Wednesday, May 23, 2012

BPP@MIT Annual Inflation Rate Falls Below 2%


The top chart above displays annual inflation rates based on daily online retail price data collected by the Billion Prices Project @ MIT, which just released data through April 30.  According to this real-time measure of major inflation trends in the U.S., inflationary pressures have been subsiding since last summer, and the annualized inflation rate fell below 2% at the end of April for the first time since early January 2010, more than two years ago.   

The bottom chart above shows that since January 2010, MIT's online price index has tracked the Consumer Price Index (NSA) pretty closely.  Despite the fact that the MIT online price index doesn't capture all of the items in the CPI, it's still a useful alternative measure of inflationary pressures in the U.S. economy, and gives us additional information about the trends in consumer prices and inflation based on real-time online retail prices.

69 Comments:

At 5/23/2012 1:19 PM, Blogger Jon Murphy said...

We probably won't see significant inflation until the PPI starts registering it. The CPI moves very closely with the PPI, and right now both are signaling low-to-moderate inflation.

 
At 5/23/2012 1:40 PM, Blogger Jet Beagle said...

According to both MIT and the federal government, consumer price levels have only increased about 5% since July 1, 2008. That's five percent over four years - or an annual compounded rate of 1.2%.

I believe these numbers. But I won't be surprised to read from a couple of regular commentors that both MIT and the federal fovernment "cooked" the numbers.

 
At 5/23/2012 1:51 PM, Blogger juandos said...

"But I won't be surprised to read from a couple of regular commentors that both MIT and the federal fovernment "cooked" the numbers"...

Well then jet b may I present this commentary to you courtesy of Zero Hedge?

Dollar Backwardation
Submitted and © by Keith Weiner of the New Austrian School of Economics

 
At 5/23/2012 2:01 PM, Blogger Larry G said...

I think a word was left out...

isn't it.....the....

New Austrian School of VODOO Economics

that's much better now...

thanks for allowing that correction..

;-)

 
At 5/23/2012 2:03 PM, Blogger Jon Murphy said...

Larry,

Nice use of ellipsis to build suspension lol

 
At 5/23/2012 2:15 PM, Blogger morganovich said...

mt question is simple:

why would bpp be a good measure of consumer inflation?

it's an unweighted morass of a billion online prices.

it vastly overweights consumer electronics, the most deflationary product on earth.

it barely represents rent, heathcare, food, education, or fuel at all. that right there is 60-70% of the average consumer basket, maybe more.

so, you take an index that focuses almost entirely on the most deflationary 30% of consumer spending and try to use to it prove CPI is right?

seems to me that if cpi looks like this index, that's proof it's badly broken.

 
At 5/23/2012 2:18 PM, Blogger W.C. Varones said...

Perfect!

Bring on QE3! It's party time!

 
At 5/23/2012 2:21 PM, Blogger morganovich said...

and jet, it's not cooked, it's measured badly in this case.

this is nothing like a representative basket. it's a measure of price moves in A basket to be sure, but that basket has very little to do with what consumes most spending if it leaves out rent, food, fuel, and healthcare.

i may be a little off, but i think that shelter is, on average, 30-40% of spending. rents are up 6% yoy nationwide.

that right there is 2% inflation and accounts for ALL of what bpp reports. add in food and healthcare and you are way above this index.

we can argue over just what the correct basket of goods is to best measure the experience of the average consumer and there probably is no perfect answer, but i think we ought to be able to agree that not including much of anything for rent, food, healthcare and fuel is not likely to produce a good or even mediocre measure of consumer expenditure.

bpp is great if you eat ipods, live in a laptop and drive a cell phone that runs on books, but i do not think that describes an average american.

 
At 5/23/2012 2:31 PM, Blogger morganovich said...

from the bpp site:

"We are constantly adding new categories of goods, but we do not cover 100% of CPI goods and services. The price of services, in particular, are not easy to find online and therefore are not included in our statistics."

no, correct me if i am wrong, but services are about 77% of US GDP.

so, bpp tells you flat out that it has incomplete coverage of less at 1/4 of gdp.

hey, if you wanna use that as an inflation metric, go for it, but to my mind, that makes it far too limited to even consider seriously, particularly if you suspect that services inflation (healthcare, education etc) runs higher than that of internet goods.

 
At 5/23/2012 4:28 PM, Blogger juandos said...

"New Austrian School of VODOO Economics"...

Ahhh, the humbling admission that larry g had an epic fail of a time in with high school economics...

 
At 5/23/2012 4:41 PM, Blogger Jet Beagle said...

morganovich: "it's an unweighted morass of a billion online prices."

I think you are mistaken. According to The Economist:

"the BPP weights its index based on the BLS basket"

 
At 5/23/2012 4:50 PM, Blogger Jet Beagle said...

morganovich: "no, correct me if i am wrong, but services are about 77% of US GDP.

so, bpp tells you flat out that it has incomplete coverage of less at 1/4 of gdp."

I believe you have misunderstood the purpose of a consumer price index. The majority of services which male up the 77% of U.S. GDP are intermediate services, not final goods and services. Examples of those services would be business accounting services, trucking of goods, retail store janitorial services, pest control services, commercial HVAC repair services, etc, etc. Such services should definitely not be included in a consumer price index.

 
At 5/23/2012 4:56 PM, Blogger Jet Beagle said...

morganovich: "but that basket has very little to do with what consumes most spending if it leaves out rent, food, fuel, and healthcare."

Why do you belive the basket leaves out rent, food, and consumer purchases of fuel?

According to the MIT News, food was the first category included in the BPP index, and the program has expanded to include real estate and other categories. It is not stated explicitly that gasoline prices are included, but I could find nothing to indicate that they were excluded.

 
At 5/23/2012 5:10 PM, Blogger Jet Beagle said...

morganovich: " that makes it far too limited to even consider seriously, particularly if you suspect that services inflation (healthcare, education etc) runs higher than that of internet goods."

Perhaps you might understand the BPP if you researched it a little more before commenting.

According to The Economist, whose writers interviewed the BPP developers:

"The data directly capture changes in the prices of retail goods only. Since most services are not widely marketed online, the BPP has to use proxy measures: it calculates health-care costs by weighting changes in the price of wages, energy and drugs."

Although their proxy measures are not truly price measures, it is not true, as you argue, that the BPP "barely represents rent, heathcare, food, education, or fuel at all."

 
At 5/23/2012 8:03 PM, Blogger morganovich said...

jet-

the only way to do good real estate comps is to compare the same unit when it rents again (as case shiller does). they do NOT do that. they use bizarre proxies and random unstandardized inputs. this is NOT data. it's garbage based on assumptions.

i also did not say that they do not include food, i said they barely include it. it's under weighed and based on very little data. most food data is NOT available online and they do not adjust it for consumption.

their healthcare "estimate" is a joke. they "estimate using wages and and energy?" seriously, how can you call that a measure of healthcare prices with a straight face?

it is ABSOLUTELY true that they do not represent healthcare and rent. they use a bunch of made up nonsense to try to substitute for it. that is not representing prices, it's called "making stuff up".

baked in are all manner of assumptions. but there are no actual facts. insurance payments are the biggest cost. they are COMPLETELY ABSENT from this index..

treating proxies like they are actual data is why climate science is such a farce.

this is the exact same thing.

they take a billion prices, but the weights they assign are not validated in any way, and lots of them are made up. the real estate and healthcare data is literally pure fiction.

also note:

"BPP does not weight a particular item’s price more than another even if it comprises a larger proportion of household spending or if the item’s price is higher than a comparable items’ price."

that right there is total death for an indexes validity. even if they did get all the goods, they scale them equally. a cd rom going from $1 to 90c counts more than a shirt going from 100 to 109 or your grocery bill going from $300 to $320 or even your rent going from $3000 to $3250.

given the huge number of low priced consumer electronic items for sale online, running an unscaled average is going to get a massive deflationary push from that.

can you seriously argue that unscaled and unweighted price data is a valid way to model a consumer basket even if it were real, as opposed to the fictions they use to "model" many items?

this is not a fixable flaw in this model as the whole point is using web spiders.

 
At 5/23/2012 8:21 PM, Blogger morganovich said...

i'm not sure i explained that well, so let me try again.

let's pretend for a minute that bpp does have rent and healthcare data (or at least that the assumptions are accurate.

your health insurance goes from $200 to $220.

you rent goes from $3000 to $3300.

a tube of chapstick goes from $1 to 90c.

a rubber ball drops from 30c to 27c.

assuming you consume all 4 of those things each month, your bill went from $3201.30 to 3521.17.

i would call that 9.99% inflation.

bpp would call it zero because:


"BPP does not weight a particular item’s price more than another even if it comprises a larger proportion of household spending or if the item’s price is higher than a comparable items’ price."

and each 10% gain was offset by a 10% drop.

now consider the online shopping universe.

it's full of low priced consumer electronics and does not have nearly as many of your big ticket expenses.

the basket that gets created is haphazard and just an uncontrolled sampling of stuff that is not scaled nor weighted by price or consumption.

why on earth, even if it did have all the data (as opposed to making much of it up) would that work?

this is why i said "it barely represents rent, healthcare, food, education, or fuel at all". even if they were in the index (itself untrue), they get swamped. a $300 hike in rent gets zeroed out by a 3c drop in the price of a superball.

i don't know about you, but that seem like a pretty questionable way to measure inflation to me.

 
At 5/24/2012 2:51 AM, Blogger PeakTrader said...

I think, some people underestimate the importance of productivity on inflation.

It's not a coincidence the quantity and quality of goods & services increased when the Baby Boomers (born between 1946-64) began to enter prime-age around 1982, the U.S. started a period of disinflation, a structural bull market began in the stock market, and international trade, along with U.S. trade deficits, began to increase around 1982.

The U.S. was able to produce and consume more with less effort, since 1982, e.g. because of labor productivity and technological improvements.

The U.S. labor force is still most productive, which is disinflationary, because Americans haven't saved enough for retirement and are living longer.

 
At 5/24/2012 8:48 AM, Blogger Hans said...

If any inflation index does not measure the service industry, then for all practical propose it is worthless.

 
At 5/24/2012 9:52 AM, Blogger bart said...

If any inflation index does not measure the service industry, then for all practical propose it is worthless.


+1


*BOOM!*



By the way, nice job morganovich.

 
At 5/24/2012 9:53 AM, Blogger bart said...

From the same Economist article:
"And since the BPP weights its index based on the BLS basket, it may still be importing official bias."



*BOOM!*

 
At 5/24/2012 10:01 AM, Blogger bart said...

And speaking of "official bias", total medical care costs from the BLS since 1968 are an average of 2.7% too low (low by 43%). Since 2000, it's an average of 2.1% too low (low by 53%).


http://www.nowandfutures.com/images/medical_care_bls_v_cms.png

 
At 5/24/2012 10:06 AM, Blogger bart said...

Even the IMF tracking of inflation is way higher than the BPP severely algo challenged methodology.

World inflation in March is 3.9%, Western hemisphere is 5.1%, Europe is 5.9%.

Yes, almost all of them are trending down - but even the IMF data shows how bogus the current BPP or BLS or Europe rates etc are.

 
At 5/24/2012 10:16 AM, Blogger Jet Beagle said...

Hans: "If any inflation index does not measure the service industry, then for all practical propose it is worthless."

Well, Hans, as i tried to point out to morganovich, the BPP does measure the services sector. It just cannot do so using direct prices for all services. As I understand it. the researchers at MIT use proxies for those services prices it cannot find online.

You and morganovich may be correct in questioning the methodology the MIT researchers use to indirectly measure services. But, unless you know what their methodology is, I don't see how you could characterize that methodology as worthless.

One important point: any consumer price index should not include prices for services which are not purchased directly by consumers. The prices of intermediate services are represented in the prices of final goods. For example, the cost of transporting beef to McDonald's is likely factored into the price Mcdonald's charges for a Big Mac. But there is no reason I can see for including that transportation cost in a consumer price index.

 
At 5/24/2012 10:36 AM, Blogger Jet Beagle said...

from the BPP website:

"BPP does not weight a particular item’s price more than another even if it comprises a larger proportion of household spending or if the item’s price is higher than a comparable items’ price."

from the Economist article:

"the BPP weights its index based on the BLS basket"

These are not necessarily contradictory statements. I interpreted the two statements hit way:

Apparently the BPP uses the weights used in the BLS basket. But, unlike the BLS, they may not weight specific items within a weight group. So the price of an 80 inch television on one website and a 55 inch television on another website would get equal weights, even though far more of the latter are actually pruchased. But it's also likely that there are far more instances of 55 inch television prices included in BPP's sampling.

 
At 5/24/2012 10:36 AM, Blogger Jet Beagle said...

This comment has been removed by the author.

 
At 5/24/2012 10:40 AM, Blogger bart said...

No. 438—PUBLIC COMMENT ON INFLATION MEASUREMENT

A recap of most of the massive issues with the BLS CPI

http://www.shadowstats.com/article/no-438-public-comment-on-inflation-measurement.


The main and most important point of Mr. Williams (we agree) is that the CPI is no longer a measure of a cost of living, but rather attempts to measure behavior with weird stuff like substitutions, geometric weighting, OER BS, failure to count all data etc etc.

My last large series of charts about medical expenses, health insurance, etc. incontrovertibly proves the very large "missing data" bias.

 
At 5/24/2012 10:43 AM, Blogger bart said...


Apparently the BPP uses the weights used in the BLS basket. But, unlike the BLS, they may not weight specific items within a weight group. So the price of an 80 inch television on one website and a 55 inch television on another website would get equal weights, even though far more of the latter are actually pruchased. But it's also likely that there are far more instances of 55 inch television prices included in BPP's sampling.



A great example of "new math" and illogic. A 55" TV is not the same as an 80" one. Even the BLS knows that.

As far as the missing services issues (proxied) from the BPP, your point about not knowing their methodology works against you too. You have no way of knowing if the BPP severely under estimates service prices.

 
At 5/24/2012 1:44 PM, Blogger Jet Beagle said...

bart: "A great example of "new math" and illogic. A 55" TV is not the same as an 80" one. Even the BLS knows that."

It's only illogic because you do not understand what the MIT folks are doing.

Based on what I've read, this is how I believe it works. BPP includes both 80 in television prices and 55 in television prices in their database. They use these prices from many different sources. They likely use many more 55 inch prices than 80 inch prices simply because fewer online websites include 80 inch televisions. They calculate the television price changes for every listing on every website, then average those to come up with one television inflation price.

What the BPP does not do is adjust their weights for actual number of purchases of 80 inch and 55 inch televisions. The more frequent listing of 55 in televisions automatically makes an weight adjustment in the final average.

 
At 5/24/2012 1:49 PM, Blogger Jet Beagle said...

What's astounding to me is that you guys criticize incessantly the government's calculation of CPI, in many cases arguing that "cooking" the number is in the best interests of the government or the Democrats or whoever is in office.

So now a private organization attempts to provide a reality check on CPI, and you guys immediately start your criticism of it - without having a clue about how this privately-derived number is calculated.

As I said in another post, the answer to every piece of information which doesn't support your worldview is likely to be:

"the number is wrong" or "worthless" or "illogical"

 
At 5/24/2012 2:01 PM, Blogger bart said...

...without having a clue about how this privately-derived number is calculated.


As I noted above neither do you, mostly because they themselves are not forthcoming with strong details of their full methodology.

And you're asking folk to believe their data and methodologies on WAY less evidence than I and others have presented on the massive issues with the BLS.

It's enough for you that they say these use *proxies* for services or that they use BLS weights - and we have zero/nada/nothing on what they actually do?

It seems from here that the onus is FAR more on you for supporting anything that the BLS or BPP says, especially given all the issues with the BPP - including that it disagrees with the IMF.


It's quite indicative from here that the BLS "worshipers" have never addressed any of the medical CPI problems that I've documented, the well known major issues of the OER, hedonics, the lack of reverse hedonics, and the same with other major issues like the CPI-U no longer being an actual measure of a stable standard cost of living, etc.

 
At 5/24/2012 2:11 PM, Blogger bart said...

Based on what I've read, this is how I believe it works.

Fair enough, but don't you see the logic issue there? The data you've read isn't even from BPP folk.

The BPP is a black box for which very little is actually known about its construction and methodologies, especially that vague/undefined concept called "proxies".

And you're relying on it. The BLS actually documents what they do and is fairly open about it.

And the people who are questioning it like myself or John Williams have hard and real evidence and logical points about its problems and severe shortcomings.

 
At 5/24/2012 2:13 PM, Blogger morganovich said...

jet-

you are totally misframing the issue.

this is not a "reality check" it's a drift into fantasy land.

let me ask you a very simple question:

let's say you take a flight once a month. it costs $1000. each time you do, you buy a dvd to watch while you fly. it costs $20.

the price of your flight increases to $1400. the price of the dvd drops to $10.

would you describe your costs as having gone up or down?

BPP would describe this situation as deflation. the 50% drop in dvd price swamps the 40% hike in airline costs even though it's only $10 and the plane ticket went up $400.

the index is, by design, not scaled not weighted.

assuming you agreed above that your costs rose (and if you do not feel that way, please explain why, because i cannot see how one could justify that response) then why do you continue to argue that BPP is a good measure of anything other than VERY short term twitches from monetary policy (which, btw is what it was first designed for)?

 
At 5/24/2012 2:23 PM, Blogger morganovich said...

in thinking about this, i have come upon what seems like a tricky issue for any inflation index.

what do you do with rapidly obsolete goods?

let's take erts madden football game as an example.

madden 2010 launches at a high price. it sells a lot of copies. when sales flag, they drop price, sales pick up etc.

where it gets tricky is when madden 2011 launches. such 2010 inventory that is left drops massively in price. however, almost none of it sells. (i know this because i used to be an investor in ERTS years ago and that is what they told me).

now, if you are BPP, you follow identical products. you know that madden 2010 is not the same as madden 2011. so you register deep deflation even if madden 2011 is priced higher than 2010 because you never compare them directly.

it seems like you need some way to compare these games at equivalent points in their lifecycles. launch price, holiday markdown, etc.

it's actually a really thorny problem and there are a great many products that do this from books to cell phones etc and cycle times keep getting faster.

i'm not claiming to know the answer, but i think it's an interesting though exercise.

 
At 5/24/2012 2:35 PM, Blogger morganovich said...

"
Apparently the BPP uses the weights used in the BLS basket. But, unlike the BLS, they may not weight specific items within a weight group. So the price of an 80 inch television on one website and a 55 inch television on another website would get equal weights, even though far more of the latter are actually pruchased. But it's also likely that there are far more instances of 55 inch television prices included in BPP's sampling."

no jet, the correct answer is that the economist got it wrong.

those ARE mutually exclusive.

"BPP does not weight a particular item’s price more than another even if it comprises a larger proportion of household spending"

is very, very clear.

they are telling you straight out that how much of your spending an item is is NOT taken into account.

you are trying to use a media article to contradict what the researchers themselves are telling us about what they did.

find anything at all from the BPP itself backing up the ecomomist, and i'd be happy to listen and if it's persuasive, change my mind, but if the researchers say they do no weight by % consumption and a news reporter who may have misunderstood says otherwise, well, i'm going to side with the researchers.

even if they did do this, failure to scale by price would make it meaningless.

getting food right as a % is of little value if a $2 hike in steak is swamped by a 2c drop in gelatin 1 gram packs.

 
At 5/24/2012 3:09 PM, Blogger morganovich said...

"Well, Hans, as i tried to point out to morganovich, the BPP does measure the services sector."

and as i pointed out to you, no BPP does not "measure" the services sector.

BPP makes a bunch of guesses about the service sectors and then fails to weight or scale them.

measurement requires looking at that which you measure. making up values to try and represent something is called "assumption" or perhaps "modeling" or, as i would put it "making stuff up".

perhaps it is because i spend my life looking at economic and financial models that i am that cynical and unimpressed with the accuracy of their output and predictive ability, but come on jet, taking wages, fuel costs, and drug costs and using them to generate a cost of healthcare number?

i have a doctor with a flashlight that can show you where those numbers come from. why should there be any sort of consistent relationship? drug prices are particularly fraught. consumers mostly make copays. the price paid by insurance cos is PPI, not cpi and further, almost no one pays list. that system is far too complex for BPP to have data on or model.

 
At 5/24/2012 3:33 PM, Blogger morganovich said...

jet-

"
As I said in another post, the answer to every piece of information which doesn't support your worldview is likely to be:

"the number is wrong" or "worthless" or "illogical""

all measurement has limitations and issues. that is incontrovertible. the question is when a method becomes so bad that the data occludes more than it reveals.

unlike the last conversation we had, this is an area in which i am extremely expert. i have spent years looking at it and read 1000's and 1000's of pages of research and data from every perspective both academic and professional.

i have worked through cpi methodology in great detail and personally modeled the effects of geometric weighting on systems and researched numerous examples where it gets the sign wrong and overweights items with declining consumption because the price move was caused by demand dropping as opposed to a dropping price increasing demand.

i have put 100's of man hours into this. it's not some passing fancy. it's a key plank in our investment process. (and it works)

why do you think we keep getting bubble after bubble and savings rates cannot recover? it's because real interest rates are negative and have been for most of the last 15+ years. note that this is EXACTLY what the austrian school predicts. there's a reason Q ratios have been above .9 for 90% of the time in the last 15 years and have departed dramatically from all historical norms since the 90's and it is monetary policy based on bad inflation assumptions.

try reading the boskin report. i think you'll find it much flimsier and devoid of empiricism than you suspect.

assumptions on inflation flow through to virtually all the stats on production. real gdp, wages, consumption, inventories, earnings, and every aspect of wealth measurement.

it's one of the most fundamental issues in economics. if you blow that measure, you are really in for it.

i think you need to take a harder look at this. what you see as dozens of unrelated issues are, in fact, one big issue. it's all the same thing. (granted, not cap utilization which is separate)

i'm happy to have this conversation with you and share such insight and opinion as i have formed, but to try to belittle it as "you just play the 'the data is BS ' card when you don't like what it says is not a productive way to engage here.

 
At 5/24/2012 4:14 PM, Blogger bart said...

i have put 100's of man hours into this. it's not some passing fancy. it's a key plank in our investment process. (and it works)

Same here, 100%


OT a bit, but I'm curious if your work has substantially diverged from Shadowstats work over the last 18+ months or so? Mine has.

 
At 5/24/2012 5:32 PM, Blogger PeakTrader said...

It seems, some people don't understand what the CPI represents.

It's a price index of goods & services based on actual consumption. When a price rises, consumers slow or stop purchases, which tends to lower the price, and when a price falls, consumers speed-up or stock-up on purchases, which tends to raise the price.

There's much less actual consumer inflation than some people believe.

 
At 5/24/2012 5:32 PM, Blogger PeakTrader said...

Also, I've had deflation with my health care, because my premiums and co-payments have been the same for years, while I'm sure the quality of health care increased.

 
At 5/24/2012 5:45 PM, Blogger bart said...

Also, I've had deflation with my health care, because my premiums and co-payments have been the same for years, while I'm sure the quality of health care increased.


Your experience is very atypical.

Here are the last 5 years of *total* medical care per person in the US.
Source: Kaiser, CMS


$7,920.44
$8,150.67
$8,398.07
$8,688.18
$9,046.44



And yes, inflation rates (especially with but also without my CPI w/o lies corrections) are down quite a bit the last two years.

 
At 5/24/2012 5:47 PM, Blogger bart said...

It seems, some people don't understand what the CPI represents.

And that's a major part of the problem with CPI. It's no longer representative of a stable standard/cost of living.

 
At 5/24/2012 5:58 PM, Blogger PeakTrader said...

Total medical care per person doesn't represent the CPI, and if the economy was static, then a static or "stable standard" CPI can be used.

 
At 5/24/2012 6:38 PM, Blogger bart said...

Total medical care per person doesn't represent the CPI, and if the economy was static, then a static or "stable standard" CPI can be used.

Read the my whole prior post again. That's not even vaguely close to what I said.

One item was that your experience with medical insurance, etc. is extremely atypical and is relatively worthless anecdotal data, as shown by the last 5 years of actual total US medical expenditures, divided by population.




Another is that medical is around 7% of the CPI-U. If it's understated by about 50%, as the actual data & comparison shows, then that makes CPI-U medical quite a bit lower than stated, same with CPI-U itself - all by itself.




If you like a CPI that changed definitions back in the 60s from something that attempted to measure real inflation to something now which 'measures changes' or 'actual consumption' regardless of real price changes, then so be it.

No wonder so many think my or shadowstats work isn't correct and too high. They have little clue what the CPI used to try and reflect.
CPI-U has little to do with actual inflation, by design of all the changes since the 60s.

 
At 5/24/2012 6:41 PM, Blogger bart said...

Repost for PT:

And speaking of "official bias", total medical care costs from the BLS since 1968 are an average of 2.7% too low (low by 43%). Since 2000, it's an average of 2.1% too low (low by 53%).


http://www.nowandfutures.com/images/medical_care_bls_v_cms.png



The BLS CPI medical category is understated by 40-50%.

 
At 5/24/2012 6:45 PM, Blogger bart said...

Total medical care per person doesn't represent the CPI, and if the economy was static, then a static or "stable standard" CPI can be used.

The economy never has been static, but when the CPI was designed to measure an actual stable standard cost of living, it worked fine.... and people living only on Social Security didn't see their purchasing power drop every year.




The substitution game of the BLS basically says that beef=chicken when the price of beef goes up more than chicken.
Peace=War too. *eye roll*

 
At 5/24/2012 6:51 PM, Blogger bart said...

Another is that medical is around 7% of the CPI-U. If it's understated by about 50%, as the actual data & comparison shows, then that makes CPI-U medical quite a bit lower than stated, same with CPI-U itself - all by itself.


oops - make that:


Another is that medical is around 7% of the CPI-U. If it's understated by about 50%, as the actual data & comparison shows, then that makes CPI-U medical *way* too low, same with CPI-U itself - all by itself.

 
At 5/25/2012 2:39 AM, Blogger PeakTrader said...

Bart, I doubt it's that uncommon my premiums and co-payments for health care have been the same for years.

Also, it should be noted, government spends a lot on health care, there are private donations, foreigners buy U.S. health care, and much of U.S. health care is indirectly financed by foreigners through international trade (because the U.S. is able to maintain trade deficits in the long-run).

The actual "out-of-pocket" spending by consumers reflected in the CPI, and excluding taxes, is much smaller.

Moreover, the quality of U.S. health care hasn't been static. For example, there was a great deal of cancer research, in recent years, that resulted in effective, although expensive, treatments.

 
At 5/25/2012 7:24 AM, Blogger bart said...

Bart, I doubt it's that uncommon my premiums and co-payments for health care have been the same for years.

As long as we going on anecdotal evidence, I know of no one at all whose total medical costs & expenditures haven't gone up very substantially over the last few years

Also, it should be noted, government spends a lot on health care, there are private donations, foreigners buy U.S. health care, and much of U.S. health care is indirectly financed by foreigners through international trade (because the U.S. is able to maintain trade deficits in the long-run).

I see. If it's government expenditures (or someone else donates money), it's automatically not part of spending by consumers. No wonder you think that total medical costs are lower than the actual facts.

That's standard BLS "logic".


The actual "out-of-pocket" spending by consumers reflected in the CPI, and excluding taxes, is much smaller.

And another non fact based issue - out of pocket expenditures are very very far from either all of medical spending, or representative of total medical costs and expenditures.


Moreover, the quality of U.S. health care hasn't been static. For example, there was a great deal of cancer research, in recent years, that resulted in effective, although expensive, treatments.

Yes, research costs a lot. That's another reason that the BLS numbers are far too low.

That's a lot like placing expenditures off budget, so that spending is "hidden".

Administrations since the early 70s have tried that bogus trick, to the point where many believe that Clinton balanced the budget - when the actual total debt grew a great deal during his terms.

Nothing but more lying "New Math".

 
At 5/25/2012 9:49 AM, Blogger Jet Beagle said...

PeakTrader: "The actual "out-of-pocket" spending by consumers reflected in the CPI, and excluding taxes, is much smaller."

That's a very important point. My health insurance cost has risen significantly for my employer. But that increased cost is not part of my expenditures. Even the recent increases in co-pays is a very tiny part of my expenditures. That's true for almost everyone covered by large employer group insurance plans.

The increased cost of health insurance to my employer is passed through to our airline passengers. At the same time, many other factors have enabled my employer to keep airfares from rising.

 
At 5/25/2012 9:52 AM, Blogger Jet Beagle said...

bart: "out of pocket expenditures are very very far from either all of medical spending, or representative of total medical costs and expenditures."

Out of pocket expenditures are the only expenditures which are relevant to the CPI, bart. Those are the only medical care expenditures which should be included in deriving the CPI.

 
At 5/25/2012 9:52 AM, Blogger Jet Beagle said...

bart: "out of pocket expenditures are very very far from either all of medical spending, or representative of total medical costs and expenditures."

Out of pocket expenditures are the only expenditures which are relevant to the CPI, bart. Those are the only medical care expenditures which should be included in deriving the CPI.

 
At 5/25/2012 10:08 AM, Blogger bart said...

My health insurance cost has risen significantly for my employer. But that increased cost is not part of my expenditures.

The increased cost of health insurance to my employer is passed through to our airline passengers.


Yep, as long as someone else pays for it, it has nothing to do with you or total medical costs... a standard and very sad part of BLS lying, cognitive dissonance and socialist "logic" etc.

At least you admit that total health insurance costs have gone WAY up.






Out of pocket expenditures are the only expenditures which are relevant to the CPI, bart.

Of course, the rest are all totally magic and no consumer/human pays for them. *eye rolls*, to the max.




The actual facts, as cited in the chart above, shows that total medical costs & expenditures per person have gone up over 45% faster than the lying/mis-stated BLS index.


Just because you can't see those costs and expenditures doesn't mean they don't exist. People, aka consumers, are the only items that use medical care... and no, pets don't pay for medical care, their owners do.

 
At 5/25/2012 10:35 AM, Blogger Jet Beagle said...

Bart,

No one is saying that health care expenditures do not matter. Of course those expenditures matter.

Health care expenditures are only relevant to the CPI where consumersa actually pay out-of-pocket for them. If you do not understand that, then you probably do not understand the purpose of a consumer price index.

It's the same with the ocean shipping expenditures which pay for transporting goods from China. Those are important expenditures. But they are not included in the consumer price index because consumers do not directly pay for them. Rather, WalMart and other retailers pay those costs and include them in determining the prices to set for retail goods in the store. The prices of retail goods are included in the CPI, but not all the expenditures used to create those goods.

Likewise, health insurance costs for my airline employer are not included in the CPI. But the fares we charge to our customers are.

 
At 5/25/2012 11:06 AM, Blogger bart said...

Health care expenditures are only relevant to the CPI where consumers actually pay out-of-pocket for them.

Completely and totally false. You and PT etc. apparently have little clue of what real consumer costs and expenditures really are, and only see what is obvious or what the BLS wants you to see about real inflation - also known as actual cost of living stable standards against which one can judge one's own situation etc.


As long as someone else pays for it, it has nothing to do with you or real total and actual medical costs... a standard and very sad part of BLS lying, cognitive dissonance and socialist "logic" etc.
I totally reject that new math and egregious bad logic. I also note that airline fares are not part of medical care, and also believe that hiding them there is just more evidence of how the BLS manages to "convince" those like you that their medical care price index is roughly 50% too low, but correct.


The actual facts, as cited in the chart above, shows that total medical costs & expenditures per person have gone up over 45% faster than the lying/mis-stated BLS index.


Just because you can't see those costs and expenditures as out-of-pocket doesn't mean they don't exist.

People, aka consumers, are the only things that use medical care.

 
At 5/25/2012 12:54 PM, Blogger misterjosh said...

"Health care expenditures are only relevant to the CPI where consumersa actually pay out-of-pocket for them. If you do not understand that, then you probably do not understand the purpose of a consumer price index."

This is bizarre and absurd. Maybe I have an overly large notion of the concept of my pocket, but every penny of the increase in medical expenditures is coming out of my pocket - either in the form of rates paid or foregone wages. Health insurance is a fundamental part of my total compensation. My cash raises have been negligible over the last few years, but i know that my compensation has been going up in the form of increased health care costs.

Every economist knows this too, so I don't understand why the BLS wouldn't include these figures in their calculations.

 
At 5/25/2012 1:43 PM, Blogger Ron H. said...

Bart: "Completely and totally false. You and PT etc. apparently have little clue of what real consumer costs and expenditures really are, and only see what is obvious or what the BLS wants you to see about real inflation - also known as actual cost of living stable standards against which one can judge one's own situation etc."

Wait! I don't see any argument that the CPI *does* reflect total medical costs. I think everyone agrees that it *doesn't* reflect the total cost of medical care, but only that portion paid by consumers directly. It is, therefore, a poor way of measuring the true, total cost of many goods and services, including such things a national defense, which consumers don't pay for directly.

I believe Jet's point is that some intermediate costs, employee health care coverage being one of them, is included in the price of the the final retail product that consumers DO buy - in his example, air travel.

Theoretically, airline tickets would be cheaper if airlines didn't pay part of their employees healthcare coverage costs.

 
At 5/25/2012 1:51 PM, Blogger bart said...

I don't see any argument that the CPI *does* reflect total medical costs.

They and a few others, including Dr. Perry, believe that either the CPI is about right or too high.

I've shown and proven with medical and a few other categories & areas that it understates inflation - severely.

 
At 5/25/2012 2:04 PM, Blogger Ron H. said...

Misterjosh: "This is bizarre and absurd. Maybe I have an overly large notion of the concept of my pocket, but every penny of the increase in medical expenditures is coming out of my pocket - either in the form of rates paid or foregone wages..."

You're absolutely correct, but the CPI doesn't show total costs in that way. While I have little confidence in the value of the CPI, it is not correct to expect it to reflect total costs by category. The Consumer Price Index reflects Prices that Consumers pay directly for goods and services, not the cost of inputs to those goods and services.

"Every economist knows this too, so I don't understand why the BLS wouldn't include these figures in their calculations."

That isn't the intent of the CPI. The total cost of healthcare is included in the final price of goods and services purchased by consumers at the retail level.

If you wish to see per capita costs by category, or costs as percentages of GDP, you need to look elsewhere.

 
At 5/25/2012 2:20 PM, Blogger bart said...

That isn't the intent of the CPI. The total cost of healthcare is included in the final price of goods and services purchased by consumers at the retail level.


The original intent of the CPI and that was changed in the 60s was to show the *real* inflation rate via a basket of goods & services etc., not to jerk around with ridiculous Orwellian BS like saying at root that beef = chicken.


And that's what I and John Williams and others have done - attempted to reconstruct that *original* CPI, so that real and actual inflation.

And the total medical care portion of CPI is, as proven, about 50% too low.

 
At 5/25/2012 2:26 PM, Blogger Ron H. said...

Bart: "They and a few others, including Dr. Perry, believe that either the CPI is about right or too high.

I've shown and proven with medical and a few other categories & areas that it understates inflation - severely.
"

I agree with you that CPI understates inflation. The problems with CPI, as I see it, are, first of all, the impossibility of accurately measuring consumer well-being and satisfaction - totally subjective values in any case, and then the use of substitution and geometric weighting in a manner designed to understates the true rate of inflation.

 
At 5/25/2012 2:49 PM, Blogger Ron H. said...

Bart: "The original intent of the CPI and that was changed in the 60s was to show the *real* inflation rate via a basket of goods & services etc., not to jerk around with ridiculous Orwellian BS like saying at root that beef = chicken."

Agreed. The intent of the CPI has changed. It now used to limit mandated COLA raises to retirees receiving Social Security benefits.

As morganovich has pointed out the CPI methodology will *always* understate real price increases.

I hesitate to use the term *real* inflation, as that depends on your reference.

I don't think anyone is arguing against your assertion that total healthcare costs have risen faster than that portion paid for directly by consumers out of pocket.

 
At 5/25/2012 3:20 PM, Blogger bart said...

I agree with you that CPI understates inflation. The problems with CPI, as I see it, are, first of all, the impossibility of accurately measuring consumer well-being and satisfaction - totally subjective values in any case, and then the use of substitution and geometric weighting in a manner designed to understates the true rate of inflation.


We're basically tracking fine.

All I'm really trying to do is work out something better than the CPI, with its many politically based issues. I believe that my CPI w/o lies is far closer to real inflation than the CPI-U.

 
At 5/25/2012 3:23 PM, Blogger bart said...

I hesitate to use the term *real* inflation, as that depends on your reference.

Fair enough, but it's impossible to use them all. And as I noted above, I strongly believe my CPI w/o lies comes way closer to measuring real inflation then the CPI-U - or the SGS CPI for that matter.


I don't think anyone is arguing against your assertion that total healthcare costs have risen faster than that portion paid for directly by consumers out of pocket.

Sure could have fooled me.

 
At 5/25/2012 6:43 PM, Blogger Ron H. said...

Bart: "Sure could have fooled me."

I believe the disagreement is with the idea that all healthcare costs should be included in CPI and labelled as such, even those that aren't paid directly by consumers, but CPI only measures those things paid for directly by consumers.

While they are, in fact, accounted for, healthcare costs paid by employers and government are included in the cost of other final products and services, but not under the heading "Healthcare".

 
At 5/25/2012 10:38 PM, Blogger Hans said...

Bart, carries the day...Many excellent points...

Bart said: "Health care expenditures are only relevant to the CPI where consumersa actually pay out-of-pocket for them. If you do not understand that, then you probably do not understand the purpose of a consumer price index."

If what you say is true, then the index is in fact broken or spreading misinformation...

So then when companies past energy cost to end users, that is also not part of the index, since it is not out-of-pocket?

The cost to carry insurance for employees has to find it's way into the CPI or the index is to narrow or inaccurate...

BTW, the health giver industry represents one out of seven dollar and is not included in the MIT's index...

The BPP is great if you are shopping at the BBC...

 
At 5/26/2012 2:35 AM, Blogger PeakTrader said...

Sure, it makes as much sense as U.S. taxpayers actually paying the $5 trillion of additional federal debt over the past four years.

The $2.5 trillion U.S. health care industry is a high quality industry, like our new Information and Biotech industries, along with continued shifts into higher-tech manufacturing (Industrial Revolution) and farming (Agricultural Revolution):

Medical-industrial complex profits: benefit or bane?
April 2, 2012

"Not only does it create jobs, but the U.S. medical industry is very alluring too. Wente compares the Canadian and U.S. versions.

Upon visiting Chicago, Wente laments, “The city’s hospitals make our hospitals look like slums. They’re gleaming, spotless and staffed with friendly, smiling people who treat patients like hotel guests. There are only two patients to a room, and if you ask for something you can get it right away. Their hospitals seem to have twice as many nurses as ours do, and three times more computers.”

Of course, there is a price to pay, one that is twice the cost of Canada’s."

 
At 5/26/2012 3:51 AM, Blogger Jet Beagle said...

hans: "The cost to carry insurance for employees has to find it's way into the CPI or the index is to narrow or inaccurate..."

Both Ron H and I have tried to explain how "the cost to carry incsurance" does "find its way into the CPI". You have apparently missed our explanation or else did not understand it.

Try to understand this:

The price Southwest Airlines pays for a 737-800 is not included in the Consumer Price Index. The fares paid by the passengers who fly on that 737-800 is included in the CPI.

Similarly, the price Southwest Airlines pays for its employees' health insurance is not included in the Consumer Price Index. Again, the fares paid by the passengers who fly on Southwest Airlines' jets are included in the CPI.

Airline ticket prices are higher than they otherwise would be - because Southwest and other airlines pay for the health insurance their employees consume and pay for the aircraft their pilots fly.

Higher health insurance costs do make their way into the CPI - but not directly because consumers do not pay for health insurance directly. Or, at least, they pay only a small part of the cost of that health insurance.

 
At 5/26/2012 7:44 AM, Blogger bart said...

I believe the disagreement is with the idea that all healthcare costs should be included in CPI and labelled as such, even those that aren't paid directly by consumers, but CPI only measures those things paid for directly by consumers.


But the CPI does *not* include only those things paid for by consumers, the OER/housing issue being the best factual and provable example. The OER has severely understated what people pay since it was invented in 1982.


While they are, in fact, accounted for, healthcare costs paid by employers and government are included in the cost of other final products and services, but not under the heading "Healthcare".

It's fine that they say that, but firstly - where's the proof? I've been looking at the job the BLS has done for years, and every time I dig in to a specific category I find more "incomplete" or just plain erroneous work and data etc.

They have nothing but assertions, where I can *prove* the issues with both medical care and housing. Those two cover almost 50% of CPI.

Another one - if they want to go all the way down to consumer prices of airline tickets, then what about also allowing for the huge losses of the airlines over the years?

What about the reverse hedonics issues of smaller seats, less leg room, TSA "interference" and many more issues that make flying much more of a drag than it used to be? Where are those covered in the BLS Transportation index? (hint: they're not - in any way, shape or form).

How many here believe that their total transportation costs have only gone only 2.9% (the most recent number from the BLS) in the last year - when comparing on an apples to apples basis, and including things like reverse hedonics etc?



And no, I haven't gone in depth into all the details of the transportation index and its very probable under valuation and under counting issues. I'm just one guy who has spent many many hundreds of hours in the area of trying to put together an inflation index, based on CPI-U, that uses represents the original concept - an index that actually tries to track a stable standard/cost of living.

 
At 5/26/2012 8:10 AM, Blogger bart said...

Since airline fares are apparently an issue, I pulled the NSA data back to 1967 from the BLS.

How many who fly now & then or frequently believe that airline fares (apples to apples, including reverse hedonics etc.) have only gone up 36% in the last 10 years, or 1.2% in the last year? Those numbers sure don't match my experiences and costs.

Total medical care has gone up about 77% in the last 10 years.

 

Post a Comment

<< Home