certainly sucks for retired folks who depend on SS checks to track inflation. And, it sucks for ALL young people who are going to have to pay (with interest) for today's tax rebates, profligate fiscal policies & because we refuse to set aside SS receipts. If you're still having trouble grasping the concept, think of Popeye's buddy Wimpie's saying: "I'd gladly pay you Tuesday for a hamburger today."
Second, what checks on the Administration's agenda have you seen, in Congress, in the Fed, in the media, anywhere? AG argued for the total repeal of Glass-Stegall, shutting off any debate before it could get going. He front-ran the Boskin Commission's recommendations to lower the CPI, shutting off debate there. And, instead of questioning Bush's highly directed tax cuts, or even deferring on the subject, he supported them. He fed the dotcom bubble & recommended ARMS as recently as '94.
It's pretty obvious to me AG was always a political animal. I think it's hard to argue otherwise. He was THE Ayn Rand groupie before he acquired the power to affect real change. BB is a first-rate, highly respected intellectual & academic who APPEARS to be apolitical. These are attributes we very much need. The fed's role is to act as an INDEPENDENT body to insure our economy remains sound, today & in the future. No question influential forces desirous of ever faster increases in our money supply are aligning. (Why does housing rhetoric these days focuses on Fed's & int. rates instead of on negligently loose lending policies & out of control GSEs?) Given a choice, I say it's not the Fed's job to do away with business cycles, it is its job to see we don't become another banana republic.
It's certainly not an earth-shattering revelation to note that there's been very little debate leading to the dramatic influence shift we've seen in our Gov't.. Nor is newsworthy to state Gov't. stats. have been politicized one too many times, everyone I know is well aware of this. It's certainly time for BB (or another respected Economist with clout) to say: we'd all benefit from Global standards for economic reporting.
Here it is in a nutshell, cb: Meaningful debate's absolutely necessary, as are meaningful Gov't stats and an independent Fed.

Posted by: bailey | March 23, 2006 at 02:25 PM

Sorry, i REALLY should proofread.

Posted by: bailey | March 23, 2006 at 02:34 PM

Well, that was a mouthful. I hope you feel better.

I must say that I am a member of many of the groups to which you seem to think are getting screwed in this country. I can't say I agree.

As far as AG dipping into politics, he has been criticized by both sides for that, however I am unwilling to believe w/o evidence that this behavior compromised the Fed in any way, other than perception.

I don't really feel like commenting on the rest of your rant, but, as this post was about CPI, I will note that you have given no reason for me to believe that CPI is overstated. Your only comments on it are more conspiracy-type stuff. Give me some specific reasons, in a non-rant, non-gov't sucks, non-AG is evil format that someone could hold up against Dave's post and come to a reasonable conclusion. Thanks.

Posted by: cb | March 23, 2006 at 03:03 PM

I personally put the burdon of proof on the conspiracy theory types because most of the those who go on and on about it are not big sophisticated investors. The big guys have more at risk and have more tools at their disposal, plus they can more easily invest in any global market they want. Given the shape of the yield curve, it doesn't look like the big guys have much problem w/ how CPI is calculated, although I'll admit yield curves are flat all around the globe.

Posted by: cb | March 23, 2006 at 03:27 PM

cb, Sorry for the soliloquy, I don't react well to sarcasm, I thought you asked for specifics about AG, & I take Dave's Blog seriously. First, I believe cpi's significantly UNDERSTATED, not overstated. Second, Many of the same big guys you refer to are buyers of FNMA & Freddie debt even though our Gov't has repeatedly said it WON'T back the debt BECAUSE they believe the Gov't. will be forced to back it and even if they're wrong, they can't afford to be the odd man out. (Some of these same "big" guys also believe cpi doesn't matter, because it's within a few points & it's not the Fed's only or even its principal inflation indicator.) On point, if the CPI were more accurately reported our Gov't would have had to reign in fiscal spending years ago - long before it rocketed as it has.
Third, Dave is Fed staff. I appreciate his explanations of Fed policy. He shares one heck of a lot on this Blog & I fine his reasoning clear & sound. But, I see no sense in arguing cpi at the micro level. For me, the cpi was NOT intended to track inflation, has been downscaled for politicical purposes to the extent it's now misdirecting nonsense that Staff did NOT opt for but must work with & rationalize. Staff's strictly the messenger on this one. (But, note, there's a lot of difference between conspiracy & just-politics.) If one demands that we argue cpi's apples/oranges, I say change the methodology to unit COUNT apples SOLD vs. oranges SOLD, or opt for any of umpteen other data selection & collection alternatives to eliminate the subjective decision-making that's integral to the cpi collection process. We're in the 21st century & I'd guess the Gov't already collects (or easily could) a heck of a lot more meaningful corporate data that's more suitable for monitoring inflation. Fourth, as far as AG, it's funny, I've NEVER heard him critically questioned for being an egalitarian. RE: the flat yield curve, I think any real dilemma remains only at the retail level. Money Center Bank/Brokerages know darned well who's buying the 10-year as they're selling it to them! So, it's implausable that the Fed doesn't know what's going on. This again leads us to the role of an independent Fed. How does BB deflate the astronomical leveraging of the last five years (see derivatives) against the wishes of his own banks and without corporate sector support or any help from this administration - without sending us into a depression? I think his answer begins with jaw-boning all large players back into the real world while his staff works on reframing the the investor-risk argument.

Posted by: bailey | March 23, 2006 at 05:25 PM

Point taken regarding the sarcasm. I've found it difficult to ween myself of the toxin.

I know you think CPI is understated, I typed too fast. OK, say AG and some politicians messed with it to the point that it's understated. For one, I doubt it's much, 50 basis points on the high end, if that. There are 7 inflation measurements that I'm aware of, and they're all pretty close. The site Dave links to at the beginning of this post has it 300-400 basis points higher, which I find ridiculous, and Dave's post does a good enough job convincing me that that individuals analysis leading to his conclusion is wrong. Two, say they did do it to limit expenditures to contracts (I assume retirement related) tied to the CPI. Personally, I don't care. Defined benefit plans were dumb to begin with, and are overly generous anyway. Obviously I don't have one :), but it's better for the country anyway.

I'm aware of Dave's employer, and take it into consideration, but I wouldn't necessarily saw he's just a messenger.

"it's implausable that the Fed doesn't know what's going on"

Agreed, BB speech Monday night was specifically about the yield curve. He mentioned many possibilities, including - clarity of future monetary policy, relative economic stability of the last 20 years, lowered inflation expectations, foreign central bank buying, pension fund buying due to changes in accounting practices, and reduction in supply. Obviously it's difficult to ascribe percentages to each factor, but it's interesting that he convincingly downplayed the role of foreign central banks, which is not what most people are saying, including many econbloggers.

I don't understand your point about Fannie and Freddie. I sell their debt all the time, and I and my clients are aware there is no explicit guarantee.

I don't follow your point about deleveraging the financial markets. I don't see why it should be done, nor do I see any language coming from anybody that it needs to be done. The beauty of the US financial markets is the innovative products that are invented that enable risk to be spread and transferred with ease. This will inevitably increase their use, which I view as a good thing, because it braces the economy to shocks. If you think BB is going to attempt to limit this, I suggest you prepare yourself for disappointment.

If you don't wish to 'jawbone' about the CPI on a micro-level, surely you don't pretend to think anybody is going to believe you when you say CPI is grossly underreported.

Here's the link to BB's Monday night speech, I liked it.

Posted by: cb | March 23, 2006 at 07:04 PM

Regarding Bernanke, I'm guessing the only significant difference between him and AG will be that he refrains from entering the political arena, as least not to the extent AG did. I suppose he's relatively more lucid when speaking to economic knowledge challenged public, and perhaps the Fed will have an explicit inflation target down the road. I suspect BB wants to downplay the Fed chairman's position, vis-a-vis AG, which I guess is your point, but I seriously doubt it will have much impact on Fed policy, if any.

Posted by: cb | March 23, 2006 at 07:29 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



Powered by TypePad